life support systems: part 4. engineering test and ... · notices this final report was submitted...

118
USAFSAIM-TR-86-*36-PT-4 PESEARCH AND DEVELOPMENT OF ANTI-G LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and Evaluation of Six Anti-G Valves (c4 Larry J. Meeker, B.S. (USAFSAM/VNS) 01 Arnold G. Krueger U ~Paul E. Love Emnily M. Gause, M.S. Robert W. Kivtz, Ph.D. CA Life Sciences Division Technology Incorporated San Antonio, TX 78216 OTIC ' August 1988 AR1181 * Final Report for Period 1 April 1981 - 31 July 1985 Approved for public i-eiease; distribution Is uniited. Prepared for~' 8 4 121 0O11 U'SAF SCHOOL OF AEROSPACE MEDICINE Human Systems Divlsidn (AFSC) 'rooks Air Force Base, TX 78235-5301 t-7,

Upload: others

Post on 05-Aug-2020

0 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

USAFSAIM-TR-86-*36-PT-4

PESEARCH AND DEVELOPMENT OF ANTI-GLIFE SUPPORT SYSTEMS:

Part 4. Engineering Test and Evaluation ofSix Anti-G Valves

(c4 Larry J. Meeker, B.S. (USAFSAM/VNS)01 Arnold G. Krueger

U ~Paul E. LoveEmnily M. Gause, M.S.Robert W. Kivtz, Ph.D.

CA

Life Sciences DivisionTechnology IncorporatedSan Antonio, TX 78216

OTIC '

August 1988 AR1181

* Final Report for Period 1 April 1981 -31 July 1985

Approved for public i-eiease; distribution Is uniited.

Prepared for~' 8 4 121 0O11U'SAF SCHOOL OF AEROSPACE MEDICINEHuman Systems Divlsidn (AFSC)'rooks Air Force Base, TX 78235-5301 t-7,

Page 2: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

NOTICES

This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio, TX 78216, under contract F33615-81-C-0600, joborder 7930-14-42, with the USAF School of Aerospace Medicine, Human SystemsD-vision, AFSC, Brooks Air Force Base, Texas. Larry J. Meeker (USAFSAMNNS) wasthe Laboratory Project Scientist-in-Charge.

When Government drawings, specdficaticns, or other data are used for any purposeother ihan in connection with a definitely Government-related procurement, the UnitedStates Government incurs no responsibility nor any obligation whatsoever. The factthat the Government may have formulated or in any way supplied the said drawings,specifications, or other data, is not to be regarded by implication, or otherwise in anymanner construed, as licensing the holder, or any other person or corporation; or ascooveying any rights or permissiorto manufacture, use, or sell any patented inventionthat may in any way be re!ated thereto.

The Office of Public Affairs has reviewed this report, and it is releasable to theNational Technical Information Service, where it will be available to the general public,including foreign nationals.

This repoit has been reviewed and is approved for publication.

LARRY J. MEEKER W. C. ALEXANDER, Ph.D.Project Scientist Supervisor

* E G. DAVIS, Colonel, USAF, MC0nder

Page 3: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

UNCLASSIFIEDSECU'ITY CLASSIFICATION OF T.S PGE ,

REPORT DOCUJMENTATION PAGE 018N.00-1,

Ia. REPORT SECURITY CLASSIFICATION Il. RESTRICTIVE MARKINGSUnclassified

Za. SECURITY CLASSIFICATION AUTHORITY 3. OISTRIOUTION/AVAILASIUTY OF REPORT

Approved for public release; distribution2b. OECLASSIFICATIONiOOWNGRAOING SCHEDULE is unlimited.

4. PERFORMING ORGANIZATION REPORT NUMBER(S) S. MONITORING ORGANIZATION REPORT NUMBER(S)USAFSAM-TR-86-36-PT-4

6.. "ME OF PERFORMING ORGANIZATION 6b. OFFICE SYMBOL 7a. NAME OF MONITORING ORGANIZATIONTechnology Incorporated I (ifp ae) USAF School of Aerospace Medicine (VNS)Life Sciences Division

6C. AOORIESS (0y, State, and ZIP CodeI 7b. ADORESS (Ctv, State. and ZIP Code)406 Breesport Human Systems Division (AFSC)San Antonio, TX 78216 Brooks Air Force Base, TX 78235-5301

8&. NAME OF FUNDING iSPONSORING lb. OFFICE SYMBOL 9. PROCUREMENT INSTRUMENT IDENTIFICATION NUMBERORGANIZATION I (/, 'f ; F33615-81-C-0600

t ADORESS (Cty, State, and ZIP Code) 10. SOURCE OF FUNDING NUMBERSPROGRAM IPROJECT TASK I WORK UNITELEMENT NO. INO. INO. CESSION NO.

62202F 7930 14 42TI. TME (Incd Securiy Caunicaboni

RESEARCH AND DEVELORMT OF ANTI-G LIFE SUPPORT SYSTEMS: Part 4Engineering Test and Evaluation of Six Anti-G Valves

12. PERSONAL AUTHOR(S) Meeker, Larry J. (USAFSAM/VNS); Krueger, Arnold G.; Love, Paul E.;Gause. Emily M.: Krutz, Robert W.

13AL TYPE OF REPORT 113b. TIME COVERED 14. DATE OF REPORT (Year, Month, Day) IS. PAGE COUNT

Final / 1988, August 11916. SUPPLEMENTARY NOTATION

17. COSATI COOES 18. SUBJECT TERMS (Conthnm on revers if necsay and idenrtfy by block number)FIELD GROUP SUB-GROUP Life-Support Equipment, Anti-G Valve Performance,23 05 Anti-G Valve Testing, Anti-G Suits13 iI

\9. ABSTRACT (Convinu, on revere if necessary and identify by block number)ix anti-G valves (i.e., valves designed to rapidly deliver output pressures proportional ,t +Gz forces actin upon

valve control elements).were evaluated for performance under conditions simulating the spectrum of Gi-onsetconditions likely to be encountered by pilots of current U.S. Air Force high-performance aircraft. ThisEngineering Test and Evaluation Task was conducted on-site at Brooks Air Force Base, Texas, using the Schoolof Aerospace Medicine (USAFSAM) human centrifuge facility. One objective was validation of uniform testingprocedures for evaluation of andi-G valves under controlled conditions of: acceleration, source pressure, valveangles, and simulated anti-G suit volumes. A second objective was measurement of actual dynamic performanceof the standard ALAR 8400A anti-G valve and five new experimental valves. The experimental valves evaluatedwere: ALAR High-flow Ready-pressure; ALAR&h-flow; Garrett Electronic/Pneumatic; Hymatic VAG- 110-022; and AAMRL Electropneumatic -ang Bangg. The battery of test conditions employed was divided intothree phases: Phase I - Maximum Flow Capacity Testing, Phase II - Dynamic Response Testing; and Phase IIl -Complex Dynamic Response Testing. Results of testing under all conditions are presented for each valve and theperformance characteristics of all six valves are compared. (/1 uC I=

20. OISTRIBUTION/AVAILABILITY OF ABSTRACT 21. ABSTRAC 1SELN RITY CLASSIFICATIONM UNCLASSIFIEO/UNLIMI TED ED SAME AS RPT OrtiC USERS Unclassified

22a. NAME OF RESPONSIBLE INDIVIDUAL 22b. TEjPHO (I I Area Code) I22C. FFIC SLarry J. Meeker 5T 12)52i-Mi1 SIIVaS

00 Form 1473, JUN 86 Previous editionsare obsolete. SECURITY CLASSIFICATION OF TIS PAGEi UNCLASSIFIED

Page 4: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

CONTENTS

1. INTRODUCTION .................................................................................................. .1

2. OBJECTIVES ........................................................................................................ 2

3. TESTING EQUIPMENT AND PROCEDURES ............................................... . 3

3.1 Testing Equipment ..................................................................................... . 3

3.1.1 General Test Configurations ........................................................ 33.1.2 Test Configuration Elements ........................................................ 5

3.1.2.1 Source Pressure (Ps) ................................................... 53.1.2.2 Valve Output Pressure (Po) ................... 53.1.2.3 Air Flow (Fv) .................................................................. 531.2.4 Simulated Suit Volume Pressure (PV) ..................... 53.1.2.5 Acceleration (+Gz) ........................................................ 63.1.2.6 Valve Angle .................................................................... 63.1.2.7. Pressure Transducers ................................................. 63.1.2.8 Flow Meter .................................................................... 63.1.2.9 Control Console Signal Conditioners ...................... 63.1.2.10 Strip-Chart Recorders ................................................. . 6

3.2 Testing Procedures ................................... 7

3.2.1 Phase I - Maximum Flow Capacity Testing ................ 73.2.2 Phase II - Dynamic Response Testing ................... 73.2.3 Phase III - Complex Dynamic Response Testing ...................... S3.2.4 Simulated Suit Volume Estimation .............................................. 10 o.-3.2.5 Data Acquisition .............................................................................. 103.2.6 Data Analysis ................................................................................... 1 .

4. RESULTS AND DISCUSSION ......................................................................... 12

4.1 ALAR 8400A Anti-G Valve ........................................................................ 13 Fr_

4.1.1 Description ....................................................................................... 13 C34.1.2 Phase I - Determination of Maximum Flow Capacity ............... 13 -

4.1.3 Phase II - Determination of Dynamic Response Capability.... 15 ton

4.1.3.1 Low Gz-Onset-Rate Conditions ................ 1 54.1.3.2 High Gz-Onset-Rate Conditions ............... 19 n/__

.ty Cods

4.1.4 Phase III - Determination of Complex Dynamic Response and/or

C apability ................................................................. 23 : l

iii IIV I

Page 5: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

4.2 ALAR High-Flow Ready-Pressure Anti-G Valve ................. 28

4.2.1 Description ....................................................................................... 284.2.2 Phase I - Determination of Maximum Flow Capacity ............... 284.2.3 Phase II - Determination of Dynamic Response Capability .... 31

4.2.3.1 Low Gz-Onset-Rate Conditions ................ 314.2.3.2 High Gz-Onset-Rate Conditions ............... 34

4.2.4 Phase III - Determination of Complex Dynamic Response

C apability ................................................................. 37

4.3 ALAR High-Flow Anti-G Valve .................................................................. 41

4.3.1 Description ...................................................................................... 414.3.2 Phase I - Determination of Maximum Flow Capacity ............... 414.3.3 Phase II - Determination of Dynamic Response Capability .... 43

4.3.3.1 Low Gz-Onset-Rate Conditions ........................ ....... 434.3.3.2 High Gz-Onset-Rate Conditions ............... 47

4.3.4 Phase III - Determination of Complex Dynamic Response

Capability ................................................................. 50

4.4 Garrett Electronic,/Pneumatic Anti-G Valve .................... 54

4.4.1 Description ....................................................................................... 544.4.2 Phase I - Determination of Maximum Flow Capacity ............... 544.4.3 Phase II - Determination of Dynamic Response Capability .... 56

4.4.3.1 Low Gz-Onset-Rate Conditions ................ 564.4.3.2 High Gz-Onset-Rate Conditions ............... 59

4.4.4 Phase III- Determination of Complex Dynamic Response

Capability ................................................................. 64

4.5 Hymatic VAG-1 10-022 Anti-G Valve ....................................................... 69

4.5.1 D escription ....................................................................................... 694.5.2 Phase I - Determination of Maximum Flow Capacity ............... 714.5.3 Phase II - Determination of Dynamic Response Capability .... 71

4.5.3.1 Low Gz-Onset-Rate Conditions ................ 714.5.3.2 High Gz-Onset-Rate Conditions ............... 75

4.5.4 Phase III - Determination of Complex Dynamic ResponseC apability ................................................................. 78

iv

Page 6: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

4.6 AAM RL Anti-G Valve .................................................................................. 83

4.6.1 Description ...................................................................................... 834.6.2 Phase I - Determination of Maximum Flow Capacity ............... 83

4.6.3 Phase II - Determination of Dynamic Response Capability .... 86

4.6.3.1 Low Gz-Onset-Rate Conditions ................ 864.6.3.2 High Gz-Onset-Rate Conditions ............... 89

4.6.4 Phase III - Determination of Complex Dynamic Response

Capability ................................................................ . 93

5. DISCUSSION AND CONCLUSIONS .............................................................. 96

6. REFERENCES ..................................................................................................... . 102

ABBREVIATIONS, ACRONYMS, AND SYMBOLS ................... 103

APPEN D IXES ....................................................................................................... 105

APPENDIX A: ALAR 8400A Anti-G Valve Experimental DataAPPENDIX B: ALAR High-Flow Ready-Pressure Anti-G Valve Experi-

mental DataAPPENDIX C: ALAR High-Flow Anti-G Valve Experimental DataAPPENDIX D: Garrett Servo-Controlled Rapid-Response Anti-G Valve

Experimental DataAPPENDIX E: Hymatic VAG-1 10-022 Anti-G Valve Experimental DataAPPENDIX F: AAMRL ("Bang Bang") Anti-G Valve Experimental Data

Fig. No.

3.1 Testing Eauigment

3.1-1 Phase I Open-Flow Capability Test Configuration ........................... 3

3.1-2 Phase II Dynamic Test Configuration ................................................. 4

3.2 Testing Procedures

3.2-1 Conditions of the Three Phase-Il "Trapezoid" CentrifugePerformance Evaluation Tests Performed at All ThreeSource Pressures, with Valve Angle Held at 00 ............................... 9

3.2-2 Conditions of Phase-Ill SACM Testing of Anti-G Valves ................. 9

v

Page 7: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

4.1 ALAR 840A Anti-G Valve

4.1-1 ALAR 8400A Anti-G Valve .................................................................... 144.1-2 Maximum flow through ALAR 8400A anti-G valve ............. 1 64.1-3 "Suit" pressures produced during acceleration/deceleration

by ALAR 8400A anti-G valve under conditions of: 150-psigsource pressure; low G onset; 10-liter volume; 00 angle ............ 1 6

4.1-4 "Suit" pressures produced during acceleration/decelerationby ALAR 8400A anti-G valve under conditions of: 150-psigsource pressure; low G onset; 10-liter volume; 150 angle ......... 1 7

4.1-5 "Suit" pressures produced during acceleration/decelerationby ALAR 8400A anti-G valve under conditions of: 150-psigsource pressure; low G onset; 10-liter volume; 300 angle ......... 1 8

4.1-6 "Suit" pressures produced during acceleration/decelerationby ALAR 8400A anti-G valve under conditions of: 30-psigsource pressure; low G onset; 10-liter volume; QO angle ............ 18

4.1-7 "Suit" pressures produced during acceleration/decelerationby ALAR 8400A anti-G valve under conditions of: 300-psigsource pressure; low G onset; 10-liter volume; 00 angle ............ 1 9

4.1-8 "Suit" pressures produced during acceleration/decelerationby ALAR 8400A anti-G valve under conditions of: 1 50-psigsource pressure; high G onset; 10-liter volume; 00 angle .......... 20

4.1-9 "Suit" pressures produced during acceleration/decelerationby ALAR 8400A anti-G valve under conditions of: 150-psigsource pressure; high G onset; 10-liter volume; 150 angle ........ 21

4.1-10 "Suit" pressures produced during acceleration/decelerationby ALAR 8400A anti-G valve under conditions of: 150-psigsource pressure; high G onset; 10-liter volume; 300 angle ........ 22

4.1-11 "Suit" pressures produced during acceleration/decelerationby ALAR 8400A anti-G valve under conditions of: 30-psigsource pressure; high G onset; 10-liter volume; 00 angle .......... 22

4.1-12 "Suit" pressures produced during acceleration/decelerationby ALAR 8400A anti-G valve under conditions of: 300-psigsource pressure; high G onset; 10-liter volume; 00 angle .......... 23

4.1-13 "Suit" pressures produced during SACM by ALAR 8400Aanti-G valve under conditions of: 150-psig sourcepressure; 10-liter volume; 00 angle ................................................ 24

4.1-14 "Suit" pressures produced during SACM by ALAR 8400Aanti-G valve under conditions of: 150-psig sourcepressure; 10-liter volume; 150 angle .................... 26

4.1-15 "Suit" pressures produced during SACM by ALAR 8400Aanti-G valve under conditions of: 150-psig sourcepressure; 10-liter volume; 300 angle .................... 26

4.1-16 "Suit" pressures produced during SACM by ALAR 8400Aanti-G valve under conditions of: 30-psig sourcepressure; 14-liter volume; 00 angle ................................................ 27

4.1-17 "Suit" pressures produced during SACM by ALAR 8400Aanti-G valve under conditions of: 300-psig sourcepressure; 6-liter volume; 00 angle ................................................... 27

vi

Page 8: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

4-2 ALAR Hiah-Flow Ready-Pressure Anti-G Valve

4.2-1 ALAR High-Flow Ready-Pressure Anti-G Valve ............... 294.2-2 Maximum flow through ALAR HFRP anti-G valve

with pressure off ................................................................................. 304.2.3 Maximum flow through ALAR HFRP anti-G valve

with pressure on ................................................................................. . 304.2-4 "Suit" pressures produced during acceleration/deceleration

by ALAR HFRP anti-G valve under conditions of: 1 50-psigsource pressure; low G onset; 10-liter volume; 0° angle ............ 32

4.2-5 "Suit" pressures produced during acceleration/decelerationby ALAR HFRP anti-G valve under conditions of: 150-psigsource pressure; low G onset; 10-liter volume; 150 angle ......... 32

4.2-6 "Suit" pressures produced during acceleration/decelerationby ALAR HFRP anti-G valve under conditions of: 1 50-psigsource pressure; low G onset; 1 0-liter volume; 300 angle ......... 33

4.2-7 "Suit" pressures produced during acceleration/decelerationby ALAR HFRP anti-G valve under conditions of: 30-psigsource pressure; low G onset; 10-liter volume; 00 angle ............ 33

4.2-8 "Suit" pressures produced during acceleration/decelerationby ALAR HFRP anti-G valve under conditions of: 300-psigsource pressure; low G onset; 10-liter volume; 00 angle ............ 35

4.2-9 "Suit" pressures produced during acceleration/decelerationby ALAR HFRP anti-G valve under conditions of: 150-psigsource pressure; high G onset; 10-liter volume; 00 angle .......... 35

4.2-10 "Suit" pressures produced during acceleration/decelerationby ALAR HFRP anti-G valve under conditions of: 150-psigsource pressure; high G onset; 10-liter volume; 150 angle ........ 36

4.2-11 "Suit" pressures produced during acceleration/decelerationby ALAR HFRP anti-G valve under conditions of: 1 50-psigsource pressure; high G onset; 10-liter volume; 30 ° angle ........ 36

4.2-12 "Suit" pressures produced during acceleration/decelerationby ALAR HFRP anti-G valve under conditions of: 30-psigsource pressure; high G onset; 10-liter volume; 00 angle .......... 38

4.2-13 "Suit" pressures produced during acceleration/decelerationby ALAR HFRP anti-G valve under conditions of: 300-psigsource pressure; high G onset; 10-liter volume; 00 angle .......... 38

4.2-14 "Suit" pressures produced during SACM by ALAR HFRPanti-G valve under conditions of: 150-psig sourcepressure; 10-liter volume; 0° angle ................................................ 39

4.2-15 "Suit" pressures produced during SACM by ALAR HFRPanti-G valve under conditions of: 150-psig sourcepressure; 10-liter volume; 300 angle .................... 39

4.2-16 "Suit" pressures produced during SACM by ALAR HFRPanti-G valve under conditions of: 30-psig sourcepressure; 14-liter volume; 00 angle ................................................ 40

4.2-17 "Suit" pressures produced during SACM by ALAR HFRPanti-G valve under conditions of: 300-psig sourcepressure; 6-liter volume; 00 angle ................................................... 41

vii

Page 9: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

4.3 ALAR High-Flow Anti-G Valve

4.3-1 ALAR High-flow Anti-G Valve ............................................................... 424.3-2 Maximum flow through ALAR High-flow anti-G valve ........... 444.3-3 "Suit" pressures produced during acceleration/deceleration

by ALAR High-flow anti-G valve under conditions of: 150-psigsource pressure; low G onset; 10-liter volume; 00 angle ............ 44

4.3-4 "Suit" pressures produced during acceleration/decelerationby ALAR High-flow anti-G valve under conditions of: 1 50-psigsource pressure; low G onset; 10-liter volume; 150 angle ......... 45

4.3-5 "Suit" pressures produced during acceleration/decelerationby ALAR High-flow anti-G valve under conditions of: 150-psigsource pressure; low G onset; 1 0-liter volume; 300 angle ......... 45

4.3-6 "Suit" pressures produced during acceleration/decelerationby ALAR High-flow anti-G valve under conditions of: 30-psigsource pressure; low G onset; 10-liter volume; 00 angle ............ 46

4.3-7 "Suit" pressures produced during acceleration/decelerationby ALAR High-flow anti-G valve under conditions of: 300-psigsource pressure; low G onset; 1 0-liter volume; 00 angle ............ 46

4.3-8 "Suit" pressures produced during acceleration/decelerationby ALAR High-flow anti-G valve under conditions of: 150-psigsource pressure; high G onset; 1 0-liter volume; 00 angle .......... 48

4.3-9 "Suit" pressures produced during acceleration/deceleration- by ALAR High-flow anti-G valve under conditions of: 150-psig

source pressure; high G onset; 10-liter volume; 150 angle ........ 484.3-10 "Suit" pressures produced during acceleration/deceleration

by ALAR High-flow anti-G valve under conditions of: 150-psigsource pressure; high G onset; 10-liter volume; 300 angle ........ 49

4.3-11 "Suit" pressures produced during acceleration/decelerationby ALAR High-flow anti-G valve under conditions of: 30-psigsource pressure; high G onset; 10-liter volume; 00 angle .......... 49

4.3-12 "Suit" pressures produced during acceleration/decelerationby ALAR High-flow anti-G valve under conditions of: 300-psigsource pressure; high G onset; 10-liter volume; 00 angle .......... 51

4.3-13 "Suit" pressures produced during SACM by ALAR High-flowanti-G valve under conditions of: 150-psig sourcepressure; 10-liter volume; 0° angle ................................................ 51

4.3-14 "Suit" pressures produced during SACM by ALAR High-flowanti-G valve under conditions of: 150-psig sourcepressure; 10-liter volume; 150 angle .................... 52

4.3-15 "Suit" pressures produced during SACM by ALAR High-flowanti-G valve under conditions of: 150-psig sourcepressure; 10-liter volume; 300 angle .................... 52

4.3-16 "Suit" pressures produced during SACM by ALAR High-flowanti-G valve under conditions of: 30-psig sourcepressure; 14-liter volume; 0' angle ................................................ 53

4.3-17 "Suit" pressures produced during SACM by ALAR High-flowanti-G valve under conditions of: 300-psig sourcepressure; 6-liter volume; 00 angle ................................................... 53

viii

Page 10: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

4.4 Garrett Electronic/Pneumatic Anti-G Valve

4.4-1 Garrett Electronic/Pneumatic Anti-G Valve .................. 554.4-2 Maximum flow through Garrett Electronic/Pneumatic

anti-G valve ....................................................................................... 574.4-3 "Suit" pressures produced during acceleration/deceleration

by Garrett Electronic/Pneumatic anti-G valve underconditions of: 150-psig source pressure; low G onset;10-liter volume; 00 angle ................................................................. . 57

4.4-4 "Suit" pressures produced during acceleration/decelerationby Garrett Electronic/Pneumatic anti-G valve underconditions of: 150-psig source pressure; low G onset;10-iter volume; 15 angle ................................................................ 58

4.4-5 "Suit" pressures produced during acceleration/decelerationby Garrett Electronic/Pneumatic anti-G valve underconditions of: 150-psig source pressure; low G onset;10-titer volume; 300 angle ................................................................ 59

4.4-6 "Suit" pressures produced during acceleration/decelerationby Garrett Electronic/Pneumatic anti-G valve underconditions of: 30-psig source pressure; low G onset;10-liter volume; 00 angle ................................................................. . 60

4.4-7 "Suit" pressures produced during acceleration/decelerationby Garrett Electronic/Pneumatic anti-G valve underconditions of: 300-psig source pressure; low G onset;10-liter volume; 00 angle ................................................................. . 60

4.4-8 "Suit" pressures produced during acceleration/decelerationby Garrett Electronic/Pneumatic anti-G valve underconditions of: 150-psig source pressure; high G onset;10-liter volum e; 00 angle .................................................................. 61

4.4-9 "Suit" pressures produced during acceleration/decelerationby Garrett Electronic/Pneumatic anti-G valve underconditions of: 150-psig source pressure; high G onset;10-liter volum e; 150 angle ............................................................... . 61

4.4-10 "Suit" pressures produced during acceleration/decelerationby Garrett Electronic/Pneumatic anti-G valve underconditions of: 150-psig source pressure; high G onset;10-liter volume; 300 angle ................................................................ 63

4.4-11 "Suit" pressures produced during acceleration/decelerationby Garrett Electronic/Pneumatic anti-G valve underconditions of: 30-psig source pressure; high G onset;10-liter volum e; 00 angle .................................................................. 63

4.4-12 "Suit" pressures produced during acceleration/decelerationby Garrett Electronic/Pneumatic anti-G valve underconditions of: 300-psig source pressure; high G onset;10-liter volum e; 00 angle .................................................................. 65

4.4-13 "Suit" pressures produced during SACM by GarrettElectronic/Pneumatic anti-G valve under conditions of:150-psig source pressure; 10-liter volume; 00 angle .................. 65

ix

Page 11: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

4.4-14 "Suit" pressures produced during SACM by GarrettElectronic/Pneumatic anti-G valve under conditions of:150-psig source pressure; 10-liter volume; 150 angle ................ 67

4.4-15 "Suit" pressures produced during SACM by GarrettElectronic/Pneumatic anti-G valve under conditions of:150-psig source pressure; 10-liter volume; 300 angle ................ 67

4.4-16 "Suit" pressures produced during SACM by GarrettElectronic/Pneumatic anti-G valve under conditions of:30-psig source pressure; 14-liter volume; 00 angle .................... 68

4.4-17 "Suit" pressures produced during SACM by GarrettElectronic/Pneumatic anti-G valve under conditions of:300-psig source pressure; 6-liter volume; 00 angle .................... 68

4.5 Hymatic VAG-1 10-022 Anti-G Valve

4.5-1 Hymatic VAG-1 10-022 Anti-G Valve ................................................... 704.5-2 Maximum flow through Hymatic VAG-1 10-022 anti-G valve .......... 734.5-3 "Suit" pressures produced during acceleration/deceleration

by Hymatic VAG-1 10-022 anti-G valve under conditionsof: 60-psig source pressure; low G onset;10-liter volum e; 00 angle ................................................................. . 73

4.5-4 "Suit" pressures produced during acceleration/decelerationby Hymatic VAG-1 10-022 anti-G valve under conditionsof: 60-psig source pressure; low G onset;10-liter volum e; 150 angle ................................................................ 74

4.5-5 "Suit" pressures produced during acceleration/decelerationby Hymatic VAG-1 10-022 anti-G valve under conditionsof: 60-psig source pressure; low G onset;10-liter volum e; 300 angle ................................................................ 74

4.5-6 "Suit" pressures produced during acceleration/decelerationby Hymatic VAG-1 10-022 anti-G valve under conditionsof: 30-psig source pressure; low G onset;10-liter volum e; 00 angle ................................................................. . 76

4.5-7 "Suit" pressures produced during acceleration/decelerationby Hymatic VAG-1 10-022 anti-G valve under conditionsof: 120-psig source pressure; low G onset;10-liter volume; 00 angle ................................................................. . 76

4.5-8 "Suit" pressures produced during acceleration/decelerationby Hymatic VAG-1 10-022 anti-G valve under conditionsof: 60-psig source pressure; high G onset;10-liter volum e; 00 angle ................................................................. . 77

4.5-9 "Suit" pressures produced during acceleration/decelerationby Hymatic VAG-1 10-022 anii-G valve under conditionsof: 60-psig source pressure; high G onset;10-liter volum e; 150 angle ................................................................ 77

4.5-10 "Suit" pressures produced during acceleration/decelerationby Hymatic VAG-1 10-022 anti-G valve under conditionsof: 60-psig source pressure; high G onset;10-liter volum e; 30 ° angle ............................................................... . 79

X

Page 12: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

4.5-11 "Suit" pressures produced during acceleration/decelerationby Hymatic VAG-1 10-022 anti-G valve under conditionsof: 30-psig source pressure; high G onset;10-liter volum e; 00 angle ................................................................. . 79

4.5-12 "Suit" pressures produced during acceleration/decelerationby Hymatic VAG-1 10-022 anti-G valve under conditionsof: 120-psig source pressure; high G onset;10-liter volum e; 00 angle ................................................................. . 80

4.5-13 "Suit" pressures produced during SACM byHymatic VAG-1 10-022 anti-G valve under conditions of:60-psig source pressure; 10-liter volume; 00 angle .................... 80

4.5-14 "Suit" pressures produced during SACM byHymatic VAG-1 10-022 anti-G valve under conditions of:60-psig source pressure; 10-liter volume; 150 angle .................. 81

4.5-15 "Suit" pressures produced during SACM byHymatic VAG-1 10-022 anti-G valve under conditions of:60-psig source pressure; 10-liter volume; 300 angle .................. 82

4.5-16 "Suit" pressures produced during SACM byHymatic VAG-1 10-022 anti-G valve under.conditions of:30-psig source pressure; 14-liter volume; 0* angle .................... 82

4.5-17 "Suit" pressures produced during SACM byHymatic VAG-1 10-022 anti-G valve under conditions.of:120-psig source pressure; 6-liter volume; 0 angle .................... 83

4.6 AAMRL Anti-G Valve

4.6-1 AAMRL Anti-G Valve ("Bang Bang") ................................................... 844.6-2 Maximum flow through AAMRL anti-G valve

w ith pressure off ................................................................................. 854.6-3 Maximum flow through AAMRL anti-G valve

with pressure on ................................................................................ . 854.6-4 "Suit" pressures produced during acceleration/deceleration

by AAMRL anti-G valve under conditions of: 1 50-psigsource pressure; low G onset; 10-liter volume; 00 angle ............ 87

4.6-5 "Suit" pressures produced during acceleration/decelerationby AAMRL anti-G valve under conditions of: 150-psigsource pressure; low G onset; 10-liter volume; 150 angle ......... 87

4.6-6 "Suit" pressures produced during acceleration/decelerationby AAMRL anti-G valve under conditions of: 1 50-psigsource pressure; low G onset; 10-liter volume; 300 angle ......... 88

4.6-7 "Suit" pressures produced during acceleration/decelerationby AAMRL anti-G valve under conditions of: 30-psigsource pressure; low G onset; 10-liter volume; 00 angle ............ 88

4.6-8 "Suit" pressures produced during acceleration/decelerationby AAMRL anti-G valve under ,onditions of: 300-psigsource pressure; low G onset; 1 0-liter volume; 00 angle ............ 90

4.6-9 "Suit" pressures produced during acceleration/decelerationby AAMRL anti-G valve under conditions of: 150-psigsource pressure; high G onset; 10-liter volume; 00 angle .......... 90

xi

Page 13: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

4.6-10 "Suit" pressures produced during acceleration/decelerationby AAMRL anti-G valve under conditions of: 150-psigsource pressure; high G onset; 10-liter volume; 150 angle ........ 91

4.6-11 "Suit" pressures produced during acceleration/decelerationby AAMRL anti-G valve under conditions of: 150-psigsource pressure; high G onset; 10-liter volume; 300 angle ........ 91

4.6-12 "Suit" pressures produced during acceleration/decelerationby AAMRL anti-G valve under conditions of: 30-psigsource pressure; high G onset; 10-liter volume; 00 angle .......... 92

4.6-13 "Suit" pressures produced during acceleration/decelerationby AAMRL anti-G valve under conditions of: 300-psigsource pressure; high G onset; 10-liter volume; 00 angle .......... 92

4.6-14 "Suit" pressures produced during SACM by AAMRLanti-G valve under conditions of: 150-psig sourcepressure; 10-liter volume; 00 angle ............................... 94

4.6-15 "Suit" pressures produced during SACM by AAMRLanti-G valve under conditions of: 150-psig sourcepressure; 10-liter volume; 150 angle .................... 94

4.6-16 "Suit" pressures produced during SACM by AAMRLanti-G valve under conditions of: 150-psig sourcepressure; 10-liter volume; 300 angle .................... 95

4.6-17 "Suit" pressures produced during SACM by AAMRLanti-G valve under conditions of: 30-psig sourcepressure; 14-liter volume; 00 angle ................................................ 95

4.6-18 "Suit" pressures produced during SACM by AAMRLanti-G valve under conditions of: 300-psig sourcepressure; 6-liter volume; 00 angle ................................................... 96

Table No,

3. TESTING EQUIPMENT AND PROCEDURES

3.2-1 Summary of Conditions Employed for Testing Anti-G Valves ........ 8

4. RESULTS AND DISCUSSION

4.0-1 Anti-G Suit Design Requirements ....................................................... 13

DISCUQ-SION AND CONCLUSIONS

5.0-1 Maximum Airflow Capacity (SCFM) as a Function ofAcceleration (Gz) for all Anti-G Valves ................... 97

5.0-2 Seconds Required to Attain Various Suit VolumePressures Under High-Onset Conditions .................. 99

5.0-3 Suit Volume Pressure (PSIG) Produced UnderLow-Onset Conditions as Function of Acceleration (Gz) ............ 100

xii

Page 14: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

5.0-4 Suit Volume Pressure (PSIG) Produced UnderHigh-Onset Conditions as Function of Acceleration (Gz) ........... 101

xiii

Page 15: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

RESEARCH AND DEVELOPMENT OF ANTI-G LIFE SUPPORT SYSTEMS:

Part 4. Engineering Test and Evaluation of Six Anti-G Valves

1. INTRODUCTION

Many U.S. Air Force (USAF) aerial combat maneuvers (ACMs) can result insevere physiological stress and hazardous performance deficits for aircrew membersdue to the high-onset +Gz acceleration encountered. Anti-G life-support equipment(LSE) helps to enable aircrew to manage physiological stress and maintainperformance.

Among the possible methods to improve tolerance to high-onset +Gz accelerationare state-of-the-art anti-G suits and anti-G valves, which together compose the totalanti-G LSE available to pilots of today's high-performance aircraft. Results of researchand development by Technology Incorporated on multiple-capstan and reticulated-foam anti-G suits were presented in Part 1 of this report. This volume (Part 4)describes characterization of current and prototype anti-G valves.

The anti-G valve (or G valve) is a special type of pressure regulator that deliversan output pressure proportional to the +Gz forces acting upon its control elements at agiven instant. The standard type of anti-G valve uses a movable mass as theacceleration-sensing element. This movable mass is aligned so that it applies forceproportional to +Gz forces to the pressure-regulating elements, which are usually aspring-loaded diaphragm and valve arrangement. At some designated +Gz threshold(usually around +2 Gz), the force exerted by the mass supplies the necessary force toopen the inlet valve. The inlet valve allows air to flow into the anti-G suit; as suitpressure increases, the back-force exerted on the diaphragm increases and tends tocounteract the force applied by the movable mass. When the suit pressure reaches alevel that causes the mass and diaphragm forces to cancel, flow ceases. A relief valvearrangement is also incorporated into the anti-G valve. This relief valve functions tolimit the maximum suit pressure, usually to within 8 to 11 psig.

Some anti-G valves incorporate an electronic, or servo, control mechanismwhereby a linear force motor controls the valve as a function of a G-related electricalsignal. Such an electronic valve offers a possibility of improved control of anti-G suitpressurization characteristics.

High sustained +Gz conditions are defined as exposure to +6 Gz and greater forperiods longer than 15 s. Newer high-performance aircraft, such as the F-1 5 and F-1 6are capable of producing +Gz conditions which approach or even exceed man'stolerance.

The inflation rates and schedules of anti-G valves have remained essentiallyunchanged since the 1940s; i.e., the valve is activated at +2 Gz and increases anti-Gsuit pressure at a rate of 1.5 psi per G to a maximum pressure of 10.5 psi. Untilrecently, this inflation schedule has been adequate (when used with an efficientstraining maneuver) to provide the increase in G-tolerance required in operational

1

Page 16: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

aircraft. With the advent of more maneuverable fighter aircraft, however, it has beenfound that the inflation schedule of the anti-G suit does not provide maximalacceleration protection, indicating a need for further research and development of anti-G valves.

The ability to screen prototype anti-G valves for performance characteristics isessential to the research, development, test, and evaluation (RDT&E) processes. Thisreport describes methods for testing anti-G valves for performance under simulatedhigh-onset and high-sustained G conditions; it also details performance characteristicsmeasured for current operational anti-G valves and several new prototype anti-Gvalves.

The term "High" G onset used in this text represents Gz-onset rates between 4and 6 Gz s-1. -The actual Gz-onset performance of the USAF School of AerospaceMedicine (USAFSAM) centrifuge is varied depending upon test requirements, and noattempt was made to standardize "High G Onset" beyond these limits.

2. OBJECTIVES

Results of work performed under Task 7, "Engineering Test and Evaluation of SixAnti-G Valves," Contract F33615-81-C-0600, are described here. This test andevaluation (T&E) effort was conducted on-site at Brooks Air Force Base using theUSAFSAM Centrifuge Facility, and monitored by USAFSAM/VNS.

One objective of Task 7 was to provide uniform testing procedures for evaluatingperformance characteristics of the anti-G valves included in this study. Performancecharacteristics of interest were the dynamic responses of the anti-G valves underclosely controlled conditions of: acceleration, source pressure, valve angles andsimulated suit volumes. Evaluation of parameters such as physical dimensions,material specifications, environmental suitability or static performance characteristicswas not within the scope of this task.

Another objective of this task was to measure and document actual dynamicperformance characteristics of the ALAR 8400A valve and 5 new or experimentalvalves. The 5 experimenta! ,alves ivaluated were:

(1) ALAR High-Flow Ready-Pressure;(2) ALAR High-Flow;(3) Garrett Electronic/Pneumatic Rapid-Response;(4) Hymatic VAG-1 10-022; and(5) AAMRL ("Bang Bang").

2

Page 17: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

3. TESTING EQUIPMENT AND PROCEDURES

The testing equipment and procedures employed here are similar, but notidentical, to those described in SAM-TR-79-30 (1). Similarity to this previousTechnical Report (TR) was desired in order to allow for comparisons with an existingdata base. The acceleration profiles used during the tests described in this report,however, differed from those employed in the previous work because much higheronset rates became attainable after completion of recent modifications to theUSAFSAM centrifuge. The onset rate is now approximately 6 G s-1 measured frombaseline Gz (+1.4 Gz) to +13 Gz, whereas the maximum onset rate used in theprevious measurements was 1.5 G s-1.

3.1 Testing Equipment

3.1.1 General Test Configurations

Two basic test configurations were employed for evaluating anti-Gvalves. One configuration was used for open-flow tests, i.e., tests in which no resis-tance to flow--such as would be provided by an anti-G suit--was encountered. A sche-matic of this configuration is shown in Figure 3.1-1. The second configuration wasidentical to the first except that a simulated anti-G suit volume and pressure transducerwere included downstream from the flow transducer as shown in Figure 3.1-2.

AIR FLOW

VALVE CONTROL

P Var XCR tacr

RECO RDER RECORDECO NTROL '-- CONTROLCO NSOLE >_ CONSOLE

Figure 3.1-1. Phase I Open-flow Capability Test Configuration. Fv - airflow; Ps TM source pressure;Po - valve outlet pressure; Gz = acceleration, XDCR - transducer; VOC - volts directcurrent; VAC a volts alternating current.

3

Page 18: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

FLOW

AIR XDCR

VALVE CONTROL SLIP RINGS

S O U C EO RGU A TR A -5 . A L V ER

VALVE(X COTO SLI RINTRS

BRUSHBRUSH

CONSL CONSOLE:

Figure 3.1-2. Phase II Dynamic Test Configuration. Fv airflow; Ps - source pressure; Po valveoutlet pressure; Gz - acceleration; Pv -volume pressure; XDCR a transducer; VDC voltsdirect current; VAC a volts alternating current.

For both test configurations, the pressure (air) source consisted of a standard "Kbottle" containing 222 SCF water-pumped breathing air at 2200 psig which wasinstalled in the gondola of the USAFSAM centrifuge. A high-flow regulator mounteddirectly on the air bottle was used to maintain the desired source pressure at the inletof the valve under test. This regulator was manually adjusted to maintain the desiredinlet pressure under flow conditions.

A remotely controlled solenoid valve was placed in the air line between theregulator output port and the inlet port of the anti-G valve. This solenoid valve wascontrolled by a relay that was activated by a 28 VDC signal sent through the slip ringsfrom a switch located at the control console.

The anti-G valve to be tested was mounted on a plate which was indexed indegrees and could be locked in place at any angular position; however, for these tests,only the 00, 150, and 300 positions were used. The plate was mounted on a specialtest stand which aligned the center of the anti-G valve with the gondola accelerometer.

4

Page 19: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

The simulated anti-G suit volume or "sink volume" used to terminate the flowthrough a valve under test consisted of a rubber bladder sewn between two layers oflow-stretch canvas material. This "flexible volume" was adjusted and optimized to rep-resent the incompressible volume required to fill a suit at 5 psig. The inherent stretchof the flexible volume was limited to an increase of 10% when the pressure wasincreased from 5 psig to 15 psig. To measure the internal pressure of the "sink vol-ume,m a small flexible catheter was inserted at the neck of the inlet hose and extendedinto the volume with the opening placed near the geometric center of the volume.

3.1.2 Test Confiauration Elements

3.1.2.1 Source Pressure (Ps)

The pressure-transducer port was located downstream fromthe solenoid valve and just upstream of the inlet port of the anti-G valve under test.This port was mounted on a section of smooth-bore copper tubing to minimize venturieffects and to provide the necessary rigidity to prevent kinking of the high-pressurehose used to plumb the test stand. Each valve was tested at three different sourcepressures; high, low, and median pressures which were consistent with the designspecifications of each individual valve were used.

3.1.2.2 Valve Output Pressure (P,)

The valve output pressure transducer port was locatedimmediately downstream from the outlet port of the anti-G valve under test. This portwas also on a section of smooth-bore copper tubing to minimize Venturi effects.

3.1.2.3 Air Flow (Fv

The flow transducer, which was located downstream from thevalve-output transducer port, was of the hot-wire type with a fast response tofluctuations in flow. The transducer was also temperature-compensated, and linearover the range of 1.0 - 40.0 standard cubic feet per minute (SCFM). The transducerwas plumbed into the system with pipe of dimensions recommended by themanufacturer to assure optimal performance.

3.1.2.4 Simulated Suit Volume Pressure (Pv.

The port for accessing the simulated suit volume pressure wasa small catheter extending from the neck of the inlet hose down into the center of thevolume. This apparatus necessitated the use of a low-compliance, low-internal-volume type of pressure transducer. The total volume of the catheter and the pressuretransducer combination was insignificant when compared to the simulated suitvolume. Suit volumes of 6, 10, and 14 liters were used.

. .. .............. . . . . - .- ,, i .= m nm m ~ lI I n

Page 20: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

3.1.2.5 Acceleration (+G,

Acceleration was measured in the +Z axis only--i.e.,perpendicular to the floor of the gondola. Measuring acceleration along other axeswas not included within the scope of this 'ask.

3.1.2.6 Valve Ancle

The valve under test was attached to a circular plate which wasindexed in degrees, such that the centerline of the valve and the centerline of the platewere superimposed. The geometric center of the circular plate was the singleattaching point between the indexed plate and the special test stand. The zero-degreeindex mark on the test stand was a line running perpendicular to the floor of thegondola. The valve could be rotated to, and firmly fixed at, any desired angle. Themounting plate was aligned as nearly as possible to the accelerometer in the gondola.For these tests only 00, 150, and 300 valve angles were used.

3.1.2.7 Pressure Transducers

The strain-gauge type pressure transducers were used formeasuring pressures. A Taber Teledyne type 176 transducer, which had a range of 0-500 psid, was used to measure source pressures. A Gould/Statham type PL131TC-50-350 transducer with a range or 0-50 psid, permanently mounted in the gondola,was used to measure valve output pressure. The simulated anti-G suit volumepressure was measured with a Statham type P6TC-20D-400 pressure transducer witha range of ± 20 psid.

3.1.2.8 Flow Meter

A Technology Incorporated linear flow meter, model LFC-20with a range of 0-40 SCFM, was used to measure airflow. The maximum error of thissystem was ±2% of reading or ± 0.5% full scale, whichever was greater; repeatabilitywas within 0.1% of flow rate measured. Response time for the sensor was < 0.07 s,and the system had a linear output of 0-5 VDC (0 to full-scale) available for recording.

3.1.2.9 Control Console Signal Conditioners

A Gould universal amplifier, model 13-4615-555604, was usedto condition each transducer output measured during testing. These amplifiers hadbuilt-in offset nulling and signal filtering. The direct current (DC) measurement rangewas 500 mV full-scale to 25 V full-scale. Attenuator inaccuracy was ±0.5% ofcalibrated step, and the nonlinearity was ± 0.1% of full-scale.

3.1.2-10 Strip-Chart Recorders

Two Gould model 2800S recorders were used to record alldata collected during these tests. The nonlinearity of these recorders was ±0.35% offull-scale; gain instability was ±0.1% of reading for 24 h and zero-line drift was ±0.1%of full-scale range for 24 h. The rise time (10% to 90% full-scale with less than 1%overshoot) was 5 ms.

6

Page 21: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

3.2 Testing Procedures

The performance evaluation tests for the anti-G valves were conducted in threephases, which are described later. All testing on the USAFSAM centrifuge wasperformed in the automated mode, using computer programs specifically written andtailored for these anti-G valve testing procedures. Automated operation providedgreater reproducibility than that attainable in the manual mode.

A test centrifuge run in which G levels were increased from baseline G (1.4 G) to11 G, held 5 s at 11 G, and then decreased back to baseline G, was defined as a"trapezoid run." This terminology arises from the shape of the actual +Gz profile tracedon a strip-chart recorder when time was recorded as one axis, which approximates atrapezoid.

3.2.1 Pnase I--Maximum Flow Capacity Testing

Phase I testing was designed to determine the maximum flow capabilityof the anti-G valve under test. The source pressure employed was the median valuefor the specific valve under test. Three trapezoidal runs were made using 0.5 G s-'onset and offset rates. Flow was the parameter of primary interest during these tests;however, the source piessure and valve output pressure were also closely monitored.The source pressure regulator was set empirically such that the pressure at the inlet ofthe valve under test remained within ±10% of the desired value.. The valve outputpressure was monitored to determine if there was any excessive restriction of airflowdownstream from the outlet port of the anti-G valve under test. The valve angle forthese tests was 00. Test conditions for each of the three phases are summarized inTable 3.2-1.

3.2.2 Phase Il--Dynamic Response Testing

Phase II testing was undertaken to determine the dynamic responsecapability of an anti-G valve. All measurements were obtained with the anti-G valveterminated in a flexible "sink volume" of 10 liters. Three 0.1-G s- 1 trapezoids were runat each source pressure (i.e., minimum, median, and maximum source pressure) withthe valve angle held at 0° . The two additional valve angles required three 0.1 G s-1trapezoids at each angle ( i.e., 150 and 300), while the source pressure was held at themedian value (Table 3.2-1). As used here, the term "output pressure" refers topressure within the suit volume, unless otherwise noted.

Three trapezoids (4.42 G s-1) were also run at each source pressure with thevalve angle held at 00 (Fig. 3.2-1). The two additional valve angles required threetrapezoids at each angle (i.e., 15° and 300), while the source pressure was held at themedian value (Table 3.2-1).

7

Page 22: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

3.2. Phase Ill-Comglex Dynamic Response Testing

Phase III testing was designed to provide a measure of the relativecapability of an anti-G valve to function under Simulated Aerial Combat Maneuver(SACM) conditions. The G profile used is shown in Figure 3.2-2. This SACM G profilewas repeated for each source pressure (minimum, median, and maximum), while thevalve angle was held at 00. The runs at minimum source pressure used the maximumsuit volume (14 liters); the runs at maximum source pressure used the minimum suitvolume (6 liters); and the runs at median source pressure used the median suit volume(10 liters). In addition, valves were tested under SACM conditions at valve angles of150 and 300 with median source and suit pressures.

TABLE 3.2-1. SUMMARY OF CONDITIONS EMPLOYED FOR TESTING ANTI-G VALVES

Gz-onset Source Suit Valve Other Valvesrate pressure volume angle parameters(Gz s-i) (psig) (liters) (0)

Phase I - Maximum flow (tr-zoidaI G omfile)

0.5 Med - 0 - all0.5 Med - 0 Eec "Offa AAMRL0.5 Med - RP "Off- b ALAR HFRP

Phase I - Dynamic response (tracezoidal G ormfilel

0.1 Min Med 0 - adHigh Min Med 0 - ad0.1 Med Med 0 - al0.1 Med Med 15 - ad0.1 Med Med 30 - adHigh Med Med 0 - adHigh Med Med 15 - allHigh Med Med 30 - all0.1 Max Med 0 - aHigh Max Med 0 - ad

Phase III -SACM Q o=fe

- Min Max 0 - al- Med Med 0 - ad- Med Med 15 - ad- Med Med 30 - al- Max Min 0 - al

a AAMRL valve normal operation/testing - electronics "On."b ALAR HFRP valve normal operation/testing - ready pressure *On."SACM - simulated aerial combat maneuver;Min, Med, Max - minimum, median, maximum values, respectively.

8

Page 23: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

I C -I A

/ '/

I ,

0 2 0 40 s so TO0o s 100 120 10 16 0

Figure 3.2-. Conditions of terePhase- " trapzof nit caleTriue prforae va uatinitests(p)ormeda all0 threedmaium(00pi source pressures, wt av nl eda ~

9 /--

- --- --

Fg.2-- 2. C diinso Phsef SAC Tetn / f "ni- vavs h - lewsrna iiu

(0 psi) meia (1 0 si) an aiu 0 si) s ourc pressures.

TtM(W9

Page 24: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

3.2.4 Simulated Suit Volume Measurement

Estimation of the simulated anti-G suit volume (the "sink volume") whichwas incorporated into test configuration two (Fig. 3.1-1) was performed by evacuatingthe sink components of the system under mild vacuum, and pressurizing to 5 psig byequilibration with a defined volume at determined initial and final pressures. Thevolume of the sink was then calculated from the pressure drop in the source volumeaccording to the following relationship:

(Po - Pr)(Vo)=V

Pi - Ps

when Po = initial pressure of known volume in psig;Pr = final pressure of known volume in psig;Ps = final suit pressure in psig;Pi = initial suit pressure in psig;Vo = known volume in in.3 (or liters); andVs = unknown suit volume in in.3 (or liters).

This relationship assumes that the temperature of the air does not change. TheVs and Vo are. the "incompressible volumes," or volumes of water required to fill thespace as opposed to the "standard air volume" or volume of air at standardtemperature and pressure which would be required to fill the volume at the subjectpressure.

3.2-5 Data Acauisition

When the electronic equipment, test stand, and control devices weresecured in the Centrifuge gondola, all required signal outputs were connected throughslip rings to the control-console patch panel. Each signal pair coming from thegondola was input in parallel to two preamplifiers, each preamplifier occupying thesame channel slot in separate but identical strip-chart recorders. The signal pairs fromthe gondola and brush recorder channels were standardized, in that they wereassigned the same parameters for all anti-G valves tested. A preprogrammable patchboard for routing the signals and control voltage was made up for the control-consolepatch panel so that the configuration was standardized for all tests.

The signal conditioners used were all of the same type and model. Signalconditioner gain was matched to the transducer output so that both were in the middleof the measurement range for the parameters monitored.

After all patching was completed, the coarse gain on all preamplifiers was setappropriately and all instrumentation was turned on (in stand-by position) for a warm-up period of at least 10 min. When the instrumentation had stabilized, all channelswere calibrated, using the strip-chart recorders as the readout. The usual calibrationpoints were "minimum" (almost always zero) and "maximum" (generally full-scale).The zero points were always calibrated first and any small offset in the transducer

10

Page 25: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

output was nulled out by using the zero-suppression controls incorporated into thepreamplifier. All pressure transducers were calibrated using the Datametrics digital-pressure calibration system installed in the gondola.

The flowmeter was calibrated by sending a DC calibration voltage through theslip rings to the preamplifier at the control console. The magnitude of the calibrationvoltage was determined from the calibration chart for the flow transducer. Thecalibration voltage was generated by a regulated DC power supply and monitored atthe gondola by a calibrated digital voltmeter.

The acceleration channels were calibrated by using the accelerometer calibratorwhich is permanently installed in the control console. The minimum calibration pointwas 1 G and the maximum was 10 G. The 1 G baseline corresponded to zerobaseline on the strip-chart recorder, thereby affording a full-scale value of 11 G.

The end products of the data-acquisition process were two identical strip-chartrecordings, each consisting of five channels of test data. Chart speeds for the strip-chart recorders were selected to afford the proper resolution, with respect to elapsedtime, for manual extraction of the required data points. One set of recorded data waskept as backup. The second set of recorded data was manually extracted from thestrip-chart recorder trace and transcribed onto standardized tabular forms for dataanalysis. The data from the strip-chart recordings were presented in final form eitheras X-Y plots or tabulated.

Access to the signal lines from the gondola to the control console was availablethrough a patch board on the back panel of the gondola interior. The signal lines wereAmerican wire gauge (AWG) #20 single-conductor shielded wire. These signal lineswere not continuous runs, but were divided into sections terminating or beginning atthe slip rings while interconnecting at terminal boards between the slip rings and thecontrol console. The average DC resistance of the signal lines was 4 ohms.

3.2.6 Data Analysis

Three iterations of each type of test run were performed. The dependentvariable during these measurements was either the open-flow capability of the anti-Gvalve, or the pressure developed by the anti-G valve interfacing with a simulated suitvolume. Independent variables were acceleration, source pressure, and simulatedsuit volumes. A special, but useful, independent variable was time (in seconds) whichis a function of the acceleration profile. The response times of the valves could beaccurately determined since acceleration profiles were standardized and runs werecomputer controlled.

For the trapezoidal runs, both ascending and descending portions were handledidentically for the purposes of extraction of data-point values. Values of open flows(Fv) and simulated suit volume pressures (Pv) were extracted from the strip-chartrecordings at intervals of 0.5 G unless there was a stepwise response, or othernonlinear response, to the smooth acceleration rate of the centrifuge. In regions ofnonlinear responses, data points were taken at sufficiently close G intervals to furnishenough resolution to accurately display the results on the final graphs. After the runwas holding steady at 11 G, data points were taken at 1.5-s intervals; unless there

11

Page 26: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

were fluctuations in the defined parameters. If there were fluctuations, data pointswere extracted at sufficiently close time intervals to allow graphing.

For the SACM runs, procedures for extracting data-point values were the same asjust described for the trapezoidal runs. Acceleration profiles for these runs were morecomplex; therefore, smaller time intervals were required for sampling.

The strip-chart recordings were grouped into sets, each set consisting of threeiterations of the same run conditions. Each iteration was inspected and comparedwithin its set to ensure that the data had been precisely recorded. Numerical valuesfor Fv and Pv for all iterations were transcribed onto a form designed for this oper-ation*. Arithmetical means for each data point were obtained. Working graphs weremade on rectilinear graph paper, plotting mean Fy and Pv vs. G; and graphs andnumerical data were cross-checked to verify that graphical values were accurate.

Data obtained from all anti-G valves are compared in summary tables (Tables5.0-1 - 5.0-4). These tables compare means of values obtained from each valve for aspecified parameter under varying G conditions. Table 5.0-1 summarizes means forairflow vs. acceleration; Table 5.0-2 shows means for time vs. volume pressure; Table5.0-3 compares means for volume pressure vs. acceleration low onset; and Table5.0-4 compares means for volume pressure vs. acceleration high onset.

For graphical presentation, both the ascending (increasing Gz) and descending(decreasing Gz) portions of the trapezoidal runs for each anti-G valve are presented onone graph. The independent variables, acceleration (Gz) and time (s), arerepresented along the x-axis. In the ascending regions, Fv and Pv were taken atintervals of 0.5 G from baseline G (1.4 G) to 11 G; when acceleration had plateaued at11 G, data points were taken at intervals of 1.5 s during the plateau phase. In thedescending regions, Fv and Pv were taken at intervals of 0.5 G from 11 G to baselineG, beginning at the instant that deceleration started and ending when baseline G wasreached. For the purposes of time evaluation, "zero" time corresponded to the instantthat deceleration began, and the end point was defined as 1.5 s after baseline G wasreached.

4. RESULTS AND DISCUSSION

The anti-G suit design requires that the suit be linearly pressurized, and containbetween 0.1 psig and 1.2 psig at 2 Gz and between 8.7 and 11.0 psig at 10 Gz.Performance characteristics of all acceptable anti-G valves must be consistent with thedesign requirements, which are restated in tabular form in Table 4.0-1.

* Copies of these forms which show all experimental data recorded for each

iteration, for all valves, are available as Appendixes A-F.

12

Page 27: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

TABLE 4.0-1. ANTI-G SUIT DESIGN REQUIREMENTS

Gz Suit pressure*(psig)

2 0.1- 1.2

10 8.7-11.0

Suit must be linearly pressurized.

4.1 ALAR 8400A Anti-G Valve

4.1.1 Description

The ALAR 8400A anti-G valve (Fig. 4.1-1) is manufactured by ALARProducts, Inc., Macedonia, Ohio. This valve employs a mass/spring system for sensingacceleration and regulating anti-G suit pressure. Source pressure (30 to 300 psig)connects to an inlet fitting on the left side of the valve; the outlet to the anti-G suit is onthe right side.

As Gz force is encountered, the mass at the top of the ALAR 8400A valve is forceddown against the spring, pushing against a diaphragm regulator assembly and valvestem which opens a flow path to the suit outlet at the right of the valve. As pressurebuilds up in the suit, back pressure against the diaphragm reduces flow until the Gzforce and the suit pressure are balanced, at which time the valve is closed. As Gzforce is reduced, the spring moves the mass assembly and diaphragm upwardopening the exhaust valve and relieving the suit pressure until Gz force and pressureare again matched. When the valve is returned to conditions of 1 Gz, the valve ventsthe suit back to ambient pressure.

The ALAR 8400A is designed to actuate (i.e., begin to apply suit pressure) be-tween 1.5 Gz and 2 Gz. The valve is fitted with a spring-loaded relief valve with suffi-cient flow capacity to limit the suit pressure to 11 psig with 300-psig source pressure.The relief valve is designed to open between 8.7 psig and 11.0 psig. An exposedbutton located at the top of the 8400A valve assembly allows the G-sensing mass to bedepressed manually to provide a functional test feature (i.e., "press-to-test").

4.1.2 Phase I - Determination of Maximum Flow Canacity

To determine maximum flow capacity of the valve, open-flow conditionswere established by omitting the terminating ("suit") volume, while leaving the normalhose attached to the output port of the flow transducer. The hose is terminated by thefemale quick-disconnect fitting used for attaching the simulated anti-G suit volume.The source pressure used for these tests was 150 psig, the selected median pressure.

13

Page 28: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

Figure 4.1 -1. ALAR 8400A anti-G valve.

14

Page 29: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

Open-flow characteristics of the ALAR 8400A valve are shown in Figure 4.1-2.Flow began at approximately 1.85 G and increased rapidly and almost linearly to avalue of 7.6 SCFM at 4.3 G; a small shoulder occurred at this point with flow steady at7.6 SCFM between 4.3 G and 4.6 G. Flow rapidly increased again, curvilineady, to amaximum value of 11.1 SCFM at 7.5 G; flow then remained steady at 11.1 SCFM from7.5 G through the duration at 11 G. Formation of the shoulder is consistent betweenruns; it is therefore presumed to be an inherent characteristic of this ALAR 8400Avalve.

4.1.3 Phase 11 - Determination of Dynamic Resoonse Caoability

The dynamic response capability of the ALAR anti-G valve wasdetermined as described in Testing Procedures, section 3.2.2. Pressure delivered bythe valve was measured at the output port and inside the simulated anti-G suit volume.As used here, the term "output pressure" refers to the pressure within the suit volume,unless otherwise noted.

4.1.3.1 Low Gz-Onset-Rate Conditions

The simulated suit volume used for all low Gz-onset measure-ments was 10 liters, and the G profile was the 0.1 G s-1 trapezoidal profile. Thepressure delivered by the valve was measured inside the simulated anti-G suitvolume. "Output pressure" as used here refers to the pressure within the suit volume,unless otherwise stated.

When tested under conditions of 150-psig source pressure, 10-liter volume, and avalve angle of 00, the ALAR 8400A responded in an almost linear fashion between thepoints where it opened and where its maximum output pressure was reached(Fig. 4.1-3). The valve opened at 2.3 G and delivered a maximum pressure of 7.1 psigat a level of 8 G. The relief valve was actuated at 7.5 G and rapidly began limiting themaximum pressure delivered into the suit volume. The average ALAR 8400Apressurization rate was 1.4 psig G- 1 over its useful range.

Angle-dependent measurements employed valve angles of 150 and 300 while thesource pressure and simulated anti-G suit volume were held at 150 psig and 10 liters,respectively. At an angle of 150, the valve opened at 2.5 G and then responded almostlinearly from 3 to 8 G (Fig. 4.1-4). At 8 G, the delivered pressure was 6.7 psig, and therelief valve actuated to effect a slowing of the pressurization rate. The pressureincreased slightly and curvilinearly from 8 G to 9.5 G, corresponding to a maximumvalue of 7.0 psig.

At an angle of 300, opening of the valve was delayed until 3 G was reached, butnormal response was not attained until 3.5 G (Fig. 4.1-5). From 3.5 G to 9.5 G, theresponse was essentially linear except for small deviations (approximately 0.2 psig).At 9.5 G the valve delivered 6.5 psig and then began limiting the pressurization rate.There was some slight wandering of the pressure while "on top" at 11 G, but thepressure essentially continued to increase slowly and reached a maximum of 7.2 psigonly after deceleration had begun and progressed to 8.5 G.

15

Page 30: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

i I I I I I I I I I I I I I I I

I 7 1 I I I I I I1

1 1.4 3 4 Il 11 1 1 10 9 8 7 6 51 * !4 4 I

I I I II l lI I , , , ,

I I I / If I II II ,I i I i I J I I I ! IrT I I I i I I I II / I ) II I I I I i J I i l , . I I I

11.4Z 3 4 5 6 7 6 9 10 I i II II I10 9 8 7 6 5 4 3 2 L4L414 G,'. 1[K. I

0- ~ C N S U e. q 9 * o - TIME(39C)~

Figure 4.1-2. Maximum flow through ALAR 8400A anti-G valve.

It 11 -I Ill

- I, ' I - { - -- I......1...I I L.I - -

- ' _ II I I III!! I.! I i JI iIiI- i. . -. I ,

I! I I 1 /' __ I I I I I I I I "

1142 34 4 67 69 IOhI II IIi 1 09 8 7 " 5 4 3 2L4L.414 G.t .I[

o o000o0o 0 oo 0 N 0 000 0 0 o 0

Figure 4.1-3. "Suit" pressures produced during accelerationldeceleration by ALAR 8400A anti-G valveunder conditions of: 150-psig source pressure; low G onset; 10-liter volume; 00 angle.

16

Page 31: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

Two other source pressure-dependent measurements were obtained with a valveangle of 00 and suit volume of 10 liters. Minimum (30 psig) and maximum (300 psig)source pressures were investigated. For 30-psig source pressure (Fig. 4.1-6), thevalve opened at 2.2 G but normal operation appeared to begin at 2.5 G. The outputpressure increased almost linearly from 2.5 G to 7.7 G, where the valve began limitingthe pressure to 6.5 psig. A slight drop in pressure (0.1 psig) occurred while "on top" at11 G. The specified operating threshold of the relief valve apparently was not reachedduring this study.

At a source pressure of 300 psig (Fig. 4.1-7), the valve again opened at 2.2 G andbegan a slow increase in output pressure until 2.5 G was reached, at which pointnormal opeiation began. Linear operation began at 2.5 G and continued until 7.6 Gwas reached. At 7.6 G, the valve rapidly began to limit the output pressure to 7.0 psig.Under these source pressure conditions (300 psig), opening of the relief valveappeared to have little effect on the output pressure of this valve.

I I I I I I i I I I i_,I I I I t ....I...4....{ I I [

= ,I i I I I I

1 3 1 1/ II I I I I I I I I ! 1

I 1.4 2 4 1 5 6 + g 9 10 II I 1 1 II 1 0 9 S 4 3 2L4 L41.4 G L'VELI. .. a 0 0 0 0(Si

Figure 4.1-4. "Suit" pressures produced during acceleration/deceleration by ALAR 8400A anti-G valveunder conditions of: 150-psig source pressure; low G onset; 1 0-liter volume; 150 angle.

17

Page 32: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

',I I ! i I I i i I i I I i

iI II

,11[I I\ T I I i

,.,,ij.....tit I ______!

L Ol 1 1 14 111 1.4#2 3 4 5 6 7 S 9 10 I I II II II 0 I o a 7 5 4 3 2L41.41.4 Gt.!V L

Figure 4.1-5. "Suito pressures produced during acceleration/deceleration by ALAR 8400A anti-G valveunder conditions of: 150-psig source pressure; low G onset; 10liter volume; 300 angle.

-I . . - I I I I

" I I I I I Ili I 111111 I !

't "/']' , II II1A2 3 4 3 6 7 S 9 10 I 1 11 1 1 1 7 6 2L4L4L4 G, LEV

fa h. I C C ' 0

Figure 4.1-6. "Suit" pressures produced during acceleration/deceleration by ALAR 8400A anti-G valveunder conditions of: 30-psig source pressure; low G onset; 10-liter volume; 0* angle.

18

Page 33: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

I tI I 1 71Si I K I I I I I I I I I I I I ! I

1 17 1 1[.N I I Ii i 1

I N I I I I I IT71I I

11.42 4 5 6 7 a 9 I0 11 1 II I 0 9 8 7 6 5 4 3 2 .4 L4L4 G. EV L

e~~~- 0 ai a *6 on ft 0 -

Figure 4.1-7. "Suir pressures produced during acceleration/deceleration by ALAR 8400A anti-G valveunder conditions of: 300-psig source pressure; low G onset: 10-liter volume; 00 angle.

4.1.3.2 High G,-Onset-Rate Conditions

The simulated anti-G suit volume used in all high-onset testswas 10 liters. The G profile for all tests was the 6 G s- 1 trapezoidal profile. Thepressure delivered by the valve was measured inside the simulated anti-G suitvolume. References to output pressure indicate pressure within the suit volume,unless otherwise defined.

When a source pressure of 150 psig and valve angle of 00 were employed(Fig. 4.1-8), the pressurization schedule (psig G- 1) significantly lagged that observedduring low Gz-onset testing, seen in Figure 4.1-3. The ALAR 8400A valve initiallybegan pressurization at 2.7 G, but oscillation was observed at the valve outlet portbeginning at 1.4 G. The valve oscillated over the interval from 1.4 G to 3 G at anamplitude of 0.7 psig peak-to-peak (p-p). During acceleration from 2.7 G to 11 G,pressurization occurred at a less-than-linear rate; at the instant that 11 G was reached,suit pressure was 6.0 psig. During the time that acceleration was plateaued at 11 G,output pressure rapidly increased to 9.6 psig.

At the instant of deceleration, the valve began oscillating at a low frequency at anamplitude of 0.2 psig; and continued to oscillate until 6 G was reached. Outputpressure remained at 9.6 psig until deceleration had reached the 10-G level and thenbegan to decrease almost linearly until it reached 2.7 psig at 2 G. The valve closed atapproximately 2 G and the pressure vented rapidly through the exhaust until baselineG was reached at 1.4 G.

19

Page 34: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

The angle-dependent studies of the ALAR 8400A during Phase II high-Gz onsettesting used valve angles of 150 and 300 while the source pressure and simulatedanti-G suit volume were kept at their median values of 150 psig and 10 liters. At avalve angle of 150 (Fig. 4.1-9), the valve began pressurization at 2.5 G; however, alow-frequency oscillation at an amplitude of 0.3 psig p-p was observed at 1.5 G. Thislow-level oscillation occurred during the interval from 1.5 G to 3 G; the valve began tooperate normally after the oscillation ceased at 3 G and 0.2 psig pressure haddeveloped within the suit volume. The pressurization rate was linear from 0.2 psig atthe 3-G level up to 10 G, although the slope of the line decreased somewhat from 6 Gto 10 G (4.7 psig); at 10 G the slope began to increase, reaching 5.9 psig at the instantthat 11 G was attained. After 1 s won topm at 11 G, the pressure increased sharply to 8.8psig, then slowly climbed to the maximum pressure of 9.3 psig by the end of the 5-sinterval at 11 G.

At the instant that deceleration began, a 0.2 psig p-p oscillation was detected atthe* valve outlet port. This oscillation continued until the 6-G level was reached. Thepressure decreased slowly from 9.3 psig to 8.8 psig during the interval from 11 G to 9G, the rate of decrease then became essentially linear from this point down to 2 G(2.5 psig). Since the valve had closed and only the exhaust function was enabled, thepressure decreased rapidly to essentially zero at baseline G.

i 1 2 j i ,

uso anleII ___"20- I ( I ",

F II.11' I I } ,_ .FF

Figure 4.1-8. "Suit" pressures produced during acceleration/deceleration by ALAR 8400A anti-G valveunder conditions of: 150-psig source pressure; high G onset; 10-liter volume; 0° angle.

20

Page 35: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

Ii I I.

I T I I

I jL " - ...!

, 4I I I-- 7 i/-, I I _ _\,7 M I I I 1 i -1 -1 7 1 .1 i I I !

11.42 3 4 6 O i i9 1 1 if 11 1O 9 6i 7 6 5 4 3 21.4L4L4 G0 LVtL, ,- r * .I , i l 1 I

Figure 4.1-9. *Suit" pressures produced during acceleration/deceleration by ALAR 8400A arti-G valveunder conditions of: 150-psig source pressure; high G onset; 10-liter volume; 15 angle.

At a valve angle of 300 (Fig. 4.1-10), the valve opened at 3.3 G; the pressurizationrate was much slower than that observed during the 00 valve angle testing (Fig. 4.1-8).When acceleration reached 11 G, pressure was 4.9 psig. During 1.5 s at 11 G,pressure rapidly increased to 8.4 psig, then slowly increased to a maximum value of9.0 psig near the end of the 5-s interval.

At the instant that deceleration began, an oscillation of 0.2 psig p-p was observedat the outlet of the valve; this oscillation continued until 7 G was reached. The pres-sure remained at 9.0 psig until the G level had decreased to 10 G; it decreased linearlyfrom 8.5 psig to 2.3 psig over the interval from 9 G to 2 G, where the valve closed andthe exhaust function allowed the pressure to rapidly decrease to essentially zero atbaseline G.

Two source pressure-dependent studies of the ALAR 8400A were conducted:one at 30 psig and the other at 300 psig. Both studies used the 0 valve angle. The30-psig source pressure study (Fig. 4.1-11) revealed that a low-level oscillation of0.2 psig p-p occurred over the interval from 1.4 G to 3 G. Pressure began to increaseat 2.8 G and at the instant that 11 G was reached, the suit.pressure had reached only4.5 psig; pressure increased rapidly over the next 15 s to a level of 7.8 psig. The maxi-mum pressure developed was 8.3 psig, which occurred at the end of the time at 11 G.The pressure remained at 8.3 psig until the G level had decreased to 9 G, at whichtime a low-amplitude oscillation and a decrease in pressure occurred simultaneously.The oscillation persisted over the interval from 9 G to 6 G. Pressure decreased at anearly linear rate over the interval from 8.0 G to 2.0 G; however, as soon as the valveclosed, the pressure dropped rapidly to essentially zero at baseline G.

21

Page 36: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

I I 8 I L__ L_ I I UWEI I I ( [ I' )'

4, C4 w P! i m ;Q q ; L, ( M P. tv I (- : I I !I IL t t I t i t t t t I _ ,____, )

4-q~ - * - . . ..- ~, . . ,, - q

Figure 4.1-10. "Suit" pressures produced during acceleration/ deceleration by ALAR 8400A anti-G valveunder conditions of: 150-psig source pressure; high G onset; 10-liter volume; 300 angle.

42 3 401 i i 4 i i 4 ,I I I II IIIIE1.... I II

'tI I /T 41 I I1 I I I,1 ) i i ) ! !I ] zII ,,. i i Ii lIII ) Ii i\ I Ii

--, _ _ _ _ _ __,_ _ _ _ _ __i_ _ _ _ _ _ _

- I I I I" ' I l I I I L.I . Ei .N ) 9f 0 1 0 N I

Figure 4.1-11. "Suit" pressures produced during acceleration/deceleration by ALAR 8400A anti-G valveunder conditions of: 30-psig source pressure; high G onset; 10-liter volume: 00 angle.

22

Page 37: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

The 300-psig source pressure testing (Fig. 4.1-12) did not reveal the oscillationsthat were previously seen during the increasing-acceleration portions of the G profile.The valve opened at 2.5 G and the pressurization rate was almost linear over theinterval from 3 G to 11 G. The pressure was 5.9 psig-at the instant 11 G was reached,but after 2 s "on top," the pressure had rapidly increased to 7.6 psig and stabilized.

During the deceleration portions of the G profile, the pressure showed nodecrease until 8.5 G was reached. A low-amplitude oscillation began at 8 G andpersisted until 6 G was reached. The pressure decreased from 7 psig (7 G) to 2.5 psig(2 G) at a nearly linear rate. After the valve had closed, the pressure rapidlydecreased to essentially zero at baseline G.

4.1.4 Phase III - Determination of Complex Dynamic Response Capability

For testing of the ALAR 8400A anti-G valve under SACM conditions, avalve angle of 00 was used for both minimum and maximum source-pressure testing.

Testing the valve under SACM conditions with a source pressure of 150 psig andvalve angle of 00 showed a considerable lag between the valve and G profile(Fig. 4.1-13). On the first high-onset run to 9 G, the "suit" pressure was 8.2 psig andslowly climbed to a maximum of 8.4 psig, which was higher than the maximumpressure on any other run during the testing of this valve. The valve had a lowoscillation (0.4 psig p-p) between 3 G and 5 G and again on deceleration between 9 Gand 6 G. This oscillation was seen only in the valve outlet pressure and not in the suit

T f i 1f T I 1 I1......... I.......I / -' , - - - - j . - -

..A,.4.H I ;zz..J T! H iz zIzz

1 1.42. 3 4 5 6 . S' S l 0' I! I I 1 I0 7- -L-- ' 3 2 14IA!.4 G 1lE -VE=!i

Figure 4.1-12. "Suit" pressures produced during acceleration/deceleration by ALAR 8400A anti-G valveunder conditions of: 300-psig source pressure; high G onset; 10-liter volume; 00 angle.

23

Page 38: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

L I1 1 1 i I II ~ ................. t t i I Ii t

- I -! 1 i tLJ 'K - -

• JJ I I II

0LLL t " ___ 2. 6, LVEL

0 0 to 30 40 50 Go TO s0 so 100 110 TIME (SEC

Figure 4.1-13. "Suit pressures produced during SACM by ALAR 8400A anti-G valve under conditionsof: 150-psig source pressure; 10-liter volume; 00 angle.

pressure. At 4.4 G the pressure was 4.3 psig and after about 9.5 s the suit pressuredropped to 3.4 psig and again oscillated between 5 G and 4.4 G. At 3 G the suitpressure was 2.1 psig, dropping off after about 5 s to 1.8 psig. Accelerating again to 5G, the pressure was 4.0 psig and leveled off at 4.3 psig after about 8.5 s. Deceleratingto 4 G produced a pressure of 3.3 psig, which after about 4.5 s leveled off to 2.9 psig.High-onset acceleration again to 9 G produced a suit pressure of 6.6 psig, which after1.5 s leveled off to 8.1 psig. A high-offset rate to 1.5 G caused a pressure of 1.6 psigwhich slowly dropped to 0.2 psig. A subsequent rapid-onset acceleration to 8 Gproduced a pressure of 4.4 psig at 8 G, and a leveling off at 7.7 psig after about 1.5 s.Decelerating from 8 G to 1.4 G caused the suit pressure to drop rapidly to 0.2 psig,which after about 1 s went to zero.

Angle-dependent studies of the ALAR 8400A under SACM conditions used valveangles of 150 and 300, source pressure of 150 psig, and simulated anti-G suit volumeof 10 liters. At a valve angle of 150 (Fig. 4.1-14), the valve began pressurization at 3 G;a low-level oscillation (0.4 psig p-p) was observed between 1.5 and 3 G. The valvebegan to operate normally after the oscillation stopped, and the "suit" pressure rose to1 psig at 3 G. A high-onset acceleration to 9 G caused the pressure to rise rapidly to5.5 psig, and after about 1.5 s "on top" at 9 G, the "suit" pressure leveled off at 8 psig.

With a rapid-offset rate from 9 G to 4.4 G, the suit pressure decreased rapidly to4 psig, and after about 9.5 s the pressure slowly decreased to, and leveled off at, 3.1psig (Fig. 4.1-14). Low-level oscillation (0.2 psig p-p) was observed again between 9G and 5.5 G. Decelerating from 4.4 G to 3 G at a slow descent rate, the pressuredropped to 2 psig and then slowly to 1.6 psig at the end of the 3-G interval.

Accelerating back up to 5 G, the pressure slowly climbed to 2 psig at 4 G and thenquickly rose to 3.5 psig. After 2.5 s into the 5-G plateau, the pressure steadily

24

Page 39: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

increased to, and leveled off at, 4 psig. At 4 G the pressure decreased to 2.5 psig.With a rapid-onset acceleration to 9 G and some low oscillation (0.3 psig p-p, from 4 Gto 5.5 G), the pressure climbed rapidly to 6.1 psig and then steadily increased to 7.9psig over the remaining 9-G interval.

Leaving 9 G with a high-offset rate down to 1.5 G, the pressure dropped to 1.4psig at 1.5 G, and after about 5 s, decreased essentially to zero. Low-level oscillation(0.2 psig p-p) was again observed between 9 G and 6 G.

With another high-onset-rate acceleration from 1.5 G to 8 G, the suit pressureincreased rapidly and linearly to 4.2 psig at 8 G and low oscillation occurred between1.5 G and 3.5 G. After about 1.5 s, the pressure increased to 7.8 psig.

Returning to baseline G (1.5 G) with a high-offset rate, the pressure decreasedlinearly to zero with low oscillation occurring between 4.5 G and 1.5 G. Thepressurization rate under these conditions (150 angle) was a little slower (0.2 psi G-1)than for the 00-angle conditions.

When a valve angle of 300 was employed (Fig. 4.1-15), test results exhibited thesame characteristics as those from the 150-angle tests, except that the pressurizationrate was still slower. The rate for the 300 angle was 0.6 psi G-1 slower than that for 150angle, and 0.8 psi G-1 slower than the 00 angle testing.

Two source pressure-dependent experiments were conducted on the ALAR8400A valve while the valve angle was held at 0°. One experiment used a sourcepressure of 30-psig (the design minimum source pressure) with a 14-liter simulatedsuit volume (Fig. 4.1-16). The other experiment used the design maximum sourcepressure of 300 psig and a 6-liter simulated suit volume (Fig. 4.1-17).

For minimum source pressure conditions (i.e., 30-psig source pressure and a suitvolume of 14 liters (Fig. 4.1-16)), the valve started to pressurize at 3 G with a suit pres-sure of 0.05 psig. A low oscillation was observed between 1.5 G and 3 G. With a high-onset acceleration from 3 G to 9 G, the pressure increased to 3.7 psig with a low oscil-lation of 0.3 psig p-p at 3.5 G. After 7 s at 9 G the suit pressure increased to 8.2 psig.Upon decreasing from 9 G to 4.4 G, the pressure dropped to 4.8 psig; a low-level oscil-lation occurred from 9 G to 6 G (0.3 psig p-p). After 9.5 s at 4.4 G, the suit pressuredecreased to 3.5 psig with a low-level oscillation from 4.4 G (0.4 psig p-p). The G levelwas then decreased to 3 G, yielding a suit pressure of 2.3 psig which dropped to2.0 psig within about 7 s. At a low-onset rate to the 5-G level, suit pressure climbedgradually to 3.8 psig then leveled off at 4.3 psig. As G level decreased to 4 G, suitpressure dropped to 3.2 psig, then 3.0 psig. Upon rapid-onset acceleration to 9 G, suitpressure rose to 4 psig, then to 7.8 psig within 7 s; a low oscillation of 0.3 psig p-p wasobserved at 4.5 G. Rapid-offset deceleration from 9 to 1.5 G resulted in the suitpressure dropping to 2.6 psig; after 5.5 s at 1.5 G, suit pressure dropped to 0.2 psig. Anoscillation (0.2 psig p-p) occurred between 9 G and 6 G. A subsequent high-onsetacceleration to 8 G produced a suit pressure of 2.4 psig which then increased to 8 psigafter 9 s; this was accompanied by an oscillation of 0.2 psig p-p between 1.5 G and 3 G.A high-offset deceleration down to 1.4 G resulted in a decrease in suit pressure to0.2 psig, with an oscillation ( 0.3 psig p-p) between 3.5 G and 1.5 G.

25

Page 40: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

W1 T'_ I I I - 0 0 1 f, LEVfL" I I I ) J~ ~ I

111111..... 3 IIVJKKLi

i ii I 17 i I I I .It - ..III[Z IIIt€

0 10 ZO 30 40 5o s 0 6o o 100 110 IME (39C)

Figure 4.1-14. "Suit" pressures produced during SACM by ALAR 8400A anti-G valve under conditionsof: 150-psig source pressure; 10-liter volume; 150 angle.

I i I I ....... - I I I I - 1 -

I I I II£ -. - - I I '

S I I II ±I.

ww 4 3.0 .4 P-**-! L, G, LM -0 10 20 30 4 5 0 70 , 0 g0 100 t0 o TIM (SEC)

Figure 4.1-15. "Suit" pressures produced during SACM by ALAR 8400A anti-G valve under conditionsof: 150-psig source pressure; 10-liter volume; 300 angle.

26

Page 41: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

I I I i i

,o I t - - - - t tI I i I -- '' J I . I I I I i'1 i I 1 j t'iF i 1 i I I ]

-I I I i I I _ I I I i I I

-4 I . i i .4I I\ 0 IS " f , LEI-' b !_ H rH I!i I L I

0 0 70 30 40 SO so 0 70 , O i 00 6 IEMC

Figure 4.1-16. "Suit" pressures produced during SACM by ALAR 8400A anti-G valve under conditionsof: 30-psig source pressure; 14-liter volume; 00 angle.

'° I 1 __ T h 1 1

Zz"rt!Ii i±- I iiEII - -

S" I i I D 7 I I. .. I, I I I ! II

I V I' I ___ ___ i_3-. . .LL. i! LIJ ILI\ IJIIIIIL., _ __i I t r i l

- J... tI I f ! l \ I ! ! i i-o1_/I ! . .l .Jt I L t; I* II I t wLJi

L-3.- - 19-. 1 4.4 _1 . 1 :. ' ?LE._v(. I

Figure 4.1-17. "Suit" pressures produced during SACM by ALAR 8400A ant:-G valve under conditionsof: 300-psig source pressure; 6-liter volume; Q° angle.

27

3-|

Page 42: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

Under maximum source pressure (300 psig) and a simulated suit volume of6 liters (Fig. 4.1.-17), the suit started to inflate at 3 G with a suit pressure of 0.1 psig;after 12 s at 3 G suit pressure was 1.4 psig; a low-level oscillation occurred between1.5 G and 3 G. A rapid-onset acceleration to 9 G resulted in a suit pressure of 6.4 psigat 9 G, which increased to 7.6 psig within 7 s at the 9-G level; an oscillation of 0.4 psigp-p was observed between 3 G and 5.5 G. During a rapid-offset deceleration to 4.4 G,the valve oscillated between 8.5 G and 6.5 G (0.2 psig p-p) and again between 5 Gand 4.4 G (0.2 psig p-p). Suit pressure decreased to 3.6 psig as 4.4 G was reached,and dropped to 3.4 psig witnin 9 s at this G level. A deceleration to 3 G produced asuit pressure of 2 psig, which decreased to 1.6 psig after 5 s. Acceleration to 5 G yield-ed a suit pressure of 3.9 psig, while dropping back to 4 G resulted in a suit pressure of3.3 psig. A rapid-onset acceleration to 9 G gave a suit pressure of 6.7 psig, and a low-level oscillation (0.4 psig p-p) between 4 G and 7 G. Rapid deceleration to 1.5 Gcaused the suit pressure to fall rapidly to 0.7 psig; after 5.5 s at 1.5 G, the pressuredecreased to 0.2 psig. A subsequent high-onset acceleration to 8 G produced a suitpressure of 7.6 psig, which stabilized at 7.5 psig within 1 s. Rapidly decreasing the Glevel down to 1.4 G caused suit pressure to drop to 0.2 psig, and produced a low-leveloscillation (0.2 psig p-p) between 4.5 G and 1.5 G.

4.2 ALAR High-Flow Ready-Pressure Anti-G Valve

4.2.1 Description

The ALAR High-Flow Ready-Pressure (ALAR HFRP) anti-G valve (Fig.4.2-1) is a modified version of the ALAS 8400A. The 8400A valve was modified toobtain a 50% increase in airflow through enlargement of orifices, and an increase infirst stage pressure from 10 psig to 19 psig through adjustments to the first stageregulator internal pressure. Initiation of airflow was at 2 G. In addition, a manualmechanical "Off/On" ready-pressure switch was added. When this switch is actuated(in the "On" position), the ready pressure causes the anti-G suit to inflate to 60% ofits air-volume capacity, and to maintain a pressure of approximately 0.2 psigwithin the suit.

4.2.2 Phase I - Determination of Maximum Flow Caoacity

Under open-flow conditions, the ALAR HFRP valve performed as shownin Figures 4.2-2 and 4.2-3. With the ready-pressure switch in the "Off" position, flowstarted at approximately 2 G (Fig. 4.2-2). At baseline G (1.4 G), the flow was 0.3SCFM; as Gz increased, flow increased rapidly to 22.7 SCFM at 11 G. A shoulder onthe curve was observed in the region of 2.6 G where flow briefly leveled off at 6.3SCFM, dropped to 6.0 SCFM, and rose again to 6.3 SCFM in approximately 0.5 s.

With the ready-pressure switch "On," flow was initiated at about 2 G (Fig. 4.2-3).At baseline G (1.4 G), flow was 3.3 SCFM; this flow was maintained as G levelincreased to 2.1 G, when flow increased until 2.7 G before leveling off momentarily at5.7 SCFM. Flow then increased rapidly to 10 SCFM at 3.6 G before slowingsomewhat; it then increased to a maximum of 22.6 SCFM at 9 G. No increase in flowwas observed as the G level was raised from 9 G to 11 G.

28

Page 43: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

Figure 4.2-1. ALAR High-Flow Ready Pressure anti-G valve.

29

Page 44: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

2 1 I I I I

1-I-1 1 - - - - - --

-- -

00 9 7 6 5 4 3 2L41.41.4 G,LEVEL

0 Q 0 - N If

0- ~ M ONq S ~ mOQ c.~ TIME(SC

Po to \ ,

Figure 4.2-2. Maximum flow through ALAR HFRP anti-G valve with pressure off.

-/ J I.. " I ]_~

4 - - - 1 -

27

114 3 4 5 6 7 6 9 10 II II II II I0 9 8 7 6 5 4 3 21 441.4 GZ LVEI.

O- q P? 0 t 0 N o oNn 0 WON o I " ToME (SEQ

Figure 4.2-3. Maximum flow through ALAR HFRP anti-G valve with pressure on.

30

Page 45: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

4.2.3 Phase It - Determination of Dynamic Resoonse Canability

The dynamic response capability of the ALAR HFRP anti-G valve wasdetermined as described in Testing Procedures, section 3.2.2. Pressure delivered bythe valve was measured at the output port and inside the simulated anti-G suit volume.

4.2.3.1 Low Gz-Onset Rate Conditions

The low Gz-onset test conditions employed were: 0.1 G s-1

trapezoidal profile and simulated suit volume of 10 liters. The pressure delivered bythe valve was measured at the output port and from the interior of the simulated anti-Gsuit volume. All tests for this phase were conducted with the ready pressure switch inthe "On" position.

When the ALAR HFRP valve was tested under conditions of 150-psig sourcepressure, 10-liter volume, and 0° valve angle, the results shown in Figure 4.2-4 wereobserved. Flow was slow initially with a suit pressure of 0.4 psig at 1.4 G whichincreased to 0.5 psig at 2 G; as G7 level increased from 2 G to 7.5 G, suit pressureincreased to 8.2 psig in an essentially linear fashion. From 7.5 G to 11 G, thepressurization rate slowed due to the relief valve cutting in. The averagepressurization rate was 1.3 psig G- 1 over the useful range of the valve.

For a 150 valve angle (and all other conditions the same), the performancecharacteristics in Figure 4.2-5 were observed. The valve responded almost linearlyfrom 2 G to 8 G, which corresponded to a suit pressure of 8 psig. The relief valve wasactivated at this point, resulting in a slowing of pressurization rate such that suitpressure increased to 10.3 psig at 11 G.

When the valve angle was 300, the valve did not open until 2.4 G was reached(Fig. 4.2-6). From 2.5 G to 8 G, suit pressure increased linearly from 0.5 to 6.8 psig.The relief valve then started to open, slowing the rate of suit pressure increase; apressure of 9.5 psig was observed when 11 G was attained, but 1 s later it increased tothe maximum of 9.6 psig.

The effect of source pressure upon valve performance under Iow-Gz-onsetconditions was explored. Holding the valve angle at 00 and the suit volume at 10liters, a minimum source pressure (30 psig) and a maximum source pressure (300psig) were employed. From 2 G to 8 G, suit pressure increased almost linearly to 8 Gwhere the suit pressure was 8.5 psig. From 8 G to 11 G, the pressure increasedcurvilinearly to a maximum of about 10.3 psig. (See Fig. 4.2-7.)

When a source pressure of 300 psig was employed (Fig. 4.2-8), the valve startedto pressurize at 2 G and pressurization rate was linear up to 7.5 G (7.8 psig suitpressure). From 7.5 G to 11 G, the rate slowed somewhat, yielding a suit volumepressure of 10.7 psig at 11 G.

31

Page 46: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

'a 1 { I~ I I\ I I .LLLJ.J I I

I NI./ _ __ tI I

un-e coniton of 5Isgsuc rsue o ne;1-ie oue 0age

I II I ! I - L

t .. . . . . .. .. . ". . . . . . . . . .____S____ --

4- IQ I I I I I I 1( 6 5 4 3 1. 4 L4 020L&

Figure 4.2-4. "Suit" pressures produced during acceleration/deceleration by ALAR HFRP anti-G valveunder conditions of: 150-psig source pressure; low G onset; 10-liter volume; 00 angle.

IF 32

II

Io 11.42 3 4 5 6 7' 6 1 OIl H II II 10 s aI 7 6 5 4 3 .1.4L41.4 GtILVt j

0 a 0P - • • -- • U • . . • ""*0O1

Figure 4.2-5. "Suit" pressures produced during acceleration/deceleration by ALAR HFRP anti-G valveunder conditions of: iSO-psig source pressure; low G onset; 10-liter volume; 150 angle.

32

Page 47: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

1 1 1 1I I I .1I I I i i . ,

_ , I L4Z 3 S 0 5 4 i :3 2 L L " L4 ., , 0 A 'i° i I I - T _

---- TI

-I At ' " 1\ I I I I I 1 1..

:11 ___ t-!/ - - I ,Ik _ i !

I L4, 3 4 5 6 70 I I III I I 1 10 9 6 7 6 5 4 3 2L4L .4.4 G Lt =E;L I

. ... . 1 P! q a -e g

Figure 4.2-6. "Suit" pressures produced during acceleration/deceleration by ALAR HFRP anti-G valveunder conditions of: 150-psig source pressure; low G onset; 10-liter volume; 30 angle.

33__! I I_ ,/ §I F,, i I i I t _

I I--. ..- ±Ji .I " 7 f'II -

!l.,II I I " ?-I

67S IO ~,I I II 1 1 43I AI I ~2

Figure 4.2-7. "Suit" pressures produced during acceleration/deceleration by ALAR HFRP anti-G valveunder conditions of: 30-psig source pressure; low G onset; 10-liter volume; 0 angle.

33

Page 48: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

4.2.3.2 High G-Onset-Rate Conditions

High Gz-onset-rate testing employed the median simulatedanti-G suit volume of 10 liters, and a trapezoidal profile of 6 G s-1. The pressuredelivered by the valve was measured at both the output port and inside the simulatedG-suit volume; however, only suit-volume pressures are discussed here.

At a source pressure of 150 psig and valve angle of 00, the pressurizationschedule (psig G -1) exhibited a significant lag behind that observed during low Gz-onset testing. Under high Gz-onset conditions (Fig. 4.2-9), the valve initially beganpressurization at 2 G, but oscillation was observed at the valve outlet port from 1.4 G.Oscillation of amplitude 0.6 psig p-p continued from 1.4 G to 9 G. From 2 to 11 G, thepressurization rate was linear; the slope of the line was such that the suit pressure was1.1 psig at 3 G and 7.2 psig at 9 G. From 9 to 1.1 G the response rate deviated slightlyfrom linearity with a suit pressure of 9.9 psig when 11 G was attained; however,pressure rapidly increased to 11.1 psig during the period at 11 G. Upon initiation ofdeceleration, the valve exhibited a low-frequency oscillation of 0.5 psig p-p amplitudewhich continued as deceleration continued down to 5 G.

Effects of valve angle upon ALAR HFRP valve performance under high Gz-onsetconditions were observed using valve angles of 150 and 300 and keeping sourcepressure and suit volume constant at 150 psig and 10 liters, respectively. For a valveangle of 150 (Fig. 4.2-10), a low-frequency oscillation with* amplitude 0.3 psig p-pstarted at 1.5 G and continued until 10 G. The valve began pressurization at 2 G andthe rate was generally linear up to 10 G: suit pressure was 0.4 psig at 2 G, 3.9 psig at6 G, and 8 psig at 10 G. Upon attaining 11 G, suit pressure was 9.3 psig; during thetime "on top" at 11 G, pressure increased to 10.7 psig within 1.5 s, and climbed to amaximum of 10.8 psig by 5 s. Upon deceleration under these conditions, anoscillation of 0.2 psig p-p occurred and continued until 5.5 G where suit pressure was6.1 psig. The depressurization rate was linear down to about 4.1 psig at 3 G, droppingoff rapidly to 0.5 psig at 1.4 G.

When the valve angle was 300 (Fig. 4.2-11), the valve opened at about 2.5 G;however, the suit pressurization rate was noticeably slower than when the valve anglewas 150. At the instant the acceleration reached 11 G, the suit volume pressure wasonly 8 psig; after 1.5 s at 11 G, pressure was at its maximum value of 9.5 psig. Asdeceleration was initiated, a low-frequency oscillation of 0.3 psig p-p amplitude wasobserved at the valve outlet port, and this oscillation continued until 5.5 G while suitpressure decreased essentially linearly from 11 G to 1.4 G.

The effects of source pressure upon performance of the ALAR HFRP valve wereexamined by testing the valve at 30 and 300 psig and 00 valve angle. At 30 psigsource pressure (Fig. 4.2-12), a low-frequency oscillation of 0.5 psig p-p in suit volumepressure occurred over the interval of 1.4 G to 8 G. The suit pressure was 5.7 psig at 8G, and 9.7 psig at the instant that 11 G was attained. A maximum suit pressure of 10.8psig was measured at the end of the interval "on top" at 11 G. The low-level oscillation(0.6 psig p-p) reccurred immediately upon deceleration, and continued until the 5-Gpoint; the suit pressure decreased linearly down to 2 G.

34

Page 49: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

I ...... I 1 .1 1\i i I

I , I ,i i A i ii I __

1 A,, I I I I I I I I I I Ii I1- A L I 1 11 1 Wi-21L4 2 3 4 5 7 S 9 1 0 II I II I I0 9 6 I 6e 4 3 I4L4 . . I..sac)

Figure 4.2-8. "Suit" pressures produced during acceleration/deceleration by ALAR HFRP anti-G valveunder conditions of: 300-psig source pressure; low G onset; 1 0-liter volume; 00 angle.

Eli 7V [ TF T\ IJ]Y]Irn3

I :z "11i~ ] -I - I ___II

I tl __ __ ___IIt

-: - - 11 , 1 T i __________

. . . . ..- . iti i .

Figure 4.2-9. "Suit" pressures produced during accelerationdeceleration by ALAR HFRP anti-G valveunder conditions of: 150-psig source pressure; high G onset; 10-liter volume; Q

° angle.

35

Page 50: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

-, - t ___ ll l i, ! I I/-' - I I I I I 1 1 7

1I I N I I 1i

.rime ___C

Figure 4.2-10. "Suito pressures produced during acceleration/deceleration by ALAR HFRP anti-G valveunder conditions of: 1 50-psig source pressure; high G onset; 1 0-liter volume; 150 angle.

' I ___ __

3t N - - i

It/ L1J i lI I} N, i

|7 a42 3 4 8 6 ? l 1 4 1 1 0 9 T 6 5 4 3 2 1. 4A4L4 1 LE VELI

# . . . . . . " - * - W i . TIMt (5IC I

a!------------------- ------- ! - 3

Figure 4.2-11. "Suit" pressures produced during acceleration/deceleration by ALAR HFRP anti-G valveunder conditions of: 150-psig source pressure; high G onset; 10-liter volume; 30° angle.

36

Page 51: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

Under 300 psig source pressure (Fig. 4.2-13), the low-frequency oscillation wasagain observed. Under ascending G conditions, an oscillation of 0.6 psig p-p occurredover the interval of 1.4 G to 10 G. A pressure of 9.7 psig was measured as 11 G wasattained, and after 1.5 s the pressure had increased to 11 psig; it then stabilized at11.1 psig. Immediately upon deceleration, the oscillation (0.5 psig p-p) recurred andpersisted until the 5.5-G level.

4.2.4 Phase III - Determination of Comglex Dynamic Resgonse Caoalility

Results of testing the ALAR HFRP anti-G valve under SACM conditions of150 psig source pressure and 00 valve angle are shown in Figure 4.2-14. On the firsthigh-onset acceleration to 9 G, suit pressure at 9 G was 8.9 psig; after about 2.4 s "ontop" the pressure had increased to 9.6 psig and it continued to increase to a maximumof 9.7 psig. A low-frequency oscillation in valve outlet pressure (but not in suitpressure) was observed on acceleration between 3 G and 9 G, and on decelerationfrom 9 G to 5 G. Under 4.4 G, the suit pressure was 5.1 psig but dropped to 3.8 psigafter 9.5 s. Under 3 G, suit pressure was 2.2 psig initially then dropped to 2 psig afterabout 5 s. Upon accelerating again to 5 G, the suit pressure rose to 4.6 psig andleveled off at 4.7 psig. The next deceleration to 4 G produced a pressure of 3.6 psigwhich then stabilized at 3.3 psig. During the second high-onset acceleration to 9 G thesuit pressure reached 9 psig and leveled off at 9.6 psig after 1.5 s. The subsequenthigh-offset rate to 1.5 G yielded a suit pressure of 2.6 psig, which slowly fell to 0.4 psig.A rapid-onset acceleration to 8 G then produced a suit pressure of 7.5 psig, which thenincreased and stabilized at 8.7 psig within about 2 s. Deceleration from 8 G to 1.4 Gresulted in a rapid reduction in suit pressure to 0.4 psig, and a low-level (0.2 psig p-p)oscillation.

The effects of valve angle upon performance of the ALAR HFRP valve underSACM conditions were explored by varying the angle (150 and 300) W -'ile holding thesource pressure and suit volume at the median values of 150 psig and 10 liters,respectively. However, results of the 150-angle experiment had to be discardedbecause of spurious signals from an improperly functioning transducer.

At a valve angle of 300 (Fig. 4.2-15), a low-frequency oscillation of 0.3 psig p-poccurred from 3 G through 9 G. Suit pressures of 1.5 psig at 3 G and 7.9 psig at 9 Gwere observed; after 6.5 s at 9 G, pressure rose to 8.8 psig. Upon deceleration, thesuit pressure decreased rapidly to 4.5 psig, hesitated for 9 s at 4.4 G, and thendropped slowly to 3.2 psig. An oscillation of 0.9 psig p-p occurred from 9 to 5.5 G. At alow-offset rate from 4.4 G to 3 G, pressure dropped to 1.9 psig, then slowly declinedto 1.5 psig by the end of the 3-G plateau. Upon subsequent low-onset acceleration to5 G, the pressure climbed slowly to 3.9 psig then leveled off at 4.5 psig within 1.5 s.Upon deceleration to 4 G, the suit pressure decreased to 3.1 psig. During the nextrapid-onset acceleration to 9 3, the suit pressure climbed rapidly to 8.1 psig andleveled off at 8.9 psig; this was accompanied by an oscillation of 0.6 psig p-p. Upondeceleration, an oscillation of 0.5 psig p-p was observed from 9 G to 5 G. At 1.5 G, theinitial pressure was 1.9 psig which leveled off at 0.5 psig after 1.5 s. The subsequentacceleration to 8 G produced a suit pressure of 8 psig, and an oscillation of 0.3 psig p-p. Upon rapid deceleration to 1.4 G, suit pressure dropped to 0.5 psig.

37

Page 52: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

I I r- 1I -II V KT - F Vi

''I I i j/J i .- 1 _

, I -

3 ILLLI....J 1 LLII 1,- I I.I... I I JtII IZ IZIt i-f-NOTE. PRESSURE OSCLLATES BETWEEN POINTS -C AN.O "Cr

1i 11.42 3 4 5 6 7 a 9 10 it It 1 11 to 9 8 7 6 S 4 3 2 1.4 L1.44 GLEYEL

T, 7 0i T It 1 R i I- 0 ! ' TIM C- N 0i 0l 0 NNi t t

Figure 4.2-12. "Suit" pressures produced during acceleration/deceleration by ALAR HFRP anti-G valveunder conditions of: 30-psig source pressure; high G onset; 10-liter volume 00 angle.

_ I I II ,, I I--~~i I -~i~iI

I I Il _ ... I...,,1.1iI1.........I.......i..I.!....

42 3 4 5 6 6 9 10 11 1 1 1 1 10 9 6 7 a 5 4 3 1i.4 4 .4 1 LEVEI L i~~**~~ 0 *E k0 Tim~eC

Figure 4.2-13. "Suit" pressures produced during acceleratlonvdeceleration by AL.AR HFRP antl-G valveunder conditions of: 300-psig source pressure; high G onset; 1 0-Iter volume; 00 angle.

38

Page 53: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

I II

I A I I I I I d I I A__

IiiTi'i. i " I I

-_ t I 1 -- IEif ___ il-,"

2 I I ) 1 I I I I I t t I It ___

J 0 I 20 30 40 5o O 70 so 90 ,00 110 Time (3 C

Figure 4.2-14. *Suit" pressures produced during SACM by ALAR HFRP anti-G valve under conditionsof: 1 50-psig source pressure; 10-liter volume; 00 angle.

4,4[ 1. f ]11T1F

ii I I II.I~ 11 I ! i iiL_ I __ __ rr ' ' L!......... I I I _

_ 1 I I I -. T itI i I

0 0 z2 30 40 7o 60 so 90 100 110 TIME SE I 3

Figure 4.2-15. "Suit" pressures produced during SACM by ALAR HFRP anti-G valve under conditions of:150-psig source pressure; "10-iter volume; 300 angle.

39

Page 54: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

The effects of source pressure upon performance of the ALAR HFRP valve underSACM conditions were examined. Minimum (30 psig) and maximum (300 psig)source pressures were used. The simulated suit volume was 14 liters for the minimumpressure and 6 liters for the maximum pressure. At a source pressure of 30 psig (Fig.4.2-16), the valve displayed pressure at 1.4 G with 0.3 psig in the suit volume; this wasaccompanied by an oscillation of 0.2 psig p-p up to 3 G. The suit pressure rose to 1.2psig at 3 G, then leveled off at 2 psig. A rapid-onset acceleration to 9 G caused the suitpressure to increase to 9 psig, accompanied by Rn oscillation of 0.3 psig p-p.Deceleration to 4.4 psig caused the pressure to decrease rapidly to 5.6 psig, with anoscillation between 9 G and 6 G. The pressure leveled off at 4.0 psig within about 9 sand then decreased to 2.3 psig at 3 G. Acceleration to 5 G produced an initialpressure of 4.7 psig, which leveled off at 4.8 psig. Subsequent deceleration to 4.0 Gyielded a pressure of 3.6 psig. The second rapid onset acceleration to 9 G produced asuit pressure of 8.2 psig which then leveled off at 9.6 psig. This accelerat9on alsocaused an oscillation of 0.9 psig p-p between 4 G and 8.5 G. Rapid deceleration to 1.5G dropped the suit pressure to 3.5 psig; this was also accompanied by an oscillationbetween 9 G and 6 G. After 5 s at the 1.5-G level, the suit pressure dropped to 0.5psig. A subsequent high-onset acceleration to 8 G resulted in the suit pressureleveling off at 8.6 G; oscillation (0.7 psig p-p) was observed between 1.5 G and 6 G. Afast-offset deceleration from 8 G to 1.4 G dropped the suit pressure to 0.5 psig whichpersisted until the end of the run. The valve oscillated (0.2 psig p-p) during this entiredeceleration.

At a source pressure of 300 psig (and 6 liters suit volume), shown in Figure4.2-17, the valve performed essentially the same as when 150 psig source pressureand suit volume of 10 liters were employed (Fig. 4.2-15). Differences were that suitpressures at the high-G levels were about 0.5 psig.lower than under 150 psig, andoscillation occurred intermittently throughout the entire profile.

-0i 1 to 3s" _ Il I ($C

Figure 4.2-16. "Suit" pressures produced during SACM by ALAR HFRP anti-G valve under conditionsof: 30-psig source pressure; 14-liter volume; 00 angle.

40

... ... = * - -a I i ml i -- H- --ll

Page 55: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

w1 _L _. L a 1 ...i. ... .. ..... EVE

I0~---

-__

0 t0 20 30 40 5b 60o 70 SO So 100 11 Ito TIE SC)

Figure 4.2-17. "Suit" pressures produced during SACM by ALAR HFRP anti-G valve under conditions of:300-psig source pressure; 6-liter volume; 00 angle.

4.3 ALAR High-Flow Anti-G Valve

4.3.1 Description

The ALAR High-flow anti-G valve (Fig. 4.3-1) is a modified form of thestandard ALAR 8400A anti-G valve (section 4.1). Modifications included enlargementof the orifices, and adjustment of the first stage regulator internal pressure and G-levelopening point. These changes were designed to: increase airflow by 50 %; increasefirst stage pressure from 10 to 19 psig; and provide initiation of airflow at 1.25 G ratherthan 2 G. The valve tested was modified to open at 1.5 G (instead of 1.25 G) to avoidactivating the valve at the USAFSAM centrifuge G threshold of 1.4 Gz.

The optimum source pressures for the ALAR High-flow anti-G valve were: 300psig, maximum; 150 psig, median; and 30 psig, minimum.

4.3.2 Phase I - Determination of Maximum Flow Capacity

Maximum flow capacity of the ALAR High-flow valve was determined asdescribed in Testing Procedures, section 3.2.1. Briefly, the median source pressure forthis valve (150 psig) and a valve angle of 00 were used for three trapezoidal runs at 0.5Gz s- onset and offset rates. The unobstructed flow of air (in SCFM) through the valve,the valve output pressure (psig), and the source pressure (psig) were measured.

41

Page 56: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

Figure 4.3-1. ALAR High-Flow Anti-G Valve.

42

Page 57: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

The relationship between Gz level and maximum flow is shown in Figure 4.3-2,where each data point is the mean of the values obtained from the three runs. Flowwas initiated when 2 G was attained; this initial flow was 1.5 SCFM. Flow thenincreased linearly with G level between 2 G and 3 G, yielding a flow of 8.6 SCFM at 3G. At this point, the rate of increase in flow began to lessen, again becoming linear--but with reduced slope-from 4.5 G to 7.5 G (12.9 to 20.1 SCFM). At 7.5 G flow beganto plateau, measuring 21.6 SCFM upon reaching 11 G and 21.8 SCFM at the end ofthe interval at 11 G. Upon initiation of deceleration, flow did not begin to decreaseuntil after 8 G was passed. Flow Was 20.7 SCFM at 7.5 G and decreased linearly to11.3 SCFM at 4 G; immediately after this point, flow jumped up slightly (11.8 SCFM at3.9 G) then began to decrease at a more rapid rate and ceased as 1.4 G was attained.

4.3.3 Phase II - Determination of Dynamic Response Capabilily

The dynamic response. capability of the ALAR High-flow anti-G valve wasdetermined as described in Testing Procedures, section 3.2.2. Pressure delivered bythe valve was measured at the output port and inside the simulated anti-G suit volume.As used here, the term "output pressure" refers to the pressure within the suit volume,unless otherwise noted.

4.3.3.1 Low G,-Onset-Rate Conditions

The 0.1 G s- i trapezoidal profile was used for all low Gz-onsettests for this valve.

Holding the source pressure at the median value of 150 psig while varying thevalve angle, produced the results shown in Figures 4.3-3 through 4.3-5. When thevalve angle was 00 , the suit pressures produced by this valve as Gz level was variedwere as shown in Figure 4.3-3. The .suit pressure was 0 psig at 1.4 G and 0.2 psig at 2G; pressure then increased linearly as G level was increased, reaching 8 psig at 7.5 G.The rate of increase in suit pressure then began to curve to yield 10.6 psig as G levelreached the 11 -G level. While at 11 G, suit pressures varied approximately 0.2 psig.With deceleration, suit pressure began to decrease when G level fell to 9.5 G andcontinued to decrease in a linear fashion until reaching 0.3 psig at 1.4 G; after 1 s at1.4 G, suit pressure was zero.

When the valve angle was changed to 150 while all other conditions were heldconstant (Fig. 4.3-4), the response of the ALAR High-flow valve was almost identical tothat at 00 (Fig. 4.3-3). The principal difference between the two valve angles was thatat 150 the maximum suit pressure at maximum G level was slightly higher than at 00valve angle (11.5-11.6 psig vs. 10.6-10.8 psig).

A valve angle of 300 produced a similar response (Fig. 4.3-5), but the rate of suitpressure buildup and the maximum suit pressure reached were reduced; themaximum G level (11 G) yielded maximum suit pressures of 9.2-9.4 psig.

43

Page 58: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

:'Ej

I ___ I

I i I I I I I I I I I.I_ I I I I 1 I I N

-A° i -tI i I i 1 i 1 I0,..i ;i iI : II I ____________

SI 7 ' 111 1• i i I !

I 4 4 5 6 7 a 9 0 1i 1 11 1 1 0 9 7 AS 5 4 3 2 IAI.AL.I4_ V I P a i - M I 4 A V 0 ! S - 'I

Figure 4.3-2. Maximum flow through ALAR High-flow anti-G valve.

/7 1 1 I I

7/II tI I I 1Ir I ll!I I \

a 7 1 I 1 I I I I I I I

i ~ . I I } I I I I _ _ _

I4 I 1 N I t " 0 I I I 0 ' S 7 i I i i li~l . . t'

L4 -3 4 5- - i -I II I I 1 10__5_4__f.4_t._(.

0 A 1 nm_ IE_ __

i2

Figure 4.3-3. "Suit" pressures produced during acceleration/deceleration by ALAR High-flow anti-G valveunder conditions of: 1 50-psig source pressure; low G onset; 1 0-liter volume;, 011 angle.

44

Page 59: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

a 2

'l - I7IiiiI - ii _______________iI I'

- 3 I I I iiii i it i ii Ii J ) i i ________

__-_ ). I 1I I.J..J.!.4I .. ______1 _ I -I IJ1 1 ! I I I I I I I I I I I * i !~ I I

4 ! 6 71 8 9 II II I I i 7 6~ S 4 3 214.4. OI..

Figure 4.3-4. "Suit" pressures produced during acceleration/deceleration by ALAR High-flow anti-G valveunder conditions of: 1 50-psig source pressure; low G onset; 1 0-liter volume; 150 angle.

_1 I I.

7 ,17 i I I I I I I 1

3L ...t - I I I I -!-I "' ) I I !I I I ________i) ! I I I I\)~ 1 1 11

_ '7 i /I I .jIIJ II I I I!-i I.,.I t.I. i i !.IJ.... I I I I _',,,.. I _ ; _

3 4 5 6S 7 8 9 10 1I 1 I 1 1 1 131 t 6 7 6 5 4 3 L4 1.4I. 01. ,fVELA N! N4 P! ra N IMr

- 10 P. 0 0 c oo

Figure 4.3-5. "Suit" pressures produced during acceleration/deceleration by ALAR High-flow anti-G valveunder conditions of: 150 -psig source pressure; low G onset; 1 0-1iter volume; 300 angle.

45

Page 60: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

17 1 - - -- T - - -I I I I 1~ -. 1 1, 1__ _ _ _ [1 1

\i I r . I i I I I I I IS i i I ~ IIIt i I ti-, I i i I i "i, i r

I i 7 "k II __ ____ I

I I I i } I I I I I I ~ iII I I

i' I I ! I I I _______

I.42 3 4 3 6 7 8 9 10 1 II II II I 1 1 0 1 6 4 5 4 3 2 L41.41.4 01 1.EVEL

IV ad C4 UI0 2 a - S2 ~O .g

Figure 4.3-6. "Suir pressures produced during acceleration/deceleration by ALAR High-flow anti-G valveunder conditions of: 30-psig source pressure; low G onset; 1 0-liter volume; 00 angle.

iI i I I I I I' I i I

I I ___________

3 1 A I I I I I" I/ i V L) I. . )I h Z! \ I I I

5 6 7 9 9I0 10 8 7 C , 4, 3 2144 a.L E'IL.

-z~ _ -__ __ __ __r7 ~ ~~~ ~ -, ,, $ ;3,aa IME =EC

Figure 4.3-7. Suit" pressures produced during acceleration/deceleration by ALAR High-flow anti-G valveunder conditions of: 300-psig source pressure; low G onset; 10-liter volume; 00 angle.

46

Page 61: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

The source pressure was varied while holding the valve angle at 00. The re-sponse profile for the minimum source pressure of 30 psig (Fig. 4.3-6) and that for themaximum source pressure of 300 psig (Fig. 4.3-7) can be compared with that for themedian source pressure of 150 psig just discussed (Fig. 4.3-3). At the minimumsource pressure (Fig. 4.3-6), the suit pressures generated as a function of Gz levelwere almost identical to those produced when the median source pressure was used(Fig. 4.3-3); the only difference being a slightly lower maximum suit pressure at 11 G(10.2-10.3 psig vs. 10.6-10.8 psig). The maximum source pressure also produced aresponse pattern (Fig. 4.3-7) which was essentially identical to those of the medianand minimum source pressures; maximum suit pressures of 10.4-10.6 psig wereobserved at 11 G.

4.3.3.2 High G,-Onset-Rate Conditons

For high Gz-onset-rate testing of the ALAR High-flow anti-Gvalve, conditions were: a simulated suit volume of 10 liters, and acceleration anddeceleration rates of 6 Gz s- 1. The pressure delivered by the valve was measured atboth the output port and inside the simulated suit volume; however, only suit volumepressures are presented and discussed here.

the effects of valve angle upon performance of the ALAR High-flow valve wereexplored by varying the valve angle (00, 150, 300) while maintaining the sourcepressure at the median value for this valve (150 psig). When the valve angle was 00,the suit pressurization profile was as shown in Figure 4.3-8. Pressurization started at 2G with a suit pressure 0.2 psig, increasing to 0.8 psig at 3 G; the increase in suitpressure was then linear up to 10.5 G where pressure was 10.2 psig. The pressurebuildup then began to slow, measuring 11.5 psig upon reaching 11 G, and plateauingat 12.2-12.4 psig during the time at 11 G. Upon deceleration, suit pressure began todecrease immediately and decreased linearly down to 3.3 psig at 2 G; it then droppedsharply to 0 psig as 1.4 G was reached.

When the valve angle was 150 (Fig. 4.3-9), the rate of suit pressurization and themaximum pressure produced were both reduced. There was an initial lag phase sothat pressure increase did not become linear until after 5 G was attained; the maximumsuit pressure produced was 10.7 psig at 11 G vs. 12.4 psig for a valve angle of 00 (Fig.4.3-8). The slopes of the deceleration phases were similar for both conditions.

When the valve angle was 300, a similar shift in the curve was observed (Fig. 4.3-10). The initial lag phase was longer than for the 150 angle (Fig. 4.3-9), with the resultthat the rate of pressurization did not become linear until 6.5 G was attained. Themaximum suit pressure produced was slightly under 10 psig (at 11 G) vs. 10.7 psig at151, and 12.4 psig at 00.

Under conditions of minimum source pressure and a valve angle of 0° , the suitpressurization was as seen in Figure 4.3-11. After an initial lag phase, the rate of pres-sure buildup was linear from 3.5 to 11 G. Pressure plateaued at 11 psig within 1.5 s at11 G and remained at this level until deceleration was initiated. During decelerationsuit pressure decreased linearly to 2.2 psig at 2 G. The pressure then decreasedfurther to 1.2 psig at 1.4 G and fell further within 3 s at baseline G to 0.3 psig.

47

Page 62: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

I I I A I I I I I , I I I I I7

I I I I I i I I I 7 1;

t~~~~~ ~~ J i / lI !I I 1itIIt !I I

I.4Z 3 4 5 6 s 1t It. II I I II 0 9 8 7 6 5 4 3 ZL.A1.4 0 .( UN ELI I I I I I I I I I

I I 1.4f I II 4I II I I1t I I I I 1 I I I , 1 I .4 l I ...V .

P! Time ?maa,

Figure 4.3-8. "Suit" pressures produced during acceleration/deceleration by ALAR High-flow anti-G valveunder conditions of: 1 50-psig source pressure; high G onset; 1 0-iter volume; 00 angle.

/ I48 { i i I I I I I-" Z IiIIiII )I I K l.--r r I F VF IK V i

-'° t I I I I I I /I I iti-' It11 i'lliii I i I I ISI .1 i~ I lIIIiI I I i I I I ISI I/..... II - I I\ IilI -

SI - -i I - ...... E...I .L....J....

IL42. 3I 4l Sl 6 8 9 I0 II II I II I0 9 l 7P 6i S 4 3t 3!1.41.4II.4 i3,1L'E€V{IL"

------------------------ N Pi 0 .. . .. !

Figure 4.3-9. "Suit" pressures producedl during acceleration/deceleration by ALAR High-flow anti-G valveunder conditions of: iSO-psig source pressure; high G onset; 10-liter volume; 15° angle.

48

Page 63: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

I I I I I i1 i1"- , Ft !i I i l l .......-....... ..*, , , I I i I i

, I L4 4 1 1 1 1 1 1 1 i I 7 1 9 8 543i 14 1S f I JI 1 I 1 1 7*J J W I j...I [l -. K] V 1 1 17T } -

ii tJ i T I

2 I I I A I i I I I I I I I I I I I I I I - I

i.42 3 4 6 6 ? £ 9 10 1 1 1 1 1 10 9 8 ? 6 54 3 2 .41.,4 .*- , , . ,, -, . 0 -. 0 -Y . , , '-

0 4 0 e ' Sac

Figure 4.3-10. "Suit" pressures produced during acceleration/deceleration by ALAR High-flow anti-G valveunder conditions of: 150-psig source pressure; high G onset: 10-liter volume; 300 angle.

I I, I I I I i / I 1 1 I I I ] I

I 1 .II I i ."q iI I I I I I I I I I I I I I I _ _ _

--II I I. I I I I....l......J... I I I I I I I It IIJ I 1 .4 iif j, ! B If I] I .I I0 ] $ ? I ______.4,e . .ll:

Figure 4.3-11. "Suit" pressures produced during acceleration/deceleration by ALAR High-flow anti-G valveunder conditions of: 30-psig source pressure; high G onset; 10-liter volume; 0° angle.

49

Page 64: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

At maximum source pressure (300 psig) and 00 angle (Fig. 4.3-12), the pressur-ization profile was more like that observed at minimum source pressure and 0° (30psig, Fig. 4.3-11), than that at median pressure and 00 (150 psig, Fig. 4.3-8). The initiallag phase persisted up to 5 G and the pressurization was slightly less linear (Fig. 4.3-12) than for the minimum source pressure (Fig. 4.3-11). Maximum suit pressureattained and the deceleration profiles were essentially identical in these two cases(minimum and maximum); however, a higher maximum suit pressure was produced bythe median source pressure (Fig. 4.3-8).

4.3.4 Phase II - Determination of Comolex Dynamic Resoonse Cacability

Performance of the ALAR High-flow anti-G valve under SACM conditionsand 00 valve angle and median source pressure is presented in Figure 4.3-13. At the3-G level, suit pressure increased to 1.5 psig within 2.2 s then continued to increase to2.4 psig by 12.4 s. Upon acceleration to 9 G, suit pressure built up to 8.9 psig andincreased to 9.6 psig within 6 s. A subsequent deceleration to 4.4 G caused suitpressure to drop to 4.8-3.9 psig. The G level was then dropped to 3 G, causing suitpressure to fall to 2.4-2.1 psig. Slow-onset acceleration to 5 G brought suit pressureup to 4.5-4.9 psig; decreasing G to 4 G then dropped pressure to 3.6-3.4 psig. Rapid-onset acceleration to 9 G resulted in an increase in suit pressure to 9.2-9.6 psig, andrapid-offset deceleration to 1.5 G brought suit pressure down to 1.8-0.3 psig. Anotherrapid-onset acceleration to 8 G produced suit pressures of 7.9-8.6 psig. During thefinal descent to baseline G (1.4), suit pressure decreased linearly to 0.3 psig.

When the valve angle was changed to 1.50 while other conditions remainedconstant, the SACM performance of this valve was as shown in Figure 4.3-14. AUlfeatures of the SACM suit-pressurization profile were essentially identical to thatobserved when the valve angle was 00 (Fig. 4.3-13). However, when the valve anglewas 300 (Fig. 4.3-15), suit pressures were generally lower throughout the SACMprotocol than for either of the first two angles.

Suit pressurization profiles for conditions of both maximum (14 liter) suit volumecombined with minimum (30 psig) source pressure (Fig. 4.3-16) and minimum (6 liter)suit volume combined with maximum (300 psig) source pressure (Fig. 4.3-17) weresimilar. Maximum suit pressures attained and rates of pressure increase anddecrease were almost identical for these two sets of conditions. Both of these figureswere also similar to that yielded under conditions of median (150 psig) sourcepressure with median (10 liter) suit volume (Fig. 4.3-13); although maximum suitpressures attained were slightly lower for the median conditions (Fig. 4.3-13).

50

Page 65: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

I I I

IAI I I I I I I I I I I I I I

-. -/jI[IT jII[I i1..~ii~~FI qh

I i IN [ I I I I

I I I I !/J I I 0 i ",, 4 I I I I

SI I/ I I I I I "I I I II'o----- -

--- - -I _ _ I _ ,,_--_

__j ot I L4Z 3 3 5 * 7 8 9 10 I I II 1 0 9 a 7 6 5 4 3 21.41.41.4 , L.' '/-

Figure 4.3-12. Suit" pressures produced during accelerationldeceleration by ALAR High-flow anti-G valveunder conditions of: 300-psig source pressure; high G onset; 1 0-liter volume; 00 angle.

'I - IN I I I I I - I

_ ", I II I

I L. I. . I I I 1 0 111 1 1 , 1 1 1 1

9.0 U 0 '.'__ [10 E JI .. ! 01 It! .L! .4 (1, LIMl I

0 10 20 30 40 s g-" 70 so so -00 (to 1i M I (SEC I

Figure 4.3-13. *Suit" pressures produced during SACM by ALAR High-flow anti-G valve underconditions of: 1 50-psig source pressure 10-liter volume; 0 I angle.

51

Page 66: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

I I I I I

7 0_1 "a, 1 -0I . =I ... .1 1 "I 1. I L/ It LIII IT1 i ii .. i...I..I......I i __ I - - TT'°i _1 ....I ...... I_ I I

10 10 Z0 30. 40 50 40 70 so s 10. 00 110 rimeC isiC2

Figure 4.3-14. "Suit* pressures produced during SACM by ALAR High-flow anti-G valve underconditions of: I 50-psig source pressure; 1 0-iter volume; 150 angIe.jIi f Z 9 Y IK V.V.' i ! ~ ...J ..J..I I

~-F-7

4

, I I I I L L 7 33 ,

I l , , ~ I _____ ..........I,,~...I .. I I I I

I o = o i Io Io Ir Io 1i. 11 ' ',, c

0 10 I0 30 40 so 60 70 so t0 100 HO0 TME (SEC

Figure 4.3-15. *Suit* pressures produced during SACM by ALAR High-flow anti-G valve underconditions of: 150-psig source pressure; 10-liter volume; 300 angle.

52

Page 67: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

I-I II I Ii i , ~

II _ _ _ _ _ I I I;JII -7-

I + __JI I I t ~ l Ii

177__ I I 1 i I 1I j / I I ii i ! I I I f \

i I -f i I I It i____Ik

1 9.0 It o l I o 1. :4 ,TI-E . L

a 0 iO ZO 30 40 so 60 s0 00 o9 100 110 TIME SEC

Figure 4.3 16. "Suit* pressures produced during SACM by ALAR High-flow anti-G valve underconditions of: 30-psig source pressure; 14-liter volume; 00 angie.

I !___i______i' -i/''*i**-r-i-;-*-i-i- I

4 1 . I . I I I t I I I I0 ,,0 2< 30 40 So so, TO so 90 100 ,,o TIME IS t !

Figure 4.3;17. "Suit" pressures produced during SACM by AlAR High-flow anti-G valve underconditions of: 30-psig source pressure; 14-liter volume, 00 angle.

,5I Ii

, it !I i -Ii I iii ____ I I \ I Ii iH I/ ________I________ __ i I I\I I I

l I I I I I I I~ -- t I I I ,I' --.. ow--n ' r---- . ... T h ' .. t .. z . 4 0 . ,,+

Figure 4.3-17. "Suit" pressures produced during SACM by ALAR High-flow anti-G valve underconditions of: 300-psig source pressure; 6-liter volume; 0° angle.

53

Page 68: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

4.4 Garrett Electronic/Pneumatic Anti-G Valve

4.4.1 Descdrtion

The Garrett Electronic/Pneumatic (Servo-controlled) anti-G valve (S. N.59364) (Fig. 4.4-1) is a poppet-type, acceleration-biased pressure regulator, which iscontrolled by a torque motor and balanced for inlet pressure. This valve ismanufactured by the Garrett Corporation, Phoenix, Arizona. This valve is designed tosense acceleration along one axis only, and to regulate pressure to an external anti-Gsuit according to a predetermined schedule which is based on the magnitude of thesensed acceleration. The valve incorporates an overpressure relief valve, a referencepressure regulator, a manual test switch, and a provision for remote accelerationsimulation.

This valve is designed so that supply air enters the unit at the inlet port and isdirected to the poppet valve (which is spring loaded, closed) and to the referencepressure regulator. The reference pressure regulator supplies a constant pressure tothe torque motor inlet nozzle; in the absence of current, the torque motor lever armblocks the inlet nozzle and vents the chamber through the vent nozzle. Thus, thespring load on the diaphragm actuator lifts the actuator off the poppet valve vent seatto cause suit pressure to be vented to the ambient environment.

As the accelerometer senses acceleration along the control axis, a voltageproportional to the magnitude of acceleration is generated. The valve electroniccontroller receives this voltage signal and provides current to the torque motor. As thecurrent increases, the motor lever moves to restrict the vent nozzle and to control thereference regulator outlet pressure flowing into the chamber. The pressure in thechamber causes the diaphragm actuator to first overcome the spring load, closing thepoppet valve vent seat; and then to begin pushing the poppet valve off the fill seat.These actions allow inlet air to flow past the fill seat and into the anti-G suit.

The suit-pressure-feedback transducer senses anti-G suit pressure and suppliesthe valve electronic controller with a voltage proportional to that pressure. The valveelectronic controller compares transducer voltage and accelerometer voltage, andadjusts the current supplied to the motor as required to maintain the specifiedrelationship between sensed acceleration and anti-G suit pressure.

A relief valve is provided to limit suit pressure to a safe level. A "Test" switchallows the unit to be tested by supplying a simulated acceleration signal to the valveelectronic controller. In addition, an electrical circuit is provided to allow externalapplication of a voltage to simulate any level of acceleration.

4.4.2 Phase I - Determination of Maximum Flow Canacity

The maximum flow capacity of the Garrett Electronic/Pneumatic anti-Gvalve was measured under open-flow conditions where the terminating suit volumewas omitted, but the normal hose remained attached to the output port of the flowtransducer. This hose was terminated by the female quick-disconnect fitting used forattaching the simulated anti-G suit volume.

54

Page 69: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

Figure 4.4-1. Garrett Electronic/Pneumatic Anti-G Valve.

55

Page 70: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

Results of the open-flow testing are shown in Figure 4.4-2. With the valvedesigned to produce ready pressure, the airflow at baseline G (1.4 G) was 8 SCFM.As the G level increased, the flow was 12.3 SCFM at 2 G and 22 SCFM at 4 G. As Glevel continued to increase, flow increased to a maximum of 28.4 SCFM at 5.7 G, thendropped to 26.5 SCFM at 5.8 G; it then stabilized, reading 25.5 SCFM at 8 G, and 25.8SCFM at 11 G. Upon deceleration, flow initially increased somewhat up to 26.3SCFM, and continued to increase to 26.8 SCFM as the G level decreased to 5.5 G;flow then began to decrease rapidly reaching 7.9 SCFM at 1.4 G.

4.4.3 Phase II - Determination of Dynamic Response Capability

The dynamic response capability of the Garrett Electronic/Pneumaticanti-G valve was determined as described in Testing Procedures (section 3.2.2). Thepressure delivered by the valve was measured at the output port and inside thesimulated anti-G suit volume. "Output pressure" refers to the pressure within thesimulated suit volume, unless otherwise noted.

4.4.3.1 Low G,-Onset-Rate Conditions

The 0.1 G s-1 trapezoidal profile was used for all low-onsettests. When this valve was tested under low Gz-onset conditions at 150-psig sourcepressure and a valve angle of 00, the results shown in Figure 4.4-3 were obtained.The suit pressure was 0.8 psig at 1.4 G. At 1.5 G the pressure began to increase andmaintained an essentially linear rate of increase up to 7 G, with a pressure of 3.3 psigat 3 G; 6.5 psig at 5 G; and 9.5 psig at 7 G. After 7 G, the rate of increase slowedsomewhat, reaching a maximum of 10.3 psig at 7.7 G and remaining at this pressureas G level increased to 11 G. The source pressure was turned off during the intervalbetween when G level reached 7.7 on acceleration and 7.5 on deceleration due toexcessive flow. Upon deceleration the suit pressure decreased linearly from 10.3 psigat 7.5 G to 0.8 psig at 1.4 G. The average pressurization rate was 1.6 psig G- 1 over theuseful range of this valve.

The effect of valve angle upon low Gz performance of the Garrett Servo-controlled valve was explored using valve angles of 150 and 300, and the mediansource pressure of 150 psig. At a valve angle of 150 (Fig. 4.4-4), suit pressure was 0.6psig at 1.4 G and started to increase at 1.5 G. From 1.5 G to 7.5 G, suit pressureincreased linearly; pressure was 3.1 psig at 3 G, 6.1 psig at 5 G, 9.3 psig at 7.5 G, and10.2 psig at 7.8 G. Suit pressure remained at the maximum value of 10.2 psig as Glevel increased to 11 G, and during the time at 11 G. As G level decreased, the suitpressure decreased linearly from 10.2 psig at 11 G to 0.6 psig at 1.4 G.

When the valve angle was 300 (Fig. 4.4-5), there was no suit pressure measuredat 1.4 G; the valve initiated flow at 1.5 G and pressure was 0.5 psig at 1.6 G. Suit pres-sure increased linearly from 2 to 8 G, with pressure readings of 2.5 psig at 3 G and 5.3psig at 5 G. At 8 G the rate of pressurization began to level off, measuring 9.5 psig at 8G and reaching a maximum pressure of 10.3 psig at 8.8 G. Suit pressure remained at10.3 psig as the G level increased to 11 G and remained there. Upon decelerationfrom 11 G to 1.4 G, suit pressure decreased linearly down to 0 psig at 1.4 G.

56

Page 71: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

I _ TT-- .L..... l I _____ J T I -1i i

I II I I N I I I, I I'" Ill I F ijIE V K'i l i I I I I I 1I t t[ I!! !.I I I i

- ' I I!!I I L I 1111 I\I IIT_ - 1 11111 11 I I I

1 - a ', I t 1 I I i I I I I I I I I

7 ; 1 .4? 4 9 t e 5 4 3 2 1. 4 . L. V'_L -

7. -' 7i- 0 0 0! rime! r (SE; il co 0 Nj a '0 6 - N (.i 0 N

Figure 4.4-2. Maximum flow through Garrett Electronic/Pneumatic anti-G valve.

3I NO !SOURCE PRESSURE TURNED OFF SETWE FONTS ' A

1 3 DUE I xozi s i iTLOW A C SO F I

W

I I // i I i IZI I II ! I l ra! - i

, _JTI "III I ] I -__! / ; I t I! I I 'I ! I\

/r I

I= i__ A ZI I I i I I I i I I I_ _

____

o .I ! I I I " t0 T-----*-

1 1.42 3 4 5 6 7 a 9 10 1I I l i IC 1 8 7 S 5 4 3 2 t4 L41.4 Gz .. EVEL I

0 0 0 0 N P7 N 0 _.- " N ai- an)

N ~ C at 0 TIM (W')N~ q C 6

Figure 4.4-3. "Suit" pressures produced during acceleration/deceleration by Garrett Electronic/Pneumaticanti-G valve under conditions of: 150-psig source pressure; low G onset; 10-liter volume;00 angle.

57

Page 72: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

The effect of source pressure upon performance of the Garrett Electronic/Pneumatic anti-G valve during low Gz-onset was determined by testing at both theminimum and maximum source pressures; i.e., 30 and 300 psig, respectively. At asource pressure of 30 psig (Fig. 4.4-6), the valve produced a suit pressure of 0.7 psigat 1.4 G. Suit pressure increased linearly as G level increased to 7 G; pressures of 1.6psig at 2 G, 6.5 psig at 5 G, and 9.6 psig at 7 G were recorded. The pressure did notclimb further after reaching 9.6 psig, but remained at this level as G increased to, andremained at, 11 G. When G level decreased from 11 to 1.4, the suit pressuredecreased linearly from 9.6 psig to 0.7 psig.

When a source pressure of 300 psig was employed (Fig. 4.4-7), a suit pressure of0.7 psig was measured at 1.4 G. The pressure immediately began to increase linearlyas G level increased to 7.5 G, reading 1.6 psig at 2 G, 6.4 psig at 5 G, and peaking at10.3 psig at 7.5 G. At this maximum source pressure, the rate of suit pressurizationlagged the rate measured at the minimum source pressure (30 psig, Fig. 4.4-6) by0.1 psig.

I t I I I I r n i i

I" I i L i I.I I ... ......../ll I \ I i I:i; _ I I I I I I I I I I I I I I I \ I I I

i / ± ij j j[ [ Ij I I - - - I 3... ,I - - !J I I i iI ! I \ __ r

II I

S11.4. 2 3 4 6 9 10 I II .1 1 0 8 7 0 " 84 2 Z14 A1.4 G, L_____ E L

0 L4.0 D ld~l C f I - N . N. o~0

N- 0 Y In 4 0 ~d TIME (SEC)

Figure 4.4-4. "Suit" pressures produced during acceleration/deceleration by Garrett Electronic/Pneumaticanti-G valve under conditions of: 150-psig source pressure; low G onset: 10-liter volume; 150angle. Source pressure turned off between points "A" and "B" due to excessive flow.

58

Page 73: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

I I I I I I

I I I I I

,_/ I I II N 1 1110 2 6 7 6 9 10 1: 1 1 1 10 9 1 7 6 t 4 3 2t4L IA G 6EE

I , 1 6 - --

SO$ ; 8 0 2 0 0-1..

Figure 4.4-5. "Suit" pressures produced during acceleration/deceleration by Garrett Electronic/Pneumaticanti-G valve under conditions of: 1 50-psig source pressure; low G onset; 1 0-liter volume; 300angle. Source pressure turned off between points "A" and "B" due to excessive flow.

4.4.3.2 High G.,-Onset-Rate Conditions

For high Gz-onset conditions, the 6 G s- trapezoidal profilewas employed. The pressure delivered by the Garrett Electronic/Pneumatic valve wasmeasured at both the valve output port and inside the simulated suit volume.However, "output pressure" refers to the pressure within the suit volume, unlessotherwise noted.

When the valve angle was 00 and the source pressure 150 psig (Fig. 4.4-8), thesuit pressurization schedule (psig G-1) under high Gz onset lagged that observedunder low Gz-onset conditions, seen in Figure 4.4-3. At 1.4 G, the suit pressure was0.6 psig (Fig. 4.4-8); as G level increased, suit pressure increased linearly. The suitpressure was 2.8 psig at 3 G, 4.4 psig at 4 G, 5.8 psig at 5 G, and 6.5 psig at 5.5 G.Under these conditions, this valve also exhibited a low-frequency oscillation (1.2 psigp-p amplitude) at the valve outlet port over the range of 1.5 G to 5.5 G. From 5.5 G to11 G, the pressurization rate slowed considerably with the result that the suit pressurewas 8.5 psig at 8 G, 8.9 psig at 9 G, and a maximum of 9.3 psig at 11 G. The pressureremained at the maximum level during the time at 11 G. As deceleration started, thepressure increased slightly until 8 G was reached; the pressure then decreased lin-early as G level decreased to 1.4 G. The valve began oscillating (0.9 psig p-p) whendeceleration started, and continued oscillating until G level had decreased to 4 G.

The effect of valve angle upon performance of the Garrett Electronic/Pneumaticvalve under high Gz-onset conditions was explored by using valve angles of 150 and300 while the source pressure was held constant at 150 psig. At a valve angle of 150

59

Page 74: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

IINII I I I I I i

'°. I

I IAI I I L I I [ II

0 I 3 4 ' 'l I I i l I l 1 06 .1 4 .a:~~ i*Ie-.1 - ]o [IE(S(-r / *\ 1 i ___

[...]I. tI'liL If If I .. ..1 ...f-" - tItIA2 3 4 5 6 7 6 9 I0 II II II II 0 $ 8 7 6 5 4 3 2 1.41.41.4 Gt'VI

i 0 . . .o n - C 0 0 0,"9r

Figure 4.4-6. "Suit" pressures produced during acceleration/deceleration by Garrett ElectronictPneumaticanti-G valve under conditions of: 30-psig source pressure: low G onset; 10-liter volume: 00angle. Source pressure turned off between points "A" and "B" due to excessive flow.

-."-. - -F! 1b 01 !I~ i

4....i.........i .....i..... I ir i I !\ I I t

I/ I I I I I I I I I I I I I I II ..j.

1 1.42 3 4 5 6 8 9 10 1 11 11 1 1 10 9 9 7 6 5 4 3 214 141.4 G.LV.-

0 0TIM W!(SC)

Figure 4.4-7. "Suit" pressures produced during acceleration/deceleration by Garrett Electronic/Pneumaticanti-G valve under conditions of: 300-psig source pressure; low G onset: 10-liter volume: 00angle. Source pressure turned off between points "A" and "B" due to excessive flow.

60

Page 75: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

i I ...... i ...i...L.....ii....L... I.....A - t I

'D '0~ In 0.a RCi TM (FT

Figure 4.4-8. Suit" pressures produced during acceleration/deceleration by Garett Electronic/Pneumaticanti-G valve under conditions of: 1 50-psig source pressure; high G onset; 10O-liter volume; 0°

angle.

7--- -. L..J...., -

I I

-' -- li i I I {!

1111 I I I T I I I

I I 11! __ __

I0

n - - - - - " E -S C 7

Figure 4.4-9. "Suit" pressures produced during acceleration/deceleration by Garrett Electronic/Pneumaticanti-G valve under conditions of: 150-psig source pressure; high G onset; 10-liter volume;150 angle.

61

... . .. -. ,,.-- ,,-,,= ,.,,, .I a - -mme • i i l-

Page 76: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

(Fig. 4.4-9), a low-frequency oscillation (of amplitude 1.4 psig p-p) was again observedas the G level increased from 1.4 G to 6 G. The suit pressurization schedule wassimilar to that just observed for the 0° angle. The slope of the increase in suit pressurewith increasing Gz level approached linearity until 7 G was attained (Fig. 4.4-9); theslope of the line then decreased, plateauing at 9.3 psig at 11 G. The suit pressure was0.5 psig at 1.5 G, 2.9 psig at 3 G, 6.8 psig at 6 G, 8.9 psig at 9 G, and 9.3 psig at 11 G.Suit pressure increased slightly during the interval at 11 G so that prior to initiation ofdeceleration, pressure was 9.5 psig. Upon deceleration, suit pressure began todecrease at 8.5 G; the decrease was then linear until 1.4 G was reached. Pressurewas 4.3 psig at 4.5 G, and 1.0 psig at 1.4 G.

For a valve angle of 300 (Fig. 4.4-10), suit pressurization began slowly, with zeropressure at 1.5 G and 0.5 psig at 2 G. The rate of pressurization as a function ofincreasing Gz level was approximately linear until 8 G was attained, although a smalldecrease in slope occurred at 4 G. Pressures.of: 2.3 psig at 3 G, 4.7 psig at 5 G, and6.9 psig at 7 G were observed. Suit pressure leveled off at 9.3 psig at 10.5 G, thenincreased to a maximum of 9.6 psig after 10 s at 11 G. Oscillation (1.2 psig p-p)occurred as Gz level increased from 1.4 G to 7 G, then ceased. During deceleration,the suit pres ire remained at the maximum level (9.6 psig) until the G level decreasedto 9.5 G; pressure then decreased linearly with decreasing G level down to 0.9 psigupon reaching 1.4 G; after 0.5 s at 1.4 G, pressure began to decrease slowly to thefinal value 0.3 psig. A low-frequency oscillation (0.75 psig p-p), starting at 9.5 G andcontinuing until 4.5 G, was again observed during deceleration; suit pressuredecreased from 9.6 psig to 4.9 psig in the presence of the oscillation.

The effects of source pressure upon high Gz-onset performance of this Garrettvalve were explored by testing at both the minimum (30 psig) and the maximum (300psig) source pressures, while keeping the valve angle at 00. When the sourcepressure was 30 psig (Fig. 4.4-11), the suit pressurization schedule was roughly linearfrom 1.4 G to 7 G, although the slope of the line decreased slightly at 5 G. Suitpressure was initially 0.6 psig at 1.4 G; it increased to 3.6 psig at 3.5 G, then 6.9 psig at6 G, and leveled off at 8.7 psig at 9.5 G. Within 5 s of reaching 11 G, suit pressureincreased to 9 psig; upon deceleration, suit pressure did not begin to decrease until Glevel had decreased to 7.5 G; it then decreased linearly to 1.3 psig at 1.4 G anddropped slowly to 0.6 psig after 0.5 s. Oscillations were observed upon acceleration(1.0 psig p-p) during the interval between 1.4 G and 3.5 G, and upon deceleration (0.9psig p-p) from 7.5 G to 4 G.

When the source pressure was 300 psig (Fig. 4.4-12), the suit pressurizationschedule was approximately linear from 1.4 G to 7 G; at this point, the slope decreasedslightly and continued in a positive linear fashion to 9 G, curving to a very gradualslope which increased throughout the interval at 11 G and the initial stages ofdeceleration. The suit pressure was 0.6 psig at 1.4 G, 8.3 psig at 7 G, and 11.3 psig at11 G which increased to 11.9 psig after 5 s "on top" (at 11 G). Upon deceleration, thesuit pressure did not change until G level had dropped to 9 G; the pressure thendecreased linearly to a value of 1.4 psig at 1.4 G. The slope of the increase anddecrease in suit pressure observed here was identical to the slopes observed for asource pressure of 30 psig (Fig. 4.4-10), up to 7 G during acceleration, and from 6.5 Gto 1.4 G during deceleration. The significant difference between these two conditions

62

Page 77: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

I j..I . ~I , ,

T T

- 11VF1 IJ I J i i }iii I I)'1i I I i r1 111 1r !i' J I _ __) I, I i i I

I I I I I I I I1l* I /I I I -I-I-7- -,--r - -. ,- -I I.. ! ,! I I -1..ii I

11.4 2 4 86 7 9 10 1 1 1 1110 9 S 7 65 4 3 2 1.4141.4 . LEVELI

__ N TIM (S - N NEC

Figure 4.4-10. "Suit" pressures produced during acceleration/deceleration by Garrett ElectronicJPneumaticantl-G valve under conditions of: 150-psig source pressure; high G onset; 1 0-liter volume;300 angle.

0- 1 I~ 1- 1- -7 -iHV

67 1F 1 1

0

11t.4 Z 3 4 5 6 7 a 9 10 11 1t I 1 1 e 1 1 10 9 6 5 4 3 2' 14 L41.4 G. L E VEL

_lIl tllm llill

0! I - I T I

Fiue-.-. "Sit prsue prdue duin aceeaio/elrainbGret ecmcPeu

Figre .411."Sit" prlesunres niin f 3 gsuc prouedungaeertodeueer hi h by Garret Elieronumatic0angle.

63

Page 78: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

of source pressure was the maximum suit pressures obtained, which were 9 psig at asource pressure of 30 psig vs. 12 psig at a source pressure of 300 psig. Low-frequency oscillations were again observed during acceleration (1.05 psig p-p) from1.4 G to 7 G, and during deceleration (0.75 psig p-p) from 9 G to 4.5 G.

4.4.4 Phase III - Determination of Comolex Dynamic Resoonse Capability

The Phase III testing under SACM conditions was performed asdescribed in Testing Procedures (section 3.2.3). Initially, median values of sourcepressure and suit volume were selected and the valve angle varied; subsequently,valve angle was maintained at 00 and the combinations of: minimum source pressurewith maximum suit volume, and maximum source pressure with minimum suit volumewere evaluated.

For the median source pressure (150 psig) and valve angle of 0 (Fig. 4.4-13), thevalve started with 0.6 psig suit pressure at 1.4 G. With a high onset lo 3 G (3 psig) thevalve showed oscillation of 0.9 psig p-p from 1.4 G to 3 G. Upon rapid-onsetacceleration to 9 G, the suit pressure increased to 9.6 psig at 9 G and leveled off at 9.7psig; oscillation occurred between 3 G and 6.5 G (1.2 psig p-p). Decreasing rapidlydown to 4.4 G and 5.1 psig, low-level oscillation (1.05 psig p-p) was present between8.5 G and 4.4 G.

Dropping off to 3 G the suit pressure dropped to 3.1 psig. Upon acceleration to 5G suit pressure increased to 6.1 psig; subsequent deceleration to 4 G resulted in a suitpressure" of 4.6 psig. With a rapid-onset acceleration to 9 G the suit pressureincreased to 9.6 psig with a slight oscillation (1.0 psig p-p) occurring between 4 G and6 G; the pressure leveled off at 9.7 psig at 9 G. With a rapid-offset deceleration from 9G to 1.5 G the suit pressure decreased to 1.5 psig, and after 6 s at 1.5 G the pressureleveled off at 0.9 psig; oscillation (1.2 psig p-p) occurred from 9 G down to 4.5 G. Witha rapid-onset acceleration to 8 G the suit pressure increased to--and leveled off at--9.8psig with oscillation (1.1 psig p-p) between 1.5 G and 6.5 G. Decelerating from 8 G to1.4 G, the pressure dropped off to 0.7 psig--with oscillation (0.6 psig p-p) from 7.8 Gdown to 1.4 G.

The angular-dependent studies of the Garrett valve during the Phase III SACMtesting used valve angles of 150 and 300 while the source pressure and simulated G-suit volume remained at their median values (150 psig and 10 liters), respectively. Ata valve angle of 150 (Fig. 4.4-14), the valve started to flow at 1.4 G with 0.5 psig suitpressure, and at 3 G the pressure was 2.8 psig; a slow oscillation of 2.9 psig p-p wasobserved between 1.4 G and 3 G.

With a high onset to 9 G the pressure in the G-suit volume increased rapidly to 9.4psig with a low-frequency oscillation of 1.2 psig p-p. After 1 s on top at 9 G thepressure increased to 9.6 psig with an oscillation of 0.9 psig p-p. A rapid offset downto 4.4 G produced a suit pressure of 4.9 psig. A slow further offset to 3 G brought thepressure to 2.9 psig, while a slow-onset back up to 5 G resulted in a pressure of 5.8psig. A slow offset down to 4 G dropped the pressure to 4.4 psig. With a high onset to9 G the pressure increased to 9.6 psig with oscillation (0.9 psig p-p) between 4 G and6.5 G; after 3.5 s on top at 9 G the pressure increased to 9.7 psig. With a high-offset

64

Page 79: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

IIV I N I 1" -f' H-K F"-' Fl P T J ' L ,,o i1 1/,1 I I t IL II \ !iZ L "z.I

- I II ' k

11 11! t, 1t !i\ l

;/ f I i I I 1 Jl 7 1 1 1 1 1 1 17 i I I I I I I I I I I I I11,4 2 11 4 6 7 0 9 10 i I I I 9 8 78 6 5 4 3 2 L4 1.414 G LEVEL

. . . . . . . . .. . .. .. "0 - . , o1N

- - - - - - , ~ , •*' M M T:ME (S C!

Figure 4.4-12. "Suit" pressures produced during acceleration/deceleration by Garrett Electronic/Pneumaticantl-G valve under conditions of: 300-psig source pressure; high G onset; 10-liter volume;0°angle.

_ 1 1 1 I I I .I I I I I r t I I I I I I I I I I

___ I I _ I I I i f I L- IIIt I I I I I

r-4 3.0 1 .01 42- 1,4 G* LEVEL

0 10 .0 30 40 50 so 70 so 90 100 IW. T C*

Figure 4.4-13. "Suit" pressures produced during SACM by Garrett Electronic/Pneumatic anti-G valveunder conditions of: 150-psig source pressure; 10-liter volume; 00 angle.

65

Page 80: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

from 9 G to 1.5 G, the pressure decreased rapidly to 0.8 psig, with a low-frequencyoscillation of 0.8 psig p-p. A rapid-onset acceleration to 8 G produced a suit pressureof 9.3 psig with oscillation between 1.5 G and 6 G; and after 3.5 s the pressure leveledat 9.7 psig. Upon deceleration from 8 G down to 1.4 G the pressure decreased to 0.6psig; oscillation (0.5 psig p-p) occurred from 8 G to 1.4 G.

At a valve angle of 300 (Fig. 4.4-15) the valve started flowing at 2 G, and at 3 Gthe pressure was 1.1 psig; oscillation of 0.8 psig p-p occurred between 1.4 G and 3 G.After 4.5 s at 3 G the pressure settled down at 2.5 psig. With a high onset to 9 G thepressure increased rapidly to 9.5 psig and after 6.5 s the pressure was 9.8 psig.Oscillation (1.1 psig p-p) was observed between 9 G and 7 G. Upon deceleration from9 G down to 4.4 G the pressure decreased to 4.1 psig with oscillation (0.9 psig p-p)between 9 G and 4.4 G. A further decrease to 3 G resulted in a pressure of 2.4 psig. Aslow onset to 5 G brought the pressure to 5 psig. Upon slow deceleration to 4 G thepressure decreased to 3.7 psig. A high onset to 9 G, with oscillation (0.9 psig p-p)between 4 G and 7.5 G, produced a suit pressure of 9.5 psig to 9.7 psig. Deceleratingrapidly from 9 G to 1.5 G the pressure decreased to 0.9 psig, and after 0.5 s settled at0.3 psig; oscillation (0.9 psig p-p) occurred between 9 G and 4.5 G. With a high onsetto 8 G the pressure increased to 9.1 psig and then dropped off to 8.3 psig within 0.5 s;oscillation (0.9 psig p-p) occurred between 1.5 G and 7 G. From 8 G down to 1.4 G thepressure decreased to 0.2 psig; oscillation (0.5 psig p-p) was again observed between75Gand2G.

Two source pressure-dependent studies were conducted in this phase on theGarrett valve. While the valve angle was at 00, one experiment combined a sourcepressure of 30 psig with a 14-liter simulated suit volume, and the other combined 300-psig source pressure with a simulated suit volume of 6 liters. At a source pressure of30 psig and a volume of 14 liters (Fig. 4.4-16), the suit volume pressure was 0.6 psig at1.4 G, and 2.7 psig at 3 G-but leveled off at 3.1 psig within 1 s. Oscillation (1.2 psig p-p) was observed between 1.4 G and 3 G. With a rapid onset to 9 G, the pressureincreased to 8.7 psig with oscillation (0.9 psig p-p) between 3 G and 5 G. Leaving 9 Gwith a high offset rate to 4.4 G, the pressure rapidly decreased to 5.3 psig; after 9 s at4.4 G the pressure further decreased to 5.1 psig. Oscillation (0.8 psig p-p) wasobserved between 8.5 G and 4.4 G. Deceleration to 3 G produced a pressure of 3.1psig; a subsequent increase to 5 G increased suit pressure to 6 psig, and anotherdecrease to 4 G dropped the pressure to 4.5 psig. With a high onset rate to 9 G thesuit pressure increased rapidly to 8.8 psig and leveled off at 8.9 psig with a smalloscillation (1 psig p-p) between 4 G and 5 G. Upon rapid offset down to 1.5 G the suitpressure decreased rapidly to 2.4 psig, and after about 6 s the pressure decreased to0.8 psig. Oscillation (0.8 psig p-p) was observed between 8.5 G and 5 G. With a highonset rate to 8 G the suit pressure climbed to 8.4 psig, and after 9 s at 8 G it leveled offat 9 psig; oscillation (1 psig p-p) was present between 1.4 G and 3 G. With a highoffset down to 1.4 G the pressure decreased rapidly to 0.8 psig, with oscillation (0.3psig p-p) between 7 G and 1.4 G.

At a source pressure of 300 psig and suit volume of 6 liters (Fig. 4.4-17) the suitpressure was 0.6 psig at 1.4 G; at 3 G the pressure was 3.1 psig and oscillation (1 psigp-p) was observed between 1.4 G and 3 G. With a high onset from 3 G to 9 G, the suitpressure increased rapidly to 11.7 psig and leveled off at 12.4 psig after about 6.5 s;an oscillation (1.2 psig p-p) occurred from 3 G to 7.5 G. Upon subsequent high offset

66

Page 81: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

I ____ JI__

_________~i _ I j I ____

L~ ~ ~ ~. _5 1 4! [GLEE

I I___ I ..~1.... I ..--- I I / )I _ -__ ! ,L ... ..... E 1

T1 7 I I I i I i i i I I I I i t I I I I I i f I i

,.4 _ i 0 1 4 1-0 [ 1.4 , L I V E L

IC0 10 20 30 40 so 60 70 8s 0 O00 11 o TIME (SEC I

Figure 4.4-14. "Suit" pressures produced during SACM by Garrett Electronic/Pneumatic anti-G valveunder conditions of: 150-psig source pressure; 10-liter volume; 150 angle.

'* I .....tt JI.....i ' j. .I I I I I,

___ l _I I I Ii 1 \.!1l 1 !\ L.Ji

H 0 10 20 0 40 50 60 70 S0 go90 c i rO TIME( SEC)

Figure 4.4-15, "Suit" pressures produced during SACM by Garrett Electronic/Pneumatic anti-G valveunder conditions of: 150-psig source pressure; 10-liter volume; 30 angle.

67

Page 82: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

I Fl 5 1i

ILL 1 I I I I I I _ I I I

17 11 1 17 j I I I_ 11 .1 .1 ..I..1 I 1

- 1 1 i I I . I

I I I I I I I I 1 I7 1 i I I I i I

ri0j 90~- 1 Ck-01 0 .0G LEVEL

o 0 20 30 400 s o 70 so 90 100 Ito TIME SEC)

Figure 4.4-16. "Suit" pressures produced during SACM by Garrett Electronic/Pneumatic anti-G valveunder conditions of: 30-psig source pressure; 14-liter volume; 00 angle.

f4 i !LT I i\ 1!DI Ki"'. I f!!1 i I , i I 17" i i I i i _ I i f!

,II 9.I i II I ll 8i 171T i I l Ii I I II I I

'.......~ .i IijiiiI I ____ I I I I 'rV- IZIt , i ,

I I 0 L 0 11 5 . . L E V E L

0 I 20 30 40 50 o 70 s0 90 00 110 TiME (SEC)

Figure 4.4-17. "Suit* pressures produced during SACM by Garrett Electronic/Pneumatic anti-G valveunder conditions of: 300-psig source pressure; 6-liter volume; 00 angle.

68

Page 83: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

down to 4.4 G the pressure decreased to 5.1 psig, with oscillation (1 psig p-p) between8.7 G and 4.4 G. Further decrease to 3 G dropped the pressure to 3 psig. Accelerationto 5 G brought the pressure to 6 psig, and the next deceleration to 4 G reduced thepressure to 4.5 psig. With a high onset to 9 G the suit pressure increased to 11.9 psigand leveled off at 12.5 psig after about 6.5 s; a low-level oscillation (1.3 psig p-p) wasobserved between 4 G and 7.5 G. With a high offset rate to 1.5 G the suit volumepressure decreased rapidly to 0.9 psig and leveled off at 0.8 psig; oscillation (0.9 psigp-p) was again observed between 8.7 G and 2.5 G. With a rapid-onset acceleration to8 G the pressure increased to 10.8 psig, with oscillations (1 psig p-p) observedbetween 1.5 G and 7.5 G; after about 8.5 s at 8 G the pressure leveled off at 12 psig.With a high offset down to 1.4 G the pressure decreased to 0.6 psig, with a 0.5 psig p-poscillation observed between 7.9 G and 1.4 G.

4.5 Hymatic VAG-1 10-022 Anti-G Valve

4.5.1 Description

The Hymatic VAG-1 10-022 anti-G valve (SN 276AS) (Fig.4.5-1) wasmanufactured by Hymatic Engineering Co., Ltd. (Redditch, England), for use by Britishmilitary personnel. This valve has an inlet filter; a nonreturn valve; a manuallyoperated two-position ("On/Off") valve; a spring-loaded G-sensitive main valveassembly; a diaphragm which incorporates a filter; an outlet connection containing acleanable biazed-in filter; and a spring-loaded relief valve (which preventsoverpressurization of the anti-G suit). The valve body, main piston, and massconstitute one matched assembly; the stop valve and stop valve body make upanother matched assembly.

This valve is designed so that placing the stop valv! lever in the "Off" positioncloses the valve to inlet airflow, thus preventing air from entering the valve body. Theanti-G suit is vented to ambient pressure through vents in the valve body, and throughports drilled to a vent in the stop valve barrel. Placing the stop valve lever in the "On"position (in the absence of G forces) prevents air from entering the anti-G suit becausethe valve piston is oriented across the flow passage. Again, the suit is vented toambient pressure through the valve body vents.

As G forces are applied to the valve, the mass pushes down the main and lowerpistons against the spring loads, resulting in closing of the bleed-air vents on the sidesof the valve body and allowing inlet air to enter the valve. The inlet air enters both thevalve outlet to the anti-G suit and the main piston (through cross holes); pressure isexerted on the lower piston, forcing it down to the lower stop assembly. The mainpiston is then suspended between the lifting forces of the upper spring pressure plusair pressure and the depressing forces of the mass acted upon by G ,"ces. Theresultant of these vectors produces a steady suit pressure; the leakage o" air past themain piston and through the bleed-air vents is equal to the inlet airflow. As G forces'4ecrease, the piston and mass are returned to the static condition (by spring pressure),.nd the suit pressure again exhausts to ambient atmosphere through the gauze-covered vents in the valve body.

69

Page 84: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

70P COVER

STOP VALVEoLEVER O

ANDV 7RN EIE AV

"LETLET

00

Page 85: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

Overpressunzation of the suit causes air pressure to act upon a spring-loadedrelief valve set on the outlet side of the anti-G valve. When this action occurs, the reliefvalve is lifted from its seat and excess pressure is vented directly to the ambientatmosphere. The relief valve contains a frictior spring which ensures smoothoperation by damping out small spring movement.

If the air supply should fail during a Gz maneuver, deflation of the pressurized suitwould be controlled by a nonretum valve adjacent to the inlet filter. A preflight check ofvalve function can be performed when engines are running by manually depressingthe diaphragm assembly in the top cover.

Performance of the Hymatic VAG anti-G valve was evaluated over the range ofsource pressures specified by the manufacturer: 120 psig, maximum; 30 psig,minimum; and 60 psig, median.

4.5.2 Phase I - Determination of Maximum Flow Capacity

The maximum flow capacity of the British Hymatic VAG-1 10-022 anti-Gvalve was measured with a source pressure of 60 psig, the median source pressurefor this valve. The Gz-onset conditions were 0.5 Gz s-1, and the valve angle wras 00.The results of the maximum-flow testing are shown in Figure 4.5-2. At baseline G (1.4G), flow was 0.4 SCFM. As the G level increased, the flowrate did not change until 2 Gwas reached when the flow was 0.75 SCFM; flow then increased rapidly, reaching 7.3SCFM at 2.4 G. At this point, the rate of increase slowed, but remained essentiallylinear up to 7 G, where the flowrate was 17.8 SCFM. Flow plateaued at 17.8 SCFMwhile G level increased to--and remained at-i 1 G for 4.5 s; flow did not change untilapproximately 6.5 G was reached during deceleration. The rate of decrease inpressure during deceleration closely mirrored the rate of increase during acceleration;a linear decrease occurred from 17.8 SCFM (11 G) to 2.5 SCFM (7.3 G). At this point,flow began to drop exponentially to 1.6 SCFM at 2 G, followed by a minimal flow of 0.5SCFM from 1.7 G throughout the time at baseline G (1.4 G).

4.5.3 Phase II - Determination of Dynamic Response Capabilily

The dynamic response capability of the Hymatic VAG-1 10-022 anti-Gvalve was determined as described in Testing Procedures (section 3.2.2). Pressuredelivered by the valve was measured at the output port and inside the simulated anti-Gsuit volume. "Output pressure" refers to the pressure within the simulated suit volume,unless otherwise noted.

4.5.3.1 Low G,-Onset-Rate Conditions

The 0.1 Gz s-1 trapezoidal profile was used for all low-onsettests.

Under conditions of low Gz-onset, median source pressure (60 psig), and a valveangle of 00, the results shown in Figure 4.5-3 were observed. As Gz level began to

71

Page 86: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

increase from the baseline of 1.4 G, there was no increase in suit pressure until after2 G; the increase in pressure was then rapid, reaching 1.5 psig at 2.3 G, whereuponthe rate of pressure increase slowed and became linear from this point up to 9 Gwhere it was 9.2 psig. The increase in suit pressure then became slightly erratic, butreached the maximum value of 10.4 psig when 11 G was attained. The pressureremained stable at 10.5 psig while 11 G was maintained. Upon deceleration, suitpressure began to decrease immediately with a linear slope which mirrored theincrease upon acceleration, except that the drop at the low end was more gradual thanthe initial jump in pressure with acceleration; at 2.5 G the suit pressure was 2 psig, at 2G it was 1.1 psig, at 1.8 G it was 0.2 psig, and at 1.4 G (baseline) suit pressure was 0psig.

The effects of valve angle upon low-Gz performance of the Hymatic VAG valvewere explored by using valve angles of 150 and 300 while maintaining sourcepressure at the median value of 60 psig. When the valve angle was 150, the resultsshown in Figure 4.5-4 were observed. Upon acceleration, suit pressunzation did notstart until G level was increased from 1.4 G to 2.4 G (0.2 psig); pressure then jumpedup rapidly to 2.1 psig at 2.6 G and increased slightly to 2.2 psig at 3 G. From this point,the increase in suit pressure became consistently linear with the increase in Gz levelup to 10.2 psig at 11 G; however, a small shoulder was observed in the region of 9.5-10 G. Suit pressure then remained stable at the maximum value of 10.2 psig while Glevel was maintained at 11 G. Upon deceleration, suit pressure began to drop at 10.5G (10.2 psig), with an essentially linear decrease down to 2 G (1 psig); pressure thendropped rapidly to 0 psig at 1.7 G.

The results obtained when the valve angle was 300 are shown in Figure 4.5-5.Pressurization of the suit was not initiated until 2.9 G was reached. Suit pressure was0.4 psig at 3 G; it then climbed rapidly to 2.6 psig at 3.3 G, whereupon it droppedslightly before beginning a nearly linear increase up to 8.1 psig at 10 G. At this pointthe suit pressure buildup faltered slightly but reached 8.8 psig as 11 G was attained.During the time interval at 11 G, the pressure continued to increase slightly, reaching9.4 psig just before deceleration was initiated. As deceleration began, suit pressuredid not change significantly down to approximately 9.7 G; from this point, pressurebegan to decrease essentially linearly while decelerating to 2.5 G (1.6 psig). At 2 G,pressure was 0.2 psig, and at 1.7 G it was 0 psig.

The effects of source pressure upon low Gz-onset performance of the HymaticVAG valve were observed by testing at both the minimum and maximum sourcepressures for this valve (30 psig and 120 psig, respectively) while keeping all otherconditions constant. At a source pressure of 30 psig, the valve performed as shown inFigure 4.5-6. Upon acceleration, suit pressure was 0 psig until after 2 G was reached;it then jumped up to 1.9 psig at 2.5 G where it began a linear increase up to 8.9 psig at9 G; it then continued to increase in a somewhat erratic fashion, reaching 10.1 psig at10.6 G, and stabilizing at 10.2 psig throughout the time at 11 G. Upon deceleration,the suit pressure did not begin to fall until approximately 9.3 G was reached; after thispoint pressures of 9.9 psig at 9.3 G, 9.5 psig at 9 G, and 9.2 psig at 8.5 G wereobserved. Suit pressure then decreased linearly with G level until 2 G (1.1 psig) wasreached; after this point, the pressure dropped to 0.7 psig at 1.9 G and continued downto 0 psig at 1.6 G.

72

Page 87: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

I II I I I i.t I i.... . I 1f Ift I I

-iii1'I ! I i I .

01 111II. J .I .I II I I I I [ I I I

11.42 3 4 3 6 7 5 9 ,o I'I I I 1 ( o i 6 i i 3 L.L4 . .-4 , L .111. .. . .* S U . . , . . . .6 6 O -

U 1. -dO - en ~ *- ~TIME (SEC)I

Figure 4.5-2. Maximum flow through Hymatic VAG-1 10-022 anti-G valve.

TF -L

I ,/ iI r, i i I I I

: Ii I I I I I I I I -f i

11 23 5 76 I 11110 t S 5 4 3 ZI44. , L!VI

Figure 4.5-3. "Suir pressures produced during acceleration/deceleration by Hymatic VAG-1 10-022 anti-Gvalve under conditions of: 60-psig source pressure: low G onset: 10-liter volume; 00 angle.

73

Page 88: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

I I I I 1____ I _ I ___ I I I I i I IT I rt1-' I i T , l

III lJI i l ' I1H III I......i..III III II I

--; I! I I K V I , iI I r~iK I lL4 2 3 4 5 a 7 8 9 O II I1 JO 1 1 1 0 S 4 3 2 1.414r.4 G1L VL

0. : : 14 1 O SII a52 f (Saw)

Figure 4.5-4. "Suit" pressures produced during acceleration/deceleration by Hymatic VAG-1 10-022 anti-Gvalve under conditions of: 60-psig source pressure; low G onset; 1 0-liter volume; 150 angle.

I J f 7

Figure 4.5-5. "Suit" pressures produced during acceleration/deceleration by Hymatic VAG- 110-022 anti-Gvalve under conditions of: 60-psig source pressure; low G onset: 10O-liter volume; 300 angle.

74

I iiii I

Page 89: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

When the source pressure was 120 psig, the Hymatic VAG valve performed asshown in Figure 4.5-7. Suit pressure did not begin to increase until G force hadincreased from 1.4 G to 2 G; after 2 G pressure jumped up rapidly to 1.9 psig at 2.5 Gbefore beginning a smooth linear increase up to 9.8 psig at 9.6 G. Pressure thencontinued to increase more slowly, reaching 10.4 psig at 11 G. During the time that Glevel was maintained at 11 G, the pressure continued to increase slightly, reaching amaximum of 10.5 psig at the end of this period. Pressure began to decreaseimmediately upon deceleration and dropped to 10 psig at 10 G, thence to 9.5 psig at 9G. From 9 G down to 2.5 G, the pressure decreased linearly to a value of 2 psig at 2.5G; pressure then decreased more rapidly to 1.1 psig at 2 G and 0 psig at 1.6 G.

4.5.3.2 High G7-Onset-Rate Conditions

For high Gz-onset testing of the Hymatic VAG anti-G valve,conditions were: a simulated suit volume of 10 liters and high acceleration anddeceleration rates. The pressure delivered by the valve was measured at both theoutput port and inside the simulated suit volume; however, only suit pressures arepresented and discussed here.

The effects of valve angle upon performance of the Hymatic anti-G valve were ex-plored by varying the valve angle (0u, 150, and 300) while maintaining the source pres-sure at the median value for this valve (60 psig). When the valve angle was 00, suitpressure did not begin to increase until G level had increased from baseline (1.4 G) to2.5 G (Fig. 4.5-8). Once pressurization began, suit pressure increased in a smoothlinear fashion up to 10 psig at 11 G; while G level was maintained at- 11 G, the pres-sure increased further to 10.5 psig withjn 1.5 s, then remained at 10.5 psig until afterdeceleration was initiated. Upon deceleration, pressure did not begin to fall until after9.5 G was attained; it fell to 10.2 psig at 9 G, then 9.5 psig at 8 G. After this point, pres-sure decreased linearly with Gz down to 3.8 psig at 2 G; it then dropped more rapidlyto 2.5 psig at 1.4 G (baseline) and after 1 s at baseline G the pressure was 0.4 psig.

When the valve angle was set to 15Q (Fig. 4.5-9), test performance under mediansource pressure was essentially identical to that at a valve angle of 00 (Fig. 4.5-8).

For a 300 valve angle and median source pressure (Fig. 4.5-10), test performancewas also similar to the other two valve angles, although in the region from 6 G to 11 G(3.8 psig to 9.3 psig) the pressure increase became nonlinear. As accelerationreached 11 G, suit pressure was 9.3 psig but jumped to 10.5 psig within 1.5 s. Upondeceleration, suit pressure began to decrease immediately and the rate of decreaseapproached linearity down to the 2-G level (3.4 psig), whereupon the decreasebecame exponential. Suit pressure was 1.8 psig as 1.4 G (baseline) was reached,then fell further to 0.8 psig within 0.5 s, followed by 0.3 psig after another 0.5 s at 1.4 G.

The effects of source pressure upon performance of the Hymatic VAG valve wereexplored by testing under conditions of minimum (30 psig) and maximum (120 psig)source pressure while holding the valve angle constant at 00. (These results can becompared to those obtained under the median source pressure of 60 psig and valveangle of 00, shown in Fig. 4.5-8.) When a minimum source pressure was used under

75

Page 90: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

C +1 ! T H "

-6 i IM E(E11.42 3 4 £ 6 7 8 9 0 1 I'I II II il 0 9J 8 7 6 5 4 3 2 L4L414 G| LEIVELJ

Figure 4.5-6. "Suit" pressures produced during acceleration/deceleration by Hymatic VAG-1 10-022 anti-Gvalve under conditions of: 30-psig source pressure; low G onset; 1 0-liter volume; 00 angle.

1 1 I I, I I

- - - I

I 14; 4 S 7 SI 9 I0 I II II II I0 9 I 7 "6 5 4 31 1.4 L41.4 1 G -,L.r L-4 N . 14 . - ' 'B @4 N '0

Figure 4.5-7. "Suit" pressures produced during acceleration/deceleration by Hymatic VAG-1 10-022 anti-Gvalve under conditions of: 120-psig source pressure: low G onset; 10-liter volume; 0° angle.

76

Page 91: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

____________ I I I

T , ; . I f ]] I

7 i, __/ _ __ j t i

' 'I I _

-' I J. I 1- [ I I I

1 1.42: 4 6 S 9 I I II II I I ' 10 J 7 6 5 4 3 2 1.41. 1.4 , LEVEL

Figure. 4.5-8 "Suit* pressures produced during acceleration/deceleration by Hymatic VAG-1 10-022 anti-Gvalve under conditions of: 60-pSig source pressure; high G onset; 1 O-liter volume; 0° angle.

- . 1 1I~ T I I 1 1I 1--1If

I I I I I I

3 . I/I I \I II.I i iLJ [ I I II Ir I A 1 _ 1 I .. II I II

I L4 2 3 4 5 iS 7 a 9 10 I II I1 II 0 9 S 7 5 $ 4 3 2 1.41.41.4 01 LEVEL

Figure 4.5-9. "Suit" pressures produced during acceleration/deceleration by Hymatic VAG-1 10-022 anti-Gvalve under conditions of: 60-psig source pressure; high G onset: 10-liter volume; 15 angle.

77

Page 92: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

these high G-onset conditions, valve performance was as shown in Figure 4.5-11. Onacceleration, suit pressure did not begin to increase until after 2.5 G, then increasedlinearly up to 2.8 psig at 5.5 G. After this point, the pressure buildup becamenonlinear, measuring 3.8 psig at 7 G, 5.2 psig at 9 G, and 7.4 psig initially at 11 G; after1.5 s at 11 G, however, the pressure jumped to 10 psig and stabilized at this level untilwell after deceleration had started. After deceleration to 8.5 G the pressure began todrop, becoming approximately linear from 7.5 G (9.1 psig) to 2 G (3.7 psig). After the2-G point, the pressure fell more rapidly, measuring 2.5 psig as 1.4 G was reached;within 1 s at 1.4 G the pressure had fallen to 0.3 psig.

At the maximum source pressure (120 psig), pressure in the suit started to buildup when the G-level reached 2-2.5 G (Fig. 4.5-12). The increase in pressure wasconsistently linear as G level increased to 11 G, measuring 2.7 psig at 5 G and 9.8psig at 11 G. After 1.5 s at 11 G, the pressure increased further to 10.5 psig where itremained throughout the interval at 11 G. Upon deceleration, suit pressure startedfalling immediately and followed a gentle convex curve as G level decreased down to2 G; pressures measured 10.1 psig at 10 G, 9.1 psig at 8 G, 7.7 psig at 6 G, and 3.8psig at 2 G. After 2 G the pressure decreased exponentially, reading 1.3 psig asbaseline G (1.4 G) was reached; after 1 s at this level pressure dropped further to 0.2psig.

4.5.4 Phase III - Determination of Comolex Dynamic Response Capability

The response pattern of the Hymatic VAG anti-G valve under SACMconditions was examined by varying the valve angle while holding the sourcepressure at the median value of 60 psig. (Suit volume was always the median value of10 liters unless specified otherwise.) Subsequently, the minimum source pressurewas combined with a maximum suit volume of 14 liters, and the maximum sourcepressure was combined with a minimum suit volume of 6 liters, while the valve anglewas held at 00.

The effects of varying the valve angle (while holding source pressure at 60 psigand suit volume at 10 liters) are shown in Figures 4.5-13 through 4.5-15. At an angleof 00, the pattern shown in Figure 4.5-13 was observed. Under 3 G, pressureincreased to 2.7 psig within 2-3 s and stabilized at this level. The first high-onset rateto 9 G produced a buildup of suit pressure from 2.7 G to 9.9 G within 2-3 s; thepressure then fell slightly to approximately 9.7 psig within 6-7 s. Rapid decelerationfrom 9 G to 4.4 G caused an immediate drop to 4.4 psig within 3 s. Furtherdeceleration to 3 G again yielded a suit pressure of 2.7-2.8 psig. A subsequent slowacceleration to 5 G produced a pressure of 5 psig; a deceleration to 4 G then resultedin a pressure of 3.9-4 psig. Another rapid-onset acceleration to 9 G caused thepressure to increase to 9.9 psig, and then to decrease slightly to 9.7 psig withinapproximately 6 s. Rapid deceleration to 1.5 G was accompanied by a sharp drop inpressure to 2.2 psig, then to 0.1 psig within 5 s. A subsequent rapid acceleration to 8G resulted in an immediate pressure increase to 8.8 psig; the pressure then drifteddown to approximately 8.6 psig within about 8 s. The return to baseline G (1.4 G)produced an essentially linear fall in suit pressure to 0.3 psig (Fig. 4.5-13).

78

Page 93: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

-" I / I I ,,

S3 4 1 1 I I I I- . - - -I I - .J ...I... . . . . - I I i I I I I

I I ____./ I ___

"__i iiJ H j__ __I i i

H 1 I I Ii iF ]l I i ! ! I i i, !

L4Z 3 4 SI 6 7 6 6 IO II It II II I0 6 8I 7 6I 5 4 3l 2I.1,41.44 GEIVJ..

Figure 4.5-10. *Suit" pressures produced during acceleration/deceleration by Hymatic VAG-1 10-022 anti-Gvalve under conditions of: 60-psig source pressure; high G onset; 1 0-iter volume; 300 angle.

, - Th7H I-I I H "IJ I C I )igr 4 "S I I I a Itll lb I \i -G

._. l l ! , I _ I I t i ",, i-i * I I I II II II 0 9 S ,5 \ 4 3 1 1

_ II p,

o N ,' c o • _ ,;,H

Figure 4.5-11. "Suit" pressures produced during acceleration/deceleration by Hymatic VAG-1 10-022 anti-Gvalve under conditions of: 30-psig source pressure; high G onset; 10-liter volume; 00 angle.

79

Page 94: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

~I I 1..........i....iIi I I i ___

I . .... ..-

3,IIzI}*-dL'*-*" 1 1F11 t II l-,i I

i . Vi i i......I.....I I I I I

~fIIZ T 111I/lI........t..1...t..I I .L.I.......ii

S!.42 a 3 4 5 6 7 8 9 10 1I II II I 10 9 S 7 6 5 4 3 L' I.44 (4 LEVEL

oi , ,~ , i -It .I W me (SEC11

Figure 4.5-12. "Suit" pressures produced during acceleration/deceleration by Hymatic VAG-1 10-022 anti-Gvalve under conditions of: 120-psig source pressure; high G onset; 10-liter volume; 00 angle.

o~ ~ ~ ~~7 111 ...!!I i I 1

II I I i I I I I I I t

a 10 20 30 40 so SQ 70 so 90 100 110 lime SEC

Figure 4.5-13. "Suit" pressures produced during SACM by Hymatic VAG-1 10-022 anti-G valve underconditions of: 60-psig source pressure; 1 0-liter volume; 00 angle.

80

Page 95: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

When the valve angle was 150, the SACM profile observed (Fig. 4.5-14) wasalmost identical to that for the 00 angle (Fig. 4.5-13). The only apparent differencebetween the two patterns was that when the angle was 150, the rapid decelerationfrom 9 G to 4.4 G caused the suit pressure to fall slightly further--i.e., to 4.1 psig vs. 4.5psig at 00.

For a valve angle of 300 (Fig. 4.5-15), the SACM profile was also similar to that forangles of 00 and 150; however, suit pressures were generally slightly lower at 30 ° .

When the maximum suit volume (14 liters) was used with the minimum sourcepressure (30 psig) and an angle of 00, the SACM suit pressurization response (Fig.4.5-16) was essentially identical to the response observed for conditions of 60-psigsource pressure, 10-liter suit volume, and 00 angle shown in Figure 4.5-13. Theconditions of maximum source pressure (120 psig), minimum suit volume (6 liters),and valve angle of 0° , shown in Figure 4.5-17, also produced a response patternidentical to that of Figures 4.5-13 and 4.5-16.

ii~ -5 i I II IIIII- 11 i

I F7 I

-, _L_ P- 4 LO 4 , LEL

I o 20 30 40 56 6o 70 0 o ioo io n T i(sic)

Figure 4.5-14. "Suit" pressures produced during SACM by Hymatic VAG-110-022 anti-G valve underconditions of: 60-psig source pressure; 1 0-liter volume; 150 angle.

81

Page 96: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

L W 3. 4,r e. E. 1 1

S4.5 -. 1 r

codtin of 60pi sorc prsue I -i oum;30age-- -44_ 0 1.4 ,LAVI.

0 10 0 30 40 0 6o 70 80 0 100 110 1me (ite)

Figure 4.5-15. "Suit" pressures produced during SACM by Hymatic VAG-110-022 anti-G valve underconditions of: 60-psig source pressure; 10-liter volume; 300 angle.

182zz - HzL:l-z

Figure 4.5-16. "Suit" pressures produced during SACM by Hymatic VAG-1 1-022 anti-G valve underconditions of: 30-psig source pressure; 14-liter volume; 00 angle.

82

Page 97: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

i J ,i -l ! - -J!l i i t tI

I

5.50 9. ..

'AO Ao kL44.4 5 p :4 0 LA0 i0 20 30' 40 50 60 i0 i0 go 100 110 TIME SEC)

Figure 4.5-17. "Suit" pressures produced during SACM by Hymatic VAG-110-022 anti-G valve underconditions of: 120-psig source pressure; 6-liter volume; 00 angle.

4.6 AAMRL Anti-G Valve

4.6.1 Description

The Armstrong Aerospace Medical Research Laboratory (AAMRL), or"Bang Bang," anti-G valve (SN 002) (Fig. 4.6-1) is a modified ALAR High Flow (SN107, see section 4.3) valve. A solenoid plunger was added to the ALAR High Flowvalve and the "Press to Test" button was relocated. The modifications were made atAAMRL, Wright-Patterson Air Force Base, Ohio.

In operation, this valve uses an electronically controlled solenoid to drive the anti-G suit pressure to maximum when the level of acceleration exceeds both +2 G and anonset rate of 2 G s-1. The maximum pressure is maintained for 2.5 s before the valvereverts back to normal operation. If the "trigger" conditions are encountered again, thesolenoid will be activated again.

4.6.2 Phase I - Determination of Maximum Flow Capacity

The maximum flow capacity of the AAMRL anti-G valve was measured at:its median pressure of 150 psig; 0.5 G s-1 ; valve angle of 00; and with electronicallycontrolled solenoid both on and off. The flow through the valve was the same for bothcases; this can be seen from Figures 4.6-2 and 4.6-3. Flow was initiated between 1.7G and 2 G, reaching 8.5 SCFM at 3 G; after this point flow increased linearly up to 7.5G, then began to plateau at a maximum level of 21-22 SCFM. Maximum flow wasmaintained throughout the time at 11 G and until 8.5 G was reached duringdeceleration. The decrease in flow was approximately linear down to 3 G (8 SCFM)whereupon it began to fall rapidly to 0 SCFM at 1.5 G (Figs. 4.6-2 and 4.6-3).

83

Page 98: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

Figure 4.6-1. MAMRL anti-G valve ("Bang Bang").

84

Page 99: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

22li ..... I 1 I 1 11 I i I- 2 0 I J..........i.....I........ I I I I I I i

! I ! tii I

! I I I Xk I t i II../.i~i.. t I i I I ...... i.....)\...i

I . ] II I I I ] I I I \ I I ,___ 11111111 !! t! I I x1i ti i

I.

1 ! ......... I/... I..... I...... I......J...... i......!...... I I I I I\ I I L21 I / i I I ...J..1..J........J...........'.. J........J........ I\ I

S, . . . . . . . .~ a ~ d0-- _

11.4 2 3 4 6 £ 7 S 9 60 I I 1 I Itl I i 9 6 " 6 5 4 3 ZL4L41.4 ( L. EE i

Figure 4.6-2. Maximum flow through AAMRL anti-G valve with pressure off.

H = Is,! € 1 I I 1 i* iH " i. IK v /z zztzL IJ F - L - I - L

111 11 1 1 1 1- ' - O9 321..44GoLVS i I I I I I IFI I I I I I i 1 I

....Jt II i I j I I 1 l I \ i I __

I i I I tIdI i I I

Figure 4.6-3. Maximum flow through AAMRL anti-G valve with pressure on.

85

Page 100: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

4.6.3 Phase II - Determination of Dynamic Resoonse Capability

The dynamic response capability of the AAMRL anti-G valve wasdetermined as described in Testing Procedures (section 3.2.2). Pressure delivered bythe valve was measured at the output port and inside the simulated anti-G suit volume."Output pressure" refers to the pressure within the simulated suit volume, unlessotherwise noted. This valve was tested under conditions of both high G-onset and lowG-onset as described in Testing Procedures, section 3.2.2.

4.6.3.1 Low G7-Onset-Rate Conditions

The 0.1 Gz s- 1 trapezoidal profile was used for all low-onsettests.

Under conditions of low Gz onset, median source pressure (150 psig), mediansuit volume (10 liters), and valve angle of 00, the valve performed as shown in Figure4.6-4. Suit pressurization did not begin until 2 G was reached; pressure thenincreased linearly to a value of 7.5 psig at 6.5 G. The rate of pressure increase withincreasing G level then began to curve to a maximum level of 10.5-10.8 psig from 10-11 G. Maximum pressure was maintained throughout the time at 11 G; upondeceleration, pressure began to decrease at about 9.5 G and decreased linearly frcm10.4 psig at 9 G to 0.4 psig at 1.5 G.

When the valve angle was changed to 150 while all other conditions were heldconstant, the valve performed as shown in Figure 4.6-5. The suit pressurization profileremained identical in all respects except that the maximum pressure attained at 150was slightly lower-10.2 psig to 10.4 psig--than at 00 (10.5 psig to 10.8 psig, Fig. 4.6-4).Further increasing the valve angle to 300, however, produced discontinuities in the suitpressurization profile, shown in Figure 4.6-6. In this case, suit pressurization was notinitiated until 2.5 G and several breaks in pressure buildup were observed; maximumpressure (10.0 psig) was attained at 10.5 G and maintained until 10.5 G was reachedupon deceleration. The decrease in suit pressure was again somewhat erratic, reach-ing 0 psig after 1 s at 1.4 G.

When the AAMRL anti-G valve was tested under conditions of low G onset,minimum source pressure (30 psig), median suit volume (10 liters), and 00 valveangle, the results shown in Figure 4.6-7 were observed. The valve opened between1.5 G and 2 G to produce a suit pressure of 0.25 psig at 2 G. Pressure increase waslinear up to 7.4 psig at 6.5 G; from this point the slope of the line decreased slightly toreach 10.4 psig at 9.5 G. The maximum pressure of 10.8 psig was attained at 11 Gand did not change until deceleration to 9.5 G; the fall in suit pressure was thenessentially linear down to 0.35-0.31 psig at 1.4 G.

Under conditions of low G-onset, maximum source pressure (300 psig), mediansuit volume and 00 valve angle, the suit pressurization profile--which is shown inFigure 4.6-8--was essentially identical to that produced by minimum source pressureand 00 angle (Fig. 4.6-7).

86

Page 101: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

- I _____,

____ I I il I I It I l ii I

i~~~~~ ~~ i/7i ' I'. \ il i

f1 - ~ -1....~.....1..I.........II I I i i i I -i- -

....I.....i I . I I iI N i l l

,L4 Z 3 4 5 6 1, ''I j, [, 'L ', , ,'o 2, .4 L41-, ., G, , ___ I

Figure 4.6-4. "Suit" pressures produced durng acceleration/deceleration by AAMRL anti-G valve under

conditions of: 1 50-psig source pressure; low G onset; 10-Niter volume; 00 angle.IJ -I - - -_-_

! I i , ! I I I I __

j I /!iiF ! !I I I i"T I I __________________

_____ I if',

1I.42 3 4 3 6? 6910 11 11 1 ft tog S a76a 5 4 3 2 I.4A.4 .G, LFVc..-77 *1 Nd ---- c s

a-d -V x S N - , * 1*_ i

Figure 4.6-5. "Suit" pressures produced during acceleration/deceleration by AAMRL anti-G valve underconditions of: 1 50-psig source pressure; low G onset; I 0-liter volume; 1S angle.

87

Page 102: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

I1 I i ~ " + I t t i i i ____

-' ! I I I I I I I I I I I I I I i

-0 i I i IiLI

='2 ,,,r CLLoLLJi IL TiiM S_' IIi/I !IiI!Ii .1 ! II I " I I i _

II I I l ill i I II I I '~ I i

-. . . . 0 = o 0 I P"eoo

Figure 4.6-6. "Suit" pressures produced during acceleration/deceleration by AAMRL anti-G valve underconditions of: 1 50-psig source pressure;- low G onset; 1 0-iter volume; 300 angle.

,, ..i I.JiI i,-1 EE !!i EI zziz ijz pt r-, b~ I I I _ __ ]...i.I I i I I

,! I I i / I t 1 ' t! 1l ! 1 1I

4__ I I-7 -/ i i I ...... !..... I.....1.....J........! .... i I t I\ I I

3 - 1

I I I ' I I I i IEV 't1i .LY i !i+- ii}- I il i__i '______1 1. 42 3 4 5 6 7 6 9 t0 II II I I' 10 1 1 7 1 1 1 9 21.41.4L4 0, LEVL

!~ A ;: OW 0 A 0N

Figure 4.6-7. "Suit" pressures produced during acceleration/deceleration by AAMRL anti-G valve underconditions of: 30-psig source pressure; low G onset; 1 0-iter volume; 00 angle.

88

Page 103: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

4.6.3.2 Hiah G7-Onset-Rate Conditions

The high G-onset conditions for the AAMRL anti-G valve were4.42 G s-1 ; and source pressures of 30, 150, and 300 psig.

For conditions of high G onset, median source pressure (150 psig), 10-liter suitvolume, and 00 valve angle, this valve produced the suit pressurization profile shownin Figure 4.6-9. During acceleration the valve opened quickly, producing a suitpressure of 0.34 psig at baseline G (1.4 G). Pressure then increased smoothly in atriphasic pattern; the slope was linear from: 1.4 G to 3 G (4 psig); 3 G to 5 G (7.2 psig);and 5.5 G to 11 G (10.8 psig). After reaching 11 G, pressure continued to increaselinearly throughout the time at maximum G level--reaching 11.6 psig after 0.6 s. Upondeceleration, suit pressure began to decrease at 10 G, and continued to decreaselinearly down to 3.6 psig at 2.5 G; the rate of pressure drop increased resulting in 2psig as baseline G (1.4 G) was reached and a further drop to 0.4 psig within 0.6 s.

When the valve angle was set to 150 while all other conditions were unchanged,this valve performed as shown in Figure 4.6-10. There was an initial lag inpressurization so that suit pressure was 0 psig at 1.4 G but 0.3 psig at 2 G. Pressurethen increased not quite as rapidly--but somewhat smoother-than when the anglewas 00 (Fig. 4.6-9). As 11 G was first attained, suit pressure was 10.4 psig but jumpedto the maximum value of 11.2 psig within 0.5 s; it was stable at this level while at 11 G.During deceleration, pressure decreased in a smooth linear manner from 10 G (11psig) to 2 G (2.8 psig); it then dropped more rapidly to measure 1.7 psig upon reaching1.4 G and stabilized at 0.2 psig after 1.2 s. Changing the valve angle to 300 producedthe suit pressurization profile shown in Figure 4.6-11. This profile is essentiallyidentical to that observed for a valve angle of 150 (Fig. 4.6-10).

With the minimum source pressure (30 psig), 10-liter suit volume, and 00 angle,data plotted in Figure 4.6-12 were obtained. Again there was an initial lag with suitpressure measuring only 0.14 psig at 2 G. From this point, pressure buildup followeda gentle curve which was really two line segments of decreasing slopes. Uponreaching 11 G, pressure was 10.4 psig but then jumped to 11.1 psig within 5 s andstabilized at this maximum level until after deceleration had started. Duringdeceleration, suit pressure decreased smoothly and linearly from 11 psig at 10 G to1.9 psig as 1.4 G was attained, then dropped to 0.4 psig within 1 s. By comparisonwith the suit pressurization profile produced at median source pressure (Fig. 4.6-9),minimum source pressure (Fig. 4.6-12) resulted in a somewhat slower, more linearincrease in pressure which stabilized within 0.5 s at maximum G level instead ofcontinuing to increase as in Figure 4.6-9.

Maximum source pressure (300 psig), 10-liter suit volume, and 00 valve angleyielded the plot shown in Figure 4.6-13. As for lower source pressures, the increase insuit pressure was multiphasic--with line segments of at least four different slopes.Maximum pressure (11.5 psig) was attained only after 0.6 s at 11 G. The drop in suitpressure during deceleration was essentially the same as for lower source pressures(Figs. 4.6-9 and 4.6-12).

89

Page 104: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

10-III I I I I I I I i I

I I I iL Li.....Jr I I I

I , i ._[I-i- I I J F I TI I i2 3 a 5 a 7 a . 0.1.1.1. 1 1.1. 11 1 a 1. i G -VC

4 .I d I! 0I r 0 I0 eI I I I; T\No! K I3 / I Ii I i I X

.......... I / ~i, I I I I I I I I I I I I I _ I I,, ,

Figure 4.6-8. "Suit" pressures produced during acceleration/deceleration by AAMRL anti-G valveunder conditions of: 300-psig source pressure; low G onset; 1 0-liter volume; 0°' angle.

t ........j iI I 7 ' I i \I ___________

I, L4 2 3_01 it t 1_ 2 L1.4 14 G2LA L

,, ~' d! Ct 1 ri MW iIIII \

Figure 4.6-9. "Suit" pressures produced during acceleration/deceleration by AAMRL anti-G valveunder conditions of: 150-psig source pressure; high G onset; 1 0-liter volume; 00 angle.

90

Hi ...........in.ni.lillli.illllill.... J .IIJ W L L. . . .. . . .

Page 105: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

I K I I I I I I I I I I I II IIN

1 1 I I I I

I I I I I I I I I m 1: I I II If 11 iIJ !' , -

6 -II!I!Ij _ I i)!I I I \

0 JJ[2 3 4 j __ 7 ] I 10 11 11 11 1-09 9 4 5 4 3 2 144 G EE

;ii.Y'lI I I lIi i I I I I i I I II i'l i i .t 1 i14;i l ,..'l .4 3 4 6 7 8 9I 10 II II II I1 I0 9I tS "t 6 5 4 3 14. 4

Cd~~IM CdS -CdQfCd" . .i C d ~ d C - Nd q. Ul * O liI l

Figure 4.6-10. "Suit" pressures produced during acceleration/deceleration by AAMRL anti-G valveunder conditions of: 150-psig source pressure; high G onset; 10-iter volume; 150 angle.

_I I flz! ztz ,- 7 .j I I I i I L _____

I ,____ 1111 I

I. L I/" I I I 'I I I I , _ I i ______

H i I !i_ I I ' I I3 ...... I I I I~ ~ - - L - - I t - -

I L42 3 4 1 6 " 0 9 ,0 11 1 [1i11 109 6 a 5' 3 2 1.4.4L4G, L. ICd~~nm CdyQ1(SECI

Figure 4.6-11. "Suit" pressures produced during acceleration/deceleration by AAMRL anti-G valveunder conditions of: 1 50-psig source pressure; high G onset; 10-liter volume; 30' angle.

91

Page 106: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

5- IX i I I I I I i

I I , r I I I I I I i i I II I I i4F 5 6 r a 9 1 1 1 1 Iit

Figre 4. . pciio/clai I Iti lI F ! I I 1!s" i i .( 1' t i I k I ...i ...i t I I ___i4..I.[.. .11 ! I I t I I I i I t I' I I . !i.......1........! I 1 1 I i : t I I i -I. i

1 1.42 3 4 5 6 7 9 10 1 I II 11 I 0 9 6 " 6 5 434 4 .4 3 L

. l O C TI Nld (SE C . .

Figure 4.6-12. "Suit" pressures produced during acceleration/deceleration by AAMRL anti-G valveunder conditions of: 30-psig source pressure; high G onset; 1 0-titer volume; 00 angle.

_ 'I I I _I I~ i III i t I_ , L .... I.....I..........1....I1 ..... L......-...."., I I Ii I1 I I I __ _!_, ...........'....L..J..i....... I I ~ I I ______

, ! i h t ! f f I I Ii _____________

I i( it ii I 111111 III _________

_________ ! _____,__________

1142 3.. .. .. 4 4 "7 S 9 IC II II 11 II 0 S S "7 6I 4, 3 2, 4 44 Gl I,,EVEL, .

o......................................... .... 7 (S" I

Figure 4.6-13. "Suit" pressures produced during acceleration/deceleration by AAMRL anti-G valveunder conditions of: 300-psig source pressure; high G onset: 1 0-liter volume; Q ° angle.

92

Page 107: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

4.6.4 Phase III - Determination of Comolex Dynamic Response Capability

Phase III testing under SACM conditions was performed as described inTesting Procedures (section 3.2.3). Initially, median values of source pressure andsuit volume were selected and the valve angle varied; subsequently, valve angle wasmaintained at 00 and the combinations of minimum source pressure with maximumsuit volume, and maximum source pressure with minimum suit volume wereevaluated.

For median source pressure (150 psig), median suit volume (10 liters), and avalve angle of 00, the SACM profile shown in Figure 4.6-14 was observed. During theinitial penod at 3 G the suit pressure spiked to 11.1 psig before settling back to 2.7-2.4psig within 5 s. A rapid onset acceleration to 9 G caused the pressure to rise rapidly to11.4 psig and then stabilize at 10.1 psig" deceleration to 4.4 G produced a rapid dropin suit pressure to 4.4 psig. Decelerating further to 3 G dropped the pressure to 2.4-2.5psig. A slow onset acceleration to 5 G resulted in the pressure building up slowly to5.1-5.2 psig, and a subsequent slowing to 4 G dropped the pressure to 3.9-3.8 psig.Another rapid acceleration to 9 G caused the pressure to increase rapidly to 11.4 psigthen stabilize at 10.1 psig, while rapid deceleration to 1.5 dropped the suit pressure to0.4 psig immediately. The last rapid acceleration to 8 G resulted in the pressure risingto 11.2 psig then stabilizing at 8.8-8.9 psig. During the slower deceleration to baselineG (1.4 G), suit pressure dropped linearly to 0.3 psig.

When the valve angle was changed to 15" while all other conditions were heldconstant, the response pattern seen in Figure 4.6-15 was observed. This suitpressurization profile was similar to that observed for a valve angle of 0 (seen in Fig.4.6-14); the differences were in the plateau pressures produced by various stationaryG levels, which were consistently lower at the 150 valve angle (Fig. 4.6-15) than at the00 angle (Fig. 4.6-14).

When the valve angle was 300, the suit pressurization profile (Fig. 4.6-16) wasalso similar to those observed for the 00 and 150 angles (Fig. 4.6-14), in that the initialspikes in suit pressure were observed upon rapid acceleration; however, thesubsequent plateau pressure levels (Fig. 4.6-16) were again lower than for even the150 angle.

When the minimum source pressure was combined with the maximum suitvolume (00 valve angle), the profile of Figure 4.6-17 was produced. Comparing theseresults to those obtained when median values of both source pressure and suitvolume were used (Fig. 4.6-14), it can be seen that the conditions of minimum sourcepressure/maximum volume caused the initial spikes in suit pressure to be dampedconsiderably, while the plateau suit pressures at stationary G levels were the same asfor the median-value conditions (Fig. 4.6-14).

When maximum source pressure (300 psig) was combined with minimum suitvolume (6 liters), suit pressurization was as shown in Figure 4.6-18. There was littledifference between this pressurization profile and that observed for the median sourcepressure/median suit volume combination (Fig. 4.6-14); the only apparent differencewas that the plateau suit pressure after the final rapid onset acceleration to 8 G wasapproximately 1 psig higher at maximum source pressure/minimum suit volumeconditions (Fig. 4.6-18) than at the median-value conditions.

93

Page 108: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

F1I. I I I I _ __

o hin I t'-L liiiI II 1 i

-; I '' I 'll ~i I 4IIl

o 0 .0 '0 0 G2 LEVEooLo____

0 10 20 30 40 s0 w0 T0 60 90 100 110 rime (SEC)

Figure 4.6-14. *Suit" pressures produced during SACM by AAMRL anti-G valve under conditions of:150-psig source pressure: 10-liter volume; 0° angle.

: I !1 ! t ! ! I I II I _______ _____ ____

1.....ILL.....L...../i 1 I t I - .

o1' I J I I I : ,

4 1 _.

0 0 20 30 40 5 60 70 50 ;0 0'0 In 1,C 11KE MEC)

Figure 4.6-15. "Suit" pressures produced during SACM by AAMRL anti-G valve under conditions of:1 50-psig source pressure; 10-liter volume; 150 angle.

94

Page 109: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

_ _ _ _ _ _ I , i Ii I I I iB ' l L! ! -1 ,_ - - - -- - - - !.IiI l I __ _ _ - i *I I i I

5 111 I I I ~ I Ii I I I !I I

i11 I I l!! IZ'.......L....J __\____

/A

,'lhVT t h I [ I/ I I iL.J} II Ij\ I ,,I

LILL4 0 -_L4 LO 400 T"S1 1.4 ~ L'

1 0 .1 0 -0-o 4 50 60 70 so 50 00 ( 10 Ito-me (sac)

Figure 4.6-16. "Suit" pressures produced during SACM by AAMRL anti-G valve under conditions of:150-psig source pressure; 10-liter volume; 300 angle.

S !L I I I I I I ! I I i I i I i

, II I ......L.. . I. . 1~. I1 1 I I I i

, I i i i i j lI\ i

7 \ /I

I ___ _____ I I

:1- I/Il 'Ih IAVIM \ _____ _

H 0 10 20 30 '0 s0 60 710 S~o 90 10 It nc0s

Figure 4.6-17. "Suit pressures produced during SACM by AAMRL anti-G valve under conditions of: 30-psig source pressure; 14-liter volume; 0o angle.

95

Page 110: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

__ 0 1 0 0 I I I I I I TIM I'

L _ - LI .J I ! i

* source pressure; I I _

5. DISCUSIONI- IIAND II I 11 7tE IF Ft

_I - "- ] ! ! t i i IIJ! 1 l I I

.I I II I , I I J . ~ ~ f I I J

Figure 4.6-18. "Suit" pressures produced during SACM by AAMRL anti-G valve under conditions of:300-psig source pressure; 6-liter volume; 0° angle.

5. DISCUSSION AND CONCLUSIONS

The prototype anti-G valves evaluated in this task were compared with thestandard ALAR 8400A valve, and with each other, on the basis of both static anddynamic response characteristics.

Table 5.0-1 compares the maximum airflow capacity (in SCFM) under static testconditions (see Test Procedures, section 3.2.1) for the various valves. The G-onsetrates were 0.5 G s-1, and source pressures were the appropriate median values foreach valve; these source pressures are specified in the Results section for each valve.

The ALAR 8400A produced a maximum airflow of 11.1 SCFM at 8 G and held thislevel without varying throughout the rest of the test. All of the five prototype valveswere also producing maximum, or close to maximum, airflows at 8 G. The GarrettElectronic/Pneumatic (servo) valve reached a maximum of 27.8 SCFM at 5.5 G; it thendropped slightly but maintained 91-93% of maximum throughout the remainder of thetest. The maximum airflows produced by the prototype valves were all considerablyhigher than that produced by the ALAR 8400A; they ranged from 18 SCFM to 28SCFM (Table 5.0-1).

The dynamic response characteristics (measured as described in TestingProcedures, sections 3.2.2 and 3.2.3) of the valves are compared in Tables 5.0-2through 5.0-4.

96

Page 111: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

TABLE 5.0-1. MAXIMUM AIRFLOW CAPACITY (SCFM) AS A FUNCTION OF

ACCELERATION (Gz) FOR ALL ANTI-G VALVES EVALUATED

(Conditions compared: 0.5 G s- 1 onset rate, median source pressure*)

Gz ValveLevel ALAR ALAR ALAR Garrett Hymatic AAMRL

8400A HFRP HF Servo VAG-110 "Bang Bang"

1.4 0.0 3.3 0.9 8.0 0.4 0.02.0 0.1 3.3 1.5 12.3 0.76 1.62.5 1.2 5.4 4.9 15.0 7.6 5.73.0 2.6 6.5 8.6 17.5 8.9 8.53.5 5.2 9.4 10.5 19.8 10.2 10.04.0 7.0 11.3 12.0 22.0 11.3 11.34.5 7.6 12.7 12.9 24.1 12.5 12.55.0 9.1 14.2 14.1 26.2 13.7 13.95.5 10.0 15.6 15.3 27.8 14.7 15.46.0 10.4 17.0 16.5 26.7 15.6 16.86.5 10.7 18.1 17.9 25.7 16.7 17.97.0 10.9 19.4 19.1 25.6 17.8 19.07.5 11.0 20.8 20.1 25.6 18.0 20.18.0 11.1 21.8 20.7 25.5 18.0 20.78.5 11.1 22.4 21.1 25.7 18.0 21.19.0 11.1 22.6 21.3 25.6 18.0 21.49.5 11.1 22.6 21.4 25.7 18.0 21.5

10.0 11.1 22.6 21.5 25.6 18.0 21.510.5 11.1 22.6 21.6 25.7 18.0 21.5

At maximum Gz level

11.0 11.1 22.6 21.6 25.8 18.0 21.6

11.0 11.1 22.6 21.8 25.9 18.2 21.7

* Median source pressure varied with valve; see text.

97

Page 112: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

The time in seconds required by each valve to produce a given suit pressure isshown in Table 5.0-2. All prototype valves produced any given suit pressure fasterthan the ALAR 8400A did. The ALAR 8400A required 3.60 s to produce a suitpressure of 9.0 psig; the "Bang-Bang" valve produced the same pressure in 41% ofthis time; the Garrett in 47%; the ALAR High Flow in 50%; the Hymatic in 51%; and theALAR HFRP in 54/.

In comparing the performances of the ALAR HFRP and ALAR HF valves (Table5.0-2), it will be noted that the ALAR HF pressurized the suit more quickly than theALAR HFRP. This result appears to be in conflict with previous testing results, and withcommon sense, for two reasons. First, in the configurations tested, the HF valveopened (started the pressurization schedule) at 1.5 Gz, while the HFRP opened at 2.0Gz; this allowed the HF valve a 0.5 Gz lead on the HFRP. Second, the HFRP modeltested was one of the later production versions, and had undergone severalmodifications to add damping in an effort to eliminate pressure oscillation ("chatter")during Gz onset and offset. These modifications limited the flow of this valve between2-3 Gz (Fig. 4.2-3), and allowed the HF valve, which was not limited in this region, tolead the HFRP valve. This combination of factors negated the inherent advantage ofthe HFRP valve.

Under low-onset conditions (Table 5.0-3), the ALAR 8400A activated at 2.5 G,whereas two of the prototype valves activated at 1.4 G, two others at 2 G, and theremaining valve activated at 2.5 G also, but with a pressure considerably higher thanthe ALAR 8400A. The ALAR 8400A reached a maximum suit pressure of 7.0 psig at7.5 G; this pressure would be equivalent to an inflation schedule of 0.93 psi G- 1. TheGarrett valve also reached a maximum pressure of 10.3 psig at this same G level (7.5G), approximating an inflation schedule of 1.37 psi G-1 . All other prototype valves didnot reach their maximum pressures until the maximum G level (11 G) was attained,and their maximum pressure levels indicated inflation schedules less than 1 psi G-1 forall these valves.

Under high-onset conditions, all prototype valves activated earlier than the ALAR8400A did (Table 5.0-4). The ALAR 8400A activated at 3 G (0.2 psig), whereas theprototype valves exhibited suit pressures of 0.6 - 4.0 psig at 3 G. The ALAR 8400Aproduced a maximum suit pressure of 9.6 psig at the end of the time at 11 G; at thestart of the 11 -G period, pressur,- was 6. 0 psig, indicating an inflation schedule of 0.55psi G- 1. All of the prototype valves tested also yielded maximum suit pressures at theend of the 11-G period; however, the pressures upon first attaining 11 G were allhigher for these valves than for the ALAR 8400A, and the apparent inflation schedulesvaried from 0.84 to 1.05 psi G- 1. The ALAR 8400A valve at 8 G, yielded 35.5% of itsmaximum pressure of 9.6 psig (attained at end of 11 -G time); as 11 G was initiallyattained, suit pressure was 64.5% of maximum. By contrast, the Garrett valveproduced 91 % of its maximum pressure of 9.3 psig at 8 3; 97% at 9.5 G; and 99% at11 G initially, while the maximum pressure was again observed at the end of the 11 -Gperiod. The other prototype valves produced pressures equivalent to 54-79% ofmaximum values at 8 G; and 89-95% of maximum values upon reaching 11 G.

98

Page 113: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

TABLE 5.0-2. SECONDS REQUIRED TO ATTAIN VARIOUS SUIT VOLUMEPRESSURES UNDER HIGH-ONSET CONDITIONS

(Conditions compared: high Gz-onset rate,median source pressure*, median volume [10 liters])

Suit ValvePressure ALAR ALAR ALAR Garrett Hymatic AAMRL(psig) 8400A HFRP HF Servo VAG-110 "Bang Bang"

0.0 0.00 ..... 0.00 --- 0.00 0.000.2 .. .0.00 . .. .. .. .0.6 ..... .. . 0.001.0 0.89 0.63 0.72 0.25 0.73 0.282.0 1.14 0.85 0.91 0.50 0.88 0.413.0 1.42 1.03 1.05 0.67 1.02 0.534.0 1.66 1.20 1.19 0.80 1.16 0.665.0 1.87 1.36 1.31 0.92 1.30 0.776.0 2.13 1.49 1.43 1.04 1.44 0.907.0 2.39 1.62 .1.55 1.17 1.56 1.038.0 2.74 1.75 1.66 1.35 1.69 1.209.0 3.60 1.93 1.79 1.69 1.85 1.469.3 ....... 2.48 -.

9.4 5.12 .. ...---10.0 - 2.22 1.93 -- 2.14 1.7510.5 -... 3.60 -11.0 - 4.00 2.12 - -- 2.3911.1 -- 4.8 ........11.3 .. .... 2.6012.0 . ... 2.55... .12.4 . .. .. 4.50 . .

*Median source pressure varied with valve; see text.

99

Page 114: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

TABLE 5.0-3. SUIT VOLUME PRESSURE (PSIG) PRODUCED UNDER LOW-ONSET CONDITIONS AS FUNCTION OF ACCELERATION (Gz)

(Conditions compared: 0.5 G s-1 onset rate,median source pressure', median volume [10 liters])

Gz ValveLevel ALAR ALAR ALAR Garrett Hymatic AAMRL

8400A HFRP HF Servo VAG-1 10 "Bang Bang"

1.4 0. 0.4 0. 0.8 0. 0.2.0 0. 0.5 0.2 1.7 0. 0.32.5 0.2 1.2 0.8 2.4 1.9 1.13.0 0.7 2.0 1.7 3.3 2.5 2.03.5 1.6 2.6 2.4 4.1 3.1 2.84.0 2.3 3.3 3.1 4.9 3.6 3.64.5 3.0 4.0 3.8 5.7 4.2 4.45.0 3.7 4.6 4.6 6.5 4.7 5.25.5 4.4 5.4 5.3 7.3 5.3 6.06.0 5.1 6.0 6.0 8.1 5.9 6.86.5 5.8 6.8 6.7 8.9 6.4 7.57.0 6.4 7.4 7.3 9.6 7.0 8.17.5 7.0 8.0 8.0 10.3 7.5 8.58.0 7.1 8.5 8.5 10.3 8.0 9.28.5 7.1 9.0 8.9 10.3 8.7 9.69.0 7.1 9.4 8.4 10.3 9.2 10.09.5 7.1 9.8 9.7 10.3 9.8 10.3

10.0 7.1 10.2 10.0 10.3 9.9 10.510.5 7.1 10.4 10.3 10.3 10.2 10.5

At maximum Gz level

11.0 7.1 10.8 10.6 10.3 10.4 10.7

Final

11.0 7.1 10.9 10.7 10.3 10.5 10.8

'Median source pressure varied with valve; see text.

100

Page 115: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

TABLE 5.0-4. SUIT VOLUME PRESSURE (PSIG) PRODUCED UNDER HIGH-ONSET CONDITIONS AS FUNCTION OF ACCELERATION (Gz)

(Conditions compared: high Gz-onset rate,median source pressure*, median volume [10 liters])

Gz ValveLevel ALAR ALAR ALAR Garrett Hymatic AAMRL

8400A HFRP HF Servo VAG-1 10 "Bang Bang"

1.4 0. 0.2 0. 0.6 0. 0.2.0 0. 0.3 0.2 1.3 0. 1.72.5 0. 0.6 0.4 2.1 0.1 2.93.0 0.2 1.1 0.8 2.8 0.6 4.03.5 0.5 1.5 1.3 3.6 1.2 5.04.0 0.9 2.0 1.8 4.4 1.9 5.84.5 1.3 2.5 2.4 5.1 2.5 6.65.0 1.7 3.0 3.0 5.8 3.0 7.25.5 1.9 3.5 3.6 6.5 3.7 7.86.0 2.3 4.0 4.3 7.1 4.2 8.26.5 2.5 4.5 4.9 7.6 4.8 8.57.0 2.8 5.0 5.6 8.0 5.3 8.87.5 3.0 5.5 6.1 8.3 5.9 9.08.0 3.3 6.0 6.9 8.5 6.5 9.28.5 3.6 6.6 7.5 8.7 7.1 9.59.0 4.0 7.2 8.1 8.9 7.7 9.79.5 4.4 7.8 8.8 9.0 8.3 10.0

10.0 4.8 8.4 9.4 9.1 8.9 10.310.5 5.3 9.0 10.2 9.2 9.4 10.6

At maximum Gz level

11.0 6.0 9.9 11.5 9.2 10.0 10.8

Final

11.0 9.6 11.1 12.4 9.3 10.5 11.6

*Median source pressure varied with valve; see text.

101

Page 116: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

6. REFERENCES

1. Thompson, R. W., C. E. Galvan, P. E. Love, and A. G. Krueger, Procedural tests foranti-G valves. SAM-TR-79-30, Vol. I, Dec 1979.

2. Thompson, R. W., C. E. Galvan, P. E. Love, and A. G. Krueger, Procedural tests foranti-G valves. SAM-TR-79-31, Vol 1, Dec 1979.

102

Page 117: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

ABBREVIATIONS. ACRONYMS. AND SYMBOLS

AAMRL Armstrong Aerospace Medical Research LaboratoryACM Aerial Combat ManeuverAWG American wire gaugeDC direct currentFv airflowG units of GzGz acceleration along the Z axis (head-to-foot)Hz Hertz (cycles per second)kHz kilohertz (1000 cycles per second)LSE Life-support equipment

p-p peak-to-peakPr final pressure of known volume in psig

PS source pressurePo valve output pressure (psig)

psi pounds per square inchpsid pounds per square inch differentialpsig pounds per square inch gauge

Pv suit pressure (psig)RDT&E Research, Development, Test, and Evaluation

SACM Simulated Aerial Combat. ManeuverSCF Standard cubic feetSCFM Standard cubic feet per minuteT&E Test and Evaluation

USAFSAM U.S. Air Force School of Aerospace MedicineVo known volume in cubic inches (or liters)VAC volts alternating currentV$ unknown suit volume in cubic inches (or liters)VDC volts direct currentXDCR transducer

103

Page 118: LIFE SUPPORT SYSTEMS: Part 4. Engineering Test and ... · NOTICES This final report was submitted by the Life Sciences Division, Technology Incorpo-rated, 406 Breesport, San Antonio,

[How to order Appendixes A-F]

RE: The USAF School of Aerospace Medicine's Technical Report Series onResearch and Development of Anti-G Life Su1oport Systems:

Part 4. Engineering Test and Evaluation of Six Anti-G Valves.(USAFSAM-TR-86-36-PT-4)

APPENDIX A: ALAR 8400A Anti-G Valve Experimental DataAPPENDIX B: ALAR High-Flow Ready-Pressure Anti-G Valve

- Experimental DataAPPENDIX C: ALAR High-Flow Anti-G Valve Experimental DataAPPENDIX D: Garrett Electronic/Pneumatic Anti-G Valve

Experimental DataAPPENDIX E: Hymatic VAG-1 10-022 Anti-G Valve Experimental DataAPPENDIX F: AAMRL ("Bang Bang") Anti-G Valve Experimental Data

In order for experimental data on this research to bereadily accessible, microfiche have been made of theseAppendixes. The microfiche are available through:

The Strughold Aeromedical LibraryUSAFSAM STINFO Office (USAFSAM/TSKS)USAF School of Aerospace MedicineBrooks AFB, Texas 78235-5301

105