engineering update and pdr plans

9 December 2002 Reiersen - 1

NCSX

Engineering Update and PDR Plans

Wayne Reiersen

PAC-6

December 8, 2002


NCSX

What has happened since the CDR

• Design improvements have been made

• Outstanding issues are being addressed

• Plans leading to the PDR have been made

• Manufacturing development and R&D plans are being implemented


NCSX

Design improvements and outstanding issues

Added poloidal breakin MC structure

Simplified MC cooling

Simplified VV port attachment

VV assembly joint tilted to avoid assembly interference,

impact on vacuum seal is TBD

Reduced number of PF coils from 6 to 5

Changed order of MC productionfrom 6(A-B-C) to 6A-6B-6C

Changed MC assembly from“1 at a time” to “3 at a time”

Adopted 2x2 conductor packin MC to minimize keystoning

Buck TF coils off CS coils rather than off separate structure

Adopt new MC set, make VV as large as possible for improveddivertor performance and flexibility


NCSX

Poloidal break added to MC structure

• Time constant w/o poloidal break was too long

– 70ms with 3 toroidal breaks

– Plasma current ramp time is 60ms

– Requirements is < 20ms

• Adding a single poloidal break and 15 toroidal breaks (none at field assembly joint) drops the longest time constant to 18 ms

• Copper cladding on the tee will be electrically isolated and interrupted every 5-10cm


NCSX

Simplified MC cooling scheme adopted

• High thermal conductivity clamps used

• Cooling tubes are attached to clamps

• Copper cladding applied to tee section over an electrical insulator – contacts clamps at attachment points

• Copper mesh or strips outside ground wrap conduct heat to clamp on outside of winding pack

Cu cladding

Cu mesh or strips

Cu clamp

4.64

0.040 x 4.92Copper Mesh

0.375 DIAStud

0.100 x 1.00G10 Shim

0.375 x 1.00Screw

6.84

0.375


NCSX

Adequate cooldown is achieved

End of 1st heating period End of 1st cooling period(15 minute cooldown)


NCSX

Adopted 2x2 conductor pack to minimize keystoning

• Degree of keystoning depends on ratio of conductor thickness to bend radius

• 2x2 conductor configuration should greatly reduce keystoning

• Thicker (32-ga v. 36-ga) strands should reduce “splintering”

• Hopefully, no compensation in the geometry of the winding surface or mechanical force to squash conductors will be required

• Keystoning tests are planned for early CY03

Use 0.25x0.313-in, 32-ga conductor4 cables per turn


NCSX

Simplified VV port attachment

1. Weld on port extension and leak check weld

Torus shell

Port extension

2. Cut port extension off and after final torus leak check, cut opening thru torus

Torus shell

Port extension, part 1

3. Prep edges (Supplier) and weld port extension in place with full penetration weld from inside ( PPPL)

Torus shell




NCSX

Assembly interference resolved by tilting assembly joint

Vertical assembly flange showing interference with mod coil during assembly operation

Solved by tilting assembly flanges 30 deg off vertical. Some VV supports also eliminated. Impact on seal is TBD.

30 deg


NCSX

Reduced number of PF coils from 6 to 5

• Adequate performance and flexibility found by combining PF3 and PF4

– New coil located in between PF 3 and PF4

– Same cross-section as PF3

• Should reduce cost (fewer coils and circuits, eliminate crown structure) and simplify assembly and maintenance (solenoid can be inserted and removed with PF3 in place)


NCSX

Buck TF coils off CS coils

• At CDR, it was shown that two options appeared viable

– Buck TF coils off a separate structure (vertical plates and horizontal disks)

– Buck TF coils off CS coils

• The second option offers significant advantages and has been adopted for the PDR

– Increased OD for CS coils, more V-s, less power required

– Simpler CS structure should translate into lower cost

Show before and after pictures


NCSX

Changed order of MC production and assembly

• Suggestion made at CDR by D. Anderson (UW) to produce all modular coils of the same type in sequence (6A-6B-6C), rather than always changing the coil type being produced (6[A-B-C])

– Clear advantages in both MC winding form production and winding

• Decision made to slide “3 modular coils at a time” instead of “1 at a time” over the vacuum vessel

– Manually manipulating the SLA model at the time of the CDR showed that trying to fit a single coil over the VV and simlutaneously mate the “wings” of the MC into the “slots” in the adjacent MC was more difficult than pre-assembling the 3 MC’s in a half field period and sliding them over the VV

– Expanding the VV to provide more room for the divertor exacerbates this difficulty

– No schedule impact on first plasma, takes VV off critical path

2.5-m99-in

2.0-m78-in

2.8-m109-in

2.1-m83-in

2.8-m109-in

2.3-m90-in


NCSX

Incorporating the new MC design and expanding the VV is the major outstanding issue

• Physics has been working hard to provide a healed MC set with increased coil-to-plasma spacing for improved divertor performance

• Key engineering metrics have been incorporated into the optimization (max coil current, min bend radius, min coil-to-coil spacing) and have been preserved or improved in candidate MC sets

• Engineering has developed tools to optimize the MC trajectory during field period assembly to facilitate expanding the VV and make more room for a divertor

• Updating the conceptual design with a new MC set and VV is the pacing item for the PDR


NCSX

VV size is limited by modular coil assembly constraint


NCSX

Algorithm developed to optimize trajectory with 6 DOF

• Sample run with M47 coils and expanded VV boundary

• Plasma fits nicely inside FW but min separation is too small

• Run to be repeated with new coils and plasma boundary


NCSX

Plans leading to the PDR have been made

• Expectations for the PDR have been defined

• Work plans leading to the PDR have been coordinated, incorporated in the project control system, and are being tracked

• Plans for manufacturing development and prototype fabrication for the vacuum vessel and modular coils are being implemented


NCSX

Expectations for the PDR

• A June PDR is planned for the VV and Modular Coils

• The conceptual design and cost and schedule estimates for all other WBS elements will be updated at that time

• PDR Requirements

– Performance requirements have been defined

– A design has been developed that fully meets those requirements

– All feasibility issues have been resolved

– Interfaces with other systems have been fully defined

– Plans for assembly, installation, and test are established

– Models and drawings have been developed, reviewed, and released at the Preliminary Design level

– A sound and stable basis exists for proceeding with Final Design


NCSX

Key activities leading to the PDR

Dec Jan Feb Mar Apr May Jun

Reconstruct conceptual design of stellarator coreComplete configuration studies around CDR designDefine MC and expanded VV geometryDefine envelopes for other core componentsDefine current waveforms for reference scenarios

Finalize plans for machine assemblyProcure SLA model to verify feasibilityDevelop and document plans for field period assyDevelop and document plans for final assemblyDevelop and document plans for integrated systems testing

Complete preliminary design of VVDefine port geometry consistent with diagnostic viewing rqmtsEstablish envelopes and locations for in-vessel componentsEstablish port allocations

This will be replaced…


NCSX

Implementation of manufacturing development and R&D plans

• 4 vendors (2 for the MC and 2 for the VV) will be brought on board in January ’03 to do mfg development and prototype fabrication

• The vendors for the production units will be selected from among those participating in this next phase

• Modular coil winding R&D is being done in-house at PPPL


NCSX

VV manufacturing development and prototype fabrication

• Manufacturing development activities will be conducted by 2 subcontractors to support the VV PDR.

– Manufacturing methods for fabricating the VV will be identified.

– Recommendations for improving the VV design and performing additional mfg development activities will be solicited.

– Preliminary MIT/QA Plans will be developed and used as the basis for budgetary cost and schedule estimates.

• The subcontractors will fabricate a full scale 20º prototype VV segment.

• The subcontractors will submit a final MIT/QA plan and a fixed price cost and schedule proposal for the production units


NCSX

20º prototype vacuum vessel segment


NCSX

MC winding form manufacturing development and prototype fabrication

• Deliverables are the same as for the vacuum vessel

• Manufacturing development activities will be conducted by 2 subcontractors to support the Modular Coil PDR

• The subcontractors will fabricate a full scale prototype (Type A) modular coil winding form (later used for winding R&D)

• The subcontractors will submit a final MIT/QA plan and a fixed price cost and schedule proposal for the production units


NCSX

In-house winding R&D

• Winding R&D at PPPL has been initiated in FY03

• Key elements

– Perform keystoning tests

– Determine winding material properties and allowables

– Develop VPI process

– Develop molding process

– VPI molded test samples in small oven (pre-PDR) and autoclave (post-PDR)

– Wind, mold, and VPI full scale prototype coil (post-PDR)


NCSX

Keystoning tests

• Winding form being designed

• Nine insulated turns using four (4) conductors per turn will be wound onto form

• “Faro” mechanical measuring arm will be used to measure tolerance build as a result of conductor keystoning


NCSX

Material tests to determine properties and allowables

• SOW being drafted identifying all necessary tests

• Mold designed and fabricated for 1st tensile specimens

• VPI of 1st tensile specimens expected this week


NCSX

Develop VPI process

• CTD 101K selected as the resin system for impregnating modular coils

• Epoxy characterization underway (cure cycle, viscosity, etc.)

• Will be used to VPI 1st tensile specimens and UT coil

• Existing small vacuum oven and viscometers will be used

• R&D is being conducted in TFTR Basement


NCSX

Develop molding process

• Wet wrap of winding pack is planned to form pressure boundary for VPI

• Molding process will first be tested on straight samples, VPI in small oven

• Molding process tested on actual coil sections next

• Larger sections and a full scale prototype will be molded after the PDR, VPI in autoclave


NCSX

Summary

• We have a better design

– Substantial configuration improvements have been made since the CDR, no new “showstoppers” have been identified

• We have a plan

– Work plans have been put in place for a June PDR and cost & schedule review

• We are moving forward

– Critical path manufacturing development and winding R&D activities are moving along well

– Incorporating the new MC design and expanded VV are the pacing items

engineering update and pdr plans

Documents