small hydrocarbon flow calibration facilities at nmij

NATIONAL METROLOGY INSTITUTE OF JAPAN

Small hydrocarbon flow calibration facilities

at NMIJ

Yoshiya Terao

APMP/TCFF Workshop


Original presentations

at 9th International Symposium on Fluid Flow Measurement (ISFFM2015),

April 2015

“Expansion of Calibration Flow Range of Small Liquid

Hydrocarbon Flow Facility at NMIJ”

by Kar-Hooi Cheong, Ryouji Doihara Takashi Shimada and

Yoshiya TERAO

“Development of prototype low-flow rate test rig applicable for

highly volatile liquids” by Ryouji Doihara, Takashi Shimada, Kar-Hooi Cheong and Yoshiya Terao


1.E-06

1.E-05

1.E-04

1.E-03

1.E-02

1.E-01

1.E+00

1.E+01

1.E+02

1.E+03

1.E+04

0.1 1 10 100 Viscosity (mPa·s)

Flo

wra

te (

m3/h

)

Fuel mixer

Fuel efficiency

measurement

Hydraulic machinery

LP

G

Gaso

lin

e

Kero

sen

e

Lig

ht

oil

Heavy

oil

Chemical plant

Tanker

Petroleum terminal

Petroleum refinery

Primary standards for hydrocarbon at NMIJ

Primary

standard for

hydrocarbon

Small

Hydrocarbon

Flow Facility

100 L/h

0.02 L/h

1 L/h


37.0 36.4

33.0 33.0 32.6 30.8 30.4 30.2 30.0 30.0

0.0

5.0

10.0

15.0

20.0

25.0

30.0

35.0

40.0

Car models

JC0

8m

od

e f

ue

l eff

icie

ncy

(km

/L)

Top 10 car models with best fuel efficiency in Japan

hybrid vehicle

mini / light vehicle

Source: http://e-nenpi.com

Background and Motivation


Main targets Automotive industry

Engine test bench

Vaporizer for

manufacturing equipment

Fuel flowmeter

Liquid mass flowmeter

Semiconductor industry


Measurement conditions:

• Real gasoline (high volatility), Light oil,

• 200 L/h – 0.02 L/h,

• Wide flow range (over 1000:1),

• Wide temperature range,

• (High pressure over 200MPa).

• Water, Alcohols, Solvents, Toxic liquids,

• 10 L/h – 0.001 L/h,

• Wide temperature range (10 – 85 ℃）,

Liquid mass flowmeter

Automotive industry

Semiconductor industry

Fuel flowmeter


Measurement principles Gravimetric calibration:

• Advantage to accuracy and to simple traceability

• Disadvantage to evaporation error at the low flow rate

range with volatile liquid

• Water is mostly used as test liquid.

Volumetric calibration:

• SVPs have been adopted in many calibration facilities for

oil (including real gasoline), because it can make a closed

pipe line system.

• Some of them have a linear encoder to output many

pulses.

Syringe pump:

• Syringe pumps are widely used in medical and analysis

fields for ultra-low flow rate.

• Fluctuation has been observed as a result of the stepper

motor driving mechanism.


Cheong Kar-Hooi’s

gravimetric system


Calibration Facility

Weighing section

(liquid collection) Test section

(DUT mounting) Flow generation section

(tank, pump)


Calibration Facility Weighing section

(liquid collection)

Test section

(DUT mounting) Flow generation section

(tank, pump)

100g weighing system

(liquid collection for

0.02~1 L/h)

2kg weighing system

(liquid collection for 1~100 L/h)

Test meterWeighing

vessel

Weighing

scale, ms

Bypass to

storage tankMeasurement point of temperature, Ttm

and pressure, ptm

Pulse counter,

Number of pulses, Ip

Pulse signal

DUTTest meterWeighing

vessel

Weighing

scale, ms

Bypass to

storage tankMeasurement point of temperature, Ttm

and pressure, ptm

Pulse counter,

Number of pulses, Ip

Pulse signal

DUT

Calibration method ・ Gravimetric (static weighing)

・ Standing-start-and-finish

thermostatic chamber

(for better stability of liquid

temperature controllable

at any value between

15℃~35℃)


Uncertainty Budget

Uncertainty sources Light oil Kerosene

Pulse count 8.2×10-5 8.2×10-5

Estimation of liquid density 2.6×10-4 3.0×10-4

Measurement of liquid mass 1.8×10-4 1.8×10-4

Dead volume effect 1.2×10-4 1.3×10-4

Relative combined standard uncertainty 3.5×10-4 3.8×10-4

Uncertainty budget for 0.02 L/h ~ 1 L/h using 100 g weighing system

Uncertainty sources Light oil Kerosene

Pulse count 8.2×10-5 8.2×10-5

Estimation of liquid density 2.6×10-4 3.0×10-4

Measurement of liquid mass 3.3×10-5 3.3×10-5

Dead volume effect 6.1×10-6 6.4×10-6

Relative combined standard uncertainty 2.8×10-4 3.2×10-4

Uncertainty budget for 1 L/h ~ 100 L/h using 2 kg weighing system


Uncertainty Budget (Kerosene)

2 kg weighing system

(1 L/h~100 L/h)

100 g weighing system

(0.02 L/h~1 L/h)

pulse count 0.008%

dead volume effect 0.001%

measurement of liquid mass

0.003%

estimation of liquid density

0.030%

pulse count 0.008%



0.018%


0.030%


pulse count 0.008%



0.018%


0.030%

Uncertainty budget for 0.02 L/h ~ 1 L/h using

100 g weighing system

Dead volume effect

The value is evaluated in relative to

the smallest amount of liquid

collection which is 10 g.

10 g of liquid collection is performed

at 0.02 L/h and it takes about 40

minutes for one collection.

Some measures to cut down the dead

volume effect:

Dead volume is made as small as

possible.

Temperature variation of liquid in the

dead volume is controlled and

stabilized (±0.025oC in one hour

duration).


pulse count 0.008%



0.018%


0.030%

Contributing factors to uncertainty of

liquid mass measurement

correction factor of weighing scale

0.003% reading of weighing scale 0.001%

buoyancy effect correction 0.001%

liquid evaporation

0.015%

leakage possibility

0.009%


Conditions for liquid collection

Liquid temperature：15℃~35℃.

Smallest liquid collection： 10 g （about 40 minutes at 0.02 L/h ）

Method of evaluation

An evaluation method thought to give the closest value to the actual

condition was performed.

Results

At 35℃, the evaporation rate is 0.0015 g per hour for kerosene and

0.0007 g per hour for light oil.

In relative to 10 g of liquid collection, this produces 1.5×10-4 (0.015

%) in terms of relative uncertainty.

Collection 100 g discharge Measurement of evaporation

for 60 minutes

Evaluation of liquid evaporation


Flow range, Q / Weighing system Volumetric flowrate Mass flowrate

1 L/h Q 100 L/h (2 kg weighing system) 0.032 %

(0.064 %)

0.010 %

(0.020 %)

0.02 L/h Q < 1 L/h (100 g weighing system) 0.039 %

(0.078 %)

0.025 %

(0.050 %)

Calibration and Measurement Capability

Ultimate uncertainty of calibration

(Calibration and measurement capability in italic, coverage factor: k=2)

Note:

(ultimate standard uncertainty)2 = (uncertainty due to facility)2 + (uncertainty due to flowmeter)2

Uncertainties due to flowmeter are as follows:

Response of flowmeter to standing-start and finish

Repeatability


Doihara’s weighing and syringe

system


Schematic diagram of the prototype test rig


Calibration procedure

• The syringe pump is calibrated using

the gravimetric weighing tank.

Calibration


Calibration procedure • Then a DUT flowmeter is calibrated

using the syringe pump.

• At a DUT calibration stage, the flow rig is able to make a closed pipe line system.

Calibration


Standing method and flying method

• The syringe pump is calibrated with a

standing start and stop method.

• What the weighing tank needs to

calibrate is only the syringe pump

(over 10 000 pulses).

• Characteristic of the syringe pump is

essentially linear to flow rate.

• Thus the standing method is not

significant problem in this syringe pump

calibration.

Standing method

Calibration


• DUT flowmeter is calibrated at a constant flow rate with a flying start and finish method.

• If the number of pulses of the flowmeter would be small, the pulse counting error can be reduced using a double chronometry pulse interpolation technique in the flying method.

Flying method

Calibration


Calibration results of the syringe

-0.01

0.00

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

0.01 0.10 1.00 10.00 100.00Flow rate [ L/h ]

Light oil: Jan. 2015

SYSY

SY

PK

V [Pulse/mL]

• They all agree within ±0.02%.

• Influence of the standing method was enough small.

Rela

tive d

evia

tion f

rom

nom

inal valu

e [

% ]

±0.02%.


Evaporation effect and correction

Condition Tank lid with

small holes

Evaporation

trap pool

Evaporation rate

[ mg/h ]

Static A ✓ ✓ 0.6

Static B ✓ 3

Static C ✓ 4

Static D 6

• The weighing tank system has some evaporation controls to reduce evaporation. – The weighing tank was surrounded by

small liquid pools.

– The weighing tank has a lid with small holes to insert tubes.

• The changing of weight from an initial weight was measured in several conditions for one hour (Static).


• The syringe pulse factor was increased at low flow rate.

• Results with evaporation correction (Static A) were similar.

-0.01

0.00

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

0.01 0.10 1.00 10.00 100.00

Flow rate [ L/h ]

Without evaporation correction

With evaporation correction (Static A)

Rela

tive d

evia

tion f

rom

nom

inal valu

e [

% ]


Evaporation effect and correction

Condition Tank lid with

small holes

Evaporation

trap pool

Evaporation rate

[ mg/h ]

Static A ✓ ✓ 0.6

Static B ✓ 3

Static C ✓ 4

Static D 6

Actual E ✓ ✓ 2 – 4

• The atmosphere inside the wind shield was ventilated when the working liquid in the weighing tank was discharged.

• The actual evaporation rate was measured in the calibration sequence.

• The measured evaporation rate (actual condition E) was between static A and static C.

Since actual evaporation rates were not constant, the

evaporation effect was corrected at each calibration.


• Corrected results (Actual E) showed almost constant value over full

flow rate range.

-0.01

0.00

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

0.01 0.10 1.00 10.00 100.00

Flow rate [ L/h ]

Without evaporation correction

With evaporation correction (Actual E)

Rela

tive d

evia

tion f

rom

nom

inal valu

e [

% ]


5mL

Position dependency of syringe pulse factor The standing stop position was varied from the cylinder end position

to 35 mL, keeping the calibration collection volume of 5 mL.

-0.01

0.00

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

0 5 10 15 20 25 30 35 40

Standing stop position from cylinder end [ mL ]

5mL

0

Rela

tive d

evia

tion f

rom

nom

inal valu

e [

% ]


10mL

Position dependency of syringe pulse factor

0 5mL

-0.01

0.00

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

0 5 10 15 20 25 30 35 40

Standing stop position from cylinder end [ mL ]

5mL collection (light oil)

10mL collection (lightl oil)

Volume domain of the cylinder has been fixed.

•Standing stop position: 5m L,

•Collection volume: 10 mL

Rela

tive d

evia

tion f

rom

nom

inal valu

e [

% ]


Calibration results of the syringe

-0.01

0.00

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

0.01 0.10 1.00 10.00 100.00Flow rate [ L/h ]

Light oil: Jan. 2015

Industrial gasoline: Feb. 2015

Industrial gasoline: Mar. 2015

Rela

tive d

evia

tion f

rom

nom

inal valu

e [

% ]

• Difference depending on kinds of liquids has not been observed.


Preliminary uncertainty budget

Uncertainty source Standard uncertainty

Combined

standard

uncertainty

Expanded

uncertainty

(k=2)

Syringe pulse count 0.002%

Mass measurement 0.008%

Density at syringe 0.010%

Change of mass in dead volume 0.007%

Random effect of syringe calibration 0.008%

Calibration for syringe pulse factor 0.017% 0.034%

Linearity and long term reproducibility

of reference syringe pulse factor 0.017%

Density at syringe 0.010%

Change of mass in dead volume 0.015%

Random effect of DUT calibration 0.005%

Calibration for mass flow rate 0.030% 0.060%

Density at DUT 0.016%

Calibration for volume flow rate 0.034% 0.068%


Specification • Flow rate:

60 L/h – 0.02 L/h (1000 mL/min – 0.3 mL/min).

• Working liquid (as the first step):

Light oil (diesel oil),

Industrial gasoline (Flash point ≥ 40 ℃ ）

• Calibration principle:

Gravimetric weighing tank,

Syringe pump.

• Temperature and pressure (as the first step):

20 ℃ (Stability ±0.08 ℃)

0 – 0.3 MPa (at test section)

• Preliminary expanded uncertainty (k=2):

0.068 % (for volumetric flow),

0.060 % (for mass flow) .


Summary

Cheong’s gravimetric system

• 0.02 L/h to 100 L/h

• Kerosene and light oil

• Liquid temperature: 15℃ to 35℃

• Open for public calibration service

• Will be peer-reviewed in 2017

Doihara’s gravimetric and syringe system

• 0.02 L/h to 60 L/h

• Applicable to volatile liquids

• Liquid temperature: 20℃

• Flexible structure

• Used for research purpose


Thank you for your attention.

small hydrocarbon flow calibration facilities at nmij

Documents