Test procedure development

Mobile Air Conditioning (MAC)

Progress update08-06-2010

Working paper No. MACTP-01-03(Geneva, 8 June 2010)

2

Contents

• Project overview• Progress made so far

• Identification of major influential parameters• Definition of test plan• Preliminary testing results

3

Project overview

Goal: To develop test conditions and procedures for MAC

• Main evaluation parameter: impact on CO2 / fuel consumption and other regulated emissions

• Procedure should be clearly discriminative of different systems

• Target accuracy and repeatability need to be clearly established

4

Project overview

• Definition of a test procedure(s) for MAC performance at type approval

• Focus on physical testing but with options to include virtual testing later.

• Procedure should provide the possibility (partial) substitution by virtual testing in the future

• Cost efficiency• Realistic representation of MAC efficiency• Use previous experience (ADAC 2007, TNO 2006)

5

MAC test conditions

• Which are the most important ambient conditions concerning the real operation of the MAC system?

• investigation of three „typical“ climates• hot: Athens• medium: Frankfurt• cold: Helsinki

6

MAC test conditions - weather, e.g. Athens% AC fuel consumption of annual AC fuel consumptiondepending on ambient temperature

=> proposal for test condition #1: 30°C/40%* *average relative humidity at 30°C

0123456789

10

0 5 10 15 20 25 30 35 40Temperature in °C

% A

C fu

el c

onsu

mpt

ion

(of a

nnua

l AC

fuel

con

sum

ptio

n) Athens

Solar radiation 700 W/m2

05

101520253035404550

<10 10…14 15…19 20…24 25…30 >30Temperature in °C

% A

C fu

el c

onsu

mpt

ion

(of a

nnua

l AC

fuel

con

sum

ptio

n)

Athens

i.e. 7.5% of the annual fuel consumption of the ac system is caused at ambient temperatures of 30°C (in Athens)

7

MAC test conditions - overview

• 15°C/75% (300 W/m2) • 20°C/65% (500 W/m2) • 30°C/40% (700 W/m2)Based on assessment of annual weather data in Athens,

Frankfurt, Helsinki.

• Remarks (results from simulation of refrigerant cycle):

• ±1 K at inlet temperature (≈ abient temperature) will result in a variation of ≈ ± 5% to 10% in cooling demand (greater impact at lower ambient temp.)

• ±3 % at inlet humidity will result in a variation of ≈ ±5% in cooling capacity

-> important for control of climatic chamber!

8

MAC test conditions - interior temperatureCondensing the results for the three evaluated cities

05

101520253035404550

<10 10…14 15…19 20…24 25…30 >30Temperature in °C

% A

C fu

el c

onsu

mpt

ion

(of a

nnua

l AC

fuel

con

sum

ptio

n)

Helsinki

05

101520253035404550

<10 10…14 15…19 20…24 25…30 >30Temperature in °C

% A

C fu

el c

onsu

mpt

ion

(of a

nnua

l AC

fuel

con

sum

ptio

n)

Frankfurt

05

101520253035404550

<10 10…14 15…19 20…24 25…30 >30Temperature in °C

% A

C fu

el c

onsu

mpt

ion

(of a

nnua

l AC

fuel

con

sum

ptio

n)

Athens

Findings: T cabin < T ambient

From simulation work we suggest: 25°C and 40%? humidity in the test cell21°C interior temperature Sensitivity tests: 30°C/40%, 23°C/50%

9

Factors to be considered in test procedure

1. Test cycle („easy to drive” for repeatable results at small fuel consumption effects)

2. Ambient temperature and humidity3. Interior temperature to be reached with MAC4. Simulation of heat from sun radiation (with heater in vehicle or via 3.)5. Settings of the MAC system6. Evaluation method for test results

Option for test procedure: Test vehicle on the chassis dynamometer with and without MAC.Difference is the additional fuel consumption from the MAC system.

Define following settings:

10

Working hypothesis for test procedure

TC3

Ta, ϕa

To CVS, exhaust gas analyser

g CO2/km

Chassis dynamometer

blower

Test programme on the chassis dynamometer for evaluation of: * different test cycles* different boundary conditions (Ta, ϕa, ma), * settings of MAC (Ti, mass flow ml,recirculating air by blower settings) * evaluation methods

Tests at TUG, KTI and LAT for reproducibility and repeatability

ma

ml

330 mm to roof

30 mm

11

0

10

20

30

40

50

60

70

0 500 1000 1500 2000 2500

Time [s]

km/h

0

1

2

3

4

5

Gea

r

VelocityGear

Bag 165 km/h ~ average speed of (NEDC+real world cycles)

Bag 2idl ing (long duration for

repeatibility)

Ti = const -> start test

For for „quick“ basic tests, preferred

MAC-2-Step cycle

NEDC Option for „not steady state test“

CADC Measurement of additional fuel consumptionfrom MAC in „real world“ cycle(not planned for type approval)

0

20

40

60

80

100

120

140

0 200 400 600 800 1000 1200Time [s]

velo

city

[km

/h]

BAG 1(UDC)

BAG 2(EUDC)

0

20

40

60

80

100

120

140

0 500 1000 1500 2000 2500 3000

Time [s]

km/h

CADC-Urban

Part 1 Part 2 Part 3 Part 4 Part 5

Prec

ondi

tioni

ng U

rban

CADC-Road

Pre

cond

ition

ing

Roa

d

Part 1 Part 2 Part 3 Part 4 Part 5

CADC-Motorway

Prec

ondi

tioni

ng M

otor

way

Part 1 Part 2 Part 3 Part 4

Test cycles

12

• Vehicle soaking:• Vehicle soaked at 25 °C for ~6 hours• Similar to emission testing pre-test soaking• Attainable at all type approval facilities

• Preconditioning phase:• Start drive cycle @ 65 km/h• Switch on MAC at desired setting*• Start measurement once desired interior temperature is

attained and stable

*settings to attain desired interior temperature need to be determined (either by trial-and-error or by manufacturer specification) and validated before start of the official test.

Soaking / preconditioning

13

Temperatures measured at cabin rear middle

0

5

10

15

20

25

30

0 200 400 600 800 1000 1200 1400 1600 1800

Time [sec]

Tem

pera

ture

[°C

]

TC1TC2TC3TC4

Chassis dyno, steady state at 65 km/hAC on, AUTO 19,5°C ventilator stage 5, heater on

Selection of the interior temperature(measurement point and settings)

Different measurement points in the vehicle and different sensor optionsPosition “TC3” seems to be best option (representative for head/shoulder)

0

5

10

15

20

25

30

0 200 400 600 800 1000 1200 1400 1600 1800Time [s]

TC3

[°C

]

TC3 (5452) mid

TC3 (5455) mid

TC3 (5465) right

TC3 (5461) mid, shielded

preconditioning necessary

Findings: *Radiation from vehicle interior makes preconditioning necessary (>600s)*T at vents outlet different and depending on MAC-comfort strategy of the

model not recommended as reference*TC3 mid, not shielded best option?

Stratification of T in cabin

Temperatures measured at air vents outlet dashboard

0

5

10

15

20

25

30

0 200 400 600 800 1000 1200 1400 1600 1800Time [sec]

Tem

pera

ture

[°C]

PT100_1PT100_2PT100_3PT100_4

Chassis dyno, steady state at 65 km/hAC on, AUTO 19,5°C ventilator stage 5, heater on

T at vents different

14

Real world measurements in the project

M-1.1) Parameterization and validation of the simulation of the sun radiation-> Park vehicles outside in sun (LAT, JRC)M-1.2) Influence of MAC in real world driving -> Measure real world cycles on the chassis dyno. Measure with PEMS after

vehicles parked outside in sun (JRC, optional) chasing of 2 similar vehicles, one MAC off, the other MAC on.

Main task: evaluation, if test procedure is representative?(due to very different climate conditions within Europe meeting exactly the “average” European MAC usage seems not to be the main goal. Covering the main influences for COP is more important)

15

Measured quantities:• Solar power density [W/m2]

(top of the vehicle roof)• Middle, front mirror [ºC]• Head, driver [ºC]• Head, front passenger [ºC]• Floor, front passenger [ºC]• Middle, by the gear lever [ºC]• Head, middle back pass. [ºC]• Trunk [ºC]

Ambient in-vehicle measurements (vehicle conditioned and then left parked under direct sunlight)

Real world measurements in the project

16

Options (results from conference with Saint-Gobain Sekurit):1) Calculate heat transmission by simplified approach:

• Measurement of direct solar transmittance (ISO 13837) of complete glazing, depending on average sun declination on each glass

• Calculating the total transmittance for complete glazing

2) Calculation of the thermal transfer:

• Calculating inner and outer surface temperatures for glazing & car body parts

• Calculating heat exchange by sun radiation, convection, conduction and emission

3) Detailed simulation with Open Source Tool (such as Energy+ for buildings)

Option 1) may be the best approach for type approval. Option 2) and 3) may be considered if in future a shift towards virtual testing is desired.

Solar Loadoptions to include effect into test procedure

17

Repeatability

1. Variability in vehicle speed variability of engine power demand

2. Variability in temperatures and humidity during the tests influence on cooling demand and COP

Influencing factors:

1) Correct for variation of vehicle speed with measured braking forces of the rollers

2) Correct for variations of Ta, ϕa, TC3 with simulated cooling demand

3) Define a sufficient preconditioning phase (if too long, DPF regeneration may happen during test phase)

Development of correction factors and procedures is in progress

18

Main open topics

•How can different settings of chassis dyno and of MAC be corrected. Can manufacturers make data for mair t2,.. for different blower positions available for the actual study to test methods?

discussion with ACEA (meeting in Graz on 03.05.2010, collection of options ongoing)evaluation of different testing options to be finalized

•How can influence of sun radiation be depictured (in a simple way)?discussion of options with Saint-Gobain Sekurit ongoing

•Which vehicle velocities (or rpm for MAC) shall be used in the tests?

19

Thank you

Thank you for your kind attention.