gtr#13: towards inherently safer hydrogen-powered vehicles

ulster.ac.uk

GTR#13: towards inherently

safer hydrogen-powered vehicles

Prof Vladimir Molkov

Recent advances by Hydrogen Safety Engineering and Research Centre

(HySAFER)

Outline of presentation

▪ Fire test (improved reproducibility)

▪ Pressure peaking phenomenon (new phenomenon)

▪ System storage container-TPRD (“coupled” approach)

▪ Concluding remarks

▪ List of relevant urgent PNR topics

2

ulster.ac.uk

Fire test reproducibility

Definitions

Bare container - ordinary (unprotected)

container

Protected container - container with thermal

protection, e.g. intumescent

paint, Ulster IP, etc.

Heat release rate (HRR) - heat release rate in a fire [kW]

Fire resistance rating (FRR) - time from burner ignition until

container rupture in a fire

(without TPRD)

4

Issues of GTR#13 fire test

▪ Poor reproducibility of the fire test in different

laboratories.

▪ No test without TPRD (a serious first

responders’ concern, EU HyResponse). EU

FireCOMP: there is a non-zero probability of

TPRD failure.

▪ No test procedure for novel thermally protected

storage containers, e.g. explosion-free in a fire

containers.

5

CFD: revealed dependence of FRR on HRR

Poor reproducibility

6

[1] D. Makarov, Y. Kim, S. Kashkarov, V. Molkov, Thermal protection and fire resistance

of high-pressure hydrogen storage, in Proceedings 8th ISFEH, Hefei, China, 2016.

[2] N. Weyandt, Analysis of Induced Catastrophic Failure Of A 5000 psig Type IV Hydrogen

Cylinder, Southwest Research Institute report for the MVFRI, 01.06939.01.001, 2005.

[3] R. Zalosh, Blast waves and fireballs generated by hydrogen fuel tank rupture during fire

exposure’, in Proceedings 5th ISFEH, Edinburgh, UK, 2007.

[4] L. Bustamante Valencia, P. Blanc-Vannet, L. Heudier, D. Jamois, ‘Thermal history

resulting in the failure of lightweight fully-wrapped composite pressure vessel for hydrogen

in a fire experimental facility’, Fire Technology, no. 52, pp. 421–442, 2016.

Heat release

rate

Fire source Fire resistance

rating

79 kW [1] Premixed CH4-air burner 16 min

170 kW [1] Premixed CH4-air burner 9 min

370 kW [2, 3] Diffusion C3H8 burner 6.5 min

4100 kW [4] n-C7H16, pool fire 6 min

Experimental validation of numerical study

Poor reproducibility

7

Poor reproducibility: way outSaturation of FRR at HRR>350 kW

8

▪ Constant HRR above 350 kW

▪ Heat flux (input) of minimum 100 kW/m2

9

Two ways to ensure reproducibility:

Type 4 (CFRP)

Steel?

HRR: saturation above 350 kW Heat flux: continuous decrease

10

Two ways to ensure reproducibility:

Parameter HRR (burner + LPG flow rate)

Heat flux (input)

Technical realisation Easy (as now) Complicated and questionable (suggested minimum is too high)

Additional cost per test No $600 per sensor in a destructive test

Location and number of sensors

Easy (as now) Where and how? How many?

Provision of required control parameter level

Easy (as now) Is it possible to provide minimum of 100 kW/m2 for type 4

Steadiness of parameter in steady-state fire

Yes No (decreases during the test)

▪ Constant HRR above 350 kW

▪ Heat flux (input) of minimum 100 kW/m2

ISO TC58: fire test without TPRDRight move: “Fire test until rupture”

11

Right approach of ISO TC58 (‘Gas Cylinders — Guidance for

design of composite cylinders — Part 2: Bonfire test issues’,

ISO/TC 58/SC 3 N 1714, 2017):

▪ ”For cylinders tested under option B – fire test until

rupture”

Engulfing fire test update

Current Engulfing fire test

Method 1E: Bare tank

Method 2E: Thermally protected tank

12

Need in thermally protected tank test

Method 1E: Bare tank (minor changes)

Fire source

Bare tank

10

0

75

Thermocouple

1650

New requirements: HRR ≥ 350 kW, TC location

25

13

Method 2E: Thermally protected tank

Thermally protected container (no TPRD)

L2=1

00

Intumescent paint initial (unreacted)

L3=7

5

Fire source

L1

Intumescent paint (expanded)

Provide functioning of thermocouples

165014

590

0 5

Time, min

Te

mp

era

ture

, °C

Till end of test

Not less than 590°C

2 hours

Maximum recorded duration of car fire is 2 hours (e.g. K.

Okamoto, et al., ‘Burning behaviour of minivan passenger

cars’, Fire Safety Journal, vol. 62, pp. 272–280, 2013).

HRR at least 350 kW

Method 2E: Thermally protected tank

15

ulster.ac.uk

Pressure peaking phenomenon

Pressure peaking phenomenonPhysics

Example

Garage 4.5x2.6x2.6 m,

“brick” vent.

Car (350 bar, D=5.08 mm):

mass flow rate from TPRD

390 g/s.

Overpressure limit for structures (10-15 kPa)

Solution: decrease TPRD

diameter + increase fire

resistance rating of

onboard storage tank!17

2

1

12

1

2

encl

S

encl

S

encl

Svent

P

P

P

PPCAV

Pressure peaking phenomenonValidation: air (no pressure peaking)

Air release 2.8 g/s (enclosure 1 m3, vent D=11 mm). Only

gases lighter than air can generate pressure peaking.

18

0 10 20 30 40 50 60 70 80 90 100

Time, s

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

Ov

erpre

ssure

, kP

a

HIWP4-044

Model prediction, CD=0.72

Steady-state pressure without peak

Pressure peaking phenomenonValidation: helium

Helium release 0.99 g/s (enclosure 1 m3, vent D=11 mm)

19

0 25 50 75 100 125 150 175 200 225 250 275

Time, s

0

0.5

1

1.5

2

2.5

3

Ov

erpre

ssure

, kP

a

HIWP4-046

MOdel prediction, CD=0.85

Pressure peak

Pressure peaking phenomenonValidation: hydrogen

Hydrogen release 0.55 g/s (enclosure 1 m3, vent D=11 mm)

200 10 20 30 40 50 60 70 80 90 100

Time, s

0

0.5

1

1.5

2

2.5

3

3.5

Ov

erpre

ssure

, kP

a

HIWP4-049

Model prediction, CD=0.85

Pressure peak

0 10 20 30 40 50 60

Time, s

0

2

4

6

8

10

12

14

Ov

erp

ress

ure

, kP

a

1 brick vent

2 bricks vent

3 bricks vent

4 bricks vent

Pressure peaking phenomenonCase 1

Ignited release in the garage:

TPRD D=2.0 mm, P=70 MPa

(107 g/s).

0 10 20 30 40 50 60

Time, s

0

20

40

60

80

100

120

140

160

180

200

220

Ov

erp

ress

ure

, kP

a

1 brick vent

2 bricks vent

3 bricks vent

4 bricks vent

Unignited release in the

garage: TPRD D=2.0 mm,

P=70 MPa (107 g/s).

Pressure peaking phenomenonCase 2

▪ Ignited release from TPRD with D=0.3 mm in a garage

2.6x2.6x4.5 m with vent 1 brick (left) or 0.5 brick (right).

▪ Onboard tank storage pressure 700 bar.

▪ Garage can withstand overpressure 10 kPa.

22

ulster.ac.uk

System tank-TPRD

24

ISO/TC 58: decoupled tank-TPRD testCould we model coupled functioning (reality)?

24

ISO, ‘Gas Cylinders — Guidance for design of composite

cylinders — Part 2: Bonfire test issues’, ISO/TC 58/SC 3 N

1714, 2017

Simulation of tank-TPRD systemInitial model development

25

0

50

100

150

200

250

300

350

400

450

500

550

600

650

700

750

800

0

5

10

15

20

25

0 100 200 300 400 500 600 700 800 900 1000

Decomposed front of the wall

Load bearing front of the wall

Pressure

Time (s)

Wal

l th

ickn

ess

(mm

)

Pre

ssu

re (

bar

)

No TPRD

0.3 mm TPRD

0.5 mm TPRD

Init

iati

ng

tim

e

180 bar

15 bar

750 bar

Concluding remarks

▪ Improvement of GTR#13 fire test reproducibility is

suggested through requirements of a burner to have

heat release rate above 350 kW. Development of a

burner requires additional PNR.

▪ Pressure peaking phenomenon could be practically

eliminated using TPRD diameter of 0.3-0.5 mm. This

would require increase of fire resistance rating (time

to rupture in test without TPRD).

▪ The use of TPRD diameter 0.3-0.5 mm increases fire

resistance (to let first responders more time to

control and eliminate “hazards”) yet doesn’t exclude

the tank rupture (preliminary result).

▪ Explosion-free in a fire tanks could be a solution. 26

List of relevant urgent PNR topics

▪ Burner design to provide GTR#13 fire test

reproducibility, including effects of wind

▪ Testing pressure peaking phenomenon at realistic

garage-like enclosures.

▪ Inherently safer tank-TPRD system with minimised

TPRD diameter (to exclude pressure peaking

phenomenon and yet to avoid tank rupture)

▪ TPRD reliability data for risk assessment

▪ Development of explosion-free in a fire tanks

▪ Vehicles in tunnels, underground parking

▪ In-situ dumping of compressed gas potential energy

▪ … 27

Thank you