Page 1

Requirements and advances in R&D related to safety and safety standards for A3 refrigerants in air conditioners

Dr Daniel Colbourne

2nd Meeting on technologies for the air conditioning industry: R-290 Experience and Perspectives Brazil Ministry of the Environment (MMA) and United Nations Industrial Development Organisation (UNIDO)

BrasĂ­lia, 28th November 2019


Introduction Hydrocarbon (HC) refrigerants presently increasingly attractive ▪ ▪ ▪ ▪

Low cost refrigerant and materials High efficiency (due to favourable thermophysical properties) Negligible global warming potential (GWP) “No reason not to use…”

On the counter-side ▪ No inherent value; no industrial backing ▪ High flammability!

How to resolve this flammability issue? ▪ Improve safety standards


Safety standards infrastructure Safety standards developed by many different bodies ― National, European, International “HORIZONTAL”

“VERTICAL” INTERNATIONAL

EUROPEAN

others

others

NATIONAL

others

others


Identification of suitable standards Evolution of key standards EN378: 2000

EN 378: 2008

ISO 5149: 1993

603352-40: 1987

603352-40: 1992

60335-2-89: 2002

ISO 5149: 2014

603352-40: 1995

60335-2-89: 2010

EN 378: 2022??

EN 378: 2016

603352-40: 2002

ISO 5149: 202?

603352-40: 2013

60335-2-89: 2019

603352-40: 2018

603352-40: 2021??

60335-2-89: 202x ???


Problems with ISO 5149, EN 378 & 60335-2-40 250 split ACs models [R290, R407C, R410A, R22 products – charge converted to R290-eq] 150 W/m2 heat load

Current limits according to standards

Practically, charge limits are prohibitive and obstructive for HCs


Barriers Why has it been so difficult to implement “favourable” requirements in safety standards for A3 refrigerants? ▪ ▪

Primarily, commercial opposition Committees and WGs stuffed with very professional, articulate, intelligent technical “lobbyists”

Totally unreasonable conduct at a time when climate change is becoming so critical


Barriers How to overcome these barriers? ▪ ▪

Match the strength of personnel Generate irrepressible technical arguments and data No need for naturals! Don’t want to change (again)! Have to be cautious! A commercial threat!

My boss told me to! Want absolute proof of safety!

Grrrrr!!!


Standards development process CDV/DIS

Initial proposal

DC

FDIS

Very long and drawn-out‌ Not much can be done to CD address the process‌


Flammable risk map

System tightness

Housing geometry

Likelihood of leak

Location of parts

Airflow

Risk of ignition Size of flammable volume

Height of system parts Leak rate

Room size

Occurrence of ignition sources Mass released

(Assume fixed in space)


Implication on risk Historically, limited investigation into the topic of safety of HCs in commercial refrigeration and air conditioners 1 – “level of safety”

▪ Charge limits based on pessimistic assumptions and lack of knowledge Basic design (“habilis”)

New requirements

Tolerable risk level

E.g., background fire frequency or dom ref/ACs: 1×10-5 y-1

Current charge limits

Refrigerant charge mass


Possible approaches include three general options IEC 60335-2-40 Edition 5.0

2013-12

Max charge +

colour inside

Household and similar electrical appliances – Safety – Part 2-40: Particular requirements for electrical heat pumps, air-conditioners and dehumidifiers

IEC 60335-2-40:2013(E)

COPYRIGHT © IEC. NOT FOR COMMERCIAL USE OR REPRODUCTION

®

INTERNATIONAL STANDARD

other requirements

Prescrip -tive

Basic options

Risk assessment

Black box/ testing

Or a combination of two or more…


Incremental black box testing

‌As in draft IEC 60335-2-40

Assumed leak rate, constant mass flow

Response of detection system optional solenoid valve in suction line

Releasable charge amount

(e.g

solenoid valve in liquid line

Calculated min airflow rate

etc.

Lots of broadly disconnected, formulae, tests, etc.!


Joined-up black box testing Assumed leak rate, constant mass flow

Why do lots of different tests, when one or two could suffice? Vision: make a hole in the system and see what happens!

Response of detection system

Releasable charge amount

Calculated min airflow rate

etc.

Protective features, eg, airflows, leak detection, etc

Unit construction characteristics Room dimensions

Cfloor ≼ LFL?

Max reft charge / min room size

Conditions, eg, leak mass flow

Constraints (e.g., max conc, etc)


Main question is hole size Ordinarily, hole size (leak mass flow) is most challenging issue ▪ Choice of hole size affects everything else…

Largest hole size found in project so far ▪ 0.4 mm2 in “uninterfered” cases ▪ 0.8 mm2 with human interference


Dispersion of leaked refrigerant Example of release into room â–Ş

High mass flow, high exit concentration, 1000 g


Dispersion of leaked refrigerant Effect of thermal gradients: two thermal persons 100% LFL

75% LFL

50% LFL

25% LFL

â–Ş Relatively small heat source results in entire space becoming well mixed â–Ş And no more flammable region across floor


Dispersion of leaked refrigerant Behaviour of mixing of a release from indoor unit with low airflow

1 min

▪ Charge of 50% × LFL × room volume = approx. 800 g ▪ Very effective mixing

6 min

½ min PV


Dispersion of leaked refrigerant ▪ Movement of personnel helps to dilute release ▪ Measurements show effect of gentle strolling following a release ▪ After approx. two to three minutes almost fully dilutes layer

300 g at 50 g/min from 1 m

Strolling

0.10

Test no. 25

Walking around

0.08

Concentration [kg/m3]

Personnel

150 g at 40 g/min from 0.2 m

0.06

0.04

LFL

#1: 1m in front, 0.7m #2: unit centre, inside

#3: unit centre, floor 0.02

#4: 0.5m in front, floor #5: 1m in front, floor #6: 0.5m to left, floor

0.00 11:34:05

11:36:14

11:38:24

11:40:34

11:42:43

11:44:53

11:47:02

#7: 1m to left, floor


Dispersion of refrigerant (from housing) Until recently standards assumed, either ▪ Quasi-homogenous mixture in room, ▪ Or severely stratified layer in room based on severely pessimistic assumptions

GIZ Proklima C4 and EULF project looked into distribution in room and associated assumptions ▪ In addition to leak hole size, effect of enclosure geometry, airflow, etc.

Mean floor conc at release end [g/m3]

Helped identify and generate new formulae, tuned more to equipment characteristics 40 fan off/low h_rel/big A_o 30

fan on/high h_rel/small A_o

20

fan off/low h_rel/small A_o

10

fan on/high h_rel/big A_o bucket case

0 0

20

40

60 80 100 max C_exit [g/m3]

120

140


New proposals for EN 378 and 60335-2-40 1) Basic measures

A) Improved sys tightness

B) Unit airflow

Continuous

With active response

2) Apply riskreducing measures

C) Limited released mass

Passive

D) Enclosure special design

E) Proof by test

With active response

In principle, for many cases requirements can be (should be) mixed and matched


New proposals for EN 378 Current proposal – in progress Flam mabil ity class

Access category a

2L

b c

2 and 3

Location classification I

II

III

���� and not more than 200 × LFL

not more than 200 Ă— LFL

���� and not more than 200×LFL

not more than 400 Ă— LFL

���� and not more than 200×LFL

���� and uncapped

���� and not more than 40×LFL

Not more than 130Ă—LFL

b

���� and not more than 65×LFL

Not more than 260Ă—LFL

c

Not more 200Ă—LFL

No charge restriction c

a

���� and not more than 200×LFL

IV

Not more than 130Ă—LFL

���� and not more than 650×LFL

No charge restriction c


New proposals for EN 378 Current proposal – in progress


New proposals for EN 378 & 60335-2-40 Measures to improve system tightness Strength pressure test Currently in -40 Leak tightness test Additional tests and measures

Improved leak tightness

Improved tightness corresponds to lower probability of given leak hole size

Reduces probability of larger holes

â–Ş â–Ş â–Ş


New proposals for EN 378 System tightness confidence level (STCL) ▪ ▪

Intended to be an improvement on “enhanced tightness refrig system” Provides option to consider small leak holes design/ type-tests

construction/ manufacturing storage/ distribution installation operation service/ repair

decommissioning


New proposals for EN 378 System tightness confidence level (STCL) Mitigation measure

Verification of physical protection against external impacts Protection against imposition of external loads Prevent ice build-up Selection of suitable materials Appropriate location and protection of parts Protection against contact with corrosive substances (e.g. water, acid) Protection against contact with corrosive substances (e.g. NH3/chlorides) [understood as “from outside� of the system] Limited vibration / pulsation Limited thermal gradients to prevent shock Strength pressure test Protection against excess pressure Correct jointing Compliance with ISO 14903 No touching of moving surfaces Limit liquid head/ use slow closing valves Tightness confirmation at manufacture Tightness confirmation at storage/distribution Tightness confirmation after installation Tightness confirmation during in-use

Clause in EN 378-2

6.2.1 [New] 6.2.3 6.2.1 [New] 6.2.3 6.2.3 [New] 5.2.1, 5.3.1 [new] 6.2.3.3.9 [New] 6.2.1

Required for STCL 1 2 X X X X X X X X

[New] 5.2.1

X

6.2.12.6 6.2.3.3.8 6.3.2 6.2.5 & 6.2.6 6.2.3 5.2.2.2; 6.2.1 [New] 6.2.11 [New] 6.2.8 6.3.3 [new] 6.3.4.2 6.3.4.5 Part 4

X X X X X (X)* X X X X X X

X X X (X)*

X X X


New proposals for EN 378 and 60335-2-40 Determining leak mass flow â–Ş Leak mass flow needed for (i) SOI test, (ii) min airflow rates and (iii) tests for finding charge limit â–Ş Assumed leak rate affects “everythingâ€? 4 đ?œŒđ?‘™đ?‘–đ?‘ž đ?‘Ąđ?‘™ = 2.6 đ?‘€áˆś đ?‘ Tightness level 1 – assumes a “leak timeâ€? for entire charge

SLTC

Tightness level 2 – assumes a leak rate through a given hole size

State

R32

R143a

R152a

RE170

R290

R600

27 °C

5.4

5.9

17

15

6.6

36

45 °C

3.1

3.4

9.6

9.0

4.0

19

đ?‘šáˆś đ?‘™đ?‘’đ?‘Žđ?‘˜

476 = đ?‘šáˆś đ?‘Ąđ?‘Žđ?‘?đ?‘™đ?‘’ Ă— đ?‘€đ?‘? đ?œŒđ?‘™đ?‘–đ?‘ž

State

R32

R143a

R152a

RE170

R290

R600

27 °C

95

84

30

24

42

9

45 °C

167

147

52

41

67

17


New proposals for EN 378 and 60335-2-40 Development of min airflow rate for room mixing Jet discharge comprising air and refrigerant

Entrained air

Impinging on opposite wall

Jet centreline ( x)

Jet

cen tre lin e

(x) Impinging on floor


New proposals for EN 378 and 60335-2-40 Validation of formulae 0.06

0.06

RACS Cf,max measured [kg/m3]

Cf,max measured [kg/m3]

RACS CRE 0.04

0.02

0.00 0.00

ҧ 𝐶𝑚𝑎𝑥,𝑡2

0.02 0.04 Cf,max calculated [kg/m3]

𝑚ሶ 𝑙𝑒𝑎𝑘 𝑚𝑟 = + 𝑉𝑟𝑚 𝛼𝑥 𝑇 𝑏𝑉𝑜ሶ 𝐴𝑜

0.06

CRE 0.04

0.02

0.00 0.00

0.02 0.04 Cf,max calculated [kg/m3]

𝑉ሶ𝑜,𝑚𝑖𝑛 =

8 𝐴𝑜 𝑚ሶ 𝑙𝑒𝑎𝑘 1/4

𝑚𝑐 𝐿𝐹𝐿3/4

0.06

𝐹1/4 1−𝐹 3/4

𝑚𝑚𝑎𝑥 = 𝐹 × 𝐿𝐹𝐿 × 𝐴 × ℎ

𝑉ሶ𝑜,𝑚𝑖𝑛 =

5 𝐴𝑜 𝑚ሶ 𝑙𝑒𝑎𝑘 1/8

ℎ𝑜

𝐿𝐹𝐿 1 − 𝐹

5/8


New proposals for EN 378 and 60335-2-40 140 g/min, 340 m3/h (xT =floor), F = 0.5 g; 560 g

140 g/min, 510 m3/h; 560 g

140 g/min, 215 m3/h, 30 s; Ao = 0.02 m2

140 g/min, 685 m3/h, 30 s; 560 g; Ao = 0.2 m2


New proposals for EN 378 and 60335-2-40 Initiation of airflow‌ Is minimum airflow sufficient to mix established stratified layer?

What should be the minimum response time to avoid formation of a flammable layer?


New proposals for EN 378 and 60335-2-40 If released mass is less than charged amount, minimum room size can be based on that released mass 1000

not leaked

leaked

For R290 AC, found approx. 15% of charge retained in system until Δp = 0.1 – 0.2 bar

R290 mass [g]

800 600 400 200 0 system off

Main issue is selection of test method and test conditions

system on


New proposals for EN 378 and 60335-2-40 Can use any means to minimise leaked amount of refrigerant, but sensors to detect leak and close valves can be used to prevent further release optional solenoid valve in suction line

not leaked

1000

leaked

R290 mass [g]

800 600unit indoor (e.g., evaporator) 400 200

solenoid valve in liquid line

0 sys on; check temp; shut valve

Retained mass

Released mass

sys off; shut valve

sys on; gas detector; shut valve

Whilst tests found approx. 10 – 20% charge retained in system under “off� mode, with shut-off valves, up to 80% charge can be retained


New proposals for EN 378 (and 60335-2-40) Surr conc test Appliance

▪ Min. 5 sampling points within 0.5 – 1.0 m of appliance ▪ Release at relevant leak rate

max 0.5 m

Sampling points

Acceptance

Offers freedom for manufacturers to enhance design of unit

0.040 0.035 0.030

Concentration [kg/m3]

▪ No concentration to exceed some % of LFL

max 0.5 m

0.025

0.020 #4

#5

#6

#7

#8

mean

LFL

90% of LFL

0.015 0.010 0.005

C_mm 0.000 0

100

200

300

400

500 Time [s]

600

700

800

900

1000


In conclusion For ‘advanced’ surrounding concentration test, simply… ▪ Make holes in the refrigerant circuit ▪ Monitor floor conc surrounding the unit ▪ Acceptance: Cmax should not exceed X % of LFL

(Similar approach now in IEC 60335-2-89 for commercial cabinets) Any charge limiting mechanism, leak detection system, airflow, etc. will function as it does… reliability can be assessed as today

Gas sensors

Offers freedom to manufacturers to enhance safe design of unit


In conclusion Reduces probability of larger holes

optional solenoid valve in suction line

indoor unit (e.g., evaporator)

solenoid valve in liquid line

e.g., R290; h = 2 m


Final remarks Resolving the problem of obstructive RACHP safety standards has been a huge headache! Oh yes

▪ Gradually, improvements have been developing; step-by-tiny-step

What a great idea! We were so wrong! Environmental advantage! My boss told me! Want absolute flexibility

With the increasing interest in A3 and other natural refrigerants and industry recognition of them as serious alternatives, support for revising RACHP standards has grown ▪ Stakeholders opposing such progress are becoming more and more ashamed ▪ Imposing opposition to revised standards illustrates these stakeholders’ unsavoury anti-climate motivation!

A variety of approaches have evolved during this journey Hopefully things will change….


Thank you for your attention!

Profile for Protocolo de Montreal - Brasil

6-Requisitos e Avanços em P&D relacionados à Segurança - Daniel Colbourne  

6-Requisitos e Avanços em P&D relacionados à Segurança - Daniel Colbourne  

Advertisement