In another thread https://www.fiatforum.com/500/460683-my-2016-500c-write-off.html jrkitching noted that it was against a Code Of Practice (COP) to sell used airbags or other SRS components. This is off topic so here's a new thread.
@typecast boy said it is not economic to repair many cars without used SRS components. While I don't think it is good idea I posted a link to a pro sell used components report
http://www.bvsf.org.uk/cms/images/PDFs/MVDA-BVSF airbag report 16 03 2016.pdf
Having read this a bit more I disagree with some of their assertions.
1/ That an electrical continuity check gives confidence a device will work, it won't e.g. if seas have failed and moisture entered or gas escaped from a hybrid device.
2/ That devices don't degrade with time, they do. apart from chemical degradation, physical changes in the propellant can have big changes in performance. e.g. if a tablet type propellant starts to break up due to vibration etc. it has a larger surface area and can burn faster creating higher peak pressures or even detonation.
3/ that airbags are no-longer classed as dangerous good after they are fitted to the car and removed. This is just plain wrong, if you ship a whole car the airbags are considered dangerous goods. As an example an aircraft passenger lifejacket is considered dangerous goods due to the CO2 cylinder. You can't ship even one on a passenger aircraft as cargo even though there is one under every seat. There are good reasons for this.


Overall I think the paper is very biased.


Robert G8RPI

My own distaste for reusing secondhand SRS components is well known; I've commented on this before, here.

The current code of practice for the classification and disposal of motor vehicle salvage is attached below. The COP states explicitly that:

"All SRS components including airbags and seat belts must be properly disposed of in accordance with the manufacturer's instructions - these components must never be resold or reused"

even for vehicles which have been classified as CAT S or CAT N.

The trade in secondhand SRS components on ebay and similar sites is widespread and goes directly against the current Code of Practice.

IMO if it's uneconomic to repair a car without resorting to used SRS components without any provenance, then the car should be scrapped.

g8rpi said:

Overall I think the paper is very biased.

Robert G8RPI

Click to expand...

I'd agree, Robert. Perhaps it's no coincidence that it's published by a group with a vested financial interest in maximising the disposal value of salvage vehicles .

I particularly dislike their claim that the observed failure rate of 1 in 200 in the recovered SRS components they tested is acceptable. These are life-critical safety systems of last resort where any failure could result in death or serious injury; I'd expect the design failure rate to be more like somewhere in the region of 1 in 10^9.

The analysis and risk management of statistically unlikely events where the consequences of failure are catastrophic is notoriously problematic and when vested interests are at stake, it is all too easy for dubious analytical methods to be used to justify a desired result. The paper you linked to is a particularly bad example of its kind.

jrkitching said:

I'd agree, Robert. Perhaps it's no coincidence that it's published by a group with a vested financial interest in maximising the disposal value of salvage vehicles .

I particularly dislike their claim that the observed failure rate of 1 in 200 in the recovered SRS components they tested is acceptable. These are life-critical safety systems of last resort where any failure could result in death or serious injury; I'd expect the design failure rate to be more like somewhere in the region of 1 in 10^9.

The analysis and risk management of statistically unlikely events where the consequences of failure are catastrophic is notoriously problematic and when vested interests are at stake, it is all too easy for dubious analytical methods to be used to justify a desired result. The paper you linked to is a particularly bad example of its kind.

Click to expand...

I'd missed the failure rate
In aviation we use 1in10^9 flight hours for hazardous events. So if we use an 11 year life that's about 1x10^5 hours so the undetected failure rate for the system at installation needs to be 1 in 10,000 (1x10^4) assuming there are no other faults during the life of the car. 1 in 200 just does not cut it.