Hacker News Reader

Comment by LegitShady | Hacker News Reader

It sounds like the tested vehicles weren't the same as actual production cars. Choosing to cheat on airbags seems like a big deal though.

https://www.reuters.com/business/autos-transportation/toyota...

>The investigation found that the airbag control units used by Daihatsu in airbag tests for some models were different from the ones used in cars sold to the public, including Toyota's Town Ace and Pixis Joy models and the Mazda Bongo.

...

>The misconduct also included false reports on headrest impact tests and test speeds for some models. The investigation found cases of misconduct were particularly prevalent after 2014 and, for one already-discontinued Daihatsu vehicle, went back as far as 1989.

View on HN

We can't trust foreign companies for various reasons. The solution seems to be to do the testing locally, not using vehicles provided by the manufacturer for testing purposes. Which seems to be what happens in Australia: https://www.racv.com.au/royalauto/transport/cars/ancap-faqs....

(The loophole would be using premium parts in early shipments, until testing has commenced, and switching components in later shipments. Not sure if we would catch that in components like airbags. We have already had Toyota recalls due to dodgy airbags.)

> We can't trust foreign companies for various reasons.

US manufacturers have their own history of cheating.

> We can't trust foreign companies for various reasons.

Removing the xenophobia:

“We can’t trust companies for various reasons.”

I guess not, but local consumer protection laws at least provide a stick to beat trust into them when found out.

The US safety test system has the same flaw- safety equipment is not randomly tested after certification.

As an example, it's an open secret that over 40% of all USDOT certified helmets fail to meet the standard: https://ultimatemotorcycling.com/2020/02/05/2020-dot-certifi...

(The EU uses a different standard which includes independent testing of random samples).

Also found a report about another problem they cheated on tests to pass:

"They said the door trim on the affected vehicles had been modified with a "notch" to minimise the risk in testing that the door interior could break with a sharp edge and cause injury to an occupant when the side airbag deployed in an accident."

https://www.reuters.com/business/autos-transportation/japans...

The strange thing about that modification is that it appears on the surface to be a legitimate safety improvement. Could the underlying test be subpar?

it sounds like it wasn't in the production vehicle.

Right, but I’m wondering if the notch was actually useful for legitimate passengers in the event of an accident.

why would that call into question the underlying test? Rather it calls into questions the design decision of the daihatsu team to know they could create a safer product but decide to not do so, and fake it on the safety tests.

Stupid question: why didn't they add that part to the actual production cars? It doesn't sound like a complex system or problem (compared to say, reducing emissions versus just cheating in emission testing) especially since they already seem like they know exactly what they had to fix.

Even then I'd understand if it was a one time thing (rush in production or design, no time to redesign) but this is systematic and across a wide range of models. I guess the other models in question had more deeply rooted design issues that can't be fixed with a less intricate tweak?

Reply on news.ycombinator.com