14 replies to this topic

[mariner]

Rarely does the design of front suspension bottom joints get a media expose but here is one from the BBC 

 

https://www.bbc.co.u...es/c20nr3zdppjo

 

I admit to very mixed views on articles like this , primarily here by the headline "JLR electric car whistleblower sacked" In fact the issue is all about the Vietnamese Vinfast electric car which has no connection with JLR except consulting work by a JLR sub. I think the idea was "clickbait" to make people think it was a JLR car that is unsafe?

 

There are clearly some issues with Vinfast EV's per US reports but a suspension link beakage on test track , while not welcome is what rough rod testing pre production is meant for I think.

 

 

Obvously the attempts to balcklist a engineer by JLR does not reflect well on them.

 

Maybe using consultants to do your design work is not ideal ! 

 

 

 

[Greg Locock]

"Maybe using consultants to do your design work is not ideal ! " The entire automotive world would grind to a halt if consultants weren't used. I was one of the few engineers in our section at Lotus Engineering who routinely worked on Lotii, mostly we did GM product and Volvos. My friend Phil was mostly Carlton, tho I did a driveline issue on that (powerhop, which for some reason I specialised in)

 

As to the suspension failures. At the start of a program an intensely dull document called DVP&R is wheeled out. This specifies what testing is done, and in this case specifically how many laps of the rough road tracks must be completed. This is accelerated testing, so you can't compare track distance vs real world. The roads are rough, in 6 weeks we can test a car to its expected life. Some of the tests make some drivers throw up.

 

So, if the part failed before its DVPR life before production it needed analysis and some design change. But if the part did not fail before its specified life then no problem.

 

Bear in mind that once a vehicle goes into production the production part might be (a) defective or (b) had design changes by somebody or other using Engineering Judgement (EJ) to avoid a re-test. Examples of by the supplier that include changing the rubber compound on a boot so it cracked before it should, cos it was cheaper, or outsourcing the entire production to a different factory, 

 

Conflating this with cars driving in reverse while unoccupied is weird, that's NOT his area of expertise.

[Slumberer]

Greg, how much more do you learn with "real world" testing, plodding round and round a test track, or even on real roads, compared to putting parts on a rig?

I know that in my work (not automotive) we always treat lab data as merely a starting point and not reflective of real word results, but we have nothing like the resources of an automotive manufacturer.

[Greg Locock]

I don't know why Test Engineering run full car rigs for durability to be honest. Many of the component level rig tests, such as on wheels, are just widely accepted industry practice, specified by SAE.

[Mark A]

Rarely does the design of front suspension bottom joints get a media expose but here is one from the BBC 

 

https://www.bbc.co.u...es/c20nr3zdppjo

 

I admit to very mixed views on articles like this , primarily here by the headline "JLR electric car whistleblower sacked" In fact the issue is all about the Vietnamese Vinfast electric car which has no connection with JLR except consulting work by a JLR sub. I think the idea was "clickbait" to make people think it was a JLR car that is unsafe?

 

There are clearly some issues with Vinfast EV's per US reports but a suspension link beakage on test track , while not welcome is what rough rod testing pre production is meant for I think.

 

 

Obvously the attempts to balcklist a engineer by JLR does not reflect well on them.

 

Maybe using consultants to do your design work is not ideal ! 

 

The original article annoyed me as it was very much trying to make people assume it was an issue at JLR.

 

I guess someone has had a word as the article title is now "Electric Car whistleblower sacked by JLR"

[Mark A]

I don't know why Test Engineering run full car rigs for durability to be honest. Many of the component level rig tests, such as on wheels, are just widely accepted industry practice, specified by SAE.

 

I've always found 6-DoF tests to be very useful, best insight into actual durability for systems with good repeatability, especially if using a proper RLD signal to drive the rig, 

 

The issue with component tests is you lose the effect from system interactions.

 

The opposite issue to answer Slumberers comment, is if you run a driven vehicle test it's not repeatable, weather can impact the speed a car can run round a track, hence the loads will change, so, for example, a test run in the summer has the potential to increase loads significantly over one run in the winter, the climatic and corrosive effects will be opposite, running a car on a MTS full vehicle rig gives that repeatability.

 

I remember a test at Bovington, 1 lap would be different to another, one driver could break the car, another wouldn't, purely by the line and speed they took. Winter the surface was softer so potholes were less damaging vs the runner when the edges were firmer.

[Greg Locock]

"Obvously the attempts to balcklist a engineer by JLR does not reflect well on them."

 

Really? If you worked for a consultancy and leaked to the press do you think you'd get a bunch of roses? Tata owns JLR.

[Magoo]

Not that I ever looked super closely, but the VinFast vehicle development timelines looked iffy to me. 

 

When the first cars arrived here in the USA and they were savaged by the press, my suspicions were sorta confirmed. 

[Greg Locock]

CR say "The company prides itself on rapid vehicle development, aiming to outpace legacy automakers with just two- to three-year timelines from design to production. That is quite ambitious, when established corporations take about twice that time. Legacy automakers also benefit from deep knowledge bases for all manner of benchmarks, from vehicle dynamics and safety equipment to small things like appropriate door weight, steering resistance, and feel for controls. "

 

That depends on what they mean. Pens down to production in 2-3 years is doable, given that you typically have protos of various forms running before the design is finished. If they mean start of production design to Job#1 in 2 years, yeahnah, and go to woe (project funding assigned to Job#1) in 3 years is pushing it unless you have a lot of carryover parts. Tooling for the IP is 18 months for a start (it's the first piece of metal ordered) and there's a queue for the machining. However it's better than it was, in the bad old days when it was fun it was not uncommon for a fundamental issue to cause delays into production for a couple of years, or even program cancellation. For instance the XJS spent the first two years (?) of powertrain NVH development with an axle stand or equivalent holding the trans in position as they couldn't get the design of the trans mount right. Which is funny, if they'd read my report on driveline boom in the Sherpa Van they'd have known what to do. NIH syndrome.

 

Oh I found the thread that kicked all this off https://www.reddit.c...vinfast_ask_me/

 

There's an odd statement

 

The knuckle snaps off in an extreme event, like hitting a pothole at 55mph under full lock breaking scenario…

 

You can tell from the spelling he's an engineer, but square edge pothole is a 30mph event, not 55! Usually after 3 runs the wheel is just hanging on

 

 

in a crash scenario the subframe is designed to decouple by making the rear fixings as weak as possible…so in a front crash it meets the regulatory requirements

 

Yup, SOP, the engine and trans (or whatever) dives under the floor rather than breaking people's legs. eg BMW X5 but it's been done for decades previously

Edited by Greg Locock, 30 December 2024 - 02:03.

[Greg Locock]

".So to avoid some durability issues we decided to increase the shore hardness...At shore hardness above 67-69 you take a bush from its linear region into its dynamic region and it starts snubbing...(also has other factors but in this case I have simplified it)...This improves the durability of the bushes, but stiffer also means more load is transferred to the vehicle itself. Isolation is reduced..."

 

Some of that is true, and some is not, the bolded bit in particular is nonsense.

[Mark A]

Just looked at the link you posted Greg.

Good grief, he said he worked for TTL and compares the Vinfast test results to JLR targets, must have taken all of 10mins for them to work out who he was.

[just me again]

Stellantis has problems with screws that sheer between the lower ball joint and lower A arm in the 3008ev front suspension!

In Denmark they have not made a total recall, but are instead fixing it in a much slower tempo

In the Video you can see a bend suspension at 45 seconds and the screws at 2min 45sec.

https://youtu.be/wS1...8gPknFQ3Txh84JO

[Mark A]

When you say 'fixing it in a much slower tempo' do you mean the speed at which the torque is applied?

This can be a legitimate solution if tightening it at a higher speed results in an incorrect clamp load due to friction etc in the joint.

[Magoo]

[cbo]

Stellantis has problems with screws that sheer between the lower ball joint and lower A arm in the 3008ev front suspension!

In Denmark they have not made a total recall, but are instead fixing it in a much slower tempo

In the Video you can see a bend suspension at 45 seconds and the screws at 2min 45sec.

https://youtu.be/wS1...8gPknFQ3Txh84JO

They have grounded the cars in Denmark. In Norway - and assumingly other countries - people can continue to drive their cars until recalled and fixed.

A bit odd - a front wheel falling off at 130 km/h while mom drives the kids home from daycare seems like a risk not worth taking for Stellantis.