Stories like this are a Rorschach test for how one views Tesla. This can be a lazy and dangerous design process that shows the company's unprofessionalism or it can be a smart hack to get the job done while minimizing costs and supply chain restraints.
I have no views on Tesla, I know a bit about the cars and the company, but not much. I honestly can't understand how anyone can see this as anything but a lazy and dangerous design process that shows the company's unprofessionalism. MAYBE it's not dangerous, but if they're doing this, what else have they done that no one has found yet? It's crazy that a car that costs this much would have something like that in place from the factory.
I recall reading that basically, unlike most other cars, where a 2018 Toyota Camry XLE can be reasonably expected to have X configuration, every Tesla is pretty much serial-number-specific from a service standpoint, because of the ridiculous number of mid-model iterations Tesla is doing on their fleet.
A big part of their hesitation to allow outside service is probably the fact that any given Tesla may have any number of slight adjustments that require slightly different parts or repair tactics, and they don't really want to publicly admit how many different variations of everything there is.
You can see in these pictures that in addition to them using some stock wood trim pieces... those wood trim pieces differ a bit from car to car. One would hope if the materials used between these cars differ, that their system knows this, in case one approach proves to be defective or dangerous, and they need to track down every car that they employed this hack in.
OMG! This isn't a joke? Part of the reason you want uniformity is so you can learn what kinds of problems your customers are going to run into and your service personnel understand how to resolve them. If every Tesla is essentially unique then they've gone straight back to the 18th century! Wow!
But what if they can iterate and fix the problems on the next car on the line. That's an advancement, as long as all the differences are well documented, it should be a step forward in car manufacturing and repair.
> But what if they can iterate and fix the problems on the next car on the line.
Cars aren't software; to do that with cars without large identical (by intent) production runs (to control for process noise) it would seem like you would need extraordinary process control (and every indication I've seen has that Tesla varies from incumbents in the opposite direction), otherwise you are chasing process noise.
> as long as all the differences are well documented
Yeah, I’m gonna stop you right there. I have serious doubts a company that sees no problem with making a quick Home Depot run for some faux wood trim in order to get a car out the door is meticulously documenting all their little hacks.
The purpose of these pieces of fake wood appears to be to spread out the stress from the strap. I have no idea whether this method is any more likely to fail than a purposely designed part or what the dangers are if this does fail, but that isn't a particularly complicated or demanding task. It is possible that a purposely designed part would provide no real benefit over this seemingly hacky solution. If that is the case, why not go this route if it makes assembling things easier?
I have no idea the risk/reward equation here. I am guessing most people voting and commenting here have no idea the risk/reward equation. My comment was only pointing at that lots of people are going to have an opinion on this that is mostly dictated by their previous assumptions about Tesla as a company.
Reminds me of that time the psychologist kept showing me pictures of my parents fighting. She insisted that they were just blots of ink. I'm not sure what she was playing at.
But that's not failing the test. The point is that you are supposed to see it as A or B, and that will give the test administrator insight into your viewpoint.
I see it from both sides, and I think a person's view of this type of hack all comes down to trust.
Do you trust that some engineers at Tesla needed a solution to a problem, modelled it out, did some materials testing and discovered that this plastic panelling was the right material for this job?
It does look bad because it's obviously not produced by Tesla, but so many parts are outsourced already in car manufacturing, this could be a solid engineering choice that's getting some bad publicity.
Of course, it could also be because some intern was tasked at solving a design problem and slapped this together overnight, and no one ever checked his work, and it went into production.
> Do you trust that some engineers at Tesla needed a solution to a problem, modelled it out, did some materials testing and discovered that this plastic panelling was the right material for this job?
That would be more credible if it was the same in all cars; it is not (see the source).
It not just not the same in each car but the pieces have rough cut edges and each piece is a different length. I would hope a production fix would at least buy a saw with a stop so that it looks standard.
Assuming this really is random consumer crap, and not just an automotive product that is designed to look like something from the cheaper end of Ikea for some reason, I really don't see how this could possibly be seen in a good light. How will this material react to the 10+ years of hot/cold cycles it'll be exposed to? What are its fire properties? Will it decay/mold (if it's an MDF-ish product, which it _looks_ like, though who knows). And so on. There are just too many unanswered questions and the whole issue could be solved by doing things properly.
And if they need to do this (due to part shortages?) what OTHER bizarre substitutions might they be making?
Again, there's an outside chance that this is a proper product that just looks really cheap-DIY-y, but I wouldn't bet on it.
First, they designed this cool octovalve thing so they could use the one heat pump for all the heating and cooling needs of the car.
Second, they have this LCC hanging off the side of it, a top heavy part mounted laterally with no support other than where it's bolted to the octovalve at its base.
Third, they realise there's a problem and need to relieve stress from the mounting point. So there we get the metal strap with a spreader.
Fourth, they realise there's no point getting plastic moulded parts specially made when they can just use bamboo, wood, or fake wood which are not only cheaper but have better vibration damping qualities than the plastic part.
No doubt they'll redesign this part at a later stage, but since it's a contract-manufactured part they need to give the contractor time to roll out the new part and they need to deplete inventory. Will the new design involve modifications to the octovalve? Perhaps the new design will simply have a bracket down the side to provide vertical support?
> This can be a lazy and dangerous design process that shows the company's unprofessionalism or it can be a smart hack to get the job done while minimizing costs and supply chain restraints.
It can be a dessert topping and a floor wax; its a hack, potentially smart depending on the constraints posed to those doing the hack, to get the job done whole minimizing costs and supply chain constraints.
It's also, quite arguably, something evidencing a lazy and dangerous design process at a higher level than the level at which it is a potentially smart hack.
Laziness at one point of a design process could imply a theme, is what I think they're getting at. I still think about this Twitter thread, for example:
> The LCC (Liquid Cooled Condenser) was probably not designed to be mounted horizontally so they added the cinch strap which needed edge protection to prevent crushing the condenser contact surfaces... enter corner molding from Home depot. Will be interesting to see if that corner molding holds up over time. The cheap stuff is mostly air, the better stuff has more plastic content. If you do 3D printing you know what "fill" means..... look how the strap crushed the plastic corner molding in this photo. If it crushes/compresses it more over time.. the strap will be lose and the condenser will shake/possibly stress crack at it's base?
I did like this sarcastic response to it:
> See OP??? The Home Depot wood trim on yours exceeds the factory spec! At least it will not get crushed by the strap because it's not cheap plastic corner trim from Lowe's like others have.
Looking at the part and where it is used, I would guess that the purpose is strain relief to keep the strap off the metal assembly.
There are a couple things I notice when I look at the shortage workaround.
1.Fake wood pieces are longer than the proper plastic piece(Why?).
2.Fake wood pieces are wrapped in green tape, proper plastic piece is not.(Speculation, metal strap edge cuts into fake wood. If so, how long will tape hold up with vibration?)
3.At least 2 different materials were used as a substitute, are these materials equivalent? Is there confidence in the supply chain of these parts that there was not a bad batch that will fail under heat or stress?
4.Overall quality of workmanship. In two of the 3 pictures, the lengths of the 2 fake wood pieces do not match. This points more to a production line workaround than a sanctioned fix.
5.The proper plastic piece has a larger area in contact with the cooler assembly than the fake wood pieces. This will result in more force being transferred to a point and possibly resulting in crushing damage to the assembly.
I work in a facility that assembles heavy off road equipment, we do sometimes use BOM parts if we encounter a shortage. Unless the part is identical in fit and function i.e. using a slightly oversized hose clamp, we will go back and rework the machine so that it conforms to our quality standards before we ship it.
The fact that it's needed in a precisely made complex machine, suggests something is wrong. I think a better question is where's the evidence this is safe?
People have had their car AC blowing out hot air which can be annoying to downright dangerous in parts of the country.
At least one has had his car refuse to start due to battery overheating. I can't for sure say THIS is the reason for any of those issues or that THIS is a dangerous bit unto itself but HVAC generally for an electric car is pretty important to get right given the battery powered system and this particular model's new "Octovalve" Thermal Management setup.
I think there are two sets of burdens of proof here.
The engineer who made this decision has a burden of proof to his boss (and maybe regulators, don't know) to show that this is safe.
The random internet commenter suggesting it is unsafe has a burden of proof to show that either it is actually unsafe, or that at least that Tesla didn't internally meet it's burden of proof.
This seems like a perfectly good solution to me. You're trying to stabilize the cooler, spread the clamping force, and avoid sharp edges and wear. The part chosen seems perfectly suitable for that, albeit slightly unconventional looking.
(I say this as someone who is short $TSLA right now [and a non-practicing Mech E]; this looks perfectly reasonable to me.)
I agree that if you dug into the engineering the math probably works[1]. It's just a weird culture to be out on the floor substituting random materials in a load-bearing, if perhaps not life-safety-critical, part.
1: Compressive strength of wood perpendicular to the grain varies but you can estimate with a healthy safety margin about 200 PSI. 100x - 1000x weaker than metals, but still not zero.
I (pretty strongly) suspect that engineers were involved, rather than this being an assembly line worker floor-level decision.
The same company that erected a factory-in-a-tent (unconventional but also appropriate) to hit aggressive production targets seems like one that would consider and execute appropriate but unconventional substitutions.
I guarantee you that the materiel is not "wood" and is indeed much weaker in all relevant parameters. And I suspect that no one at TESLA has any real handle on the compressive strength of this material.
