Front drive unit failure at 40 miles

[Deleted member 6231]

Not-so-early Tesla Model S's had lots of drive unit failures, too.

I know there aren't that many trucks built yet, but I find it interesting that most serious failures so far seem to be only one vehicle reporting it. If it is a single failure of each type, then it sorta falls under the sh!t happens category, with something like a ~ 1/2000 chance (which is high, but not astronomical).

It would be interesting to see the internal report from Rivian. I'm curious, was it a motor manufacturing fault? A fault on Rivian's side? Something happened in shipping?

The rear drive units in my 2013 and 2015 Tesla Model S failed the day after I took delivery.

Sponsored

 

[Zoidz]

I think they call it "turtle" mode. Funnily enough, the stuff that will actually break an EV and prevent it from moving are not detected by software … if you snap a half shaft, all it could do is report the wheel sensor is faulty and ABS would be disabled. : )

I get what you're saying, but this example is something I have experience with in software in factory production lines. A good programmer can detect a snapped shaft quite easily. You normally get a zero to slow pulses from the rotation sensor and moderate to high inverter current when starting from a stopped state. With a broken half shaft on a vehicle, you would get the same wheel pulses on all four as the vehicle starts to move, but low current from that motor with normal current from the others. If the wheel was spinning due to low traction, you would get fast pulses and low current on one wheel compared to the others.

In production lines, we have to program for these anomalies for product, equipment and life safety. Get this wrong, and you have a 24" wide 1/8" thick steel strip flailing out of a steel rolling mill. Not pretty. Here's what happens if your software does not detect such problems.

 

[the long way downunder]

I get what you're saying, but this example is something I have experience with in software in factory production lines. A good programmer can detect a snapped shaft quite easily. You normally get a zero to slow pulses from the rotation sensor and moderate to high inverter current when starting from a stopped state. With a broken half shaft on a vehicle, you would get the same wheel pulses on all four as the vehicle starts to move, but low current from that motor with normal current from the others. If the wheel was spinning due to low traction, you would get fast pulses and low current on one wheel compared to the others.

In production lines, we have to program for these anomalies for product, equipment and life safety. Get this wrong, and you have a 24" wide 1/8" thick steel strip flailing out of a steel rolling mill. Not pretty.

Sure, it can be coded, if there's a sensor or even an interpolation as you describe. But the thing that stopped this R1 was a so-called "plausibility" error or a sensor code that may or may not be real. The simplest case would be when the R1 goes into 12 mph limp mode if a gear tunnel door is ajar … if the driver says "no, it's not, the sensor is broken" or "yes, the door just got ripped off on a tree trunk" the vehicle stays in limp mode, tough luck, you're 200 miles from civilization … good luck … better to make you drive for 10 hours than risk the unknown. It really must be possible to override all the "nanny" modes and all the error messages. If there are workarounds
(e.g. three cold starts or "Up, up, down, down, left, right, left, right, B, A, Start") … those should be published in the vehicle, not hidden in some "let me talk to my supervisor" roadside support call that might not even be called give or take cellular coverage, phone battery or the imperative/emergency to move the vehicle, not work through cost-saving corporate phone menu systems … "press 3 to continue English" … )

 

[the long way downunder]

The rear drive units in my 2013 and 2015 Tesla Model S failed the day after I took delivery.

there's been more than a few S Plaid owners report complete failures (from computers to drive modules or software errors with unknown causes) … hopefully the ex-Tesla people at Rivian are treating quality seriously … my R1 is marvelously well built and recording all these defects should not be taken as a representative sample … people with problems come to forums, given the production numbers for Q1 and Q4, so far, so good. Could be better, could be far worse and still be better than Tesla or Ford …

 

[ironpig]

Coincidentally, today, I was wondering about this very issue, what happens if there is a motor failure. Specifically, I was wondering if the vehicle would/could disconnect the failed motor and allow one to proceed, using only the good motors. It sounds like that is a definite maybe.

Been interested to learn more about this as well. Since we know it can use fewer motors in conserve mode I had hoped maybe it would be intelligent enough to perform with just from or tears in a failure.

But I’m sure they just have it brick itself to be safe.

One downside to the quad motor set up is that we have twice as many motors that can fail. Time will tell if this is a bigger issue. A lot of us with the early 2013 and 2014 Model S had our electric motors replaced.

 

[ironpig]

Not-so-early Tesla Model S's had lots of drive unit failures, too.

I know there aren't that many trucks built yet, but I find it interesting that most serious failures so far seem to be only one vehicle reporting it. If it is a single failure of each type, then it sorta falls under the sh!t happens category, with something like a ~ 1/2000 chance (which is high, but not astronomical).

It would be interesting to see the internal report from Rivian. I'm curious, was it a motor manufacturing fault? A fault on Rivian's side? Something happened in shipping?

My 2014 Model S driver unit failed and was replaced in the first year. For what it’s worth, that repair was relatively quick and straight forward. Never had any issues after that. But supply chain issues could plague Rivian If they need to replace a lot of them.

 

[kylealden]

The executives instructing the developers are exceeding their legal boundaries by disabling the vehicle to suit corporate policy.

You might be interested in reading up on product liability law. "Allow the end user to do whatever they want" may be not only unsafe, but illegal or subject them to massive civil liability. A clickthrough dialog isn't going to change that.

Rivian has a responsibility to take reasonable steps to keep you from killing yourself in a vehicle that is unsafe to drive. "Right to repair" and "completely ungoverned performance" are very different things.

 

[Zoidz]

(e.g. three cold starts or "Up, up, down, down, left, right, left, right, B, A, Start") …

We need a cheat code sticky!

 

[zipzag]

So....if this happens at say....60 mph.....what happens??

Probably the same. It will let you pull over and then it dies. Most/all EV makers strive to have this sort of failure procedure.

 

[johnnylawson]

I can chime in with my 2013 RAV4 EV which is a model s drive unit in reverse. I've had it replaced at 50k and 60k miles. Luckily under warrenty both times. Got 108k on it now and holding my breath until R1S is delivered so I can off load the RAV.

It's a well known issue with those model s drive units and leaking coolant getting into them - at least on the RAVs.

So happy about the huge Rivian warrenty that will take away my concerns about the motors and drive units.

 

[Monkey]

Percentage of delivers with major issues continues to rise. Gotta be above 5% now.

New manufacturer making an entirely new truck and ramping production. Yeah, it happens. Tesla, VW/Porsche, GM... They all went through this with each new EV model. It's part of the adventure of being one of the early adopters. I got to see my first Hummer EV in the wild the other day (since the reveal prototype at Barrett-Jackson). At my GMC dealership while dropping off one of our company trucks. It was sitting there in one of their new EV bays. I asked the service writer if they were prepping it for delivery and he said "unfortunately no... battery issues." At least they caught it in PDI, but that means someone is missing their delivery. I guess that's better than driving off with it and having it towed back.

 

[MountainBikeDude]

This drive units are Rivian designed or off the shelf???

Bosch

 

[godfodder0901]

Bosch

Only the motors are Bosch. The drive units are Rivian in-house.

 

[SASSquatch]

They did offer to rent me a car. I declined the offer as this is just a fun truck and not mission critical to my day to day.

+1 for the use of "mission critical"

I see you...

 

[cardad]

I'm not thinking of something elegant like disengaging a faulty motor, I'm thinking with one corner of the vehicle ripped off in a collision or while off-roading, resting on three wheels (or two) with the other corner dragging on the ground and no coolant, the vehicle must be designed to respond to the driver and keep going if at all possible.
The executives instructing the developers are exceeding their legal boundaries by disabling the vehicle to suit corporate policy. This "right to repair" disaster is getting worse. Tesla has talking about the robotaxi as being something the customer pays to "own" but never owns, only leases and is obligated to return after some period, no matter what, non-negotiable. It's analogous to people being forced to rent because they don't "qualify" to get a mortgage … they can somehow qualify to pay $2000 to rent, but not $2000 to pay a mortgage and eventually own their home … somehow "sentenced" to be in perpetual debt and own nothing. (I'm not the first to observe.) Hopefully Rivian doesn't end up like Tesla.

Dubious analogy here with renters and qualifying for a mortgage. Seems like a strange dig to throw into your commentary on a car. Just like renters can find landlords with low underwriting standards you can also find lenders (non-QM and owner financing) where the underwriting standards will be more flexible for a higher cost. This isn't that uncommon and it's a question of whether someone is willing to pay a premium to "own" when they could rent for much less cash out of pocket and maintain their liquidty. Not sure you understand the market here to be commenting... as you can just as easily be held hostage by the equity in your house (i.e. throwing money into it for repair, maintenance, upgrades, HOA, etc. etc.) vs. renting and having more flexibility on how your cash is being utilized.

Sponsored