PCS failure rates by year...is it just a matter of time?

[Metalwolftensei]

Is there a way to see if the PCS is failing if your household setup limits you to 24 amps? Since ownership I have always limited my charging to +24..

Sponsored

 

[Outdoors]

Is there a way to see if the PCS is failing if your household setup limits you to 24 amps? Since ownership I have always limited my charging to +24..

Easy suggestion would be to check plugshare.com for an AC charger nearby. A DC fast charger wouldn't be what would help you if one is trying to test the AC side of the PCS.

Toggle the filters for what you are after amp wise.

 

[Sjohnson20]

Is there a way to see if the PCS is failing if your household setup limits you to 24 amps? Since ownership I have always limited my charging to +24..

not a guarantee but it might say something in the service mode if it had a PCS error.

 

[TyPope]

I think we all need to file nhtsa especially those that had the issue. Grok says 40% of owners are having issues… https://www.nhtsa.gov/report-a-safety-problem

We could end up stranded in unsafe conditions if our vehicles won’t charge while traveling. We shouldn’t be waiting to either fix it free after it happens or pay $7-6000 dollars when it happens after warranty. Sounds like it’s not an if but a when situation.


My truck hasn’t failed yet, but I’m genuinely concerned it might. I’m hearing from a lot of other Cybertruck owners that their Power Conversion System has failed, and now that people are getting past the 50,000-mile warranty mark, repairs are reportedly costing over $6,000.

From what I understand, when it fails the truck suddenly can’t charge, which essentially makes it unusable. That’s what worries me most. If this happens at 50,001 miles, you’re suddenly responsible for a $6,000 repair on what appears to be a known issue.

I’m getting close to that mileage, and I’ve seen reports of failures happening anywhere from 5,000 miles to much higher. It makes me uneasy knowing this could happen at any time or even worse wrote after my warranty ends!!!

I honestly wish there were a way to proactively test or address it before the warranty expires.

Does this seem like something that should qualify as a recall if it’s a known issue that renders the vehicle unable to charge and effectively unusable? What would you recommend?

I've only heard of a couple having issues. Maybe a handful. It's out of my control so I'm not going to worry about it. Will that bite me in the ass someday? Maybe. That's a problem for future me. Today me is enjoying my truck.

 

[Known-Background]

I've only heard of a couple having issues. Maybe a handful. It's out of my control so I'm not going to worry about it. Will that bite me in the ass someday? Maybe. That's a problem for future me. Today me is enjoying my truck.

Zen

 

[btcrealm]

Can confirm my August 2024 Cyberbeast PCS controller failed on me March 10th. It was plugged in and charging (Tesla L2 standard charger) when it failed.

 

[Known-Background]

This should go class action

 

[koolio]

Does anyone have data that shows when to expect it to fail?

I’m wondering is it number of charge sessions, amperage of the session, and duration? I’m guessing it’s not mileage dependent per se, but one would expect higher mileage = more charging sessions thus more “wear” on the PCS before failure.

Asking because I’m wondering when this will occur on my Nov 2024 built AWD. I only have 10k miles and 90% of charging is at home (mostly at 24-32amps bc I use Charge on Solar, but also at 48a when needed).

 

[noobla3]

Does anyone have data that shows when to expect it to fail?

I’m wondering is it number of charge sessions, amperage of the session, and duration? I’m guessing it’s not mileage dependent per se, but one would expect higher mileage = more charging sessions thus more “wear” on the PCS before failure.

Asking because I’m wondering when this will occur on my Nov 2024 built AWD. I only have 10k miles and 90% of charging is at home (mostly at 24-32amps bc I use Charge on Solar, but also at 48a when needed).

I charge at 24 amps at home also. I think most of the reports are charging at 48 amps.

 

[Jager]

No one outside Tesla knows the frequency of PCS failure or any causal pattern(s) that might exist.

The anecdotal examples on the forum here are all over the place. It's been suggested that higher L2 charging amperages might be at play, presumably because of the additional heat that would involve. But it's a pretty straightforward engineering exercise to understand component temperature at various L2 power levels and Tesla's engineers have historically demonstrated a very high degree of technical acumen. So that kind of obvious engineering mistake doesn't make a lot of sense.

And AFAIK Tesla service centers are not quizzing customers with failed PCS units as to what their charging habits have been; and while the vehicle permanently logs AC vs DC kWh's charged, it doesn't, AFAIK, log the actual power levels.

Given all that, I'm disinclined to think higher L2 power levels are the problem. If I could charge at higher than my present 32 amps (40-amp branch circuit), I would.

My guess - and that's all it is - is that the PCS2 failures are related to heat-related component failure, but not associated with L2 power levels.

I might be entirely wrong...

 

[Eka]

No one outside Tesla knows the frequency of PCS failure or any causal pattern(s) that might exist.

The anecdotal examples on the forum here are all over the place. It's been suggested that higher L2 charging amperages might be at play, presumably because of the additional heat that would involve. But it's a pretty straightforward engineering exercise to understand component temperature at various L2 power levels and Tesla's engineers have historically demonstrated a very high degree of technical acumen. So that kind of obvious engineering mistake doesn't make a lot of sense.

And AFAIK Tesla service centers are not quizzing customers with failed PCS units as to what their charging habits have been; and while the vehicle permanently logs AC vs DC kWh's charged, it doesn't, AFAIK, log the actual power levels.

Given all that, I'm disinclined to think higher L2 power levels are the problem. If I could charge at higher than my present 32 amps (40-amp branch circuit), I would.

My guess - and that's all it is - is that the PCS2 failures are related to heat-related component failure, but not associated with L2 power levels.

I might be entirely wrong...

I think it is the AC power components on the bottom of the PCB. The power components may have been hanging on to the bottom side and their solder remelted as the top side chips were soldered in the reflow oven. This allowed them to become less level as they are just hanging on the melted solder. The fix is to have the power components on the last side of the PCB soldered. The other issue is they might not have enough copper in the PCB to draw heat away. A thicker copper layer is the solution. If somebody could get me the chip ID, or power and it's internal construction, I could rough guess the PCB thermal area needed. I thought they were too close, but advances in power FETs. Using a PCB inductor means they need a high switching frequency. That could add lots of thermal load if the gates are not properly driven.

Video in this post shows evidence the components are not level.
https://www.cybertruckownersclub.co...-it-just-a-matter-of-time.55072/post-30778754

(I might have designed and made more than a few high power PCBs in the past.)

 

[mongo]

I think it is the AC power components on the bottom of the PCB. The power components may have been hanging on to the bottom side and their solder remelted as the top side chips were soldered in the reflow oven. This allowed them to become less level as they are just hanging on the melted solder. The fix is to have the power components on the last side of the PCB soldered. The other issue is they might not have enough copper in the PCB to draw heat away. A thicker copper layer is the solution. If somebody could get me the chip ID, or power and it's internal construction, I could rough guess the PCB thermal area needed. I thought they were too close, but advances in power FETs. Using a PCB inductor means they need a high switching frequency. That could add lots of thermal load if the gates are not properly driven.

Video in this post shows evidence the components are not level.
https://www.cybertruckownersclub.co...-it-just-a-matter-of-time.55072/post-30778754

(I might have designed and made more than a few high power PCBs in the past.)

The FETs dump heat to the liquid cooled cold plate. AC side looks like ST SCT040HU65G3AG HU3PAK parts.
https://www.st.com/resource/en/tech...g-and-thermal-behavior-stmicroelectronics.pdf

 

[Eka]

The FETs dump heat to the liquid cooled cold plate. AC side looks like ST SCT040HU65G3AG HU3PAK parts.
https://www.st.com/resource/en/tech...g-and-thermal-behavior-stmicroelectronics.pdf

From the PDF file:

If components must be placed on both PCB sides, the side with HU3PAK must be processed last, as wetting forces could not hold the HU3PAK package during a second reflow.

As suspected.

The package looks to be designed to dump heat out via a heatsink/cold wall on top of it like Tesla uses, versus using the PCB copper traces. The copper traces only seam to be used for current flow and signals.

Section "3: Heat sink attachment to package" tells me using a thermal compound that can be totally squished out is electrically bad. Can this be the right package? Where is the needed electrical isolation coming from? The Munro video shows use of what looks like a caulk style gap filling thermal compound. They don't reliably provide electrical isolation. Also many of them are not as good conductors of heat as some of the thermal pad style. OK, looked at the Munro video again. There is a black surface on the electronics side of the cold plate. It might be the electrical insulator.

While it may seam like a very high heat conductivity compound that stays fluid or semi fluid could be used, they often contain electrically conductive particles. In an environment where vibration is common, they eventually end up everywhere. Been there, done that, suffered the consequences. Poor PCB became an arc path as it vaporized...

 

[mongo]

From the PDF file:

As suspected.

The package looks to be designed to dump heat out via a heatsink/cold wall on top of it like Tesla uses, versus using the PCB copper traces. The copper traces only seam to be used for current flow and signals.

Section "3: Heat sink attachment to package" tells me using a thermal compound that can be totally squished out is electrically bad. Can this be the right package? Where is the needed electrical isolation coming from? The Munro video shows use of what looks like a caulk style gap filling thermal compound. They don't reliably provide electrical isolation. Also many of them are not as good conductors of heat as some of the thermal pad style. OK, looked at the Munro video again. There is a black surface on the electronics side of the cold plate. It might be the electrical insulator.

While it may seam like a very high heat conductivity compound that stays fluid or semi fluid could be used, they often contain electrically conductive particles. In an environment where vibration is common, they eventually end up everywhere. Been there, done that, suffered the consequences. Poor PCB became an arc path as it vaporized...

Not sure on what material they used between the package and cold plate.

 

[btcrealm]

This was just posted in the other thread about this.

https://www.notateslaapp.com/news/3...rive-inverters-how-to-check-if-youre-affected

Not exactly the same thing as not being able to charge but still the same components maybe? Most of ours could still drive.

[EDIT]
I had reposted this from someone else post thinking it was related, apparently unrelated and old. It just references PCS failure.
[/EDIT]

Sponsored