How Do I Responsibly Report a Powerwall 2 Hardware Safety Defect?

[mudder]

Hello, first time poster with an outrageous claim, so I'll cut right to the chase:
I have identified a hardware design defect with the Tesla Powerwall 2 that violates IEC 62109-1, which Powerwall is certified to adhere to. The violation isn't borderline, and it poses an electrocution hazard under certain conditions.

I have attempted to responsibly report this issue to Tesla, but their support system is non-responsive/inept/terrible. Does it even exist at all? Anyone have any tips on how to report this issue responsibly to Tesla, before I publicly disclose it? It would appear that Tesla's bounty program doesn't apply to hardware, but I've emailed their team anyway... and received no response. It's just impossible to get any response from them.

So I guess roast me?...
...or maybe let me know how I should get in touch with Tesla's hardware team. Surely there are better avenues than tweeting @elonmusk?

More about me:
-I'm an electrical engineer specializing in power conversion.
-I'm also technically a master electrician, as it simplifies some paperwork.
-As a hobby, I've developed (and now sell) a 10.4 kWh drop-in lithium conversion kit for the G1 Honda Insight. The original gas engine is retained, too, so it's still technically a hybrid.

 

[Matt-FL]

Maybe try calling their support number: 1 888-518-3752. The only trick I can provide in addition is to call from a number not associated with a Tesla account. You get different prompts where you can (actually) reach a real human being. Not possible if you call from a Tesla associated account (at least not that I have found).

 

[MP3Mike]

So I guess roast me?...
...or maybe let me know how I should get in touch with Tesla's hardware team. Surely there are better avenues than tweeting @elonmusk?

Or you could just report it to CPSC and hope they take it up with Tesla: Home - SaferProducts

 

[Vines]

Hello, first time poster with an outrageous claim, so I'll cut right to the chase:
I have identified a hardware design defect with the Tesla Powerwall 2 that violates IEC 62109-1, which Powerwall is certified to adhere to. The violation isn't borderline, and it poses an electrocution hazard under certain conditions.

I have attempted to responsibly report this issue to Tesla, but their support system is non-responsive/inept/terrible. Does it even exist at all? Anyone have any tips on how to report this issue responsibly to Tesla, before I publicly disclose it? It would appear that Tesla's bounty program doesn't apply to hardware, but I've emailed their team anyway... and received no response. It's just impossible to get any response from them.

So I guess roast me?...
...or maybe let me know how I should get in touch with Tesla's hardware team. Surely there are better avenues than tweeting @elonmusk?

More about me:
-I'm an electrical engineer specializing in power conversion.
-I'm also technically a master electrician, as it simplifies some paperwork.
-As a hobby, I've developed (and now sell) a 10.4 kWh drop-in lithium conversion kit for the G1 Honda Insight. The original gas engine is retained, too, so it's still technically a hybrid.

One way to verify that you are correct is to bring your concerns to the listing organization that is responsible for the listing.

Despite your credentials, it is not always obvious how products maintain compliance with given standards unless you are part of the technical committee responsible for the standard, or were part of the discussions with the NRTL.

Which section of the standard do you feel the device is out of compliance with? Is this concern in relation to charging the batteries from the grid?

 

[mudder]

Thanks for the advice so far. My intended goal here is to raise this issue directly with Tesla, so that I can understand how exactly they certified this product. As you may know, many standards governing bodies allow in-house certification. For example, when I worked at National Instruments, we had an in-house UL-approved certification team. This allowed us to test our products in hours/days, instead of weeks.

An established member here on teslamotorclub has contacted me privately to discuss the particulars. If they agree the issue might exist, they plan to forward my findings to their Tesla contacts.

 

[CrazyRabbit]

Just state what you found, you tried.

 

[gsmith123]

Just state what you found, you tried.

I'm also a curious hah.

The motivation for reporting software security defects privately to companies is to they can fix it before attackers use the flaw and cause harm to people.

That motivation doesn't apply here at all, does it? It's a hardware flaw. I assume the risk here is under certain circumstances it poses a hazard to users, but this isn't an attack vector that a malicious actor could take advantage of.

Given that: why not write up a factual, non-clickbaity analysis and share it here? You've done your due diligence to let Tesla fix it preemptively. Sharing it poses zero downside to users of the Powerwall and it motivates Tesla to fix it. The only drawback is Tesla might be.. annoyed? People share hugely misleading FUD about Tesla all day long on the internet. They'll be OK

 

[Proppilot]

If it is a real hazard then you should report it to the certifying body and the electrical safety authority. Relying on a manufacturer (any of them) to react quickly to something is futile.

 

[mudder]

In my (as yet unanswered) correspondence with Tesla, I mentioned I would publish my findings on JAN09, with or without their feedback. Whether or not it made sense, I suppose this was a professional curtesy. Regardless, since I've already made that offer, I intend to stick to it... even though I doubt they respond.

I'd love to collect more data from anyone who has disassembled their Powerwall 2 unit. If you'd like to help out, please post photographs of both the inverter PCBA and the aluminum bracket in the following orientation:

 

[father_of_6]

Other posters have made some sensible suggestions. Mine - not so much.

I suggest you do something to *really* get their attention... like electrocute an elephant with your powerwall.

 

[jjrandorin]

Other posters have made some sensible suggestions. Mine - not so much.

This particular niche of TMC is still pretty darn courteous, quite a bit like "old internet", and I actually work very hard to keep it that way, so that people are comfortable sharing here but there is not a lot of "signal to noise ratio" stuff in this small corner of the site.

I say that in response to your post (which I intend to leave for right now) as "guidance" that I dont intend to let this spin up into anything resembling what goes on in the rest of TMC.

 

[mudder]

I suggest you do something to *really* get their attention... like electrocute an elephant with your powerwall.

Not as interesting, but I was able to capture a pretty cool corona discharge when I applied a 950 volt test signal across the defect area. FYI: 950 volts is several times less than the required overvoltage handling requirements.

This particular niche of TMC is still pretty darn courteous, quite a bit like "old internet", and I actually work very hard to keep it that way, so that people are comfortable sharing here but there is not a lot of "signal to noise ratio" stuff in this small corner of the site.

I say that in response to your post (which I intend to leave for right now) as "guidance" that I dont intend to let this spin up into anything resembling what goes on in the rest of TMC.

I certainly miss the "old internet". I enjoy the internet's few remaining "small corners".

 

[mudder]

Update:
Based on the feedback I've received so far, I went ahead and filed a formal complaint with UL's Market Surveillance team. They specify a two business day response time, so we'll see what happens next week.

 

[cali8484]

Are you the one that made the PW2 teardown video showing the ridiculously small gaps between the huge bus bars and the enclosure with painter's tape holding things? It's eye opening to see what Tesla did to make the cheapest home battery per kwh.

 

[mudder]

Yes, that's my video.

 

[taphil]

Why would anyone disassemble their own PW to corroborate your claims?

 

[jjrandorin]

Why would anyone disassemble their own PW to corroborate your claims?

I didnt get the impression that there was any implication or request for anyone to do that. The statement was "if you would like to help out....xxxxxxx".

We have some members here who work in the industry, and who knows what product they might or might not have access to, or want to share (or not).

 

[mudder]

Why would anyone disassemble their own PW to corroborate your claims?

as @jjrandorin proposed, I did not ask anyone to "disassemble their own Powerwall".

Update:
2024JAN05 @ 00:21: I sent my complaint to UL.
2024JAN05 @ 08:19: UL responds that the Tesla Powerwall 2 was independently verified to conform to UL standards by a 3rd party, and then provided the 3rd party contact information, along with the 3rd party approval number. What great service!
2024JAN05 @ 14:37: I forwarded my complaint to 3rd party.

 

[mudder]

2024JAN05 @ 19:47: Intertek (the 3rd party I mentioned previously) is "assembling a team to investigate the claim".

I'd love to collect more data from anyone who has disassembled their Powerwall 2 unit. If you'd like to help out, please post photographs of both the inverter PCBA and the aluminum bracket in the following orientation...

Why would anyone disassemble their own PW to corroborate your claims?

Someone else has agreed to peek inside their PW to corroborate my claim. They're not wanting to post publicly, but have allowed me to post their pictures so they can remain anonymous. I should have those pictures soon.

 

[mudder]

The motivation for reporting software security defects privately to companies is to they can fix it before attackers use the flaw and cause harm to people.

That motivation doesn't apply here at all, does it? It's a hardware flaw. I assume the risk here is under certain circumstances it poses a hazard to users, but this isn't an attack vector that a malicious actor could take advantage of.

Consider the following hypothetical:
Suppose that you discover a serious design defect that could cause death or serious injury. You have a full time job and very little free time, but you go out of your way to cross reference the various (far from free) standards to verify the issue actually exists; fully document the issue, including an in-depth video, detailed pictures, and a full writeup; and then spend far too many hours attempting to report this issue in a way that is useful to the responsible parties.

Suppose that in doing the above, you prevent just one person from dying from electric shock. In doing so, you prevent the company that designed the product from having to pay out a $5M wrongful death lawsuit. A hardware bounty program encourages engineers to endeavor/persist/waste time going down rabbit holes they might otherwise have no obligation/desire/need to follow.