openinverter forum

This is a bit of a long shot, but I thought I'd ask. I have a gen2 charger that was sold to me as faulty. After installing the open source logic board, it does indeed seem that one of the three charging modules has failed. When enabled, it correctly reports the AC and DC voltages, but enters a fault state as soon as charging begins. I intend to use this charger with its two working modules anyway, but I am interested to know 3 things:

1) Is there a common fault that is likely to be to blame here?
2) Am I likely to be able to disassemble / remove the module without causing damage?
3) Could this be repairable?

The reported state when attempting to use the bad module (1) and good module (2) is as follows:

Do you have the partnumber of that module?
I might have an idea.

Also if you have a savvycan log; i could check it for you. The modules send more info out over it's faults.

catphish wrote: ↑Sat Dec 11, 2021 4:31 pm
2) Am I likely to be able to disassemble / remove the module without causing damage?

Really hard, it's all glued in, I don't think anyone has taken one apart like that, which is why we X-rayed one when we wanted to look inside.

EV_Builder wrote: ↑Mon Dec 13, 2021 1:02 pm Do you have the partnumber of that module?

The part number of the charger is 1014963-00-C - or did you mean the individual module? I can open it up and check if that helps.

EV_Builder wrote: ↑Mon Dec 13, 2021 1:02 pm Also if you have a savvycan log; i could check it for you. The modules send more info out over it's faults.

I will make a CAN log and send it over ASAP. If you can help me decode it, that will be great.

Yes; the module itself is what I need.

I will open it up and have a proper look tomorrow. In the meantime, here's a CAN log from a working charge (modules 2+3) and a failing charge (modules 1+2+3).

Any idea where the part number can be found? There's a photo here I took previously, but nothing obviously visible. https://i.imgur.com/VgQKGv7.jpg

Normally its on the right part of the module, under the plastic, just visible under the bridge.
If you keep the powerlines (DC and AC) on the right side.

Maybe they relocated that...

Ok i checked your logs.

The pre-charge sequence fails.
So when the module is switched on it tries to reach battery voltage inside. That fails.

Maybe the module is faulty (albeit i don't see any signals from that) or the DC connection is bad (like not stable), or maybe the AC connection??

EV_Builder wrote: ↑Mon Dec 13, 2021 10:22 pm Ok i checked your logs.

The pre-charge sequence fails.
So when the module is switched on it tries to reach battery voltage inside. That fails.

Maybe the module is faulty (albeit i don't see any signals from that) or the DC connection is bad (like not stable), or maybe the AC connection??

Thank you for looking. I am fairly sure the module is faulty. I have some questions though:
1) What leads you to the conclusion it is failing to reach DC voltage?
2) Do you have savvycan schema files? If so, could you share them, so that I can see the data more easily?

So if I understand correctly, the module is correctly identifying the voltage on the DC bus, but is failing to generate this voltage internally?

In any case, if the module can't be removed from the charger, it seems unlikely I'll be able to fix it. I'll follow up with the model number from the modules tomorrow.

catphish wrote: ↑Mon Dec 13, 2021 11:38 pm Thank you for looking. I am fairly sure the module is faulty. I have some questions though:
1) What leads you to the conclusion it is failing to reach DC voltage?

The module sends info over CanBus and it fails with a pre-charge error state.

2) Do you have savvycan schema files? If so, could you share them, so that I can see the data more easily?

They are not mine and I promised not to share.
I'm trying to help you out; could have stayed silenced...
If you want them you can find them in the Tesla modify corner of the internet. Or maybe other owners step by....

So if I understand correctly, the module is correctly identifying the voltage on the DC bus, but is failing to generate this voltage internally?

In any case, if the module can't be removed from the charger, it seems unlikely I'll be able to fix it. I'll follow up with the model number from the modules tomorrow.

Yes; it's failing to match pre-charge VDC.
I think you can remove it. The question is how hard it is, how todo it and if everything survives.

Double check the connections because I don't see HW errors. That leads me to think something simple gave or is wrong.

Thanks. The model number on the modules is: 1026696-00-C REV01
I don't believe there is a wiring problem, though I am not easily able to access the wiring inside the charger. The error definitely seems to just be the DC precharge failure.

I have another oddity error with this charger:
If I first attempt to charge at 2A DC using modules 2 & 3, I only get a few milliamps. If I then attempt to charge with module 2 or module 3 individually, I get 2.5A, and then switching back to 2&3 simultaneously, i get 2A as expected. This is very odd behaviour. I didn't take a CAN capture of this sadly, but I might. Seems more like a software issue in this case.

I had a long running problem with a Tesla charger, where it failed after a few second when I tried to run all three modules. However, I could only get any 1 of the 3 modules running at a time. It I tried to run all three again, all would fail. Then I got any 2 of the 3 to run. Once I improved the ground connection of the overall charger, all three worked perfectly.

catphish wrote: ↑Tue Dec 14, 2021 1:31 pm I have another oddity error with this charger:
If I first attempt to charge at 2A DC using modules 2 & 3, I only get a few milliamps. If I then attempt to charge with module 2 or module 3 individually, I get 2.5A, and then switching back to 2&3 simultaneously, i get 2A as expected. This is very odd behaviour. I didn't take a CAN capture of this sadly, but I might. Seems more like a software issue in this case.

This issue was fixed in the software by Johannes. The reason was that some parameters were read from module 1 only, so if module 1 wasn't enabled, they would be assumed to be zero. This has now been fixed so that the parameters are read from the first enabled module.

Module 1 is still broken, but the other two work as they should.