Tesla Large Drive Unit Support Thread

Topics concerning the Tesla front and rear drive unit drop-in board
Locked
Volta1r3
Posts: 22
Joined: Sun Jan 17, 2021 9:17 pm

Re: Tesla Large Drive Unit Support Thread

Post by Volta1r3 »

Howdy everyone,

A couple of nights ago, I installed the community logic board into my LDU and did a forward/reverse function check on a regulated power supply. I was able to confirm forward and reverse by increasing the current, giving me VERY little incremental movement (set to 40V but dropped due power regulation). Altered params were set to: UDCMIN=0, UDCSW=30 at that time.

Yesterday, I put it on an deep cycle AGM Pb 48V nominal pack (8x6V batteries in series, 224Ah @ C20) with param: UDCMIN=0, UDCSW=42 (pack voltage at ~51V fully charged). My test bench was switch on, put in forward, pulsed a start signal, and then something magical happened... smoke at the power input (jumper cables) of the +/-B contactors. I got a slow rotation, smoke, a jerk, then nothing. The 400A (rated for 80V) fuse had blown (failed continuity test and visibly blown). The fuse was replace and tried again with an additional param. change to UDCNOM=48. Negative contactor closed, then pre-charge, but when the start signal was sent, the pre-charge contactor connected then dropped back out. At that point I read an OVERCURRENT error and status of UdcBelowUdcSw (udc= 6.4 on a 50.9V measured pack). That's when I got really excited and set my UDCMIN,UDCSW, and UDCNOM @ 0 resulting in same error/status.

EDIT: I had meant to say positive contactor, not pre-charge!

Today, I plan to run in manual mode.

I'm worried I may have a bridge across the PCB pads (possibly in the underlying layers?) of the ampseal connector, as I had difficulty soldering pads on the 7-10 pins (up waaay to late to be soldering). I took continuity readings of said pads after wicking up solder and removing/replacing the damaged connector to find continuity between some of the mid pins on the board.

Does anyone know if each pin on the ampseal is supposed to be isolated or if there is some kind of common node shared by a few of the middle connection points, or even the pinout of that connector? Maybe they are the encoder pins.. hopefully running in manual mode will hopefully verify this!
User avatar
johu
Site Admin
Posts: 5684
Joined: Thu Nov 08, 2018 10:52 pm
Location: Kassel/Germany
Has thanked: 153 times
Been thanked: 960 times
Contact:

Re: Tesla Large Drive Unit Support Thread

Post by johu »

Not sure about your other issues, but I know where to find pinout :) https://github.com/damienmaguire/Tesla- ... UT_NEW.pdf
Support R/D and forum on Patreon: https://patreon.com/openinverter - Subscribe on odysee: https://odysee.com/@openinverter:9
Volta1r3
Posts: 22
Joined: Sun Jan 17, 2021 9:17 pm

Re: Tesla Large Drive Unit Support Thread

Post by Volta1r3 »

I guess I shouldn't say pinout lol, but more so the "pinin" at the board side connector.

Duh, they correspond to the order of the pins on the board :|, maybe it's time for another coffee...
Volta1r3
Posts: 22
Joined: Sun Jan 17, 2021 9:17 pm

Re: Tesla Large Drive Unit Support Thread

Post by Volta1r3 »

After some troubleshooting with Johannes, the issue was found to be a cold solder joint on a couple of the pins of the IGBT driver connector. This was found by checking for ~5V on the six fault pins found on IC10 (IC3 on V3 schem and V3 PCB pdf). A quick reflow of solder on these pins and the motor was able to spin. The RPM was shown to be very slow and clunky only pulling ~4A, as seen in the following video:

https://drive.google.com/file/d/1Mup6Fy ... sp=sharing

When the encoder cable is removed the unit seems to be running a little smoother as seen here:

https://drive.google.com/file/d/1MzgGux ... sp=sharing

I will be addressing this next. Thank you for your help with this Johannes!
P.S.Mangelsdorf
Posts: 753
Joined: Tue Sep 17, 2019 8:33 pm
Location: Raleigh, NC, USA
Has thanked: 89 times
Been thanked: 92 times

Re: Tesla Large Drive Unit Support Thread

Post by P.S.Mangelsdorf »

Volta1r3 wrote: Thu Feb 11, 2021 2:01 am
When the encoder cable is removed the unit seems to be running a little smoother as seen here:

https://drive.google.com/file/d/1MzgGux ... sp=sharing

I will be addressing this next. Thank you for your help with this Johannes!
Sounds like the encoder wires may be flipped. Known issue caused by Tesla changing the wiring harness.
If at first you don't succeed, buy a bigger hammer.

1940 Chevrolet w/ Tesla LDU - "Shocking Chevy" - Completed 2023 Hot Rod Drag Week
User avatar
Jack Bauer
Posts: 3563
Joined: Wed Dec 12, 2018 5:24 pm
Location: Ireland
Has thanked: 1 time
Been thanked: 87 times
Contact:

Re: Tesla Large Drive Unit Support Thread

Post by Jack Bauer »

I'm going to need a hacksaw
Volta1r3
Posts: 22
Joined: Sun Jan 17, 2021 9:17 pm

Re: Tesla Large Drive Unit Support Thread

Post by Volta1r3 »

Thank you for the info guys. I've switched encoder signal A & B, but the unit spins same as before when the encoder was removed and ran. After switching back to the original A/B wiring, I was able to see signal A fluctuate around 2.5V, while signal B showed < 200mV (so ~0V) while at full throttle. There is ~ 5V that can be seen at the encoder lowpass filters R20 & R21 (R36 & R37 on V3 PCB/Schem). I'm going to keep poking around to find the issue with B.
Volta1r3
Posts: 22
Joined: Sun Jan 17, 2021 9:17 pm

Re: Tesla Large Drive Unit Support Thread

Post by Volta1r3 »

I checked continuity from R21 to the ampseal 23 pin connector pin, and it rang out. So, I checked between the ampseal 23 pin connector pin to the encoder, and that rang out. So, I know at this point there is a solid connection from the MCU to the encoder, voltage is getting to the board edge connector, the resistor (R21) value is within tolerance, encoder GND and encoder B/A are isolated from each other, but I can't get a voltage between the ampseal connection and the encoder on B.

Maybe there is something going on in the encoder?

EDIT: I was checking voltage of R21 and comparing to voltage at R20 when I noticed that R21 was in the mV range. So maybe I'm not getting an output from the MCU? I tried flipping the wires again and the spin is smooth, but pretty slow. Maybe I should try more voltage than 48V...

Here's a video of the motor running with encoder A & B swapped:

https://drive.google.com/file/d/1O0mS4Z ... sp=sharing

Here's one with the encoder removed:

https://drive.google.com/file/d/1O29ata ... sp=sharing

There is a slight difference I didn't notice before, maybe I just need to adjust parameters and increase the power?
User avatar
johu
Site Admin
Posts: 5684
Joined: Thu Nov 08, 2018 10:52 pm
Location: Kassel/Germany
Has thanked: 153 times
Been thanked: 960 times
Contact:

Re: Tesla Large Drive Unit Support Thread

Post by johu »

No, don't increase power. A properly working motor will spin up even with little power.
The encoder works like this: it has two channels A&B which are inverted to each other. When A is high, B is low and vice versa. But they have to be solid signals, either 5V or 0V not something in between. So check which one is high, then turn the motor a bit until the other one is high.
People had issues with bad crimps on the encoder plug on the back of the motor.
When in manual mode with ffslipspnt=0 and ampnom=0 plot speed and turn the motor by hand. It must show a couple of rpm then.
Support R/D and forum on Patreon: https://patreon.com/openinverter - Subscribe on odysee: https://odysee.com/@openinverter:9
User avatar
muehlpower
Posts: 570
Joined: Fri Oct 11, 2019 10:51 am
Location: Germany Fürstenfeldbruck
Has thanked: 12 times
Been thanked: 96 times

Re: Tesla Large Drive Unit Support Thread

Post by muehlpower »

I had the same problems, even swapping the two encoder signals didn't help. After set to single encoder in the menu it runns smooth.
Volta1r3
Posts: 22
Joined: Sun Jan 17, 2021 9:17 pm

Re: Tesla Large Drive Unit Support Thread

Post by Volta1r3 »

johu wrote: Fri Feb 12, 2021 8:34 am So check which one is high, then turn the motor a bit until the other one is high.
I powered the motor/controller in normal operation, encoder A had shown ~5V and B had shown around ~20mV. I then rotated the motor by hand, A dropped down to <200mV while B went a little over 100mV. When stopped A stayed low, then eventually came back up to ~5V.
johu wrote: Fri Feb 12, 2021 8:34 am When in manual mode with ffslipspnt=0 and ampnom=0 plot speed and turn the motor by hand. It must show a couple of rpm then.
When ran in manual mode, I was able to get a low rpm to plot when turning the motor by hand.
muehlpower wrote: Fri Feb 12, 2021 6:56 pm I had the same problems, even swapping the two encoder signals didn't help. After set to single encoder in the menu it runns smooth.
I don't see the encmode param listed in the version I'm using.

In the Parameters wiki, encmode shows default value set to single channel encoder. Is there a way to confirm that encmode value is set to quadrature?
User avatar
johu
Site Admin
Posts: 5684
Joined: Thu Nov 08, 2018 10:52 pm
Location: Kassel/Germany
Has thanked: 153 times
Been thanked: 960 times
Contact:

Re: Tesla Large Drive Unit Support Thread

Post by johu »

So something wrong with channel B. Can you check the crimp for that channel and for a short on the logic board chan B to ground?
Yes setting encmode to single would run it with just one channel.

I have hidden that parameter on Tesla drivetrains because it doesn't make sense to change normally. You can bring it back by typing

Code: Select all

flag encmode ~hidden
in the custom command box (and press send)
Support R/D and forum on Patreon: https://patreon.com/openinverter - Subscribe on odysee: https://odysee.com/@openinverter:9
Volta1r3
Posts: 22
Joined: Sun Jan 17, 2021 9:17 pm

Re: Tesla Large Drive Unit Support Thread

Post by Volta1r3 »

johu wrote: Sun Feb 14, 2021 6:45 pm So something wrong with channel B. Can you check the crimp for that channel and for a short on the logic board chan B to ground?
I agree.

4 pin is OEM and crimp looks good.

I'm getting continuity between a quick connect on the wire harness side of 23 pin ampseal connector (originally used for switching A & B signals) for B channel and ground from power supply, but not getting continuity between main ground (pin 12) and encoder B (pin 7) at board level. I disconnect board level ampseal connector and no continuity can be found at the point upstream of the 23 pin connector that was showing continuity before. I also tried to check board level pin against power supply ground and no continuity can be seen.

If you would prefer, we can go through support to investigate further.
Volta1r3
Posts: 22
Joined: Sun Jan 17, 2021 9:17 pm

Re: Tesla Large Drive Unit Support Thread

Post by Volta1r3 »

Got her going in single mode for now:

https://drive.google.com/file/d/1eVJx-a ... sp=sharing

Still going to focus on getting the second channel running!
User avatar
johu
Site Admin
Posts: 5684
Joined: Thu Nov 08, 2018 10:52 pm
Location: Kassel/Germany
Has thanked: 153 times
Been thanked: 960 times
Contact:

Re: Tesla Large Drive Unit Support Thread

Post by johu »

Some say dual channel mode gives you better performance, I can't tell as I have only ever been using single channel mode. Nothing wrong with using it but definitely worthwhile checking the B channel.
Support R/D and forum on Patreon: https://patreon.com/openinverter - Subscribe on odysee: https://odysee.com/@openinverter:9
User avatar
muehlpower
Posts: 570
Joined: Fri Oct 11, 2019 10:51 am
Location: Germany Fürstenfeldbruck
Has thanked: 12 times
Been thanked: 96 times

Re: Tesla Large Drive Unit Support Thread

Post by muehlpower »

I have observed that sometimes a very high speed is outputed (on CAN) when the motor stands still. I think that comes from the fact that the encoder is on the edge and only swings back and forth. This cannot happen with an A / B encoder and the direction of rotation can be determined.
User avatar
johu
Site Admin
Posts: 5684
Joined: Thu Nov 08, 2018 10:52 pm
Location: Kassel/Germany
Has thanked: 153 times
Been thanked: 960 times
Contact:

Re: Tesla Large Drive Unit Support Thread

Post by johu »

Yes that's correct and is an issue in Polo, too. Tried to mitigate by looking at the last 5 samples and discarding spikes but that didn't always help.
Support R/D and forum on Patreon: https://patreon.com/openinverter - Subscribe on odysee: https://odysee.com/@openinverter:9
Permag
Posts: 15
Joined: Thu May 09, 2019 9:20 am

Re: Tesla Large Drive Unit Support Thread

Post by Permag »

At start up lasterr showing OVERCURRENT... any help with this is greatly appreciated.

Where do I find the schematic for IC10?
Elektro-Funken RSR/#P85DPorsche.
Youtube video:
https://www.youtube.com/watch?v=fHaCl6OBNRQ&app=desktop
User avatar
johu
Site Admin
Posts: 5684
Joined: Thu Nov 08, 2018 10:52 pm
Location: Kassel/Germany
Has thanked: 153 times
Been thanked: 960 times
Contact:

Re: Tesla Large Drive Unit Support Thread

Post by johu »

Probably bad connection to one of the gate drivers or actual IGBT failure. Check IC10 in schematic for fault signals, they should all read 5V
Support R/D and forum on Patreon: https://patreon.com/openinverter - Subscribe on odysee: https://odysee.com/@openinverter:9
User avatar
Boxster EV
Posts: 434
Joined: Fri Jul 26, 2019 9:32 pm
Location: UK
Has thanked: 35 times
Been thanked: 26 times

Re: Tesla Large Drive Unit Support Thread

Post by Boxster EV »

Helpful overview video of the LDU. There’s some errors in the info, like Dexron 6 is actually the correct lubricant. However generally interesting.

Porsche 986 powered by a Tesla large drive unit. Backwards. Build documented here and Instagram @tesla_porsche here.
rjmcdermott81
Posts: 107
Joined: Thu Dec 03, 2020 5:03 pm

Re: Tesla Large Drive Unit Support Thread

Post by rjmcdermott81 »

Boxster EV wrote: Sat Feb 20, 2021 10:52 pm Helpful overview video of the LDU. There’s some errors in the info, like Dexron 6 is actually the correct lubricant. However generally interesting.

I’ve watched some of his other videos in the past. Really awesome that he takes the time to document and share with everyone. This is really helpful thanks for sharing.
KimchiKimKim
Posts: 69
Joined: Wed Mar 03, 2021 1:36 am
Location: California
Has thanked: 4 times
Been thanked: 4 times
Contact:

Re: Tesla Large Drive Unit Support Thread

Post by KimchiKimKim »

Hey Guys, I'm thinking about buying a used V3 LDU drive board unit I found off of a local guy (He bought it directly from EVBMW). I was curious if I could add the ESP chip and interface using the web GUI. I have been looking around in the form for a while before posting and I couldn't seem to find anything. If so, do I have to use an ST Link to flash the updated firmware onto the board and something else to configure the ESP Chip? Thanks in advance!
User avatar
johu
Site Admin
Posts: 5684
Joined: Thu Nov 08, 2018 10:52 pm
Location: Kassel/Germany
Has thanked: 153 times
Been thanked: 960 times
Contact:

Re: Tesla Large Drive Unit Support Thread

Post by johu »

ESP is programmed like this: viewtopic.php?f=5&t=8
If it is a recent board, it will plug right in.
Support R/D and forum on Patreon: https://patreon.com/openinverter - Subscribe on odysee: https://odysee.com/@openinverter:9
KimchiKimKim
Posts: 69
Joined: Wed Mar 03, 2021 1:36 am
Location: California
Has thanked: 4 times
Been thanked: 4 times
Contact:

Re: Tesla Large Drive Unit Support Thread

Post by KimchiKimKim »

I appreciate the quick response! It is a Version 3 board without the connector I believe as on Git hub I see that the V4 LDU board was the first to integrate the ESP on the actual board.
Permag
Posts: 15
Joined: Thu May 09, 2019 9:20 am

Re: Tesla Large Drive Unit Support Thread

Post by Permag »

johu wrote: Sat Feb 20, 2021 4:47 pm Probably bad connection to one of the gate drivers or actual IGBT failure. Check IC10 in schematic for fault signals, they should all read 5V
Where do I find the IC10 schematic?
Elektro-Funken RSR/#P85DPorsche.
Youtube video:
https://www.youtube.com/watch?v=fHaCl6OBNRQ&app=desktop
Locked