Model 3 High Voltage Controller

[nkiernan]

Hi Amzoo, have you any photos of the cell tap boards in place on the battery modules and the full BMS setup you have? Looks great. Are you selling those tab boards somewhere?

[amzoo]

Don't have pictures of the full BMS setup as I was in a rush to get things working before moving across the state. Give me a couple days to look through boxes to find them again and I'll take pictures of them sitting on the battery modules later this week.

I never populated them with the Molex Connectors because I was in a rush to get things running so used header pins. If you notice on the boards, there are multiple slots for the same cell taps. Different battery BMS boards have slightly different locations, so I could either populate all the pogo pins and have you solder the jumpers to your configuration, or I can provide the pogo pins not attached.

Either way it's easy enough for me to set them up completely if you take a picture of all your battery boards. I have plans to list them publicly once I get the Tesla BMS working. Anyone working on the Tesla BMS who contributes to this thread can have them for price of parts + shipping.

[nkiernan]

Sounds good, cheers

[amzoo]

Full pinout attached. Check the PDF for full page of connected components.

[amzoo]

It seems like you're all reading data off the Charge Port CAN. Did I get my pinout wrong or is vehicle CAN not being used for monitoring values?

[nkiernan]

It seems like you're all reading data off the Charge Port CAN. Did I get my pinout wrong or is vehicle CAN not being used for monitoring values?

I've been using CAN ID's 292, 312, 332, 401 and 712 for now. They don't all show on any one CAN bus, so I've just been swapping between the CP and vehicle CAN as needed

[sarasota]

I am working on replicating the steps of getting voltage data from message 401 captures as shown in the example below and could use some help. Have read through everything here multiple times and have tried a few different ways. It is clearly described here how to do this with message 332 captures , but not for message 401. How exactly does one take the CAN message 401 and convert it to the 3 voltage readings. If anyone has a spreadsheet with the formulas that would be awesome, or just a written description of the process from hex to binary to decimal to voltage, thanks.

[nkiernan]

See if this helps:

TM3 CAN 401.xlsx

(12.91 KiB) Downloaded 1039 times

[wimboone]

Didn't realize this thread existed! Adding myself into this effort. I have a US long range pack with all penthouse components.

Connector:
I've seen two different part numbers for this, haven't talked to any suppliers yet
Sumitomo TS Sealed Series
6189-6934 https://kinkong-connector.com/ru/produc ... D2081.html#
6189-7077 http://prd.sws.co.jp/components/en/deta ... s=61897077
(Screenshots attached if URLs are down)

Simple 3D STEP files for the 6189-7077 are attached as well. If I can't find a supplier (once I confirm BMS functionality), then I'll model the connectors fully and 3D print them.

Here is the link for the connector on aliexpress

https://nl.aliexpress.com/item/10050034 ... 6850%22%7D

[amzoo]

Have any of you taken a look at the Ingenext product for battery/motor control?
https://ingenext.ca/products/tesla-mode ... controller

[amzoo]

Hi Amzoo, have you any photos of the cell tap boards in place on the battery modules and the full BMS setup you have? Looks great. Are you selling those tab boards somewhere?

Finally back on this after a hellish end of a year for me. Here is the cell tap board on one of my modules. The Molex connectors I spec'd for these boards are backordered, so I'm planning on soldering wires directly to the board.

[bielec]

I am about to get a model3 battery from a breaker and I want to verify that it is ok when I get it home. I think the easiest way is to connect a bluetooth can dongle to pins 15+16 and 12v to pins 8+9 and the dongle then use "Scan my Tesla" to read the cell voltages via bluetooth.

My question to the group is, will this work? Is this all that is required to make the battery pack spew out can data? Will "ScanMyTesla" work with only the battery and without the rest of the car being there?

Thanks for eventual help.

[nkiernan]

I am about to get a model3 battery from a breaker and I want to verify that it is ok when I get it home. I think the easiest way is to connect a bluetooth can dongle to pins 15+16 and 12v to pins 8+9 and the dongle then use "Scan my Tesla" to read the cell voltages via bluetooth.

My question to the group is, will this work? Is this all that is required to make the battery pack spew out can data? Will "ScanMyTesla" work with only the battery and without the rest of the car being there?

I am not familiar with ScanMyTesla or its functionality, but can confirm that I was able to access the battery pack connector under the rear seat on a TM3 in a breakers yard, connected a 12V supply and CAN lines (as per previous pin out in this thread) and used an Arduino with CAN shield to read the 96 cell voltages to check all was ok. So maybe prepare for both and test ScanMyTesla but be sure you can get the answer with the Arduino

[bielec]

Thank you so much for that info, so I now know that I can get can data out. I will try first with ScanMyTesla and if that doesn't work I have an Arduino Due which I can use. I shall report back when the battery arrives.

[bielec]

Well, my 77Kwh model 3 battery arrived. Decided the quickest test was to open the penthouse and check the pyro fuse and the overall battery voltage. The overall voltage was 182.0+182.2 = 364.2 average 3.79v per cell. The pyro fuse was open and had voltage across it which I presume to be leakage. Using a 1M ohm meter there was 27v across the fuse suggesting 27uA current. What is the significance of the blown fuse? Is this done during a collision? The car was accident damaged but the battery is ok.

[bielec]

I have just displayed voltages for all 96 cells using "ScanMyTesla" via a Bluetooth OBD dongle connected to connector X098 using the pinout posted in this thread. This is a 77KWh 2020 AWD model 3 battery from a breaker - out of the car.
This must be one of the quickest ways to confirm the state of a wreck battery. I had a 20 day sale or return arrangement with the supplier.
A few notes...
CAN lines were terminated with 120 ohms
The Vehicle CAN pins were only producing msg, 0x132 = battery Volts and Amps, nothing else. I suspect this is because my pyrofuse is open?
I used Chargeport CAN which had msg 0x401 and many others but NOT msg 0x332
I initially used a current limited lab PSU to provide 12v on pins 8 and 18 with 0v on pin 9. The current drawn was about 60mA.

Next is to fix the pyrofuse. My application will never draw more than 25A so I am considering replacing it with a DC 1000v 30A fuse or MCB.

Edit. Just replaced the Pyrofuse and the vehicle CAN port still outputs only msg 0x132. I must be missing something.

[muehlpower]

Well, my 77Kwh model 3 battery arrived. Decided the quickest test was to open the penthouse and check the pyro fuse and the overall battery voltage. The overall voltage was 182.0+182.2 = 364.2 average 3.79v per cell. The pyro fuse was open and had voltage across it which I presume to be leakage. Using a 1M ohm meter there was 27v across the fuse suggesting 27uA current. What is the significance of the blown fuse? Is this done during a collision? The car was accident damaged but the battery is ok.

I got my battery with a working pyrofuse and measured about 390V. I have disconnected the various connectors and disassembled the battery without short circuits or other incidents. Then I found out that the fuse was broken. I assume that the controller blows the fuse when certain connections are loosened because it suspects a broken wire.
With my next battery I will remove the fuse first!

[bielec]

I now have all messages on the vehicle bus. I needed 12v on the stud terminals, + on the outer one. I was pleased to see 0x401 was present on this 2020 pack. A CanDue/Savvycan log is attached.
I tried to close the contactors by sending the two forms of msg 0x221.....
0x221 41 11 01 00 00 00 9c c8
0x221 40 40 05 00 00 00 9c c8
..... as indicated on the EVTV video by Jack Rickard.
But, no joy!
I had 100ohms across the HVILpins 1 and 3 of X098
I had 12v on the battery studs from the dcdc
and 12v on pins 8 and 18 of X098
I varied the timing of sending 0x221 but Jack suggested around 100mS.
When I sent them very fast, the canbus stopped spewing data once but I suspect that was my laptop hanging in Savvycan?
The chargeport and front and rear drive connectors are not fitted - the battery is out of the car - do these have another HVIL on the small pins which might prevent closure?
Can anyone suggest what I might try next?

[bielec]

Attempting Contact Closure.
I noticed there was a reference in the DBC file to "penthouse lid open" and mine is open for changing the Pyrofuse, so I replaced it and all the screws. I also fitted the chargeport cables and linked the sense pins on the two motor connectors. Then tried sending 0x221 messages but still no contact closure! So I looked at the messages 0x232 and 0x20A and these change when the 0x221's are sent. I noticed that the changes are timing dependent - the rate must be faster than 40mS or so to work. I used 5mS. I attach a log file while this is happening.

I note that in msg 0x232 starts to show "pack set request close" so my 0x221's are being noticed.
msg 0x20A shows "DC link allowed to energise" and "Closing allowed" and there is a "Pyrofuse test" state which stays set.
This might be important because I have replaced the Pyrofuse with a 1000v 10A fuse link and thus the pyrofuse ignition element is absent. This may be causing a failed test. Could anyone with a working Pyrofuse possibly help by measuring the resistance across the element so I could fit a resistor to spoof the test?

This is all very specific to contact closure. Is the administrator happy that it sits here in this thread or should I start a new topic?

[bielec]

I replaced the pyrofuse with a DC circuit breaker and put a 10 ohm resistor across the pyro element terminals in an attempt to fool the bms into thinking the pyro fuse was ok. The contactors still did not close !

I studied the results in msg 0x20A while sending 0x221 and now can see that the hvil is ok because there are reports as follows...

0x20A C9 11 09 92 10 00 before sending 0x221 this is spewed out

after sending 0x221 messages at 5mS intervals

0x20a C9 11 09 12 38 01 the 01 means hvil ok and 38 = pyrotest +dc link allow + closing allow

this is sent for 30 seconds then it reverts to the first - even though still sending 0x221, so the closure is aborted.

I confirmed the hvil by unscrewing the two bolts of connector x098 then I saw...
0x20A c9 11 09 12 10 05 the 05 means "penthouse lid open"

So, no closure yet and running out of ideas. Starting to think of hard wiring to the main terminals inside and running cables to my own contactors outside. Not elegant!

[nkiernan]

Good work bielec

[bielec]

After careful examination of Jack Rickards video (at x0.25 speed) I now see that his software sends many different versions of 0x221 while I have sent only 2. This may be it. I have counted at least 11 versions on the video and presumably this is a startup sequence to enable contactors. The order is probably important but I will try some combinations today. If I get stuck, a can log from a car while closing contactors would be handy.

[bielec]

I found 15 different versions of 0x221 on the video and there may be more. The screen updates so slow that I cannot see a continuous stream so I have no idea of the correct sequence. I attach the log file of the message order I used and this did not cause closure, however there was a double pulse in current consumption which might have been a half-hearted attempt? I do not have sufficient information to figure out the sequence so a working car can dump would be the best way forward. Or does anyone out there have the sequence?
edit. I found some logs online but they had presumably older format of 0x221 which did not begin with 40 or 41. I need one which begins 40 or 41.

[Doig5710]

Not much help for the coding side but maybe try a resistor between 1.8 - 2.2 ohm, all airbags in vehicles are that resistance that I have come across.

[bielec]

Not much help for the coding side but maybe try a resistor between 1.8 - 2.2 ohm, all airbags in vehicles are that resistance that I have come across.

Thanks for that. I tried 10 ohms then a fine wire for zero ohms maybe 2 ohms would be worth a go