Model 3 High Voltage Controller

Topics concerning the Tesla front and rear drive unit drop-in board
collin80
Posts: 110
Joined: Sun Aug 30, 2020 3:28 pm
Location: United States, Michigan
Been thanked: 4 times
Contact:

Re: Model 3 High Voltage Controller

Post by collin80 »

nkiernan wrote: Mon May 24, 2021 1:03 pm Latest capture of the vehicle CAN with the four modules connected up to make full pack and a second CAN capture with 0x500 introduced after approx 10 seconds for comparison. As Collin said, the two extra ID's do show up here too

240521 - TM3 HV Controller - Vehicle CAN - Full Pack Connected 0x500.csv

240521 - TM3 HV Controller - Vehicle CAN - Full Pack Connected.csv
I loaded the 0x500 connected file. It seems like you're dropping a lot of frames for some reason. For instance, 0x332 is sometimes only coming through every 3-5 seconds. It is sent much faster than that. The 0x7D2 message that gets unlocked by sending 0x500 should be occurring every second. That's certainly not happening. It could be because the 0x70A message is spamming the bus. All of these units seem to do that. I think it's looking for the other devices in the car and spams the bus with that message to reach out to them... somehow? You don't see 0x70A being constantly sent in the full car. It gets sent sometimes for a little while but then it stops again. In your captures the 0x70A message is 3/4 of the traffic! 0x132 happens a lot too but not nearly as often.

What are you using for capture? I'm almost positive that I was getting much more consistent results when I captured with a PeakCAN adapter but, alas, I did not save any captures.

Anyway, 0x332 shows this:
Min cell: 2.99v (cell 60)
Max cell: 3.15v (cell 45)

So, that pack is quite discharged. It shows about 7% SOC too.
nkiernan
Posts: 505
Joined: Mon Feb 24, 2020 8:59 pm
Location: Ireland
Has thanked: 276 times
Been thanked: 64 times

Re: Model 3 High Voltage Controller

Post by nkiernan »

collin80 wrote: Mon May 24, 2021 11:21 pm
I loaded the 0x500 connected file. It seems like you're dropping a lot of frames for some reason. For instance, 0x332 is sometimes only coming through every 3-5 seconds. It is sent much faster than that. The 0x7D2 message that gets unlocked by sending 0x500 should be occurring every second. That's certainly not happening. It could be because the 0x70A message is spamming the bus. All of these units seem to do that. I think it's looking for the other devices in the car and spams the bus with that message to reach out to them... somehow? You don't see 0x70A being constantly sent in the full car. It gets sent sometimes for a little while but then it stops again. In your captures the 0x70A message is 3/4 of the traffic! 0x132 happens a lot too but not nearly as often.

What are you using for capture? I'm almost positive that I was getting much more consistent results when I captured with a PeakCAN adapter but, alas, I did not save any captures.

Anyway, 0x332 shows this:
Min cell: 2.99v (cell 60)
Max cell: 3.15v (cell 45)

So, that pack is quite discharged. It shows about 7% SOC too.
Thanks for looking at the file Collin. I am using a Macchina M2 running M2RET with Savvycan. I'll tidy up the routing and make sure I'm not getting any interference or resistance issues.

Thanks for pulling out that info, I was a bit concerned about the pack as I knew it was well down but relatively new to these batteries and how bad is too bad! So was keen to get some charging sorted and to make sure this BMS and some balancing was working...so this thread has been helpful, and I have a a good bit more to figure out.

I have Damien's PCS controller and getting it set up so should be able to do some charging shortly. Connecting the PCS to the HV controller probably won't do anything for 0x70A traffic, so part of that solution is probably going to need some vehicle CAN mimic at some stage.
User avatar
fredd90
Posts: 87
Joined: Tue Oct 20, 2020 7:36 pm
Been thanked: 1 time

Re: Model 3 High Voltage Controller

Post by fredd90 »

Now it's happening more stuff when I change to pt-can

Very annoying 0x70A, send like 1000 messages/second.

Also tested to connect +30 but nothing remarkable happened to can output

Not really sure how to get the correct info of 0x332, even if your explanation was good.


This is my recent readout from 0x332 approx 1 message/sec

0x332 DD 1F 00 00 14 2F
0X332 14 03 74 74 74 74


I have saved 1 previous canlog, now its time to test if the battey get balanced under 2h, then take a new can log in about 2 hour to see if there been some difference
Attachments
IMG_20210525_162823.jpg
IMG_20210525_162816.jpg
collin80
Posts: 110
Joined: Sun Aug 30, 2020 3:28 pm
Location: United States, Michigan
Been thanked: 4 times
Contact:

Re: Model 3 High Voltage Controller

Post by collin80 »

fredd90 wrote: Tue May 25, 2021 2:26 pm Not really sure how to get the correct info of 0x332, even if your explanation was good.

This is my recent readout from 0x332 approx 1 message/sec

0x332 DD 1F 00 00 14 2F
0X332 14 03 74 74 74 74
The message is multiplexed which is why you're getting such different data in the frame. It should look to be alternating back and forth between two types.

The bottom 2 bits are the multiplexor.
0 = Thermistor data 1 = voltage data

For voltage data the next 12 bits are max voltage, then starting at the low bit of byte 2 there's a 12 bit number for minimum voltage. Both are in 2mv scale. Then the cell numbers for each are in the next two bytes. Working it out in your case:

The one that starts with DD is the voltage. 0x1FDD shifted down 2 places (to get rid of the multiplexor bits) is 0x07F7. Multiply by 0.002 and you get 4.078V. Of course, what does not make sense is that 0000 * 0.002 is still 0. This means it thinks you have a set of cells at 0V but I doubt that. 0x14 is cell group 21 (0x14 + 1) and 0x2F is cell group 48.

So, it seems to make sense to me except that it claims you have cells at zero volts. Maybe it will say that if it can't communicate to all the cells properly? I'm not sure...
nkiernan
Posts: 505
Joined: Mon Feb 24, 2020 8:59 pm
Location: Ireland
Has thanked: 276 times
Been thanked: 64 times

Re: Model 3 High Voltage Controller

Post by nkiernan »

fredd90, had just laid this out before I saw Collin explained above, will share anyway in case it helps:
0x332 Explained.png
User avatar
fredd90
Posts: 87
Joined: Tue Oct 20, 2020 7:36 pm
Been thanked: 1 time

Re: Model 3 High Voltage Controller

Post by fredd90 »

collin80 wrote: Tue May 25, 2021 4:06 pm
fredd90 wrote: Tue May 25, 2021 2:26 pm Not really sure how to get the correct info of 0x332, even if your explanation was good.

This is my recent readout from 0x332 approx 1 message/sec

0x332 DD 1F 00 00 14 2F
0X332 14 03 74 74 74 74
The message is multiplexed which is why you're getting such different data in the frame. It should look to be alternating back and forth between two types.

The bottom 2 bits are the multiplexor.
0 = Thermistor data 1 = voltage data

For voltage data the next 12 bits are max voltage, then starting at the low bit of byte 2 there's a 12 bit number for minimum voltage. Both are in 2mv scale. Then the cell numbers for each are in the next two bytes. Working it out in your case:

The one that starts with DD is the voltage. 0x1FDD shifted down 2 places (to get rid of the multiplexor bits) is 0x07F7. Multiply by 0.002 and you get 4.078V. Of course, what does not make sense is that 0000 * 0.002 is still 0. This means it thinks you have a set of cells at 0V but I doubt that. 0x14 is cell group 21 (0x14 + 1) and 0x2F is cell group 48.

So, it seems to make sense to me except that it claims you have cells at zero volts. Maybe it will say that if it can't communicate to all the cells properly? I'm not sure...
Wow thanks for great explanation👍😀

Just harvested all the cells voltage via canbus,
92 cells between 37,45 to 4,076
4 cells with 0,000volt, the last cellfrom each "batterypack".

Total voltage with reference of canbus: 360vdc
From my fluke 391vdc, verified with another fluke. Same result.

Pack 1 and 2 195,5vdc
Pack 3 and 4 195,5vdc
Attachments
IMG_20210525_210048.jpg
nkiernan
Posts: 505
Joined: Mon Feb 24, 2020 8:59 pm
Location: Ireland
Has thanked: 276 times
Been thanked: 64 times

Re: Model 3 High Voltage Controller

Post by nkiernan »

fredd90 wrote: Tue May 25, 2021 7:11 pm
Wow thanks for great explanation👍😀

Just harvested all the cells voltage via canbus,
92 cells between 37,45 to 4,076
4 cells with 0,000volt, the last cellfrom each "batterypack".

Total voltage with reference of canbus: 360vdc
From my fluke 391vdc, verified with another fluke. Same result.

Pack 1 and 2 195,5vdc
Pack 3 and 4 195,5vdc
fredd90, what way did you pull out the cell voltages and pack voltage, had you a script or manually? Use 0x401?
collin80
Posts: 110
Joined: Sun Aug 30, 2020 3:28 pm
Location: United States, Michigan
Been thanked: 4 times
Contact:

Re: Model 3 High Voltage Controller

Post by collin80 »

fredd90 wrote: Tue May 25, 2021 7:11 pm Wow thanks for great explanation👍😀

Just harvested all the cells voltage via canbus,
92 cells between 37,45 to 4,076
4 cells with 0,000volt, the last cellfrom each "batterypack".

Total voltage with reference of canbus: 360vdc
From my fluke 391vdc, verified with another fluke. Same result.

Pack 1 and 2 195,5vdc
Pack 3 and 4 195,5vdc
Ouch... So it wasn't a lie, there really are cells at 0V. That's a bummer. What is interesting is that it said that cell 48 is the lowest and, sure enough, you show 0V at 48. So, at least it wasn't lying about the value or the location.
User avatar
fredd90
Posts: 87
Joined: Tue Oct 20, 2020 7:36 pm
Been thanked: 1 time

Re: Model 3 High Voltage Controller

Post by fredd90 »

nkiernan wrote: Tue May 25, 2021 8:47 pm
fredd90 wrote: Tue May 25, 2021 7:11 pm
Wow thanks for great explanation👍😀

Just harvested all the cells voltage via canbus,
92 cells between 37,45 to 4,076
4 cells with 0,000volt, the last cellfrom each "batterypack".

Total voltage with reference of canbus: 360vdc
From my fluke 391vdc, verified with another fluke. Same result.

Pack 1 and 2 195,5vdc
Pack 3 and 4 195,5vdc
fredd90, what way did you pull out the cell voltages and pack voltage, had you a script or manually? Use 0x401?
Did i manually through 0x401, took about 30 min.
collin80 wrote: Tue May 25, 2021 10:40 pm
fredd90 wrote: Tue May 25, 2021 7:11 pm Wow thanks for great explanation👍😀

Just harvested all the cells voltage via canbus,
92 cells between 37,45 to 4,076
4 cells with 0,000volt, the last cellfrom each "batterypack".

Total voltage with reference of canbus: 360vdc
From my fluke 391vdc, verified with another fluke. Same result.

Pack 1 and 2 195,5vdc
Pack 3 and 4 195,5vdc
Ouch... So it wasn't a lie, there really are cells at 0V. That's a bummer. What is interesting is that it said that cell 48 is the lowest and, sure enough, you show 0V at 48. So, at least it wasn't lying about the value or the location.

If that really are cells at 0v thats really sad, but is there any explanation how it could be the last one of each pack?

I got mesaurement of 391v total. at battery terminals, if i divide 391 with the (healthy) cells(92pcs) that would be 4.25v/cell.



Would be really nice if you nkiernan could do a manual controll of 06, 0F, 18, 1E
Isaac96
Posts: 656
Joined: Sat Oct 05, 2019 6:50 pm
Location: Northern California, USA
Been thanked: 1 time
Contact:

Re: Model 3 High Voltage Controller

Post by Isaac96 »

Those are interesting readings. To me it seems unlikely that you'd have such a variation between cells -- usually that only happens if the pack has been severely abused. And specifically the last cells all being at zero also seems strange. That and your measured voltage not matching the BMS reported voltage.

It gets more interesting looking at your list of voltages where there are only really 2 different levels. 3.749 and 4.075.

What's even better is that 4.075 * 96 = 391.2.

Edit: A number.
nkiernan
Posts: 505
Joined: Mon Feb 24, 2020 8:59 pm
Location: Ireland
Has thanked: 276 times
Been thanked: 64 times

Re: Model 3 High Voltage Controller

Post by nkiernan »

Isaac96 wrote: Wed May 26, 2021 5:27 am Those are interesting readings. To me it seems unlikely that you'd have such a variation between cells -- usually that only happens if the pack has been severely abused. And specifically the last cells all being at zero also seems strange. That and your measured voltage not matching the BMS reported voltage.

It gets more interesting looking at your list of voltages where there are only really 2 different levels. 3.749 and 4.075.

What's even better is that 4.075 * 96 = 391.2.

Edit: A number.
That's what I wondered too. Would it be more likely the CAN messages or readings are an issue such as the delicate aluminium wires back to the BMS pads are damaged coincidentally at the last cell on each module (strange). How many cells difference are there between the long and short range battery modules, nothing to do with firmware for a certain pack (55kWh vs 75 or 90)?
nkiernan
Posts: 505
Joined: Mon Feb 24, 2020 8:59 pm
Location: Ireland
Has thanked: 276 times
Been thanked: 64 times

Re: Model 3 High Voltage Controller

Post by nkiernan »

fredd90 wrote: Wed May 26, 2021 4:27 am
Would be really nice if you nkiernan could do a manual controll of 06, 0F, 18, 1E
fredd90, can do. Are these from 0x401 or which one do they relate to?
User avatar
fredd90
Posts: 87
Joined: Tue Oct 20, 2020 7:36 pm
Been thanked: 1 time

Re: Model 3 High Voltage Controller

Post by fredd90 »

Isaac96 wrote: Wed May 26, 2021 5:27 am Those are interesting readings. To me it seems unlikely that you'd have such a variation between cells -- usually that only happens if the pack has been severely abused. And specifically the last cells all being at zero also seems strange. That and your measured voltage not matching the BMS reported voltage.

It gets more interesting looking at your list of voltages where there are only really 2 different levels. 3.749 and 4.075.

What's even better is that 4.075 * 96 = 391.2.

Edit: A number.
Maybe there are some mesaurement fault if the active cell balancing is ongoing?
You would have a way to simulate that the ignition is on to stop active balancing

The positive thing is like you said, the total bms mesurement should be closer to the real value, probably has something to do with reading value @ balancing
nkiernan wrote: Wed May 26, 2021 6:28 am
Isaac96 wrote: Wed May 26, 2021 5:27 am Those are interesting readings. To me it seems unlikely that you'd have such a variation between cells -- usually that only happens if the pack has been severely abused. And specifically the last cells all being at zero also seems strange. That and your measured voltage not matching the BMS reported voltage.

It gets more interesting looking at your list of voltages where there are only really 2 different levels. 3.749 and 4.075.

What's even better is that 4.075 * 96 = 391.2.

Edit: A number.
That's what I wondered too. Would it be more likely the CAN messages or readings are an issue such as the delicate aluminium wires back to the BMS pads are damaged coincidentally at the last cell on each module (strange). How many cells difference are there between the long and short range battery modules, nothing to do with firmware for a certain pack (55kWh vs 75 or 90)?
Probably there are some strange things going on when the active balancing is on, this is the 75kwh package two 23cells and two 25 cells

Can it be that bms boards are made for even number? then the last "odd" number cell is mesaured a different way?
nkiernan wrote: Wed May 26, 2021 6:32 am
fredd90 wrote: Wed May 26, 2021 4:27 am
Would be really nice if you nkiernan could do a manual controll of 06, 0F, 18, 1E
fredd90, can do. Are these from 0x401 or which one do they relate to?
Yes the message i got info from is 0x401
nkiernan
Posts: 505
Joined: Mon Feb 24, 2020 8:59 pm
Location: Ireland
Has thanked: 276 times
Been thanked: 64 times

Re: Model 3 High Voltage Controller

Post by nkiernan »

fredd90 wrote: Wed May 26, 2021 4:27 am
Would be really nice if you nkiernan could do a manual controll of 06, 0F, 18, 1E
Fredd90, got some interesting results looking at mine to compare with what you have! Assuming I have the conversions right.
TM3 Cell Voltages - 260521.GIF
Some stray low voltages in there I wasn't happy to see, but when I total these 96 values I get 272.194V. Measuring the actual pack voltage with a fluke meter gives me 294.5V. There are about 8 really low values there, and if they were averaging the same as other cells, that would bring the total up to around the measured value!

Hoping this is just a CAN message issue for some reason! Looking at a couple of the groups in more detail, group 13 stays consistent but group 3 bounces between zeros and these values.
TM3 Cell Voltages1 - 260521.GIF
TM3 Cell Voltages2 - 260521.GIF
Edit: Just to note, this is for a short range 55kWh pack
User avatar
fredd90
Posts: 87
Joined: Tue Oct 20, 2020 7:36 pm
Been thanked: 1 time

Re: Model 3 High Voltage Controller

Post by fredd90 »

Nice work👍. Did you also do it manually?

That is almost like the active balancing shuts off voltage measuring?
Some of Damiens video about just the bms balancing, did he expect the active balancing to be around 250mA if i remember correctly??
If the voltage measuring happens during balancing there would be a huge volt drop due to thin pcb leads.

One of my "zero" cell also fluctuated between 0v and 4v.

Gonna dig a bit deeper tomorrow after 5-6 hour of active balancing 👍
collin80
Posts: 110
Joined: Sun Aug 30, 2020 3:28 pm
Location: United States, Michigan
Been thanked: 4 times
Contact:

Re: Model 3 High Voltage Controller

Post by collin80 »

fredd90 wrote: Wed May 26, 2021 8:34 pm Nice work👍. Did you also do it manually?

That is almost like the active balancing shuts off voltage measuring?
Some of Damiens video about just the bms balancing, did he expect the active balancing to be around 250mA if i remember correctly??
If the voltage measuring happens during balancing there would be a huge volt drop due to thin pcb leads.

One of my "zero" cell also fluctuated between 0v and 4v.

Gonna dig a bit deeper tomorrow after 5-6 hour of active balancing 👍
In theory, the system works by temporarily disabling balancing as it is reading voltage.

But, something sure seems to be messing up the voltage readings. Remember that I saw the data in 0x332 also show the same 0V as people are getting with the 0x401 messages. So, it seems the BMS really does think those cells are dead when it measures them... sometimes. Other time it seems fine?!

I wonder if there is some sort of delay being caused by the BMS freaking out about the lack of CAN traffic? That shouldn't affect the readings you wouldn't think. Or, maybe with the cabling out of the car there are problems with data transmission over the ISOSPI lines?
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: Model 3 High Voltage Controller

Post by muehlpower »

nkiernan wrote: Wed May 26, 2021 7:54 pm
fredd90 wrote: Wed May 26, 2021 4:27 am
Would be really nice if you nkiernan could do a manual controll of 06, 0F, 18, 1E
Fredd90, got some interesting results looking at mine to compare with what you have! Assuming I have the conversions right.

TM3 Cell Voltages - 260521.GIF

Some stray low voltages in there I wasn't happy to see, but when I total these 96 values I get 272.194V. Measuring the actual pack voltage with a fluke meter gives me 294.5V. There are about 8 really low values there, and if they were averaging the same as other cells, that would bring the total up to around the measured value!

Hoping this is just a CAN message issue for some reason! Looking at a couple of the groups in more detail, group 13 stays consistent but group 3 bounces between zeros and these values.

TM3 Cell Voltages1 - 260521.GIF

TM3 Cell Voltages2 - 260521.GIF

Edit: Just to note, this is for a short range 55kWh pack
where do the values ​​come from? 0x401 from the CP CAN or can thay also be found on the vehicle CAN. What was the value of D1 in the respective measurements!
User avatar
fredd90
Posts: 87
Joined: Tue Oct 20, 2020 7:36 pm
Been thanked: 1 time

Re: Model 3 High Voltage Controller

Post by fredd90 »

Now tested with hvdc wiring in place, 391vcd to hv controller, but still the same readings.
The shunt was also connected, battery "battery always connected pin" got 196vdc

The value of D1 does not change due to missing values. I found these values in CP can not in vehicle can.

I have add my excel file and schematic of hv wiring.

-------------------------------------------------------------------
EDIT:, looking trough vehicle can, found 0x2D2 80 9D F5 9B 00 00 00 00 0.0900
1 hour later 0x2D2 80 9D 18 9C 00 00 00 00 0.1000

Maybe max min volt bms? scale 0,0001 bit 4 and 3 9BF5 : 39925 bit 2 and 1 9D80 : 40320
1 hour later, bit 4 and 3 9C18 : 39960 bit 2 and 1 9D80 : 40320

Maybe there is some movement in the lowest cell,
Attachments
tesla model 3 hv.pdf
(355.03 KiB) Downloaded 158 times
0x401.xlsx
(11.09 KiB) Downloaded 128 times
nkiernan
Posts: 505
Joined: Mon Feb 24, 2020 8:59 pm
Location: Ireland
Has thanked: 276 times
Been thanked: 64 times

Re: Model 3 High Voltage Controller

Post by nkiernan »

muehlpower wrote: Thu May 27, 2021 8:18 am
where do the values ​​come from? 0x401 from the CP CAN or can thay also be found on the vehicle CAN. What was the value of D1 in the respective measurements!
For my values, I got them from 0x401 on CP CAN bus with fredd90's help on how to read them. Image below shows the corresponding CAN data:
TM3 Cell Voltages3 - 260521.GIF
On the Vehicle CAN bus then, reading 0x332 I get below output for min and max cell voltages and cell numbers (same result Collin was able to read from my csv files earlier in this thread), and this shows a mismatch in results:
TM3 Cell Voltages4 - 260521.GIF
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: Model 3 High Voltage Controller

Post by muehlpower »

I am sure that only the values ​​where D1 = 0x2A are correct. With D1 = 0x15 all values ​​are dummy values. In your case the blocks 0x1C - 0x1F. Unfortunately, the strange values ​​are also marked as correct.
nkiernan
Posts: 505
Joined: Mon Feb 24, 2020 8:59 pm
Location: Ireland
Has thanked: 276 times
Been thanked: 64 times

Re: Model 3 High Voltage Controller

Post by nkiernan »

muehlpower wrote: Thu May 27, 2021 11:50 am I am sure that only the values ​​where D1 = 0x2A are correct. With D1 = 0x15 all values ​​are dummy values. In your case the blocks 0x1C - 0x1F. Unfortunately, the strange values ​​are also marked as correct.
You've read the same values from 0x401 and 0x332 from your own full pack yes? Do you get accurate consistent data from your own? You also pulled out these direct from the SPI data for comparison?

I checked the rest of the 0x1C to 0x1F values and they're consistent throughout with that dummy info. Was trying to make a link between the 75kWh pack (23bricks/25/25/23) and my short range pack to see if there was an explanation. Online says the 50kWh pack is also 23/25/25/23 with fewer cells per brick. My modules are the same length, 55kWh pack seemingly!? Is it correct to count the bricks corresponding to the BMS ribbon strip pads as per photos here? If so, I get 25 bricks per module across all four?
27.gif
26.gif
25.gif
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: Model 3 High Voltage Controller

Post by muehlpower »

nkiernan wrote: Thu May 27, 2021 12:53 pm
muehlpower wrote: Thu May 27, 2021 11:50 am I am sure that only the values ​​where D1 = 0x2A are correct. With D1 = 0x15 all values ​​are dummy values. In your case the blocks 0x1C - 0x1F. Unfortunately, the strange values ​​are also marked as correct.
You've read the same values from 0x401 and 0x332 from your own full pack yes? Do you get accurate consistent data from your own? You also pulled out these direct from the SPI data for comparison?

I checked the rest of the 0x1C to 0x1F values and they're consistent throughout with that dummy info. Was trying to make a link between the 75kWh pack (23bricks/25/25/23) and my short range pack to see if there was an explanation. Online says the 50kWh pack is also 23/25/25/23 with fewer cells per brick. My modules are the same length, 55kWh pack seemingly!? Is it correct to count the bricks corresponding to the BMS ribbon strip pads as per photos here? If so, I get 25 bricks per module across all four?

Here is my first log with all 4 modules. Only 0x2A for D1, and no strange values! Volt readings show 92V on the short modules and 100V on the long modules. It is a long range from a 2018 model 3. The BMS boards have connections on both sides, the later ones only had it on one side. I forgot the connection between module 2 and 3, but it still shows all values. Seems to be redundant!
210527full.png
210527full.csv
(1022 Bytes) Downloaded 102 times
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: Model 3 High Voltage Controller

Post by muehlpower »

nkiernan wrote: Thu May 27, 2021 12:53 pm
muehlpower wrote: Thu May 27, 2021 11:50 am I am sure that only the values ​​where D1 = 0x2A are correct. With D1 = 0x15 all values ​​are dummy values. In your case the blocks 0x1C - 0x1F. Unfortunately, the strange values ​​are also marked as correct.
I checked the rest of the 0x1C to 0x1F values and they're consistent throughout with that dummy info. Was trying to make a link between the 75kWh pack (23bricks/25/25/23) and my short range pack to see if there was an explanation. Online says the 50kWh pack is also 23/25/25/23 with fewer cells per brick. My modules are the same length, 55kWh pack seemingly!? Is it correct to count the bricks corresponding to the BMS ribbon strip pads as per photos here? If so, I get 25 bricks per module across all four?

it seems that there are 2 different short-range variants. if you look in classified, physically short "shortrange" and physically long "shortrange" are offered there. short has 24/24, long 23/25.
are your baatteries and HV controler from the same car?
nkiernan
Posts: 505
Joined: Mon Feb 24, 2020 8:59 pm
Location: Ireland
Has thanked: 276 times
Been thanked: 64 times

Re: Model 3 High Voltage Controller

Post by nkiernan »

muehlpower wrote: Fri May 28, 2021 8:32 am
it seems that there are 2 different short-range variants. if you look in classified, physically short "shortrange" and physically long "shortrange" are offered there. short has 24/24, long 23/25.
are your baatteries and HV controler from the same car?
Supposed to be from the same car yes, and they all came together but the penthouse had been opened. Some investigation and rechecking of CAN side to do this weekend.
User avatar
fredd90
Posts: 87
Joined: Tue Oct 20, 2020 7:36 pm
Been thanked: 1 time

Re: Model 3 High Voltage Controller

Post by fredd90 »

muehlpower wrote: Thu May 27, 2021 4:04 pm
nkiernan wrote: Thu May 27, 2021 12:53 pm
muehlpower wrote: Thu May 27, 2021 11:50 am I am sure that only the values ​​where D1 = 0x2A are correct. With D1 = 0x15 all values ​​are dummy values. In your case the blocks 0x1C - 0x1F. Unfortunately, the strange values ​​are also marked as correct.
You've read the same values from 0x401 and 0x332 from your own full pack yes? Do you get accurate consistent data from your own? You also pulled out these direct from the SPI data for comparison?

I checked the rest of the 0x1C to 0x1F values and they're consistent throughout with that dummy info. Was trying to make a link between the 75kWh pack (23bricks/25/25/23) and my short range pack to see if there was an explanation. Online says the 50kWh pack is also 23/25/25/23 with fewer cells per brick. My modules are the same length, 55kWh pack seemingly!? Is it correct to count the bricks corresponding to the BMS ribbon strip pads as per photos here? If so, I get 25 bricks per module across all four?

Here is my first log with all 4 modules. Only 0x2A for D1, and no strange values! Volt readings show 92V on the short modules and 100V on the long modules. It is a long range from a 2018 model 3. The BMS boards have connections on both sides, the later ones only had it on one side. I forgot the connection between module 2 and 3, but it still shows all values. Seems to be redundant!
210527full.png210527full.csv
Good to know that the Bms system is redundant, sad to admit that my pack have some problem, the battery cells is probably good because of the high voltage measured by my fluke multimeter. 391,5v.

Is there a slight chance of cutting bottom of penthouse and measure? evshop.eu sells model 3 "bms replacement" boards for orion 2 bms, "Just replace original boards and solder" they said.


My battery was from a scrapyard, the donor car looked nearly unharmed, and has around 1600km on the odometer. roof, windshield, hood was damaged.
Could this really messed up bms readings from the end of 4 batterypacks :?
Post Reply