[BogdanT]

I'm currently working on a Volt pack with SimpBMS, configured as 2 strings in parallel.
Does anyone know if the canbus connections to the slaves can be swapped?
eg: Original wiring goes through modules 30-24-24-18, with 30 being the first and 18 the last.

The reconfiguration means now that I have 30+18 in the front, and 24+24 in the back, and for simplifying the wiring I would like to connect them in the order 30-18-24-24.

Would that work or there's something in the last slave (18) that would need to keep it the last in the chain?

 

[BogdanT]

@swoozle do you remember if the slaves were spitting the voltages at the same time as receiving the balance command?

I'm playing around with the SIMP code and as soon as it enters "balance" mode (frames 300 and 310 populated) there are no CAN messages coming from the slaves.

 

[BogdanT]

I meant to say as soon as I send 300 (populated) + 310 (populated) + 200 -> no more CAN responses from the slaves. If I send just the 200 frame, they are responding. Probably I'm doing something wrong with Tom's code, I'll give it more thought tomorrow.
But thanks for the input, that's what I wanted to confirm that the cells are still spiting voltages on CAN when they receive the 'next' 200 frame after queueing balance commands.

 

[BogdanT]

@swoozle Back to you with a question... In your original excel file, I think there is a typo on frames 461 and 471
- the bytes to activate the cells seems to be reversed (when compared to the other 8-cell groups). Marked with yellow - reversed; green -assumed correct.

Can you confirm the typo or is that group somehow 'special'?

 

[BogdanT]

We're you able to further check the part on queueing "all" cells for balancing?

I'm working on fixing the SIMP balancing, which appears broken and would also like to implement that check/workaround.

Question would be:
Is that restriction applying on whole BICM or "module within BICM"? Eg. For BICM4 can you queue 17(of 18) cells or just 11(of 12) from one module + 5(of 6) from the other module?

 

[BogdanT]

It seems like it works. So what happened on my side: while trying to fix the SIMP bms code, I ended up with 3 cells on BICM4 at 3.90V and all the others in the full pack at 3.94

After clearing all the bugs, I left it to balance, effectively triggering all the other cells (except the 3 on BICM4).
Next morning all the cells were within 10mV.

The code loop was updated to trigger 4 balance requests x 200ms then 1 sampling request (blank frames 300/310) - not sure if this is needed to get correct voltage but seemed reasonable to do.

It's not a very accurate test but it deed seem to work Discharge time was around 3hours / 10mV.