Is a Leaf BMS OK with the Battery being split in the wrong place?? [SOLVED]

[Alibro]

Hi Guys
I'm basically on the home straight for my EV conversion and have spent a lot of time recently hooking up BMS wires in the battery packs at the front of the car. I am planning to use a Nissan Leaf BMS to monitor the pack with Leafspy and may also be able to use it for some level of control but just yesterday something occurred to me.
I know the Leaf BMS is designed to have the pack broken at cell 48 as it has an extra connection for the negative of cell 48 (IIRC) and I assume this is to maintain the full pack voltage at the BMS when the HV safety breaker is pulled.
In my case I have 40 cells at the front of the car and 56 cells at the rear and plan to use contactors in the rear battery box to break both pos and neg lines of the pack.
My question then is will the BMS have a problem with me breaking the pack at cell 56 instead of 48?

Thanks

[JaniK]

When you are driving/charging the pack is complete and BMS works. No problem there.

When car is parked and batteries split, there should be no consumption and no real need for the BMS to stay alive?

I plan to cut off all power when my car is parked.
Including the BMS. What would it do anyways when battery packs are split and no power can flow out? Well you could send a message to yourself that " I am BMS, I have consumed your 12V battery to empty, come feed me them electrons!"

My idea to start the car from hibernation:
1. Power to contactor between packs only
(race car type emergency cutoff)
2. Power to BMS (on/off toggle)
3. BMS controlled precharge + main contactor
4. Inverter power on from BMS
5. Gear select and go.

Only problem would be that the cell voltage rises after you have charged to 100% and the BMS is shut down so it can't balance/discharge the high cell via the resistor.

This can happen if battery warms up after charging.
(Car out in winter, charges to full outside, drives 10m to warm garage with 100%)

(In RC cars the battery is only balanced when charging and only "BMS' is the low voltage cutoff when discharging/driving. If I am not wrong its same for EV?)

I think the EV battery is top balanced when charging and when you are driving, the lowest cell voltage read by BMS will determine the point where the inverter should be switched to "regen only mode."

I can say that OEM leaf lets you drive the car until the battery has uumpf for just slowing down and parking only. But it will let you tow the vehicle in "turtle mode" and charge with up to 30kW power via regen.. trust me I have tested this twice

[johu]

The Leaf BMS has galvanic isolation between cell 48 and 49. So if the breaker is placed there no current can flow via the BMS and destroy it. When you split the pack somewhere else the BMS provides a current path when the breaker is open. We have destroyed many Leaf BMSes at clipper cab because they split the pack at 24/25 and 72/73 and then somehow we managed to run current.

Conclusion: not recommended

[Alibro]

The Leaf BMS has galvanic isolation between cell 48 and 49. So if the breaker is placed there no current can flow via the BMS and destroy it. When you split the pack somewhere else the BMS provides a current path when the breaker is open. We have destroyed many Leaf BMSes at clipper cab because they split the pack at 24/25 and 72/73 and then somehow we managed to run current.

Conclusion: not recommended

Thank you Johu, this is what I feared.
So I was wrong about the reason but right about the possible outcome, the extra neg at cell 48 is to protect the BMS, NOT to maintain the pack voltage with the manual disconnect pulled as I said.
The layout is what it is and changing it at this late stage is not an option therefore I will just push on.
Maybe I can leave the BMS disconnected most of the time and once in a while (with the pack live) connect it up to check the cells are still in balance.
It is not an ideal plan but for now the best I have unless anyone can suggest a better one.

[Alibro]

When you are driving/charging the pack is complete and BMS works. No problem there.

When car is parked and batteries split, there should be no consumption and no real need for the BMS to stay alive?

I plan to cut off all power when my car is parked.
Including the BMS. What would it do anyways when battery packs are split and no power can flow out? Well you could send a message to yourself that " I am BMS, I have consumed your 12V battery to empty, come feed me them electrons!"

My idea to start the car from hibernation:
1. Power to contactor between packs only
(race car type emergency cutoff)
2. Power to BMS (on/off toggle)
3. BMS controlled precharge + main contactor
4. Inverter power on from BMS
5. Gear select and go.

Only problem would be that the cell voltage rises after you have charged to 100% and the BMS is shut down so it can't balance/discharge the high cell via the resistor.

This can happen if battery warms up after charging.
(Car out in winter, charges to full outside, drives 10m to warm garage with 100%)

(In RC cars the battery is only balanced when charging and only "BMS' is the low voltage cutoff when discharging/driving. If I am not wrong its same for EV?)

I think the EV battery is top balanced when charging and when you are driving, the lowest cell voltage read by BMS will determine the point where the inverter should be switched to "regen only mode."

I can say that OEM leaf lets you drive the car until the battery has uumpf for just slowing down and parking only. But it will let you tow the vehicle in "turtle mode" and charge with up to 30kW power via regen.. trust me I have tested this twice

I don't think the BMS does much in the way of balancing anyway so if I find the pack is getting out of balance I will need to manually intervene.
For me the BMS is more about monitoring than balancing and from what Johu has said splitting it in the wrong place will probably kill the BMS.

[Alibro]

OK stupid suggestion time.
What if the connections going to the BMS either side of the proposed break at cell 56 were passed through a relay so that with the ignition off the pack is disconnected from the BMS it that point but when the ignition is turned on it is connected again.
Provided the Relay was timed such that it disconnected before the main battery contactors opened and reconnected after the contactors closed then there should never be a time that the BMS is the sole path of HV.

What do you think?

[JaniK]

OK stupid suggestion time.
What if the connections going to the BMS either side of the proposed break at cell 56 were passed through a relay so that with the ignition off the pack is broken to the BMS but when the ignition is turned on they are connected again.
Provided they were timed such that they disconnected before the main battery contactors opened and reconnected after the contactors closed then there should never be a time that the BMS is the sole path of HV.

What do you think?

This is basicly the "same result" as using a HV contactor to break the HV pack in 2 parts.?

The BMS only senses 1 side of cell, in your case if one VW module to module cable is disconnected, the bms will read 0V for one cell in one of the modules. The next cell should have a normal reading.

If you cut 1cell tap from each pack to BMS with a relay, you will have 2 cells reading 0V

If go this route, I would propably use a small fuse in the cable from the relay cut cell pack to BMS. Just in case.

I don't think the BMS does much in the way of balancing anyway

Her you see how much in 40km driving on highway the cell voltages vary on OEM 2014 Leaf. The BMS balances them when charged. When the battery gets more empty the differences are larger. E: This 40km took 6,32kWh

Starting with freshly charged battery.

After 40km of driving.

[2fast4u]

There is no reason not having the service switch between cell 48/49 even if the packs are split differently. With a fixed cable between your packs the split can be wherever you want as long as the service breaker is at the correct place.

[janosch]

Yeah, don't put the disconnects in another place, I have done that and it was stupid and I burnt more than one BMS that way.

[Alibro]

Yeah, don't put the disconnects in another place, I have done that and it was stupid and I burnt more than one BMS that way.

Thanks Janosch, are we right to say it will be OK so long as I use a relay to break the connection to the BMS? If the comments above are correct the relay could be at cell 48 regardless of where the split in the packs is.
My theory is by doing so the BMS will not have full pack voltage across it therefore any oddities from charger/heater etc. cannot harm it.
If the relay is fed through the disconnect tamper switch along with the contactors then when any HV is pulled everything should be disconnected at the same time.

For the moment it is a moot point as I am pushing on with the build without the BMS but I would like to have it in the near future if possible.

[Alibro]

This is the rear pack pretty much complete now

Cell 48 is the top left module so not a suitable place to mount the disconnect as there would still be voltage on the cable albeit a small voltage.

This is a closeup of the contactors etc.

It may be overkill using a manual disconnect alongside a contactor but I figured it would do no harm.

[janosch]

My theory is by doing so the BMS will not have full pack voltage across it therefore any oddities from charger/heater etc. cannot harm it.

Explain this in more details please. I don't think this is a good plan, but maybe I am misunderstanding.

Where would the relay be and where would the maintenance disconnect be?

The disconnect needs to be at 48/49 and that's the end of it I am afraid. Sorry to bring bad news

If you put it anywhere else you will have a bad time and release smoke sooner or later (for example when the contactors are welded & you are removing the BMS and momentarily touching the contactor again).

I posted the 40kWh PCB in the wiki a while ago, there is also a link to the MAX chip datasheet: https://openinverter.org/wiki/Nissan_Leaf_BMS

[Alibro]

You probably know better than me but the theory is while the HV battery is whole then the BMS will be OK. The problem is when the Battery is split in the wrong place then the BMS becomes the HV path for any device with residual voltage still connected like HV heater, charger or inverter.
If a relay or some other form of switch is in line with the BMS connection at cell 48/49 and is powered by the voltage going to the contactors and through the HV anti-tamper circuit and then should the HV be broken then the BMS would be disconnected at cell 48/49.

I guess the concern would be if the relay or solid state relay would switch fast enough to save the BMS.

[nkiernan]

I'm not really familiar with the Leaf BMS setup, but I had to deal with something similar for fuse and isolator locations using an Orion2 BMS with Tesla Model 3 batteries. There's a section in the Orion2 BMS documentation that explains what and why that helped me understand better. May not be relevant to Leaf, but might be of some help. I would be nervous of the relay approach mentioned.

See Safety Disconnect and Fuse Position section in doc linked below:

https://www.orionbms.com/manuals/pdf/or ... manual.pdf