It looks like one of my pgroups is dead…
This is what it looks like after running flat:
The board actually seems to work ok… But I’m I playing with fire?
What should I do here. Is there nothing for it but to get a new battery? Is it possible to just cut out the bad pgroup, rearrange the sense wires and run 17s?
You could be playing with fire. Do you know a battery builder or a skilled forum member nearby? Where are you based? Where did the pack come from?
Don’t cut out the P group. Your charger voltage will be for 18s still. You would be better to replace the P group than simply remove it.
You don’t actually know the P group is zero volts simply by looking at the app either. Until someone’s put a meter across the cells independently, I don’t know with certainty that it’s not a BMS failure, or wiring / solder joint failure of a balance interconnect.
The fact that your board still goes fine suggests either the P group is shorted out, or the P group is fine and the BMS/BMS wiring needs repaired or replaced. If vesctool shows 4V higher than the BMS app, then more likely you have a BMS/balance wiring fault.
You shouldn’t be using or charging the battery with a fault like this though. If you’re not competent then do as Ben suggests, take it to someone whom is and get it sorted.
That’s a great summary. I would definitely start with measuring the total pack voltage. If it’s closer to 72V than to 68V, it’s likely either a BMS or a balance wire issue. If it’s 68V, you likely have a bad P-group.
Nice summary there’s just one thing I would like to add.
Changing out a P group is not a safe option if the battery has been used for a while already. I would personally say around 10 cycles as the limit. This is because there would be some difference between the new and the old cells and the battery pack would be much more prone to getting out of balance.
Depending on what charger you have, you might be able to open it up and just tune down the voltage to 17S with a potentiometer. If it’s possible to do that then that’s what I would probably do, assuming the p group is actually dead.
Removing the P group from the pack and properly rewiring in the bms is not an easy work so if you don’t feel confident at doing it find an expert. Note, you have to consult the BMS’s product page to see how you need to rewire the BMS for a lower S count if it supports it, or if it doesn’t support it then you also need a new BMS to go 17S, but most if not all LLTs support it.
@GnomeMeansNo emphasizes on this, charging cycles on this battery with this fault can cause fires or take a small issue and make it into a big issue. The advantage i have found with smart bms is that you can see these things early and make small repairs before they kill a pack by ruining a p group.
It MAY be repairable but it absolutely needs to be benched. Lots of good advice here already, all of it needs to start with taking the deck out of rotation and opening it up.
I agree it’s not ideal but as long as BMS is doing its job correctly, it’s not unsafe either. I would just turn the balancing off and accept the fact that BMS won’t allow the battery to get fully charged as the replaced P-group gets out of balance with the original P-groups.
Not ideal but there’s really no good alternative other than disposing the whole battery, which I would consider wasteful.
Well, the alternative is going to 17S! Just take out that one cell group, rewire the BMS, and modify the charger voltage.
It’s a bit more work than replacing the P group but in my opinion it’s the safer call.
I think the biggest thing, imho, when replacing pgroups with unmatched cells for a repair is it makes charging and discharging a more active process. You now have to monitor the pack when charging above the cv transition to make sure it balances and then also when discharge goes below the ?20%? Soc to make sure the esc doesn’t over discharge that or other groups too far.
It becomes a problem for some to always be aware of the packs new requirements to keep it healthy so i hesitate to suggest it as a solution without big caveats and warnings. 
Edit, these are possible problems, but if you’re asking here for advice I assure you are not expert in diagnosing battery issues in the first place and don’t want to set someone up for failure
edit - misunderstood your comment. Deleted.
Hey Angus - sorry to see it man. This sucks.
Yeah, charging this is 100% playing with fire. Trying to charge an overdischarged p-group is how battery fires are started. (dendrites/internal shorts = fire boom)
For remedy, you could remove the two problem p-groups and run the board at 16s.
LLT smart bms’s typically have a range of s counts they can support. I would look up your model and see if it can support 16s. If it can, you need to splice the balance wires in the correct way to get it to balance for a 16s battery.
You also would need a 16s charger. You would have a lower top speed, but you could swap out the motor pulley to 18t or 20t to get the speed back if you want that.
orrrr you could buy my nazare supersport that is up for sale 
Thanks for the responses so far folks.
To answer some questions:
It’s a Lonestar Supersport.
Does vesc show the same voltage as LTT or is it 4v higher:
I can’t see the voltage in the android tool but since it’s at 47% I’m going to assume it’s 4v low.
Can the charger be tuned down to 17s. These are the YEWY chargers it came with:
Not sure if there’s a Pot inside them to change the voltage.
Can the BMS be changed to support different s-sizes:
It would seem so.
It would seem to me that even if the chargers can’t be changed to run at 17s, that would actually be ok as long as the BMS can be set for 17s since it cuts off the charging when the pack tries to go over voltage.
Any flaw in that logic?
If the BMS would fail to stop the charging for whatever reason it might burn your house down.
Good. Definitely check the product page for a wiring diagram though, the balance wires definitely need rewiring if you are going to run it at a lower s count.
If you do not do surgery on that pack to remove the dead p-groups (9 and 10), you are likely to start a battery fire.
Once, you remove the dead p-groups, you need to get an experienced battery builder to either re-wire your balance leads to support 16s, or install a new 16s BMS.
If you just modify the app to 16s, it’s going to give you incorrect readings. It will do nothing to make anything safe.
I cannot stress this enough - the BMS does not protect you at all in the scenario you have.
Your battery has p-groups that are beyond DEAD. There is no bringing back those p-groups. The BMS does not protect you in this situation. A BMS works to keep a healthy pack well balanced.
You have a pack with two p-groups that are dead af, and are a fire hazard.
You know those kaly xlr’s that burned down people’s homes? this is basically what you have on your hands.
Do not ride this board. Do not charge this board. Put it somewhere that you don’t mind burning down until you have someone dispose of the p-groups 9 and 10.
Yup I hear you shadow, I’m not proposing that I keep the dead pgroup wired in - I’m talking about bypassing it to cut it out of the circuit and rearranging the balance wires so it’s completely out of the system even if it’s physically in the pack still.
I think there’s only one dead pgroup. The reason one or two others that look a bit out of balance is that the pack does not charge high enough for balancing to kick in any more. That’s what I’m thinking anyway - I’m hoping they will come back into balance once the LTT is set to 17s and actually hits the balancing level.
18s8p on the loanstar ss is a pcb style pack iirc with 6 8p groups per section and modifications are kind of a pain but not impossible. Open and take some voltage readings from the pcb - it will be the center cell group
Ok, sweet. It also depends what generation of lacroix battery you have here.
Originally, lacroix did single p group packs connected by wire. These are easy to replace or remove a single p group in.
In the last couple years, lacroix moved to PCB battery packs, with 6 p-groups on a single pcb. These are more difficult to replace or remove a single p-group.
Looking at your first picture where p-group 10 is drained 200mv lower than the other groups (and then 4.018v at “full”), I assume that p-group is also toast. Either loose welds or cells gone bad.
It’s possible that the p-group 10 charging or discharge is being affected by 9, but tbh I am not confident either way. It’s likely to be dead, but it might be ok.
Yeah now that you mention it I see P group 10 doesn’t look good either. My guess would be that there’s a loose weld on one of the cells and the other cells are doing all the work in that group. I think it would be wise to cut that one out too and go to 16S. That’s a much more common voltage anyways, so it you need to buy a new charger it will be easier to find one.
Just FYI, regarding voltage on the andriod vesctool, scroll to one screen above the main dash screen and it’s there.
