Nikola+ charging issue (BMS?)

[siffleur]

Here’s a series of events

0. The red lights at the front of the nikola started flashing instead of the colored lights around the circumference of the wheel. This normally indicates an error with the dc-dc converter board and sometimes solves itself. (See: 

)

1. The red light stopped blinking after 2 days

2. Three days later (7 calendar days at 70% battery with two days of ~1mi riding each) the battery stops charging at 93v.

3. I test múltiple chargers, same issue. They are outputting the correct voltage. 

4. I open up the wheel and everything seems okay. Connections look good

5. I disconnect one battery at a time, and the battery on the opposite side of the board seems to be the issue. I can charge the other pack with that one disconnected, but with just that pack connected, the charger stays green (indicator doesn’t turn green to indicate current flow). Furthermore, if I have both packs connected and momentarily disconnect the balance leads (4-wire clip connector, not XT60/30), the charger will kick in.

6. This sounds like a bms issue with that pack. Are there things I can do to test it? Replace it? What do you recommend in this instance? Is it worth disassembling and testing each cell individually? I’m into rc, and am an electrical engineer so am more comfortable with this stuff. 

7. Some folks seem to suggest charging individually to balance charge and sometimes the packs recover. Is this true? How do you balance charge the pack of its bms isn’t letting you charge it at all?

8. Does anyone have a pinout of the balance leads? I’m curious what each wire indicates. 
 

thanks for your help!

looks like folks from this thread may be able to help provide insight:

@RagingGrandpa @alcatraz @mike_bike_kite  @Chriull

[Chriull]

6 hours ago, siffleur said:

This sounds like a bms issue with that pack. Are there things I can do to test it? Replace it?

Could be. I'd bet on some dead/bad cell groups.

For testing one would have to open the pack and measure the voltages. Most likely some cell groups have low voltage and have to be replaced.

To finally determine if it's the bms or the cells "original" fault thourough anaoysis and testing  of the board is necessary.

6 hours ago, siffleur said:

Some folks seem to suggest charging individually to balance charge and sometimes the packs recover. Is this true? How do you balance charge the pack of its bms isn’t letting you charge it at all?

If it makes sense depends on how deep these cell groups got discharged. Carging individually is only possible after opening the pack.

6 hours ago, siffleur said:

Does anyone have a pinout of the balance leads? I’m curious what each wire indicates. 

There are no balance leads. Just the single cells after opening the pack.

Imho best to sell it to some diy guy who can use the good cells and get a new pack.

[RagingGrandpa]

On 11/29/2022 at 10:11 AM, siffleur said:

I disconnect one battery at a time, and the battery on the opposite side of the board seems to be the issue. I can charge the other pack with that one disconnected, but with just that pack connected, the charger stays green (indicator doesn’t turn green to indicate current flow). Furthermore, if I have both packs connected and momentarily disconnect the balance leads (4-wire clip connector, not XT60/30), the charger will kick in.

That's conclusive: some cell group in your left-side pack is bad. 
Do not ride the EUC with that pack installed, it will be releasing heat quickly if current flows...

The 4-wire connector between packs is what I referred to as the "charge-stop trigger link." The pins are redundant- it is only one signal, a voltage between red and black, which will be driven high by the BMS in any pack that has a problem, and serves to prevent all other packs from recharging.

Open the shrink and measure individual cell voltages in the left-side pack. 
If any cell groups are below 2V, the pack is garbage, dispose of it. 
If some groups are low but still above 2V, you could attempt recharging them individually using test leads and a single-cell charger or bench power supply. (This process is called "balancing voltage manually.")
Once all cell voltages are equalized, retry whole-pack charging.

Avoid temptation to operate the Nikola on one pack only. The cells and interconnects are not suitable for bearing the full ~60A of controller current from one pack alone, and you risk a cutout during high-speed riding. If you must ride, enable the L2 beep as a reminder (22mph).

 

Edited November 30, 2022 by RagingGrandpa

[alcatraz]

There are likely bad cell groups and that pack is likely compromised and should not be used. However, in the hands of an ELECTRICAL ENGINEER, there's no reason that pack can't still be used for many many years. You just can't give the pack away to someone else.

Just attach balance leads and balance the groups that are too low for the bms to balance. I don't mean charge sub 2.5v groups. I mean charge the groups that are above 2.5v but won't balance with the bms alone. There are a couple of reasons why that would happen. You can use a simple 1S charger 2-3 amps and hook it up to the balance lead.

Mismatched cells is not as bad as dead cells.

[siffleur]

Update: here's photo of inside the pack. looks like I was fortunate that some chips let the smoke out but didn't let the fire out. 

https://ibb.co/Km8qSNF

https://ibb.co/5sy2zMm

 

all of the cells measure 3.9v within ~30mv, but one measures 3.58v. Withe the burnt components on the BMS, I don't think its worth trying to rebalance, but any insight you have would be appreciated.@alcatraz @RagingGrandpa @Chriull

Edited December 1, 2022 by RagingGrandpa
(img url)

[Chriull]

4 hours ago, siffleur said:

Update: here's photo of inside the pack. looks like I was fortunate that some chips let the smoke out but didn't let the fire out. 

Chips die fast - there is no real fire.

Edit: Seems like a PCB track vaporized? The other side of the PCB would be interesting to see?

Could it be, that there were some hot nickel strips, too? If welds/nickel strips/li ion cells got overheated they should be replaced.

Orherwise the voltages seem fine - could maybe even be balanced (within some cycles) by a full functional bms again.

Edited December 1, 2022 by Chriull

[RagingGrandpa]

On 12/1/2022 at 12:16 AM, siffleur said:

Withe the burnt components on the BMS, I don't think its worth trying to rebalance

Correct, don't use it as-is.

But it's good news, because it is practical to replace the BMS without discarding the cells.
It's a careful soldering job, reasonable for any technician to attempt.
(But if you don't understand which connections do what, don't mess with this- you'll short something and destroy your cell strips in the process. Take it to a battery repair person.)

You'll need a replacement Begode BMS PCB. Newer versions of it are still compatible with your cells and EUC, nevermind that its components are arranged differently now.
https://www.aliexpress.us/item/3256803819068033.html 

 

Edited December 13, 2022 by RagingGrandpa

[Paradox]

5 minutes ago, RagingGrandpa said:

But it's good news, because it is practical to replace the BMS without discarding the cells.
It's a careful soldering job, reasonable for any technician to attempt.
(But if you don't understand which connections do what, don't mess with this- you'll short something and destroy your cell strips in the process. Take it to a battery repair person.)

@RagingGrandpa @siffleur  I found this on a fellow EUC riders channel. 

 

[RagingGrandpa]

1 hour ago, Paradox said:

I found this
https://www.youtube.com/watch?v=Kjt08ogLSFk

Cringe-worthy... insulate the center row of balancing leads before lifting the PCB; if they touch their neighbors, they vaporize 

I appreciate the welding blanket, but why are there puppies on the floor and not a 5gal bucket of water?  

 

Edited December 1, 2022 by RagingGrandpa

[alcatraz]

This is good news in a way. The cells are not dead. 

I think you've just got a dead balancing circuit. (I'm not an electrical engineer.) This pack is potentially a great candidate for attaching balance leads.

What I'd do is to see if the overcharge protection works or not. Use a 1S charger and charge that weak group 100mV higher than the rest and then observe it when you charge the pack. It shouldn't go over 4.25v without the bms cutting charging and switching the charger light to green.

If that works, then you are pretty much 99.99% protected. You just need to manually balance charge that group every few months. 

If it didn't work you can try to manually balance charge the group to the same as the rest and observe it after discharging and charging. If its tendency is to lead the other groups when discharging, and trail them when charging the pack, you're safe. It means there's more internal resistance in that group and that will keep you safe when charging/discharging blindly. You just need to manually balance the group every few weeks/months (more frequent depending on how much you ride).

The reason why I feel I can speak on it relatively confidently is because I have a wheel with those symptoms. It won't balance but I'm either protected by the weak group's characteristics AND/OR the over voltage protection confirmed works. 

Just remember that an unmatched pack is a problem and you need to monitor the pack voltages. As the pack ages, you might discover accelerated deviation in voltages. For me this didn't happen. 3 years and no accelerated deviation. My pack however is a bit oversized for the wheel current draw. If you push your wheel you could a bit easier overheat those weaker cells. So observation is the key. Open the wheel up after a hot summer month and look for overheating signs. In winter just check the voltages. 

Swapping the BMS might sound like a solution but it's not a guaranteed fix. You might want to first charge the group and ride a bit. If those cells deviate again within just a ride or two, it's possible that the balancing circuit is too weak. A new bms will then maybe not save the pack. Some might say, "ok - the pack is done" and that's not exactly true either. Original bms' have very weak balancing circuits. They're only designed to take out small amounts of imbalance, and extend the life by a few years. With a powerful bms you can extend that by many more years. Is it FUN to ride a pack for 10years? Not really from a performance standpoint as the pack will slowly develop a lot of sag etc. But is CAN work fine at let's say 90-80% of maximum range and currents of a new pack. So tone down your riding style a smidge and enjoy all the extra years.

Edited December 4, 2022 by alcatraz

[Chriull]

7 hours ago, alcatraz said:

Swapping the BMS might sound like a solution but it's not a guaranteed fix.

The bms is broken and is to be replaced.

Without knowing what's broken it's not safe to use it.

Of course no one knows the state of the cells, the weldings, nickel strips, etc by now.

[alcatraz]

I mean shelling out the money and time to replace it doesn't guarantee a working pack. He might want to check before investing, if that group is still fairly matched or not.

[Chriull]

9 minutes ago, alcatraz said:

I mean shelling out the money and time to replace it doesn't guarantee a working pack. He might want to check before investing, if that group is still fairly matched or not.

Seems they are still quite in range:

On 12/1/2022 at 6:16 AM, siffleur said:

all of the cells measure 3.9v within ~30mv, but one measures 3.58v

 

[alcatraz]

Yeah the cells are definitely not trash. But they could still be too much to handle for an original bms. They could also be fine, of course. It's hard to know for sure unless you do some monitoring for a ride or two.

Bleed resistor balancing is pretty awful against one weak group. All 23 other groups need to hit the overcharge protection countless times to bring it up. If that group is unmatched, the rest of the pack has to "suffer". If you're going to swap bms, then at least help it back to the other groups with a 1S charger first, and give that new bms a fighting chance.

[RagingGrandpa]

The BMS failed. (Its circuits burned open!)
After it failed, cells became imbalanced.

You must replace the BMS with an original Begod BMS. 
And perform one-time manual balancing, to equalize the cells that were imbalanced because of BMS failure.

Replacing with an alternative BMS would defeat the coordination of charge-stop with the other, original battery pack. That's unsafe.

 

Edited December 5, 2022 by RagingGrandpa

[alcatraz]

Did the cell group voltage lower because of the bms failure? Or did the bms fail because the cell group failed? It's not unwise to analyze a bit before taking action.

If an original bms is an option then go that route. If it probably won't work, then you need to consider other options. 

On 12/6/2022 at 12:19 AM, RagingGrandpa said:

Replacing with an alternative BMS would defeat the coordination of charge-stop with the other, original battery pack. That's unsafe.

Are you interested in a third party bms?

[RagingGrandpa]

12 hours ago, alcatraz said:

did the bms fail because the cell group failed?

No.
Because: no cell group has failed. The lowest group was just 0.3V below the others. Imbalanced, but uninjured.
And because the burned section of the BMS is not near the load resistors.

It's clear to me- that BMS had a defect.

In any case, retrofitting 3rd-party BMS requires consideration of how to ensure that recharging will stop in both packs, when there is an overvoltage cell in just one of them. It is an important safety feature and needs to remain effective.

[alcatraz]

You brought up third party bms' as a solution and now arguing against it. What's with that? Hehe.

A repair shop would definitely measure that cell group's internal resistance/impedance before doing anything. If it differs significantly from the rest, then replacing with an original bms won't work.

Original bms' are designed to work with new matched cells, degrading together. Once they mismatch, it will fail quickly. I'm just surprised at the jump to swap it without checking first that a new one would work. 

I'm just informing the OP. Use the info as you please.

[siffleur]

@alcatraz@RagingGrandpa I'm thinking thru my battery hygiene if there are things I can learn from this. I charge at two places: work and and home.

At work, I charge normally: charger is on, I plug charger in and charge, once I see the green light, I wait to balance, then unplug.

At home, I have a charger on a smart plug. It turns on for 4 hours overnight. When I come home, I plug it in when the charger is off. It sparks as the voltage equalizes. Then it sits with the charger unplugged for several hours before and after it charges when the smart plug is not on.

Do you guys know how these gotway chargers (or EUC chargers / chargers in general) behave when they have voltage on the battery side and not the power source side? Would that be detrimental to battery health? Happy to pose this question to the larger community in a dedicated thread as well.

[Paradox]

1 hour ago, siffleur said:

At home, I have a charger on a smart plug. It turns on for 4 hours overnight. When I come home, I plug it in when the charger is off. It sparks as the voltage equalizes.

If you turn your smart plug on for a moment and connect your wheel then turn your smart plug off you eliminate the spark. 

[alcatraz]

I just got a smart plug myself and there's a switch on it. Maybe you can just temporarily turn it on/off when connecting? (like Paradox said)

[RagingGrandpa]

On 12/9/2022 at 9:58 PM, alcatraz said:

You brought up third party bms' as a solution

Hmm, I didn't mean to... no harm, but what made you think that?

 

On 12/10/2022 at 10:05 AM, siffleur said:

Do you know how (EUC chargers / chargers in general) behave when they have voltage on the battery side and not the power source side?
Would that be detrimental to battery health?

A charger attached to the EUC and not connected to AC is practically harmless.

A small parasitic current will flow to it (less than 1W of power). It's simple to measure and confirm with a handheld ammeter, if you're familiar. It would take months for this tiny load to affect the state of charge of the large EUC battery. And many modern EUC's (LiTech packs; Leaperkim; etc) do not allow any current to flow out of the charging port, preventing parasitic drain entirely.

It won't cause imbalance.

 

Edited December 14, 2022 by RagingGrandpa