diy solar

diy solar

One cell spiking while under charge

ScottWCO

New Member
Joined
Feb 4, 2021
Messages
55
Hi gang.

I've got two LiFePO4 banks I built last year out of EVE 280Ah cells, using JBD BMSs.

I've really had no troubles with these at all, except twice - once a few months ago while on a trip in my van build-in-progress, and then again just a little while ago while testing this out after some system revisions and getting everything reinstalled and connected again.

What's happening is with one of the batteries, when put on a good charge (in this case, shore power via a Victron MP 3k inverter/charger, set at 15amps), cell #1 will spike up. The BMS will try and burn off excess to balance it, but it's too much. I took it off the charge quickly. This also happened a few months ago, I disabled charging when I saw this going on then as well.

I've done some other charging - a few days after the first time this happened late last year, I was also testing out my solar panels and SCC. I had it all hooked up and charge came in fine, and I don't recall seeing any issues. I'm going to hook it up again in a few days and test it out again and watch more carefully to be sure.

Recently I rebuilt my batteries, took them out of the janky plywood enclosures I'd made, got them fitted in something better. So all the connections were redone, but the cells were kept in the same configurations within each pack.

I figure there's two possibilities here - 1) BMS issue, 2) cell issue. I'm leaning more toward cell issue, as I don't see the BMS having much to do with charging other than allowing/disallowing it. I plan to pull that battery out, disassemble it, mix up the cell positions, put it back together, try it again and see if the spiky behavior follows the cell.

Two screenshots - one at rest, one under charge.
 

Attachments

  • IMG_3546.PNG
    IMG_3546.PNG
    503.1 KB · Views: 25
  • IMG_3545.PNG
    IMG_3545.PNG
    525.5 KB · Views: 26
First thing to do is verify the BMS cell voltages against your reference DVOM.
Report back with your findings.
 
What is the bulk/absorption voltage setting on the Multiplus?
Is the charge set to terminate on float or tail current?
What is the float votlage setting on the Multiplus?
 
I think you have the BMS misconfigured.
Cell 0 shows a voltage of 3.662 volts but the charge fets are still closed.
This is not good.
You BMS should open the charge fets when the first cell exceeds 3.65 volts.
 
I think you have the BMS misconfigured.
Cell 0 shows a voltage of 3.662 volts but the charge fets are still closed.
This is not good.
You BMS should open the charge fets when the first cell exceeds 3.65 volts.
I'll have to do some screenshots of the settings once I'm back at the van (I've got it in a friend's shop nearby).

I'm pretty positive I saved the settings from one and used them on the other. One battery works fine, one is exhibiting issues.

"Charge fets are still closed" - where do you see that? I'm looking at both screenshots and not seeing anything obvious. Unless you're referring to the screenshot I mentioned showed everything at rest? I had disabled charging to be safe at that point.
 
I'll have to do some screenshots of the settings once I'm back at the van (I've got it in a friend's shop nearby).

I'm pretty positive I saved the settings from one and used them on the other. One battery works fine, one is exhibiting issues.

"Charge fets are still closed" - where do you see that? I'm looking at both screenshots and not seeing anything obvious. Unless you're referring to the screenshot I mentioned showed everything at rest? I had disabled charging to be safe at that point.
The image where the battery voltage is 13.81 volts.
In the top left corner it shows 26.48 amps going into the battery.
Just below that we see the toggle widget is in the on position.
Which stands to reason because there is current going into the battery.
But the toggle should be in the off position and no current should be going into the battery.
 
Yes, first screenshot is with charge going in and the first cell spiking.

Then I disabled charging via the BMS and took the second screenshot.
 
Yes, first screenshot is with charge going in and the first cell spiking.

Then I disabled charging via the BMS and took the second screenshot.
The BMS should have disabled charging.
Your BMS is not protecting your battery so I suspect it is mis-configured.
 
The BMS should have disabled charging.
Your BMS is not protecting your battery so I suspect it is mis-configured.
Ahh, I see where you're going now. Yeah I'll get screenshots of the settings on both packs and post them when I can.
 
I plan to pull that battery out, disassemble it, mix up the cell positions, put it back together, try it again and see if the spiky behavior follows the cell.
Did you do any logical pairing/selection of the cells in each battery? Or random?

If you put the stronger cells (higher capacity) in one battery and the weaker ones in the other, it will naturally charge more evenly.

Did you perform a top balance?
 
Did you do any logical pairing/selection of the cells in each battery? Or random?

If you put the stronger cells (higher capacity) in one battery and the weaker ones in the other, it will naturally charge more evenly.

Did you perform a top balance?
I got all eight cells at the same time. They all tested the same voltage upon receipt, so I just took four for one pack, four for the other, so essentiall random.

I did a good top balance when I first assembled the packs. The four cells in each pack all are very close at rest or under light load. It's only under a good amount of charge that the one pack acts wonky. Other pack is fine.
 
It's only under a good amount of charge that the one pack acts wonky. Other pack is fine
First i will have to admit that all the weird chit with cells seems to happen to cell in position 1.

You should label your cells (i like letters A, B, C... because the cell positions are numbers). Note this wonky (possibly low capacity) cell and take notes. This will be vital in making sense of changes you make and the results your changes yield.
 
So here's the BMS settings for each pack, three screenshots each. B1, B2. B2 is the trouble pack. I do see that I have a cell undervoltage release value different between them, I'll have to get that resolved, but that shouldn't be pertinent here.

That said, if I have anything obviously set wrong here, I'd very much like to know. I bought the BMSs from Overkill Solar and I used their suggested settings.

Tomorrow I'm going to pull that pack out, disassemble it, and reorder the cells inside and see if that makes a difference.
 

Attachments

  • IMG_3548.PNG
    IMG_3548.PNG
    635.9 KB · Views: 17
  • IMG_3553.PNG
    IMG_3553.PNG
    760.4 KB · Views: 14
  • IMG_3552.PNG
    IMG_3552.PNG
    851.4 KB · Views: 13
  • IMG_3551.PNG
    IMG_3551.PNG
    636.8 KB · Views: 10
  • IMG_3550.PNG
    IMG_3550.PNG
    759.4 KB · Views: 12
  • IMG_3549.PNG
    IMG_3549.PNG
    859.9 KB · Views: 16
Hi gang.

I've got two LiFePO4 banks I built last year out of EVE 280Ah cells, using JBD BMSs.

I've really had no troubles with these at all, except twice - once a few months ago while on a trip in my van build-in-progress, and then again just a little while ago while testing this out after some system revisions and getting everything reinstalled and connected again.

What's happening is with one of the batteries, when put on a good charge (in this case, shore power via a Victron MP 3k inverter/charger, set at 15amps), cell #1 will spike up. The BMS will try and burn off excess to balance it, but it's too much. I took it off the charge quickly. This also happened a few months ago, I disabled charging when I saw this going on then as well.

I've done some other charging - a few days after the first time this happened late last year, I was also testing out my solar panels and SCC. I had it all hooked up and charge came in fine, and I don't recall seeing any issues. I'm going to hook it up again in a few days and test it out again and watch more carefully to be sure.

Recently I rebuilt my batteries, took them out of the janky plywood enclosures I'd made, got them fitted in something better. So all the connections were redone, but the cells were kept in the same configurations within each pack.

I figure there's two possibilities here - 1) BMS issue, 2) cell issue. I'm leaning more toward cell issue, as I don't see the BMS having much to do with charging other than allowing/disallowing it. I plan to pull that battery out, disassemble it, mix up the cell positions, put it back together, try it again and see if the spiky behavior follows the cell.

Two screenshots - one at rest, one under charge.

First check for a bad connection. Probably the most common problem.

It is possible that a good Top Balance may correct the spiking cell. See attached Top Balance Tutorial document.

If the top balance doesn't fully correct the spiking (runner cell) possibly an active balancer. It is important to only enable the active balancer when the the cells are above 3.4 volts. This can be accomplished by simply plugging & unplugging the Heltec active balancer or solder a switch on the Run solder pad. Some automate the process by switching the Run solder pad by a remote device.

Heltec Active Balancer Link.


Another way to Top Balance With BMS Connected (preferred by some since BMS protecting cells)

First ensure all connections are good on each cell of the battery and the BMS is correctly wired to the Lifepo4 cells that are wired in series.
Set BMS for a 3.65 volt high voltage disconnect (Cell Over Voltage).

Wire an alligator clip to each end of a 2 ohm resistor. Charge entire battery at a higher current to get cell to start its run. Then charge entire battery at a low current and clip the alligator clips on the 2 studs of the runner cell observed on the BMS app. Watch the runner cell bleed down on the BMS. Repeat as necessary on each runner cell while slowly increasing voltage up to around to 3.60 or so volts for each cell as read on the BMS until all cells are balanced at a vey low current.

As long as the BMS is connected and set for 3.65 volt per cell high voltage disconnect this method should safely top balance a battery with the cells wired in series.

With a cell at 3.60 volts the 2 ohm resister will bleed about 1.8 amps from the cell. I = E/R 3.6Volts/2ohms = 1.8 amps. This is usually sufficient to bleed down a runner cell when charging the battery at a low current. If runner cell still going into over voltage lower battery charging current.

2 ohm 100 watt resister link.
https://www.amazon.com/dp/B07CVRH6DF
 

Attachments

  • Top Balance Tutorial .pdf
    369.2 KB · Views: 12
Last edited:
So it seems resolved!

When I got the lid off that one pack, I found the hut holding the pair of 6awg cables from the last cell positive to the box lid terminal wasn't super-tight. Got that resolved, did a little testing, all seemed well. Put the pack back in my van and got it hooked up to the main system, threw shore power charging on, and all went well - no more cell #1 spiking.

This morning I got my other battery back in place as well, same test, everything went just fine.

Gotta make sure those connections are good and tight!
 
I've been following this thread and it has solved my near identical problem. I just want to add my note, and thanks to all contributors - in particular Dynoman, who's notes on the active balancer really educated me!
My setup is 8 x 300ah cells 4s2p with JBD. When I bought the cells last year from a recommended guy here in the UK, he offered me an active balancer and just said "fit one of these, it's better than the JBD one". I just took him at his word and fitted it, but I left the JBD balancer switched on. After a few weeks or so I started to notice a runaway cell at position 2. But it wasn't really affecting overall performance that much, and I was on the road (RV installation) and the performance of my new 600ah lithium pack was so amazing compared with my previous LA setup, that I just made a mental note to look into it next time I was working on the pack - to get at the pack to either change a cell position or even check the connections was quite a big job - my battery box is quite crowded.
But then a couple weeks ago I came across this thread and decided to dive in and see what I could do. I had just experience an issue with my b2b not delivering its full potential, and that turned out to be a loose connection - in fact I have just discovered the hard way that thousands of miles of travel can loosen connections - I now know to check everything every few months as a matter of routine.
Just like ScottWCO, I found that the end connection of the pack was loose - only in my case it was the negative side. The nut turned up at least 2 more turns - but everything else on the pack was still good and tight. And then I remembered Dynoman's comment and wondered whether my balancer setup was wrong - so I read as many comments and reviews of the balancer as I could, and several guys said to disable the JBD balancer if an active balancer was used. I don't know what mine is, whether it is a Heltec or a clone, but it looks the same. So I disabled the JBD balancer in the app.
Well the difference was immediate - as the pack was being charged up from shore power from below 50%, there was no longer cell 2 creeping up and reaching 3.65 before the rest. When charge finally shut off at the preset 14.2v, cell 2 hadn't hit 3.65 as it had previously. But the delta was a bit off. But then with charge finished, the active balancer did its thing - it hadn't before - and within 4 hours all 4 cell packs were at a solid 3.55.
So my problem looks fixed and I have learnt lots.
Again many thanks to all in this thread.
Ron B, UK, currently in Turkey.
 
I can't speak to active balancers, I've not used one, but it makes sense that two balancers might fight each other. An active balancer would be far better than the weak little one the BMS itself offers. Glad to hear you got your issue sorted out too!
 
with no balancer only the bms my 4 16spackshave ll fallen out of balance to the point the pack only charges at 54.1 vice 54.4 while not the worst, it is obvious i need to match the cells a bit closer this spring. at the DIY level perfection will not happen.. remember used cell cannot e shipped out of a country as hazmat rules apply. this means that all the folks...99.9 telling you they are selling you brand new cells are full of shit.. just acknowledge the fact that the cells you bought are used and halfway at least through life.. and get on with life... thats why you can afford them.
 
Back
Top