Is BMS supposed to run until overvoltage when charging?

I recently built a 12v LIFEPO4 battery with 4 3.2v CALB cells from AliExpress and the BMD from Overkill (basically the Will Prowse DIY 12v battery). I discharged the battery to zero during a capacity test and then hooked it up to my charge converter (Progressive Dynamics PD9160ALV) and watched it go. The XiaoXiang app correclty showed the current going in to the battery and the voltage of each cell. It reached "100ah" on the app a little early based on the current going in (60 amps/hr) and time it took (1hr+15min), but I assume it was just showing 100AH and 100% because the voltage had reached 3.4 per cell. (? correct)
Anyway, it kept charging as I assumed it would since I knew 100AH had not actually gone in to the battery. At around the 1+40 mark the cells were now starting to climb in voltage and showing 3.53-3.59. Also, the current going in had slowly started to drop from 60 down to about 20amps until the BMS shut down due to over voltage of one of the cells hitting 3.65.
I assume the BMS was doing what it had to do because the battery had hit full charge. But once it did that, the tracking of "remaining capacity" in the app went from 100 (after slowly climbing during the charge cycle) to zero. It looks like the BMS now doesn't know the power that is currently residing in the battery, which I assumed it should track.

1) I had assumed with the BMS and the app, I could track power going in to as well as out of the battery, and the app would tell me how much "life" was left in the battery at all times. Is that not true?
2) If it SHOULD track this, how do I prevent it from erasing the data once a full charge is attained and the BMS cuts off charging due to one of the cells hitting the over voltage at 3.65?
3) 100 AH went in to the battery, yet an hour after I removed the charge converter the cells seem to be resting around 3.35v which also seems wrong. Is this correct or could something be set up wrong?
4) I have done a full charge and a full discharge cycle of the system, so I have confirmed that I have 100-105AH in the batteries, so I feel like it's either a BMS hardware or software issue / charge converter issue / or just a lack of understanding and what I am seeing is normal.

Thank you for any help you can provide.

Not going to try to address all your questions, but did you calibrate the charge and load currents? My BMS didn't track AH correctly til I did this.

Bob B said:

Not going to try to address all your questions, but did you calibrate the charge and load currents? My BMS didn't track AH correctly til I did this.

Click to expand...

Not sure how to calibrate that on the bms. Are the directions somewhere? I did notice that while the software said 60ah were going in, my stand alone meter registered 57-58. Using a rate of 57/he I still come to 102 going in, so I know the cells are taking full capacity.

wstcoaster07 said:

I recently built a 12v LIFEPO4 battery with 4 3.2v CALB cells from AliExpress and the BMD from Overkill (basically the Will Prowse DIY 12v battery). I discharged the battery to zero during a capacity test and then hooked it up to my charge converter (Progressive Dynamics PD9160ALV) and watched it go. The XiaoXiang app correclty showed the current going in to the battery and the voltage of each cell. It reached "100ah" on the app a little early based on the current going in (60 amps/hr) and time it took (1hr+15min), but I assume it was just showing 100AH and 100% because the voltage had reached 3.4 per cell. (? correct)

Click to expand...

Maybe. It depends on the algorithm that sets 100%.

Anyway, it kept charging as I assumed it would since I knew 100AH had not actually gone in to the battery. At around the 1+40 mark the cells were now starting to climb in voltage and showing 3.53-3.59. Also, the current going in had slowly started to drop from 60 down to about 20amps until the BMS shut down due to over voltage of one of the cells hitting 3.65.

Click to expand...

That point where the current started tapering is where the charger went from constant current (Bulk) to constant voltage (Absorb). I would guess that the CV set point on your charger was set for 3.6 volts.

I assume the BMS was doing what it had to do because the battery had hit full charge. But once it did that, the tracking of "remaining capacity" in the app went from 100 (after slowly climbing during the charge cycle) to zero. It looks like the BMS now doesn't know the power that is currently residing in the battery, which I assumed it should track.

Click to expand...

The BMS must have volatile memory which was lost when the BMS went offline. The solution is to lower your absorb voltage on your charger to something like 3.4 volts and it will probably start to taper before that one cell hits 3.65. That way your Charger should shut off before the BMS, which is the way you want it to happen anyway.

1) I had assumed with the BMS and the app, I could track power going in to as well as out of the battery, and the app would tell me how much "life" was left in the battery at all times. Is that not true?

Click to expand...

It will if you set your charger and don't force the BMS to shut down.

2) If it SHOULD track this, how do I prevent it from erasing the data once a full charge is attained and the BMS cuts off charging due to one of the cells hitting the over voltage at 3.65?

Click to expand...

As mentioned earlier, set your charger to a lower Absorb (CV) setting

3) 100 AH went in to the battery, yet an hour after I removed the charge converter the cells seem to be resting around 3.35v which also seems wrong. Is this correct or could something be set up wrong?

Click to expand...

This is the normal resting voltage of LFP cells. You did not lose any capacity.

4) I have done a full charge and a full discharge cycle of the system, so I have confirmed that I have 100-105AH in the batteries, so I feel like it's either a BMS hardware or software issue / charge converter issue / or just a lack of understanding and what I am seeing is normal.

Click to expand...

The only issue that I see is everything will work better with a different voltage setting on your charger. Try that and cycle a few times and come back if your understanding (or lack thereof) is any different than I tried to explain.

Are you using the Iphone or Android app?

With the Iphone App I have, if you go into config there is a current calibration tab at the bottom of the screen. Maybe @OverkillSolar will check in and give you some info.

Yeah I was hoping it wasn't the charger but thought it might be. I don't think the progressive charger is adjustable which means I'm either buying a different converter for the rv or living with the issue. I figured they were the standard for rv chargers so didn't think it would be an issue.
Thanks for everyone's help. Keep any and all ideas coming.

wstcoaster07 said:

Yeah I was hoping it wasn't the charger but thought it might be. I don't think the progressive charger is adjustable which means I'm either buying a different converter for the rv or living with the issue. I figured they were the standard for rv chargers so didn't think it would be an issue.

Click to expand...

In my mind an adjustable charger is worth the money because I can set the charge voltage to a point that I only use 80 to 90% of the cells capacity and therefore get more life out of my cells. You can always keep it for redundancy.

The only workaround is something I did on an EV when I was using power supplies before I spent money on a really good programable charger. I used a programmable relay to cutoff the power to the charger at a user programmable setting. The one i used was $30 or $40 dollars but recently I have seen them for less on Amazon. You have to read the specs carefully to make sure they are DC rated and for the range you want to control. Mine was not very accurate at 100 volts DC, but that was still 50 millivolts at the cell level and it was consistent.

That workaround has some issues. One it really turns your charger into a single phase bulk charger. Two because it has no taper of amperage there is not the time to get balancing done for the 30 minutes to an hour that some constant voltage phases take to complete.

Ampster said:

In my mind an adjustable charger is worth the money because I can set the charge voltage to a point that I only use 80 to 90% of the cells capacity and therefore get more life out of my cells. You can always keep it for redundancy.

The only workaround is something I did on an EV when I was using power supplies before I spent money on a really good programable charger. I used a programmable relay to cutoff the power to the charger at a user programmable setting. The one i used was $30 or $40 dollars but recently I have seen them for less on Amazon. You have to read the specs carefully to make sure they are DC rated and for the range you want to control. Mine was not very accurate at 100 volts DC, but that was still 50 millivolts at the cell level and it was consistent.

That workaround has some issues. One it really turns your charger into a single phase bulk charger. Two because it has no taper of amperage there is not the time to get balancing done for the 30 minutes to an hour that some constant voltage phases take to complete.

Click to expand...

Do you have one you recommend for my 100ah battery bank? (used in a travel trailer)

Also, if the battery is rated at 3.65 for a full charge, should my over voltage cutoff be set at 3.65 or slightly higher since 3.65 should be expected? I have seen some sites list voltages as high as 4.2!?

wstcoaster07 said:

Do you have one you recommend for my 100ah battery bank? (used in a travel trailer)

Click to expand...

Sorry, my experience is with EV conversions and those chargers are $600 because they have more bells and whistles need for an EV such as public charging interface, mutliple charge profiles and CANBUS communication with BMS.
There are a lot of RV people here and I am sure there are good recomendations.

Also, if the battery is rated at 3.65 for a full charge, should my over voltage cutoff be set at 3.65 or slightly higher since 3.65 should be expected? I have seen some sites list voltages as high as 4.2!?

Click to expand...

No, that was the point of that long dialogue. You can do that but if you rely on the BMS to shut down you have no other back stop in case it fails. If you stop the charger at 3.4 or 3.5. then if the charger fails to stop the BMS will shut down the pack
For longer life I only take LFPs to 3.4 volts per cell.
The battery you saw was probably NMC chemistry, similar to Nissan Leaf and Tesla. I am currently running an old Nissan Leaf pack and I only take those cells to 4.05

Ampster said:

No, that was the point of that long dialogue. You can do that but if you rely on the BMS to shut down you have no other back stop in case it fails. If you stop the charger at 3.4 or 3.5. then if the charger fails to stop the BMS will shut down the pack
For longer life I only take LFPs to 3.4 volts per cell.
The battery you saw was probably NMC chemistry, similar to Nissan Leaf and Tesla. I am currently running an old Nissan Leaf pack and I only take those cells to 4.05

Click to expand...

Yeah, that makes sense. I emailed progressive dynamics to see if I (or they) can modify my output voltage to 14.1 (or any voltage less than 14.6). I would be surprised but who knows. In the mean time I am in search of a replacement for hopefully under 250.

Bob B said:

Maybe @OverkillSolar will check in and give you some info.

Click to expand...

The BMS does not count coulombs, it only assumes the state of charge based on voltage. I hate that it does that.

I have 3 ideas in the works to address it.
1. a coulomb counter module that works with the existing bluetooth module.
2. a new Wifi module that includes the coulomb counter.
3. a completely new BMS design.

Anyway about the issue at hand, check the voltages of all 4 cells at the top of the charge cycle.
I suspect your cells need to be top balanced. more info here:
You could also lower your charger's upper voltage limit to 14.0v, which is still near 99% full on the LiFePo4

Thanks. Yeah, I had incorrectly thought that if I got the new calb cells they would be close enough without needing to buy anything to balance them with. Looks like I'll be buying a power supply.

Thanks for all of the help everyone.

OverkillSolar said:

The BMS does not count coulombs, it only assumes the state of charge based on voltage. I hate that it does that.

Click to expand...

@OverkillSolar Hmmm.... I don't like that either.

I have a BMS that came with a battery I bought pre-built that uses the same App .... but must be a different BMS. It initially uses voltage to determine SOC .... (the parameters that determine voltage SOC can even be adjusted) ... but after fully charging, it starts using Amps in vs Amps out to determine SOC. If I write a new parameter to the BMS, it reverts back to a voltage assumption but starts tracking coulombs again after doing a full charge. I am able to calibrate the amps in and amps out with this BMS ..... I haven't had the nerve to tear it apart to see what BMS is actually in there. All I know is that it is a 100A BMS and uses the same APP as the one you are selling. ( The app identifies the BMS as SOLMAX LITHIUM .... SOLMAX-SP04S005-L4S-100A) I tried doing a Google search to see if I could find the Mfg to buy this BMS directly, but wasn't able to find it.

Bob B said:

@OverkillSolar Hmmm.... I don't like that either.

I have a BMS that came with a battery I bought pre-built that uses the same App .... but must be a different BMS. It initially uses voltage to determine SOC .... (the parameters that determine voltage SOC can even be adjusted) ... but after fully charging, it starts using Amps in vs Amps out to determine SOC. If I write a new parameter to the BMS, it reverts back to a voltage assumption but starts tracking coulombs again after doing a full charge. I am able to calibrate the amps in and amps out with this BMS ..... I haven't had the nerve to tear it apart to see what BMS is actually in there. All I know is that it is a 100A BMS and uses the same APP as the one you are selling. ( The app identifies the BMS as SOLMAX LITHIUM .... SOLMAX-SP04S005-L4S-100A) I tried doing a Google search to see if I could find the Mfg to buy this BMS directly, but wasn't able to find it.

Click to expand...

You know, I have never had one unit in service long enough for this to happen.

So maybe it does count coulombs eventually?

Can anyone confirm this?

wstcoaster07 said:

I recently built a 12v LIFEPO4 battery with 4 3.2v CALB cells from AliExpress and the BMD from Overkill (basically the Will Prowse DIY 12v battery).

Click to expand...

Please provide a link to the BMS.

My amazon page with Will's affiliate link: https://amzn.to/3aD2PQa

All I can add is that my $600 Orion Jr. BMS does count coulombs. My Skybox inverter also does and I can set a percentage of discharge that will cause it to go back to the grid when in TOU mode. It still has a low voltage disconnect as the final backstop but the percentage discharge is useful in managing my pack usage versus allowing me to self consume as much of my own power and avoid consuming power at peak rates. I don't think that is as critical in an RV as it is for me because I am not always in the home but I want o optimize my cost savings while I am gone.

OverkillSolar said:

You know, I have never had one unit in service long enough for this to happen.

So maybe it does count coulombs eventually?

Can anyone confirm this?

Click to expand...

A few of us had a discussion about this on another forum a couple of months ago .... I'll see if I can find it.

IFthe BMS can measure Amps and Volts it would be the software not the hardware that was doing the Coulomb counting.

wstcoaster07 said:

Yeah I was hoping it wasn't the charger but thought it might be. I don't think the progressive charger is adjustable which means I'm either buying a different converter for the rv or living with the issue. I figured they were the standard for rv chargers so didn't think it would be an issue.
Thanks for everyone's help. Keep any and all ideas coming.

Click to expand...

well, you could control the progressive using a relay type voltage controller. let it control the feed (A/C side). You can find them with timers built-in. a bit of observation. it could be controlled