Is it normal for my BMS to report over voltage cells?
- Thread starter delvxe
- Start date
John Frum
Tell me your problems
- Joined
- Nov 30, 2019
- Messages
- 15,233
I just checked back to the start of the thread and you mention that you are charging at 300-400 watts when the bms disconnects the battery. that doesn't sound good.
Current should be approaching 0 with the battery that full.
Don't increase the hi cell cutoff.
Current should be approaching 0 with the battery that full.
Don't increase the hi cell cutoff.
Ah. I see. Not connected. While the manual notes this exists it doesn’t mention anything about connecting.
John Frum
Tell me your problems
- Joined
- Nov 30, 2019
- Messages
- 15,233
Suggest op try another charge profile with lower voltage.Could it be that the Renogy is targetting a higher voltage than spec'd? I have a memory of that being mentioned
Consult manual.
I just ran a test.
- set the BMS cell over voltage higher so I was pretty sure it wouldn't trip
- put my multi-meter on the battery
- started the van to charge and monitored both BMS and multi-meter
- charging continued at 300-400 watts until my multi meter read 14.38 where it started tapering off quickly. Charging went to zero when meter read 14.42 (BMS read 14.39).
- BMS consistently read .03 volts higher than my meter.
Attachments
gnubie
Solar Wizard
- Joined
- Sep 20, 2019
- Messages
- 3,847
The gel profile is likely fine especially since you can configure the BMS as you see fit regarding balancing voltage threshold. I charge my own lifepo4 batteries at the same per cell voltage and it's also the point that Victron uses as their default for lifepo4.
Thanks for finally checking the battery voltage under charge. TBH that looks like the Renogy itself is doing the right thing. If I had known that the images you posted before were when you were using a different charger I would haven't pushed this point as hard but considering how all the figures looked it was still a good possibility.
Thanks for finally checking the battery voltage under charge. TBH that looks like the Renogy itself is doing the right thing. If I had known that the images you posted before were when you were using a different charger I would haven't pushed this point as hard but considering how all the figures looked it was still a good possibility.
John Frum
Tell me your problems
- Joined
- Nov 30, 2019
- Messages
- 15,233
How high did the highest cell get?
Did the current approach 0?
Last edited:
KenDan
New Member
I have a similar situation on my new LifePO4 16S battery bank. I'm using the aluminum cased cells from China with an ANT BMS.
If you have a cell or cells which go higher and reach the cell protection cutoff voltage, it will open the circuit and stop the charger. This can happen when the total voltage of the pack is still somewhat below the bulk charging voltage.
Typically, this happens when you have a cell(s) that has a higher voltage than the others as you near the end of the charge. The charge power can still be a few hundred watts because the overall battery pack voltage is just low enough that it is tapering down but still being charged at a moderate rate.
In my case, I have perfectly balanced cells until I reach a nearly charged state (just under 54 volts). At that point I have two cells which will shoot up in voltage faster than the others. One of those cells can hit 3.6 v/cell and trigger the BMS overvoltage protection, while most of the others are still around 3.35. This "imbalance" only exists at a very high state of charge. When the charger stops at night, the BMS balancer will bring the high cells down to 3.35 to match the others, but the next morning when the charger turns on again to try to "top off" the charge to 54 volts total, those same two cells will shoot higher.
This does not seem to be a faulty cell / internal resistance problem. Under load all the cells seem to stay well balanced - none of the cells are sagging on discharge. I have tried lowering the bulk charge and float charge voltages slightly and it does seem to help a little, but the pack still always suddenly goes out of balance as full SOC is approached. I have come to the conclusion that those two cells have a differently shaped "knee" at the top end of the charge curve.
If you have a cell or cells which go higher and reach the cell protection cutoff voltage, it will open the circuit and stop the charger. This can happen when the total voltage of the pack is still somewhat below the bulk charging voltage.
Typically, this happens when you have a cell(s) that has a higher voltage than the others as you near the end of the charge. The charge power can still be a few hundred watts because the overall battery pack voltage is just low enough that it is tapering down but still being charged at a moderate rate.
In my case, I have perfectly balanced cells until I reach a nearly charged state (just under 54 volts). At that point I have two cells which will shoot up in voltage faster than the others. One of those cells can hit 3.6 v/cell and trigger the BMS overvoltage protection, while most of the others are still around 3.35. This "imbalance" only exists at a very high state of charge. When the charger stops at night, the BMS balancer will bring the high cells down to 3.35 to match the others, but the next morning when the charger turns on again to try to "top off" the charge to 54 volts total, those same two cells will shoot higher.
This does not seem to be a faulty cell / internal resistance problem. Under load all the cells seem to stay well balanced - none of the cells are sagging on discharge. I have tried lowering the bulk charge and float charge voltages slightly and it does seem to help a little, but the pack still always suddenly goes out of balance as full SOC is approached. I have come to the conclusion that those two cells have a differently shaped "knee" at the top end of the charge curve.
John Frum
Tell me your problems
- Joined
- Nov 30, 2019
- Messages
- 15,233
Top balance( and bottom balance for that matter) is cell synchronization at one specific point. For our application that is usually full charge.
The cells will not necessarily be synchronized at any other point.
I don't think think your pack is appropriately top balanced.
The cells will not necessarily be synchronized at any other point.
I don't think think your pack is appropriately top balanced.
Last edited:
I agree with @smoothJoey on this one. I think those two cells have slightly more energy in them, or a slightly smaller capacity, than the other ones. I do not think this is a serious problem though. If you wanted you could bleed some energy out of one of these manually with a resistor or some other load just to see what happens. Again this is not pack threatning but you could be losing more than a few Amp hours per cell on the lower ones since they do not get fully charged.
KenDan
New Member
I think you guys are both correct. However, "appropriately top balanced" does not seem to be easily achievable with a BMS/balancer alone. Equal voltage isn't necessarily good enough for cells with such flat discharge curves.
I parallel connected the cells and left them sit for days after I fully charged them for the first time. They we all identical, but had settled down to around 3.35 volts.
The same thing happens now with the BMS balancer connected. The balancer will kick in to equalize the cell voltages perfectly after the charger turns off when the sun goes down. I can leave the inverter off so that the batteries remain at a full SOC. But, come the next day when the charger turns back on again it attempts to push the voltage very slightly higher because the overall pack voltage settled slightly overnight. The cell voltage imbalance then returns. The high cells can be as much as 0.2 higher than the others (3.55v vs. 3.35v)
I have tried slightly reducing the bulk charge to (55.8v) and float charge voltage to (53.8v) on the LV5048 and it showed a slight improvement.
On my Ant BMS, I just increased the "start balancing voltage" from 3.25v/cell to 3.32v/cell. My idea here is to prevent the cell balance function from operating unless the pack has pretty much reached full SOC. I believe that if the cell balancer cuts in too early, I am chasing my tail.
I'm going to try using the resistor as suggested to manually force those two cells to a slightly lower voltage at a full SOC and see if it works, and if it has a lasting effect.
John Frum
Tell me your problems
- Joined
- Nov 30, 2019
- Messages
- 15,233
You could try this
If your bms has the option, set it to only balance during charge and only balance at 3.4 volts per cell and above.
If that doesn't start the cells converging after a few cycles then decrease the charge current so they stay charging longer.
If that still doesn't start them converging adjust the tail current down to keep them even longer at the top.
You don't want to do that forever though so when they sufficiently converge at full charge you can revert the changes.
I would still leave the pack so that it only balances during charge though.
You keep mentioning to only balance during charge. But why wouldn't you balance in storage as well. If you dont balance in storage the cells will stop being balanced when charge shuts off due to over voltage.
The voltage with wich to start balancing is also tricky. I think it may need to be adjusted if you look down at 3.4 and he wrong cells are too high. Not saying it is a bad number just to verify it is not too low.
John Frum
Tell me your problems
- Joined
- Nov 30, 2019
- Messages
- 15,233
My idea is to get the cells to converge at a single point.
If the battery is charged to 3.65 volts per cell that is where the cells should be synchronized in my opinion.
If the balancer is still working when their is no charge current or when float voltage is present then the batteries will balance at full resting and/or float voltage which means they will be potentially less syncronized at 3.65 volts per cell.
Yes that number is a bit of a guesstimate based on the idea of converging on a single point.
Last edited:
Hedges
I See Electromagnetic Fields!
- Joined
- Mar 28, 2020
- Messages
- 20,827
It seems to me (being a guy with EE background who never worked with lithium cells) that the charger (alternator) voltage setting and charge current both matter.
As people have said, with lower max voltage applied, avoids out of balance cells exceeding max voltage allowed by BMS at higher voltages.
Also, BMS attempts to balance cells, either with passive or active circuitry. But it will have a maximum rate at which it can do so. Therefore, if charge current is reduced, I think there would be more time for BMS to bring up charge of the low cells.
Alternators generally have a voltage setpoint. Is this same alternator as charges your starting battery? If so, its voltage and temperature coefficient was set for lead-acid starting battery.
You could put one or more diodes (of appropriate current and heatsinking) in series with wire from battery to alternator to add voltage drop.
You could design a linear regulator circuit to provide desired voltage rather than following alternator voltage. It would have to be low dropout or no dropout. PMOS on high side could work, but MOS has trouble with higher wattage in linear mode but I've used them far below spec current. NMOS has same issues and would require boosted gate voltage. PNP is something else I would consider.
Some inverters communicate with BMS and adjust their charge settings accordingly.
As people have said, with lower max voltage applied, avoids out of balance cells exceeding max voltage allowed by BMS at higher voltages.
Also, BMS attempts to balance cells, either with passive or active circuitry. But it will have a maximum rate at which it can do so. Therefore, if charge current is reduced, I think there would be more time for BMS to bring up charge of the low cells.
Alternators generally have a voltage setpoint. Is this same alternator as charges your starting battery? If so, its voltage and temperature coefficient was set for lead-acid starting battery.
You could put one or more diodes (of appropriate current and heatsinking) in series with wire from battery to alternator to add voltage drop.
You could design a linear regulator circuit to provide desired voltage rather than following alternator voltage. It would have to be low dropout or no dropout. PMOS on high side could work, but MOS has trouble with higher wattage in linear mode but I've used them far below spec current. NMOS has same issues and would require boosted gate voltage. PNP is something else I would consider.
Some inverters communicate with BMS and adjust their charge settings accordingly.
That is a fair point Im just thinking that shutting down balancing as soon as charging stops will make balancing take forever. I guess it would also depend on where your resume charging set point is. Do you cut off charging at 3.65 and resume when cells drop to 3.4 or do you resume at 3.5.
I have my LTO cells cut off at 2.65 and start balancing at 2.5 I balance in all 3 states charge,discharge and resting I also expect the natural sink in the voltage to help balance. I do not resume charging until 2.4 volts however with the idea of giving balance more time to work. I am not worried about balancing too long under discharge because if the load is high enough to significantly drop the cell voltage it will go below the 2.5 balance level fairly fast,
I do however totally like the option to cut charge current. It would be nice if charge current at the top end was lower than balance capacity I think that would allow the balancer to work faster. Although it also becomes a game of wack a mole since all cells are getting say 1 amp and the balancer can only balance 1 cell at a time. If you have 3 high cells the balancer needs to bleed all of them and will probably keep switching around.
I definitely am not saying there is a right or a wrong way to do this.
Maybe we should spin this to its own thread as its interesting. I also have some other points I have been trying for months to figure out how to articulate.
Ampster
Renewable Energy Hobbyist
That is the best scenerio. It is less common in the solar charging environment because the goal is often to get the most solar energy from the sun into the cells.
