diy solar

diy solar

JBD BMS in parallel

drbytes

New Member
Joined
Oct 21, 2021
Messages
113
Hi.

I have 304 and 280ah batt bank, wired in parallel, both have a JBD BMS, the one with the relay : Jaibaida AP20S003S 7~20S.
I'm having an issue with the 280ah; it'll trigger an over-voltage disconnect (one cell will peak to 3.65) and the relay will disconnect the battery and because the 304 is supplying power to the loads the 280's relay will not re-engage as long as the 304 is supplying power. Only when the 304 is completely drained and hits it's low voltage disconnect will the relay on the 280ah re-engage at which point it dumps 100's of amps into the 304 while also supplying power to the load.

I saw this vid, the guy there has a similar re-enage problem albeit on the under-voltage disconnect :
Anybody here that has tackled this problem?

If this is the way it is I'll have to get ridd of this bms, it unnecessarily stresses both batteries.
 
From your description seems like the BMS is doing its job. Sounds like the 280Ah battery pack needs to be top balanced. Also 3.65V is the maximum charging voltage recommended for LFP cells they will tolerate short excursions up to 3.8 or even a bit higher. I would set the High Voltage Cut Out on both BMS's a little higher than 3.65V to give some "head room". Perhaps also reduce the charging voltage a little, the cells will charge to 95%+ at 3.5V per cell. Its not necessary to charge at higher voltages that push some of the cells over the HV Cut Out limit.
 
Well,... No, its not doing it's job, it disconnects on high voltage, that's fine (there are reasons why it hit that, one of the reasons is that I'm going to top balance and trying to cramm all the amps in there before all the cells are placed in parallel later on) -- however, it should cut back in after the timeout expired and supply power to the loads.

See, I heard the high voltage disconnect and since I was curious to see if it would recover I waited to see when it would cut back in. It didn't, it kept letting the 304 supply power. The over voltage release should work.

Other than that, yeah, cells max at 3.45 but then you need to top balance once/twice a year to keep things tight.
 
Thanks for the clarification. You are correct, once the voltage drops below the HV Cut Out it should close the contactor and reconnect the battery to the system. Seems like these kinds of BMS problems are discussed quite frequently on this forum. Due to these issues, I went with a ZEVA BMS (which most have never heard of) so far so good. I've even taken a dive into the murky waters of a DIY BMS. Been more trouble than its worth but I have a partially working prototype and may actually have a fully functional prototype soon. Will have to wait and see if the Firmware can be debugged to my satisfaction.
 
Those are conservative charge numbers. Maybe too conservative? My JBD BMS won't start balancing until the charge voltage is above 3.4 volts for a cell. Your 55.2 volts is above 54.4 volt threshold for cell balancing, so it should be balancing. But 55.2 volts is a long way from the 58.4 volt high voltage disconnect for the battery. So, as you've surmised already, the BMS is cutting of the charge due to a cell exceeding the high voltage threshold.

I'll ask the obvious question: Were these cells Top Balanced according to the guide used here on the forum?

I would also look at tweaking the High Voltage Reconnect threshold so that the BMS can allow charging sooner.
 
Update : it happens every day now.

The 280 is charged before the 304, the charge controller keeps charging because the 304 pulls the total voltage down -- it doesn't see that the 280 is full and the voltage is being pulled low because of the 304. The 280 goes into overvoltage protection. Eventually the 304 reaches 99% soc which coincides with the voltage being satisfied to turn off the charge controller.

What happens then; the 304 is being discharged while the 280 doesn't reconnect to the load because it doesn't see a draw (304 satisfies the load).

To solve it I need to flip the breaker on the 304 which instantaneously energizes the 280 relay and it connects to the load. I then reconnect the 304 via the breaker and there is a massive dump from the 280 into the 304 (depending on how long the 280 has been disconnected and how much AH has been depleted in the 304)

This is some BS.
 
Update : it happens every day now.

The 280 is charged before the 304, the charge controller keeps charging because the 304 pulls the total voltage down -- it doesn't see that the 280 is full and the voltage is being pulled low because of the 304. The 280 goes into overvoltage protection. Eventually the 304 reaches 99% soc which coincides with the voltage being satisfied to turn off the charge controller.

What happens then; the 304 is being discharged while the 280 doesn't reconnect to the load because it doesn't see a draw (304 satisfies the load).

To solve it I need to flip the breaker on the 304 which instantaneously energizes the 280 relay and it connects to the load. I then reconnect the 304 via the breaker and there is a massive dump from the 280 into the 304 (depending on how long the 280 has been disconnected and how much AH has been depleted in the 304)

This is some BS.
I'm having trouble making sense of this. You have a 304Ah battery and a 280Ah battery wired in parallel. By definition they should both be at the same voltage. But you are saying the 304Ah battery "pulls the total voltage down" but the 280Ah "goes into overvoltage protection". How is this happening?

You said that one of the cells in the 280Ah pack is hitting 3.65V causing the OVCO, but since you are only charging to 3.45V per cell, you've got some pretty serious imbalance between your cells. If the 304Ah battery is holding the total voltage down, the 280Ah battery should also be seeing less than the 55.2V you have set for your charger. That makes the out-of-balance problem sound even worse.

As @HRTKD asked: Did you do a top balance of these cells before hooking them up as a battery?
 
The 280 is charged before the 304, the charge controller keeps charging because the 304 pulls the total voltage down --
I suspect what is happening is that one of the BMSs is going offline because a cell hit overvoltage. At that point the packs are no longer on parallel and their voltages could diverge.
 
Yes one cell in the 280 battery I ordered from Docan Technologies isn't behaving very nice. Multiple top balance efforts have been made.

Why does the JBD not respect the over voltage release timings ? I guess we'll never know or we can conclude that a single relay for charge/discharge disconnect logic is quite crap and there should be two relays.

This really is super basic stuff, reconnect the relay after the timeout I've set.
 
This is where I don't get relay based BMS. If the relay trips the whole battery is disconnected. FET based controls charge & discharge separately. If you trip on high voltage any loads will still draw the battery down. You don't want the BMS tripping but the real world springs surprises or you may be tuning your settings.
 
Last edited:
Back
Top