• Have you tried out dark mode?! Scroll to the bottom of any page to find a sun or moon icon to turn dark mode on or off!

diy solar

diy solar

JBD 7s-21s BMS acting weird

When I'm doing this should I charge it first to 100% SOC and then disconnect it so that there's no load?
If you want to go this route:
The best method would be to keep the cells charging at low current. If you do it without a load, the resting V declines and depending on the load of the light bulb, may give you false readings.

Car bulbs can have 2 filaments, low ~.25A, high ~ 1.0A and both ~1.25A at our 3.4X Voltages.
These BMS should have 43 Ohm resistors for each the balancing circuit. (80mA @3.2V)


You'll want to take the highest cell close to the Lowest cells' projected value ( remember the low cell will gradually rise in V since it's charging). Once you remove the bulb, that cell's V will be a bit higher. Then proceed with next highest cell etc.
 
Last edited:
Definitely no alarms or events that I can see with the newest JBDTools. I also wrote my own program to fetch data from the BMS and I don't see any messages that JBDTools doesn't show. The BMS simply decides to sleep and opens the contactor. It doesn't wake again without me unplugging it and plugging it back in.

I also tested my spare BMS (that I pulled from this system) using a power supply and resistors to spoof an 8S pack. I didn't leave it alone long enough to time out, but I did use JBDTools to tell it to sleep. It immediately dropped the contactor and would not close it again until I unplugged it and plugged it back in. I do not see any button on JBDTools to wake the system.

When I left this time I plugged a jumper into the switch port and enabled the power switch feature. I'll see if that will keep it from sleeping.

Parameter settings:
Parameter.PNG

Battery info page while it was sleeping (note that the current is zero with the contactor open):
PackInfo.PNG
 
Definitely no alarms or events that I can see with the newest JBDTools. I also wrote my own program to fetch data from the BMS and I don't see any messages that JBDTools doesn't show. The BMS simply decides to sleep and opens the contactor. It doesn't wake again without me unplugging it and plugging it back in.

I also tested my spare BMS (that I pulled from this system) using a power supply and resistors to spoof an 8S pack. I didn't leave it alone long enough to time out, but I did use JBDTools to tell it to sleep. It immediately dropped the contactor and would not close it again until I unplugged it and plugged it back in. I do not see any button on JBDTools to wake the system.

When I left this time I plugged a jumper into the switch port and enabled the power switch feature. I'll see if that will keep it from sleeping.

Parameter settings:
View attachment 261461

Battery info page while it was sleeping (note that the current is zero with the contactor open):
View attachment 261462
Is this on a phone?
 
I have 3 16s pack using the 300a JBD from Battery Hookup. I had an instance recently where one of the balance leads had an intermittent connection, for reasons I won't get into, and that pack shut down. Checking a day or two later noticed the imbalance on the packs and had a oh crap moment. Luckily after tugging on the leads I found the one that was having the issues.

I fixed the issue, but had to wait until the pack were close to the same voltage and the differential was nominal. Yes I had to shut off the other two packs to force the pack that had the issue back on. Then turn on the other two packs via the disconnect switch. All is well, first issue in more than 18 months since commissioning the packs.

I feel though the pack shutting down and being forced to turn back on via manually intervention is a good thing. I don't want potentially 500 amps jumping into a pack all at once with voltage imbalance if the bms decides to turn on. The issue I had is a prime example of why it should be manual intervention.
To OP:
Anyways hope this helps, as other pointed out my guess is that cell is way to far out of balance. Try to get the pack cell voltages within 0.050 mv differential give or take a little at 0.5C charge/ discharge. I have 280 amp hour cells and that differential would have me worried. I would also lower your max voltages a touch and raise you min voltages a touch. Other gave great suggestion on how to manually lower that cell. Good luck!
 
What is needed to connect to the BMS in that way? Does the Bluetooth link wo k for that?
The BT dongle is driven with TTL-232. I unplug the dongle and plug in a 5V TTL RS232 converter. No need to use the +12V wire at the plug since it isn't needed to power the converter.
 
No new info here except that I shorted the switch wires and enabled the switch function in software. No joy.

Jiabaida never replied to my email.

I'm ready to rewire the contactor to let my microcontroller drive it and only open it if the BMS reports that it wants to open either the charge or discharge FETs. It is absolutely ridiculous that JBD created a disconnect that you cannot control or disable.

I'm also planning to snitch the firmware and try patching it to not disconnect the contactor based on the timeout, but that might take a bit longer to accomplish.

Needless to say, I won't buy another one of their products in the future. They say that a load will keep it from disconnecting. Wrong. They say that getting data from it will keep it from disconnecting. Wrong. They don't say that using the switch interface will keep it from disconnecting, but I tried that also and it still disconnects.
 
I went out to the cabin today and found that the older BMS had released the contactor again. I decided to see what kind of current I might have when I reboot it since it was about 0.4V different from the other battery, so I used my meter to test the current across the two (open) contactor terminals. Before the meter could show much current at all, the contactor closed.

Now I know that the thing is indeed sleeping even though JBD says it won't do that if you are requesting data. The plan now is to figure out which pin the processor uses to determine that current is flowing and lie to it so it won't drop the contactor based on the sleep timer expiring. With any luck they reduce the measurement to a discrete via a comparator, and all I have to do is tie a pin high.
 
Stick a resistor across the contacts, bit of ohm's law and done. Aim for maybe 5ma, if that doesn't work try 50ma.
 
Definitely no alarms or events that I can see with the newest JBDTools. I also wrote my own program to fetch data from the BMS and I don't see any messages that JBDTools doesn't show. The BMS simply decides to sleep and opens the contactor. It doesn't wake again without me unplugging it and plugging it back in.

I also tested my spare BMS (that I pulled from this system) using a power supply and resistors to spoof an 8S pack. I didn't leave it alone long enough to time out, but I did use JBDTools to tell it to sleep. It immediately dropped the contactor and would not close it again until I unplugged it and plugged it back in. I do not see any button on JBDTools to wake the system.

When I left this time I plugged a jumper into the switch port and enabled the power switch feature. I'll see if that will keep it from sleeping.

Parameter settings:
View attachment 261461

Battery info page while it was sleeping (note that the current is zero with the contactor open):
View attachment 261462
Can you please tell me where you got that version of jbdtools?
 
Stick a resistor across the contacts, bit of ohm's law and done. Aim for maybe 5ma, if that doesn't work try 50ma.
That is one option I'm exploring. I have the old one at home on my bench. I rigged up an 8s resistor array with 24v across it. I hope to figure out quickly what works best.
 
I'm cross-posting this from another JBD thread:

Ok, so I have been playing with the 6S-20S BMS that I replaced. I rigged up a resistor network and power supply to run it on the bench thinking it is controlling an 8S battery. Next I started probing the board to find the signal that turns off the contactor.

The CPU uses GPIO PC4 (pin 29) to control the contactor. The issue is that the CPU will put itself into sleep mode soon after it drops the contactor if it isn't busy responding to data requests. When it sleeps, it also tells most of the power circuits to sleep as well.

I have successfully jumped the 5v power to PC4 and kept the contactor up indefinitely as long as I'm also asking the BMS for data. My installation does that anyway, so my plan is to have my data receiver tell the BMS to hold the contactor up as long as voltages are OK and neither of the charge/discharge FETS are off. It will do this with a discrete signal driving an optoisolator and junction transistor.

Hopefully more to come real soon. Wouldn't it be great if JaiBaiDa simply allowed us to turn off the sleep feature?
 

diy solar

diy solar
Back
Top