Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
JK-B1A24S / JK-B2A24S Active Balancer
#1
I bought this active balancer a while back but never got around to actually use it. It's a step up from my current active balancer which has no ability to configure it. So I'm posting up all I have on it at the moment and will continue to update it. My current balancer works fine but I thought it's time to try something new. At under $100 it was worth trying it. I'm in no way recommending it yet but it sounds like a promising unit to balance your packs at a pretty good rate. Most balancers promise 1A but rarely give you any more than 10% of that rate. This one doesn't get there but I'd say it gets almost 60% there.

https://www.jkbms.com/product/jk-b1a24s/

First there's a couple versions of it, ranging from 1A to 10A. It also has versions that has Canbus and RS485. I bought mine before they released those versions, which is why I never started using it because I needed it to be able to log the data into a database.
https://www.aliexpress.com/item/4000252216391.html

This active balancer works by transferring the energy into a supercapacitor. It then transfers that energy into another battery. Since it can't transfer from battery to battery directly this essentially halves the transfer rate. So if you were looking to drain a battery at 1Ah you will need a 2A version. It claims that it can transfer this at 1A (for the 1A version) but so far I've seen it transfer to the supercap at around 600mA and then from the supercap to battery at around 800mA. This takes about 30 to 50 seconds each to charge the supercap and then discharge. I haven't thoroughly tested it so maybe it could be a connection issue. 

I tested it using just 4 18650 cells

Below is when I started at a 300mV difference
 

Two hours later it was brought down to ~10mV difference (ignore my celllog8s, I had a bad connection so it shows 39mV). This is pretty impressive for me anyways. The unit was not even warm to the touch, just because the energy is essentially transferred to another cell instead of being dissipated via resistors as heat as most of the other passive balancers do.


It does come with a bluetooth app which allows you to monitor it, as well as program it for the minimum voltage cutoff, the differential voltage cutoff, voltage calibration, number of cells (up to 24) and max current.

The most important thing for me was the minimum voltage cutoff so it doesn't drain the battery and a differential voltage cutoff so it doesn't constantly balance and only balance when the cells start to vary. The power consumption is next to nothing and my 4-cell setup test has not drained at all once it achieves the balance.

Now for the logging part. This is where I was stuck since my version that lacked a canbus or rs485 serial output I was unable to use it. It has a Bluetooth Low Energy on it but no way to interface to it until now. A user on github by the name of jblance posted up a version of it that would talk to it and output the data through mqqt to an influxdb.

https://github.com/jblance/jkbms

I'm still a little stuck on it because I need to modify it to match my current bms data so I won't lose it, but the author must be some wiz with all the classes and proper way of programming it that it's almost too much for me to understand. But I've been able to get some data out through the gatttool and hcitool command so I'm actually getting somewhere now. So it will be a matter of time before I make my own easier to understand python script that will communicate with the bluetooth.

Below is the documentation for the Canbus  and I'm sure there's a serial rs-485 version out there as well)
https://secondlifestorage.com/showthread...7#pid58487

Hope this helps anyone that is looking at this balancer.
Redpacket likes this post
Reply
#2
Quick update. Got the jkbms python script running through grafana and did some tests with some good results.

Took 4 18650 and had a difference of over 650mV between the highest and lowest cell from 3.3V to 4.15V.
 

The balancer took about 3.5 hrs to bring it down to under 25mV (the threshold setting)
 

Total voltage stayed the same, so not much energy lost during the balance.


My next test would be to see the sustained transfer on one cell but will need a coulumb counter for that.
Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)