Heltec (JK) 200A Smart BMS with 2A Active Balance
- Thread starter ArthurEld
- Start date
Sanwizard
Solar Wizard
- Joined
- Feb 2, 2021
- Messages
- 2,715
Calibrating current is the active balance function. You typically want to only run this at a higher state of charge.
GXMnow
Solar Wizard
- Joined
- Jul 17, 2020
- Messages
- 2,710
Actually on the Heltec/JK-BMS with active balancing, this is not the case.
The "Calibrating current" is a way to calibrate the internal current measurement to a more accurate reading. This is done by supplying a steady charge or discharge current. I did mine at about 50 amps. The higher the current, the more accurate it will be able to calibrate. Once you have a steady known current flow, you then enter the true actual current and it will then apply a calibration scale factor to make the internal current reading match the value to stated. Be careful to make sure you don't enter a bad value. If the current is already reading very close, you don't need to do this, but if it is off more than 3%, you can likely do better. This will effect the amp hours used and charged, so the more accurate you can get it, the better it will track your battery state of charge. I am not sure, but I think it may also scale the current reading for the over current protection.
The voltage calibration works about the same way, but it works best at no load. Measure the true battery voltage with a know accurate meter. Enter that value, and it will scale the internal volt meter readings to match. Keep in mind, it does not measure he full pack voltage. It just measures each cell, and then adds them up. If the number of cells has not been set, it will read wrong, and if a balance lead is not working correctly, the reading will also be off. I figured this out when one of my balance leads failed and the pack voltage was showing almost 8 volts low. The failed balance lead showed up as two cells at zero volts.
Sanwizard
Solar Wizard
- Joined
- Feb 2, 2021
- Messages
- 2,715
Well damn! I learn something new every day! Thanks for that overview, as my assumptions were obviously not correct.
Sure wish the manual went over the settings!
mrzed001
Voice of reason
Been dropping tidbits in the comments (you know, you've seen them). He could save himself a lot of time and frustration just doing a search here 
GXMnow Told me of that trick... but use a 9v battery....It's easier to do..
mrzed001
Voice of reason
I think he does this on purposeBeen dropping tidbits in the comments (you know, you've seen them). He could save himself a lot of time and frustration just doing a search here
The show must go on ... ? ?
GXMnow
Solar Wizard
- Joined
- Jul 17, 2020
- Messages
- 2,710
What functions can be gained with BMS can communication?
When charging, the BMS can report an accurate state of charge. Not a bad thing to have, but is it essential?
One odd missing feature is charging sources that don't have a provision for stopping charge at low temperatures. This is another thing that you should have to stop charging LFP cells if they drop below freezing temps. The JK-BMS will open the battery at low temp, but the XW will see it as a battery failure. The XW has a battery temp sensor, but it won't stop charging due to cold, only from over temp. I hope they fix that in a software update.
When charging, if one cells starts to go too high, it can tell the charge source to reduce the current. This is a good function, but it is actually covering poor cell matching and/or balancing. Without this function, the BMS might end up disconnecting the battery, and you will have to find out why it did and fix it.
When discharging, the BMS can again report the accurate state of charge. Again, this is a good function to have, especially with LFP cells as the voltage barely changes, so true state of charge can't be determined by only a voltage reading. If you are trying to ensure you have 50% remaining, then this may be needed.
If a cells starts to discharge too low, even though the whole pack voltage is fine, the the BMS can report the problem, but in most cases it will just have to shut it down. This does not help much over a stand alone BMS shutting off due to a low cell fault. In both cases, there is a balance or matching error in the battery.
Having the BMS able to talk to the charger/charge controller/inverter is certainly not a bad thing, but is it really needed?
If you are using LFP cells, I can see where having the BMS communication could help a bit, as the state of charge is hard to determine. In my case, I am using NMC cells. Their voltage changes enough and very linear, so using voltage control is not a problem at all. If the cells are not well matched, then again BMS communication would be good, but it is just correcting for another problem. LFP cells that are not well balanced and matched is where you would really need the BMS to have more control. I did a lot of reading online, and I determined that I was not worried about not having the BMS talk to my Schneider XW-Pro. If I do ever switch to LFP cells, I may look for a BMS that can talk, or I will add the Schneider Battery Monitor to at least give good state of charge data to the system.
When charging, the BMS can report an accurate state of charge. Not a bad thing to have, but is it essential?
One odd missing feature is charging sources that don't have a provision for stopping charge at low temperatures. This is another thing that you should have to stop charging LFP cells if they drop below freezing temps. The JK-BMS will open the battery at low temp, but the XW will see it as a battery failure. The XW has a battery temp sensor, but it won't stop charging due to cold, only from over temp. I hope they fix that in a software update.
When charging, if one cells starts to go too high, it can tell the charge source to reduce the current. This is a good function, but it is actually covering poor cell matching and/or balancing. Without this function, the BMS might end up disconnecting the battery, and you will have to find out why it did and fix it.
When discharging, the BMS can again report the accurate state of charge. Again, this is a good function to have, especially with LFP cells as the voltage barely changes, so true state of charge can't be determined by only a voltage reading. If you are trying to ensure you have 50% remaining, then this may be needed.
If a cells starts to discharge too low, even though the whole pack voltage is fine, the the BMS can report the problem, but in most cases it will just have to shut it down. This does not help much over a stand alone BMS shutting off due to a low cell fault. In both cases, there is a balance or matching error in the battery.
Having the BMS able to talk to the charger/charge controller/inverter is certainly not a bad thing, but is it really needed?
If you are using LFP cells, I can see where having the BMS communication could help a bit, as the state of charge is hard to determine. In my case, I am using NMC cells. Their voltage changes enough and very linear, so using voltage control is not a problem at all. If the cells are not well matched, then again BMS communication would be good, but it is just correcting for another problem. LFP cells that are not well balanced and matched is where you would really need the BMS to have more control. I did a lot of reading online, and I determined that I was not worried about not having the BMS talk to my Schneider XW-Pro. If I do ever switch to LFP cells, I may look for a BMS that can talk, or I will add the Schneider Battery Monitor to at least give good state of charge data to the system.
Sanwizard
Solar Wizard
- Joined
- Feb 2, 2021
- Messages
- 2,715
I have MPP Solar, and it uses the Pylontech protocol. I dont know what Protocol the JK BMS uses, but I have 4 of them, so I think I need to lab this out. I already ordered the Seplos BMS, as I know they support Pylontech, so I have a backup plan, and their $1000 Mason DIY kits are just too cool to pass up.
I am happy Andy is now testing the JK/Heltec, so we will know if it works with Victron soon I hope.
I get that it’s not needed but it’s very useful. I thought the same as you and I’ve been using “dumb bms’s” for years now. The benefits of having the BMS communicate with the inverter is worth it.
1. It gives a much more accurate state of charge reading from the inverter without need another shunt.
2. I can set exactly what percentage I want my batteries to charge or discharge to.
3. The BMS will tell the inverter to slow down charging or discharging depending on cell voltages.
I was looking to pick up one of those Mason kits. They still going for 1k?
mrzed001
Voice of reason
Did someone try this ?
GitHub - jblance/mpp-solar: Python package to communicate to MPP Solar PIP-4048MS inverters (and similar)
Python package to communicate to MPP Solar PIP-4048MS inverters (and similar) - jblance/mpp-solar
github.com
Currently has support for:
- MPP-Solar and similar inverters, e.g.
- PIP-4048MS
- IPS-4000WM
- Voltronic Axpert MKS 5KVA Off-grid Inverter-Chargers
- LV5048
- JK BMS
- JK-B2A24S (HW version 3.0)
- JK-B1A24S (HW version 3.0)
- Victron VE Direct Devices:
- tested on SmartShunt 500A
- Daly BMS
Sanwizard
Solar Wizard
- Joined
- Feb 2, 2021
- Messages
- 2,715
Yup, DMI inc shows it running on his youtube channel. I bought a couple Pi4's and installed it a few months back, but I was not able to gey the MQTT portion running with the database for some reason. Its probably better by now.Did someone try this ?
GitHub - jblance/mpp-solar: Python package to communicate to MPP Solar PIP-4048MS inverters (and similar)
Python package to communicate to MPP Solar PIP-4048MS inverters (and similar) - jblance/mpp-solar
Currently has support for:
- MPP-Solar and similar inverters, e.g.
- PIP-4048MS
- IPS-4000WM
- Voltronic Axpert MKS 5KVA Off-grid Inverter-Chargers
- LV5048
- JK BMS
- JK-B2A24S (HW version 3.0)
- JK-B1A24S (HW version 3.0)
- Victron VE Direct Devices:
- tested on SmartShunt 500A
- Daly BMS
I took the easy way out and downloaded the Solar-assistant.io code, and that has been working flawless with the Victron shunt providing SOC across my banks.
I ultimately want to simplify and enable the inverters themselves to have the SOC, and just leave the BMS to protect the cells, so I bought a couple 200amp Seplos BMS's to test with MPP Solar LV6548. The LV has a combined CAN/RS485 port. @houseofancients uses a Pi to translate the RS485 to CAN, bit I am going to just try different tcable pinouts to see if it works. If not, I will use the conversion method.
In the end, I want a set and forget solution, like the grid, but off-grid.
I will be building enough Battery and solar to replace the need for grid power, and just use that if needed for backup, and then a whole house natural gas generator for long stretches of bad weather with grid out.
Sanwizard
Solar Wizard
- Joined
- Feb 2, 2021
- Messages
- 2,715
Funny that Andy on his offgridgarage youtube channel finally found this BMS also. I also have 4 of them. Its been interesting watching many folks build out their Batrium configs, electrodacus, and relay based BMS's, then adding balancers. They just seem overly complicated to me after using this BMS. It simple and just works, which is kinda nice.
I have high hopes for Seplos also, as even the new prebuilt batteries Will has been testing seems to use the Seplos BMS.
But for now, I totally agree with you @ArthurEld, these JK/Heltec BMS's are awesome. Only one complaint, and that is the need for extra voltage to boot it.
ArthurEld
Solar Wizard
@upnorthandpersonal is the one that got me interested in these BMSs.
I am very happy with them which is why I have 4.
It does seem like they should do something more user friendly to turn them on.
I actually like that feature: it means the BMS is off by default, and can turn itself off and stop depleting an already empty battery. Starting it with a 9V battery (even adding the circuit to the system) is painless really. Over the past year, I think I had to restart a BMS once - and that was because I was messing with something. Under normal conditions, the BMS just stays on.
GXMnow
Solar Wizard
- Joined
- Jul 17, 2020
- Messages
- 2,710
If your battery is not fully charged, at least 5 volts less than your absorb voltage, then the charger should just kick it on. But since I had been balancing up my pack, it was at 56 volts when I connected the BMS. So my maximum safe 58.8 volts from the charger did nothing. MY first "cheat" to make it power up was connecting the charger to the 12th cell+ and the P- out of the BMS. That worked just fine.
Similar threads
