diy solar

diy solar

Heltec (JK) 200A Smart BMS with 2A Active Balance

Oh, another question, does yours run really hot? Mine is rated at 150A continuous, 300A surge. I was pulling ~40A last night, and the MOS temperature was around 71C, which is well below the 90C default cut off in the app, but the thing was so hot, I couldn't even touch it for more than a second.

That seems way more than it should be. What's your ambient? Andy's didn't get over 30C or so when he did his stress test:


I would say something is definitely wrong... Please share any details/pictures you can.
 
Are you sure that both the Charge and Discharge FET switches are on in the App? Mine barely gets warm at 50 amps. It measures just about 1 mohm in the proper on state. So 50 amps would be creating just 2.5 watts of heat, but if the charge FET's are off, then it has 0.6 volts of drop on top of this. If you take 0.6 volts x 50 amps, then you get another 30 watts of heat. That is more than 10 times the power going in to heat it up.
 
Are you sure that both the Charge and Discharge FET switches are on in the App? Mine barely gets warm at 50 amps. It measures just about 1 mohm in the proper on state. So 50 amps would be creating just 2.5 watts of heat, but if the charge FET's are off, then it has 0.6 volts of drop on top of this. If you take 0.6 volts x 50 amps, then you get another 30 watts of heat. That is more than 10 times the power going in to heat it up.

Aha! Just went back to check the video @lincomatic posted and the app shows the charge MOSFET is off while discharge MOSFET is on. However, this seems abnormal since the BMS should automatically do this once even a small current is flowing. I wonder if there are some conflicting settings in the app or something.
 
You have to make sure you turn them both on manually in the app. Once you set them on, they will go off if there is a fault, but will turn back on when things go back to the release settings. But if it is manually set to off, it will stay off.
 
But if it is manually set to off, it will stay off.

That's not my experience, and neither Andy's when he tested that.

Video - you can see the discharge switch (which is manually turned off) go to the on position once some charge current (1.3A or so) is flowing:

 
Last edited:
I will admit, I never actually tried to push much charge or discharge current without the manual switches on. I did do a few low current tests to see it would block on but not the other. I never once saw it actually switch on by itself. At one point, I have my 10 amp charger running into it, and when I flipped off charge, the current dropped to zero, and the charger output voltage went up to 58.8 while that battery bank was still at about 57 volts. I also did the reverse, I was pulling about 6 amps out, and it dropped to zero when I turned off the discharge switch. From the scene in Andy's video, it looks like the BMS is closing the opposite mosfet switch when it sees a reasonable current flow as a protection mechanism to reduce BMS heating. But does it actually turn on the mode? I don't think it does. When he lowered the current, it went back to showing the voltage drop again. In any normal use, you need to have both switches on to make sure it is not adding a diode voltage drop in either direction.
 
That seems way more than it should be. What's your ambient? Andy's didn't get over 30C or so when he did his stress test:


I would say something is definitely wrong... Please share any details/pictures you can.
Both of mine never go over room temp. Not even warm.
 
From the scene in Andy's video, it looks like the BMS is closing the opposite mosfet switch when it sees a reasonable current flow as a protection mechanism to reduce BMS heating. But does it actually turn on the mode? I don't think it does. When he lowered the current, it went back to showing the voltage drop again. In any normal use, you need to have both switches on to make sure it is not adding a diode voltage drop in either direction.

It turns it on and off based on the settings. That is, if you disable discharge, but enable charge, your discharge will be turned on when enough charge current flows and turns it back off once the current drops again. Likewise, if you disable charge but enable discharge, it should turn on charge once enough discharge current flows, and turn charge back off once it drops. At least, that's what it should do. From the video @lincomatic posted, it seems the latter one is not happening for some reason.
 
Are you sure that both the Charge and Discharge FET switches are on in the App? Mine barely gets warm at 50 amps. It measures just about 1 mohm in the proper on state. So 50 amps would be creating just 2.5 watts of heat, but if the charge FET's are off, then it has 0.6 volts of drop on top of this. If you take 0.6 volts x 50 amps, then you get another 30 watts of heat. That is more than 10 times the power going in to heat it up.
Are you kidding me? Why do they give you separate manual switches if it's bad to use them? That's a really stupid design. The reason I had the charge FET turned off is because I was discharging through a new inverter whose settings I'm not familiar with, and I wanted to make sure that it wouldn't inadvertently start charging the battery during my discharge test. OK, I will try the discharge test again tonight, and see if turning on the charge FET keeps the temperatures down. I think you're probably right about the source of the heat, because this morning, when I was charging, and pushing 50A into the battery, I had both FETs turned on, and it wasn't heating up at all. I was wondering why charging would generate less heat than discharging.
 
Are you kidding me? Why do they give you separate manual switches if it's bad to use them? That's a really stupid design. The reason I had the charge FET turned off is because I was discharging through a new inverter whose settings I'm not familiar with, and I wanted to make sure that it wouldn't inadvertently start charging the battery during my discharge test. OK, I will try the discharge test again tonight, and see if turning on the charge FET keeps the temperatures down. I think you're probably right about the source of the heat, because this morning, when I was charging, and pushing 50A into the battery, I had both FETs turned on, and it wasn't heating up at all. I was wondering why charging would generate less heat than discharging.

Because the BMS should automatically control those switches even if they're set manually. I wrote this recently:


Now the question becomes: why isn't it doing this in your case...
 
Because the BMS should automatically control those switches even if they're set manually. I wrote this recently:


Now the question becomes: why isn't it doing this in your case...
Thanks for posting the explanation. OK, I was finally able to test this today. Sure enough, if I turn off the charging MOSFET while the discharge MOSFET is on, and large current is being drawn, the MOS temperature increases rapidly. Turning the charging FET back on instantly causes the temperature to fall again. The BMS doesn't attempt to override the setting.

On the other hand, when charging, if I turn off the discharge FET in the control screen, the home screen still shows that the discharge FET is on, so the BMS overrides it.

Thanks a lot for solving this issue for me. Unfortunately, despite making sure all of the connections are tight on two separate 16S batteries w/ two separate BMSs, I'm still seeing the current and power on the home screen wandering so much that the readings are basically useless. As a cross check for the connections, I used a FLIR camera, and didn't see anything heating up, even when pushing 50A out, so the connections are definitely OK. Oh well, I guess I can live with that.
 
Twice now I've had an odd issue which I think is the BMS.

I've got a 16s pack using ETC 277ah cells and am about 100 cycles in, the pack was top balanced for a good week before it went in and i have balancing on.

This morning, with the heater on and the kettle boiling, the BMS logged a low voltage alert on a particular cell and disconnected the loads. It was pulling about 100amps, but the battery was about 70% full so the voltage sag still had the pack voltage above 51v as a whole, just 2.9v on one particular cell. Temp monitoring said about 15deg on the pack.

I went and rebooted the inverter (no solar at this time, it was early) and then loaded it all back up and grabbed the multimeter to test the cell under load and didn't see anything below 3v on the multimeter, but more like 3.1v as the BMS is reporting low 2.9v on that one cell. Using a fluke 325 which only has 1decimal place so it could be rounding up and could also not be updating fast enough for me to see real voltages.

That said, I had this same issue last week, but on a different cell and you'd think if I had a bad cell it would be consistent.

I've also noticed, that each time I reboot the BMS, the cell balance wire lead resistance values are increasing. Is there an underlying issue here? All connections look good/no corrosion etc.
 
Twice now I've had an odd issue which I think is the BMS.

I've got a 16s pack using ETC 277ah cells and am about 100 cycles in, the pack was top balanced for a good week before it went in and i have balancing on.

This morning, with the heater on and the kettle boiling, the BMS logged a low voltage alert on a particular cell and disconnected the loads. It was pulling about 100amps, but the battery was about 70% full so the voltage sag still had the pack voltage above 51v as a whole, just 2.9v on one particular cell. Temp monitoring said about 15deg on the pack.

I went and rebooted the inverter (no solar at this time, it was early) and then loaded it all back up and grabbed the multimeter to test the cell under load and didn't see anything below 3v on the multimeter, but more like 3.1v as the BMS is reporting low 2.9v on that one cell. Using a fluke 325 which only has 1decimal place so it could be rounding up and could also not be updating fast enough for me to see real voltages.

That said, I had this same issue last week, but on a different cell and you'd think if I had a bad cell it would be consistent.

I've also noticed, that each time I reboot the BMS, the cell balance wire lead resistance values are increasing. Is there an underlying issue here? All connections look good/no corrosion etc.
Could be loose connections, or bad connections. Try adding some MG Chem silver conductive paste or pen on the connections, and then review your resistance on the cell sense wires. Mine went from .4ohm to .1ohm just doing that. I previously used No-Ox-ID-special, but the MG Chem works better.
 
I did remove cell 5 and re torque it, the first cell that did this, but i didn't get a chance to do cell 7 this morning.

Should i be expecting to see the sense wire values change real time? it only seems to change if the BMS is rebooted and it only ever goes up so far.
 
I did remove cell 5 and re torque it, the first cell that did this, but i didn't get a chance to do cell 7 this morning.

Should i be expecting to see the sense wire values change real time? it only seems to change if the BMS is rebooted and it only ever goes up so far.
Not sure. Every time i played with the connections I turned everything off for safety, so rebooted the BMS after.
 
OK wtf.

Put some new washers on all my cells, just steel, nothing exotic, little bit of carbon grease also because that's what the store had, noticed a few connections were just a little loose so gave those a twig up with the crimper also now I'm missing half my cells even though I have tested continuity between each balance lead terminal and the plug into the BMS and it's fine.

I can also confirm I have a voltage reading on each pin on the BMS loom plug when referenced to the main battery neg. As you test each pin, voltage climbs by the cell value all the way up to cell 16 and full pack voltage as you'd expect, there's no bent pins either.

Pack voltage is perfect also. What gives?

246739135_410974464020833_7424912989227818966_n.jpg
 
Last edited:
Back
Top