Sorry to hear it sounds like one of these has failed at just 80 amps. I never expected them to truly do 200 amps for long term, but 80 should not be a problem.
Yeah, I figured sizing at 150% of sustained capacity would be OK. I’m suspecting the unit was defective and in contact with Heltec to get it sorted out. I have a second unused unit but want guidance from them about what sustained current I should be able to pass for 4 hours with passive cooling only before testing it.
My 200 amp rated JK (Heltec) BMS has only been used up to 80 amps so far. It ran that current for about 1 hour and was a little warm. I have it bolted to a 5.25 inch x 19 inch rack panel, so that does help pull some heat out. I then added a heat sink on the outside of the panel to help pull out heat, and I had a heat pipe CPU cooler on the face of the BMS for a while, but the thermal adhesive gave up and it pulled away. I will need to add a clamp of some sort to hold up the weight. I have not added a fan yet.
Adding a better heat sink would obviously be a way to improve the capability of an all-in-one BMS like this - I just had mine passively cooled by ambient air.
I need an 8S BMS that can charge at 80A for 3-4 hours a day and can discharge at 80A for 3-4 hours a day and if that kind of use requires careful attention to heatsink design and possibly active cooling, I’m thinking the Chargery architecture may be a better way to go. They move the best-generating ‘switch’ into it’s own box with active cooling such as what you find on an inverter or a charge controller. Leaving the sensing and ‘smarts’ it it’s own relatively cool unit.
For sustained use at 80A of higher, that architecture is starting to appeal yo
Now that my system is up and running, I only run about 30 amps while the grid is up. Right now the sun is shining, and the battery bank is charging at just 28 amps. The rest of my solar power is running my house. In the evening, I discharge the battery bank at just 29 amps to zero out my grid use during the peak rate time for 5 hours. In this usage, the BMS only reaches about 3-5 degrees C above the ambient temp in my garage. Measuring the voltage drop from the ends of the cables, the total resistance through the BMS with the 6 inches of wire on both end works out to just under 1 milliohm. At 80 amps, that is generating 6.4 watts of wasted heat out of the BMS and the wiring.
So you measured 1 mOhm @ 29A meaning a voltage drop of 29mV @ 29A and you are projecting that you’d measure an 80mV voltage drop at 80A (but have not confirmed that with measurement), right?
In any case, it’s pretty clear that by the time I used my DMM to measure the BMS voltage drop, something was wrong.
The primary purpose of my test was to check for cell voltage droop under 0.3C discharge and to measure battery capacity under sustained 0.3C discharge. I was checking cell voltages and temperatures constantly but not really focused at all on the BMS. Since I had it and had never tested it, I thought I might as well connect it and figured I test LVD at the end of the discharge and HVD after charging back to 100%. But it never occurred to me that a 300A BMS would struggle with only 80A of current, so I didn’t really focus on the BMS until an hour into the t
That is not a lot, but it is enough to need some air flow to keep it cool. I just grabbed a snap shot right now. This is cold weather for us at just 8C (46F) right now, and it was colder overnight. My garage has 2 common walls with the house so it does not get quite as cold as outside. The battery temp is reporting at 13 C, the Schneider battery temp sensor, which is current just hanging next to the battery cables also shows 13C. So that is probably the air temp in the garage right now. The BMS FET temp is showing 16C at my 28 amp charge current.
Smart BMSs with reporting definitely have an advantage over dumb BMSes like the one I have when it comes to detecting a problem before it’s become a failure.
I was running my test using space heaters which warmed the room to 24C (75F) which didn’t help with cooling, but it gets that hot in the summer so if a BMS like this needs enhanced cooling to manage 80A for 4+ hours continuous, better to understand that earlier than later.
As for the number of cells....
Officially, they say it supports 14S to 24S, but I did a few tests on the unit, and found it worked perfectly even at just 4S as long as you provide it with 40 to 100 volts to power the BMS electronics. The setup app even let me enter 4S and it functioned properly and reported no errors. The only "issue" is the need for a separate power supply which is going to be less efficient. The board itself does appear to run at lower voltage, but the DC to DC converter they built on the board needs 40 volts. It may be possible to supply it after that converter, but it would require a bit of experimenting to see what it would take. It does not seem to measure the full pack voltage. It looks like it just adds up all of the cell voltages. I had it powered at 42 volts, but it only reported the 4S pack at the correct 15.6 volts.
Inspecting my failed BMS, it looks like an entire P- copper busbar was poorly-soldered - I can slip a piece of paper under it most of the length. So whether that could have led to the high resistance and heat buildup I experienced or not, I can’t say, but it certainly didn’t help.
After this experience, I will either move to a Heltec Smart BMS like yours or switch to Chargery...
The need for a seperate power supply to operate at 8S is a PITA but not a showstopper - how much power does it consume?
heltec-bms.com
