BMS source, Bafangusadirect

Joined
Nov 10, 2014
Messages
222
Location
Dallas
I acquired a Bafangusadirect 52/17.5 shark pack of uncertain heritage that wouldn't charge to see if there was something that could be done with it. 49.4 v across output with out load, zero with. BMS has a burned Mosfet , maybe2 and cells appear good.
It appears that both the P or B terminals have big solder blobs that may have arced to the first mosfet or shorted out a trace, I suspect the adjacent P.
What are good sources for 14S BMS? The original is labeled a 2019-05-06SP-M14-020-A02 30A
I contacted Bafangusadirect, who did not have the in stock.
If BMS are pretty much generic, are there brands or sources to avoid?
If the suspected short was not the culprit, any good troubleshooting resources for things that could do this?
 

Attachments

  • IMG_6329.JPG
    IMG_6329.JPG
    386.5 KB · Views: 675
onemorejoltwarden said:
I acquired a Bafangusadirect 52/17.5 shark pack of uncertain heritage that wouldn't charge to see if there was something that could be done with it. 49.4 v across output with out load, zero with. BMS has a burned Mosfet , maybe2 and cells appear good.
When a battery won't charge, then that means the charge FET has turned off or failed. (assuming you are using the charge port to charge with.) I don't see a visible problem with the charge fet (the one by itself), though you'd have to test it to verify it's operation.

Most likely the pack has a cell-level problem that has caused it to turn off the FET to prevent damage to the cells that could cause a fire. That's usually a cell that is way too low, or way too high.

You say "cells appear good" but don't specify what tests you've done to verify this. What are the voltages of each cell group, starting with the most negative, to the most positive? Use the multimeter on 20DCVolts, black lead on the battery negative, red lead on the first balance wire at the BMS connector itself. Note that voltage down, then move black lead to where red lead is, and red lead to the next balance wire. Repeat until you've measured all the way to the pack positive (which should be the same as the last balance wire).
 
Thanks Amberwolf . Sincerely appreciate the help
I checked the cell packs at the BMS balance plug , originally, but couldn't measure , calculate and record at same time without shorting something ,but I got the wrap off and all the cell packs/groups are about 3.6v.
Seems like a burnt BMS, specificly that toasted Mosfet, though with paralleled packs, dead cells could sure hide .
I found a 14S 52v BMS from vruzend.com but was hoping to try and spot the original reason for the failure
as I would hate to fry a new board. Bafangusa wanted 2x what Mica is asking ($75BUD/39 vruzend) and he seems like a guy that would carry good stuff, and BUD is out of BMS.
I think it could be a short or arc from the P lead to the mosfet, but that's just speculation as there is no clear connection or continuity between the mosfet and the blob of solder at P. It could have arced. I dont have any board diagrams to help track, so was just looking for things to check.
I wish I had the battery's original story, it was just billed as dead.
70 nice Samsung cells waiting to get back in the game , so far.
I will : 1) make sure it was wired right originally, 2) check charger. (it's been OK with other 52v bat)
Is there a "what causes batteries and BMS to fail " thread that I have missed?


Thanks again and thanks for the tip on how to support the board, it has been a great resource
 
Fwiw, I'm on my third BMS for my 13s pack. The other two literally caught fire while attempting to turn off. Cells were fine.

I think the root cause of this was "crappy BMS". One of them had 1Mohm resistors on the FET gates, meaning they probably switch off over a few hundred ms, thus absorbing a crazy amount of energy.

I now have a Daly 13s 40A one. So far it's successfully shut off when a cell went below 2.8V without catching fire. It seems useless at balancing though.

Sadly, the world is full of fire hazard BMSs from our favorite country to buy cheap crap from. Also sad is that there really aren't many readily available not rubbish ones.
 
Could you identify the chips on the right of your pic for me? I'm currently designing a BMS of my own, hoping to overcome this nonsense... They look similar pinout.
 
mxlemming said:
Fwiw, I'm on my third BMS for my 13s pack. The other two literally caught fire while attempting to turn off. Cells were fine.

I think the root cause of this was "crappy BMS". One of them had 1Mohm resistors on the FET gates, meaning they probably switch off over a few hundred ms, thus absorbing a crazy amount of energy.

I now have a Daly 13s 40A one. So far it's successfully shut off when a cell went below 2.8V without catching fire. It seems useless at balancing though.

Sadly, the world is full of fire hazard BMSs from our favorite country to buy cheap crap from. Also sad is that there really aren't many readily available not rubbish ones.
if you just replace the crappy mosfets with good ones you are bascially there. it will cost you probably more then what the entire bms costed you tho...
 
mxlemming said:
Could you identify the chips on the right of your pic for me? I'm currently designing a BMS of my own, hoping to overcome this nonsense... They look similar pinout.

They are Mosfets, and do not have labels. I assume 30A, because the board is labelled that way and is sold as such by a reputable dealer. Pretty indirect eval ,eh?
The wiring diagrams for the BMS boards in general have the symbol and I would assume they are tested like transistors.
I am searching for the board schematics,but am just going to replace it with a VRUZEN board based on Micah's reputation rather than knowledge of specifics, getting a bit ahead of myself. By the time they arrive I hope to be smarter.

Thanks for the stories above.
So many of the BMS are sold by price without specs, that it is good to have things to check.

I also found a decent YT to use in inspection of the "dead" battery : https://youtu.be/o8msnl_NqJQ
and one very good one where they analyse, then build a BMS: https://youtu.be/rT-1gvkFj60 that explains things well.
On BMS wiring ind internal function with sample schematic: https://youtu.be/OG-UUXEOZ8E
 
onemorejoltwarden said:
mxlemming said:
Could you identify the chips on the right of your pic for me? I'm currently designing a BMS of my own, hoping to overcome this nonsense... They look similar pinout.

They are Mosfets, and do not have labels. I assume 30A, because the board is labelled that way and is sold as such by a reputable dealer. Pretty indirect eval ,eh?
The wiring diagrams for the BMS boards in general have the symbol and I would assume they are tested like transistors.
I am searching for the board schematics,but am just going to replace it with a VRUZEN board based on Micah's reputation rather than knowledge of specifics, getting a bit ahead of myself. By the time they arrive I hope to be smarter.

Thanks for the stories above.
So many of the BMS are sold by price without specs, that it is good to have things to check.

I also found a decent YT to use in inspection of the "dead" battery : https://youtu.be/o8msnl_NqJQ
and one very good one where they analyse, then build a BMS: https://youtu.be/rT-1gvkFj60 that explains things well.

Pretty sure they're not MOSFETs. The tssop20 things not the d2pak MOSFETs.
 
I'll have someone above my level take a look.
Fortunately the B&L gemscope is up
they are labeled :
(First line has weird # , but:
OM M (in a circle)1OP0
CRSS052N08N
3M8B8219-4


ED:
oops, Don't know left from right. No wonder I need an electric assist.
How about .\\ (M?) 017 B425A
 
Given the above info, the way the pack assists to have volts but under load gives nothing, id suspect the control chips are wrecked. If the MOSFET was the failure I'd expect it to conduct more.

It's possible when the FET failed it took the control chip with it.

Don't waste your time trying to fix it, buy a new BMS. They're generic. 50$ just isn't worth your time on this.

My Daly one at least doesn't burn, they make a 14s model and flippy recommended some brand in another thread with bluetooth and higher balance current.

I'm currently in the market for a 20s4p p42a pack BMS, I'd be interested in his recommendation...
 
A clever url leading buyers to believe they are dealing directly with Bafang. Bulls pizzle!
 
Just received a VRUZEND BMS.
The DALY BT BMS looks great with cell monitoring on iPhone app for ~$60.
though this pack is a 52v and they are all 48. Can get with UART, BT or CAN, which is nice for balancing
Sounds like the way to go in the future.

I'll post my outcome as the project progresses. Thanks
 
I replaced the BMS with the 30A 52v from VRUZEND. Bat has apparent good cells , good charge, good balance ˜60% charge voltage 50+
Easy install on to the Hailong 17.5 pack.
On detailed inspection the fault was in the BMS where a solder blob on P- shorted out to the first FET and burned out the first two FET chips as in the pic in the OP. Somebody had real poor QA. This was factory soldering.
I don't have the story , but I wager it was either DOA and replaced or failed on first use.
BMS died so that other may live.
I'll contact BUD and see if they remember the story and can give any details.
 
Aside from a close visual inspection of solder points, would there be a way to test a BMS prior to hooking it to a bank of cells?

I guess a regulated power supply on a trustworthy circuit breaker and put the BMS in a bucket of sand outside...
Makes charging your battery in a BBQ grill seem like a reasonable solution.
 
Back
Top