First pack build - BMS issue

sunkmail

New member
Joined
Apr 26, 2019
Messages
10
Hi everyone,

I'm making my first attempt at building a pack:

High quality, but used SamsungLI-iON 18650's arrange as4s

Input from a AC powered Supply set to 16.8VDC and 1.0A max

The Cells all tested good on anObus charger and have similar mAHr results
Cells currently holding about 3.8V each.

BMS: cf-4s30a-a


I have it wired up as per the wiring diagram and everything is good until I insert the last cell.


When un-powered, the first 3 cells total about 11.8V.
When the last cell in inserted, the TOTAL pack voltage falls toabout 11.2V

The cells are all correctly orientated.

When measured individually, the cells are all good, and Total around 15V,so I believe the problem is in the BMS.



Any suggestions woulld be appreciated, Thanks!
 
It is hardly possible to make any analysis statement without picture, circuit, voltages, currents, and so on.

What i can say is either the BMS is bad or you have wrong connections. But, the hardly does any help.

:D
 
Finally able to get some pictures ...


Cells: Samsung ICR18650-26J

BMS: CF-40S30A-A



Wiring arrangement:

This is to test out the BMS, so I rigged up a system that uses a 4x cell holder, while still allowing the balance leads to connect to each cell...

image_rbmalp.jpg


Power is applied through a barrel connector to the P+ & P- solder pads. (Problem occurs before applying power)

image_szkatl.jpg


Balance lead B-and Pack negative connectedto the B- Pad.
Pack Positive connected to the B+ Pad.

image_bhejca.jpg


Balance leads are soldered on to nickel strips and firmly held in contact by cell holder springs.

All connections read good on a continuity checker. Voltages on the Balance Cable are the same as when reading on the individual solder pads on the PCB.


What's Happening:

When inserting the cells the Balance leads show appropriate voltages for B1-B3.

The Total Voltage at B3 is 11.8V (All Cells are around 3.8V)

If I measure at B+ (Pack +), without the 4th cell inserted, I read 11.2V.

If I insert the 4th Cell, the B+ voltage remains the same. (11.2V), when it should be reading approx. 15.6V

The Cells are all good, and I have tried multiple cells with the same results.



Any suggestions, other than 'get a new BMS'?? :)
 
How is the cell holder wired from cell to cell? Picture of that wiring could help as well.

But, just based off the above, it would seem to me that you have cell 3 and cell 4 wired in parallel instead of series. That'd be the only explanation I can see, other than your bms is busted.
 
The holder does a back and forth arrangement:


image_mbmvdl.jpg


(The negative of each cell goes to the spring)
 
did you connect B- to the B1- cell, I can't see on the photo

and also B+ to B4+ ?

the balance wires and the extra wire need to go to the pack

best regards

Stefaan

image_zedglc.jpg
 
Yes,

B- is connected to the Negative lead of the pack.

B+ is connected to the Positive lead.

B4 Balance lead is also connected to B+ on the PCB.

If I insert B4 in the holder so only the negative side is touching, I get the expected 15V (or so).


The weird 11.2V on B+ is present with only the first 3 cells installed and doesn't change when the 4th cell is fully inserted.

I'm thinking there is probably a short somewhere on the board, but I can't see it visually and I don't have a schematic.
 
removed. Nonsense from me.
 
sunkmail said:
Yes,

B- is connected to the Negative lead of the pack.

B+ is connected to the Positive lead.

B4 Balance lead is also connected to B+ on the PCB.

If I insert B4 in the holder so only the negative side is touching, I get the expected 15V (or so).


The weird 11.2V on B+ is present with only the first 3 cells installed and doesn't change when the 4th cell is fully inserted.

I'm thinking there is probably a short somewhere on the board, but I can't see it visually and I don't have a schematic.

why did you make that connection?
 
StefaanDeRidder said:
sunkmail said:
Yes,

B- is connected to the Negative lead of the pack.

B+ is connected to the Positive lead.

B4 Balance lead is also connected to B+ on the PCB.

If I insert B4 in the holder so only the negative side is touching, I get the expected 15V (or so).


The weird 11.2V on B+ is present with only the first 3 cells installed and doesn't change when the 4th cell is fully inserted.

I'm thinking there is probably a short somewhere on the board, but I can't see it visually and I don't have a schematic.

why did you make that connection?
Because it is correct.
 
the original picture is not very clear, but i can not see 2 connections (minus) to the pack, 1 coming from B- on the PCB and 1 from the balance lead (black wire)


same goes for the positive side (from B+ on pcb) and 1 from 4th balance lead.

I would also try to put the balanceleads on the spring side (- of cell) to have better contact.

try to follow the exact same schematic, it might seem some wires are redundant, but i think the small voltagedrop could cause the BMS to go eratic.

just a guess
 
StefaanDeRidder said:
the original picture is not very clear, but i can not see 2 connections (minus) to the pack, 1 coming from B- on the PCB and 1 from the balance lead (black wire)


same goes for the positive side (from B+ on pcb) and 1 from 4th balance lead.

I would also try to put the balanceleads on the spring side (- of cell) to have better contact.

try to follow the exact same schematic, it might seem some wires are redundant, but i think the small voltagedrop could cause the BMS to go eratic.

just a guess

The connection goes over the wires.
 
Ok, I haven't seen it mentioned (i could be blind), but the PCB has 4 large pads for B1 - B4, what are these for? These pads look to be separate from the balance wires.
 
I have the same bms in stock, ill look tomorrow.
 
Korishan said:
Ok, I haven't seen it mentioned (i could be blind), but the PCB has 4 large pads for B1 - B4, what are these for? These pads look to be separate from the balance wires.

I believe they are for direct soldering if you don't want to use the removable balance leads. They are electrically identical to the associated balance leads.
 
sunkmail said:
Korishan said:
Ok, I haven't seen it mentioned (i could be blind), but the PCB has 4 large pads for B1 - B4, what are these for? These pads look to be separate from the balance wires.

I believe they are for direct soldering if you don't want to use the removable balance leads. They are electrically identical to the associated balance leads.
Thats what i believe as well, but still have to prove.
 
I have an update ...

I ordered a couple of other BMS's that will be taking their time to get here. In the meantime, I figured I didn't have anything to lose, so I started playing around with my soldering iron.

Since the problem seemed to be focused on the 4th cell, I reflowed every component involved in that cell's management.

IT WORKED! (mostly)

I didn't do any testing along the way, so I have no idea where the problem actually was, but it appears to have been a cold solder joint or similar.

It still was a little flakey on the connection to the charge/discharge lead. At first it didn't want to connect the power to the cells - I was checking if anything was getting hot and found that if I pressed in one corner everything started working as expected.

So, I decided to re-solder ever connection of every part. It was a bit of a pain with the tiny SMD parts (I think they are mostly 0402, maybe 0603 - I try to stay at 0805+, if possible)

So, with that problem sorted, I turned my attention to making the balance leads, and the pack as a whole, less of a mess....

I don't have any pictures at the moment, but I was able to solder my own leads onto the wire that joins each cell to the next. I then attached the BMS to the back of the cell holder with some double sided tape and soldered the new 'balance leads' directly to the various connection points around the board.

(I removed the wired assembly entirely - figuring that i wasn't going to be separating the BMS from the cell holder - Even if I change cells, its just swapping out in a holder.)

I haven't had a chance to do any more testing since completing the wiring - Hoping to get some time this week to confirm its all happy.


Moving forward -
If I can't find any other issues, I plan to connect a DC-DC regulator with a 12V output and use this pack as a backup for my router.
(I'm fairly certain the router has a wide input range, but since it's technically owned by the service provider, I'm not going to open it to check.)

The existing supply says it needs a 3A supply, but I doubt that would be a constant load.

My cells: Samsung ICR18650-26J
Discharge:
520mA Standard
5200mA Maximum

Anyone see a problem with this arrangement?
 
sunkmail said:
The existing supply says it needs a 3A supply, but I doubt that would be a constant load.
[...]
Anyone see a problem with this arrangement?

great update. congrats on the fix!

looking forwards to more pics too ; P

your plan sounds all very sane to me.

you may get good value out of getting a breakout power lead and measuring the ACTUAL draw of the modem... which should stay pretty constant, is my gut feeling. be useful to know ...

and while I'm here ... YES i do agree with Cherry67 et al that the solder pads are IDENTICAL to the balance/fly leads ... they have been on the BMS's Ive experienced and frankly ... I cant think of what else they even could be..
 
DCkiwi said:
you may get good value out of getting a breakout power lead and measuring the ACTUAL draw of the modem... which should stay pretty constant, is my gut feeling. be useful to know ...

and while I'm here ... YES i do agree with Cherry67 et al that the solder pads are IDENTICAL to the balance/fly leads ... they have been on the BMS's Ive experienced and frankly ... I cant think of what else they even could be..

I was planning on doing a breakout of some kind to see the current for the exact reason you mention - 3A just seems like overkill. I may add a large Cap on the output to act as a reservoir for occasional spikes, just in case.

I still have to find a suitable barrel jack though - as mentioned, I don't actually own the hardware, so I can't just snip it off the existing power supply.
It's a fairly small one, probably about 1/2 the size of the really common size used on Arduinos and the like -but I haven't actually measured it yet.

From what I can tell, yes, you and Cheery67 are both correct
- continuity tests show those pads are connected directly to the associated balance lead.
I think they are there for exactly this use - Hardwiring to the PCB without a connector.
(For my case, it just makes more sense to do it this way - If it was a spot welded pack, I think I'dprefer the detachable leads)

Thanks for the feedback!

I'll try to remember to get some pictures when I get some time. :)

Cheers!


PS:

When I bought the holder, The example image was similar tothis: (Note: this image is from a different seller, found randomly online)


image_tfgovj.jpg

I figured this arrangement would allow me to wire anything up how I wanted (Parallel, Series, 2 cells of each, whatever). However, the seller sent what I ended up with- All cells in series, rather than wired individually -and had relisted the item with the proper picture with only 2 power wires. Unfortunately, I didn't take a close enough look at them when they first arrived, so I missed the window to get a refund. My own fault, I know.

As I ordered several to try to make a larger pack/wall, This little project is also a chance for me to experiment with how I may be able to use these 'incorrect' parts the way I intended. I think my new arrangement (Pictures coming) will work - I's just a lot of extra work I didn't really want to be doing. oh well. :p
 
sunkmail said:
I was planning on doing a breakout of some kind to see the current for the exact reason you mention - 3A just seems like overkill. I may add a large Cap on the output to act as a reservoir for occasional spikes, just in case.

I still have to find a suitable barrel jack though - as mentioned, I don't actually own the hardware, so I can't just snip it off the existing power supply.
It's a fairly small one, probably about 1/2 the size of the really common size used on Arduinos and the like -but I haven't actually measured it yet.

ok, I respect your decision not to cut into 'their' power supply/lead, BUT ... where are you based?
(it might be wholly your property when contract is paid out, but I won't go on, or press you on this point).

your profile location reveals nothing (*hint hint*) ... BUT ... regardless of where you are / local supply situation... check out your hardware stash ... you got one, right? if not, time to start one ... this includes chopping off barrel connectors when a PSU fails (start grabbing those around you) ... which happens amazingly often ... eventually, you will end up with every connector you need. Keep/maintain a good collection.

right now, all you need is something that can plug in. check out your local recycler ... your local IT shop ... do the dumpster diving there ... you CAN find a lead that matches. keep looking!

NB for this project to work you will absolutely need a PSU plug that fits. Non-negotiable (according to your plan/words) so I think this should be a massive priority, right?

and yes, they all have 'standard' names/measures, but for now, I recommend just grabbing them all and learning as you go. no point focussing on these details for now ... you need a stash of connectors, as this kind of problem is annoying at best ... beat it! ; )

(PS) dont bother about extra caps. if your 12V regulator is good enough, it is just that, good enough. augmenting it with extra design features is unlikely to ... be fruitful. 3A is no sweat. even a budget BEC from an FPV/drone will do that ... ALL switchmode/DC-DC convertors are generally pretty decent these days (if you don't mind the health hazard, that is).

(Parallel, Series, 2 cells of each, whatever)

as for that ... you are talking 2p. important to get your p's and s's right, and to speak clearly ... (*hugs*)


hoping this helps. don't stop ... building OR sharing your progress/questions. you are in the right place. : )
 
Back
Top