Missing resistors

[Kurt]

I am in the proses of reconfiguring my headway battery pack to fit into my new pannier bag.This involved the removal and relocation of the BMS .Just by chance I had a good close look at it and discovered something that looked out of place. The BMS is the one headway supply with there battery packs . Mine is for a 36v battery. It seems that most of the components are separated into 12 groups and there is provisions on the board for 16 I take it that for the 48v packs.

Anyhow when looking closely at the tiny surface mount resistors. I have circled the first 3 groups in yellow. The first 3 groups each have 4 resistors. The next 9 groups a couple have the same 4 resistors but some only have 3 and in one case just 2 resistors I circled in red were there was a missing resistor.Everything thing on the board seems very consistent and repetitive except the little resistors that seem all over the place with some missing.

I am sorry my camera dost have a good macro setting. Everything seems to work ok on the battery while charging and discharging but i am just wondering if some idiot has forgot to solder a couple of resistors on there.I can take more pics and explain the BMS A little better if need be.

 

[voicecoils]

WEIRD

Does every channel have 1 surface mounted resistor missing or do some channels have one missing and others don't? Can you see how the resistors are wired together? If they're all in series or all in parallel you could take your multi-meter and check of each bank of resistors had matching resistance overall. Just a thought. Does seem a bit odd to me. Perhaps the guys building the gary/fetcher BMS can comment?

I think you're correct about the circuit board being for 16 channels and only 12 are filled in. That's what I suspect mine are like too but I can't see them under my blue shrink wrap.

 

[Kurt]

Each channel has provisions for 6 surface mount resistors.I am not so concerned that all 6 pads on the board aren't filled with a resistor. Its the inconsistancy between each channel that has me conserned.

The first 3 channels on the right that I circled in yellow in my picture have 4 resistors on each channel. The remaining 9 channels to the left are inconsistent. Some have 4 resistors others 3 and some only 2 resistors on each channel.

My guess is that 4 of the 6 pads on each channel should have a resistor on it and a couple of the channels are simply missing a resistor or two.

I was thinking the channels that had only two resistors perhaps used resistors of a higher value and added up to the same resistance as the channels with 4 resistors.That didnt seem to be the case as the numbers were the same.

Are the resisters part of the balancing function of the BMS?

I hate soldering surface mount at the best of times. If I can find out for sure that things are not correct and its not ment to be like this. I wouldn't mind making it headways problem not mine. But then again sending things around the world over a couple of resistors seems silly.The battery has about 30 cycles on it and everything seems ok but I would like to get to the bottom of this and fix it if necessary before putting any more cycles on it.

edit,
after looking at the back of the board I can see that it is two sided and there are some resistors on the back that line up directly under the resistors on the top. They are covered with some kind of hard plastic glue.Its hard to know if there is a consistent amount of them or if some bridge through to to top half of the board. I starting to think the only way I am going to know if all is as it should be is test the BMS with a multi meeter. Not sure were to start.

Kurt

 

[voicecoils]

I can't really help you with the component level stuff on the board. I don't know much about how to diagnose it myself.

But, I reckon they've just soldered to the back when it was too hard or something went wrong soldering to the front. Overall resistance is probably correct.

If the pack is working fine I doubt there's any problem. If there is a problem I think John @ http://www.stealthelectricbikes.com.au will sort you out. The packs are under warranty at this point too still.

You've hit LVC and it worked properly right? I've hit LVC on my pair of 36v10ah packs just once so far. According to the CycleAnalyst I got over 11Ah out of them (wired in series) so I'm quite happy with that!

 

[Jeff]

Anyhow when looking closely at the tiny surface mount resistors. I have circled the first 3 groups in yellow. The first 3 groups each have 4 resistors. The next 9 groups a couple have the same 4 resistors but some only have 3 and in one case just 2 resistors I circled in red were there was a missing resistor.Everything thing on the board seems very consistent and repetitive except the little resistors that seem all over the place with some missing.
]

Kurt, from what I can see of the circuitry, it's fairly clear that headway has used the same TL431 shunt regulator circuit as others. In their implementation, they included additional pads for trimming the vclamp threshold voltage. Resistors will vary in their values, and this circuit requires somewhat critical trimming for each channel. I used this same method of paralleling resistors for trimming in a vclamp circuit.
A pot for each channel was used in the Goodrum vclamp circuit implementation. Although it lends for easy adjustment, the same issues are cause for drift.

Regards, Jeff

 

[fechter]

I agree it looks like some kind of voltage trimming arrangement.
On our BMS, we used 1% resistors on the dividers and get around 20mv in variation between cells at the most. Perhaps they found it cheaper to use 10% resistors and trim it later. Seems like a lot of extra work to me.

 

[Kurt]

Thanks for that guys,

Looking at it now the fact that there are so many resistors missing. There is to much variation to be a simple slip up in when soldering the BMS. What you are saying about trimming the voltage with the resistors sound more logical .

I think I will just get on with reconfiguring my pack and put the BMS back to use.

Kurt.