Simplest BMS

[Skippic]

Jeremy, does your pack charge the same way every time? I don't understand, where the charge from the different cells goes... If they are leveled at full charge, then you take out say 5Ah from all of them, the voltages might be different due to their respective capacities, but once recharged those 5Ah, they should be more or less the same.

[Jeremy Harris]

It's the "more or less" that turns out to be the problem. If you could guarantee that all the shunts would turn on at exactly the same instant, then your system would work fine. TBH, this is the mistake I made with my first BMS, I assumed that because the pack was balanced this is what would happen.

The reality is that tiny effects, like the temperature of one cell being different to another, or one having self discharged by 1% more than the rest, will cause one cell to activate it's shunt before the others. Once this happens the series resistor for the shunt that's on will develop a voltage across it, equal to the difference in voltage between the charger and the all the cell voltages, and that will cause the voltage across the first cell to turn on to rise.

If you're lucky, then it won't rise too far to do harm before the other cells in the pack catch up and their shunts turn on. If you're unlucky the voltage will rise and kill the cell before that happens. It's a bit like Russian roulette, you have no way of knowing which way it'll go.

My case was a less likely to do this than yours, because I used a power darlington transistor to increase the shunt current capability of the TL431 to 300 mA. The higher shunt current should have meant that the other cells would come up to voltage quickly. What happened was that the first shunt to turn on caused the whole charge current to flow through its shunt, which smoked it. I then added a series resistor to the shunt, thinking that this would reduce the dissipation in the pass transistor, pretty much as you've done. This resulted in a cell going to over 5 V............

The answer was to make the thing more complicated. In the end I added optocouplers to each shunt and diode OR'd them to a FET switch that turned off when the first shunt turned on. This FET normally shorted out a big power resistor in the negative feed from the charger, so when it turned off the difference between the charger voltage and the total cell voltage was dropped across the resistor, rather than one of the shunts. This works fine, it's a bit crude and simple, but it's proved to be faultless over the two years I've been using it on my electric boat. The circuit is below - bear in mind it's for a 4 cell LiFePO4 group pack that's only 13.2 V nominal, so some changes would be needed to the gate resistors on the FET to cope with a higher pack voltage and the power resistor the FET shunts would need changing for a different set up. You could add as many cells as you wished otherwise, just adding more shunt circuits.

4 cell charge control.jpg

(158.13 KiB) Downloaded 3 times

[Skippic]

You say you have LiFePO4, maybe it's different for Lipo. I'll test it with my battery and maybe with GCinDC's and we will report back.

[Jeremy Harris]

You say you have LiFePO4, maybe it's different for Lipo. I'll test it with my battery and maybe with GCinDC's and we will report back.

No.I'm afraid it's the same with my LiPo packs on the bikes. They do the exactly the same thing, one cell group reaches cut-off before the others, unfortunately.

[Skippic]

Please don't take my word for it, try it, as I (and others, like Gary Goodrum, Richard Fechter and BobMcCree) have done. You will find that one cell will, at some point, go way over voltage with a simple system like this. Maybe not on the first charge or two, but at some point in the life of the pack it will happen and you'll perhaps blow a cell.

Can you please post some links for this? I'm not saying I don't believe you, just wanted some more detailed info.

[Jeremy Harris]

I'm not sure I could easily find the full story of all the BMS developments on here, as it was a few years ago now and the BMS threads that were started by either Bob or Richard (I think) got to be really long, with all the good stuff buried in them.

If you want to go looking for this stuff, then it starts in around 2008/9 I think, and most of the good stuff will have been posted by Bob McCree, Richard Fechter or Gary Goodrum, as they were the guys who worked on this for months.

[GCinDC]

...I'll test it with my battery and maybe with GCinDC's and we will report back.

just tuning in... glad to hear that my new bms is ready.

...You will find that one cell will, at some point, go way over voltage with a simple system like this. Maybe not on the first charge or two, but at some point in the life of the pack it will happen and you'll perhaps blow a cell.

but maybe i better order a back up pack tho? and increase the insurance on the garage?

how's your bms running now, skippic?

[Skippic]

Mine is connected, but I haven't measured the cells yet. I don't really expect to see great results, since my pack was always balanced (I always bulk charge).

I'm running a test on the old 5s2p pack you gave me. Yesterday I balance charged them together to 4.1V, then charged the first cell on both to 4.22V. After that I disconnected the packs to get 2 x 5s. I connected the first pack to the BMS. For now the first cells on the respective packs are 4.17V an 4.18V. So there is a certain voltage drop from yesterday on both (I measured the 4.22V just after charging). There is an additional drop of 0.01V on the BMS pack currently at 4.17V and discharging at 3.3mA. To see the effect it has on the pack I'll keep monitoring the voltage difference.

[Skippic]

So some 24h after the beginning of the experiment the first cells are 4.16V with BMS and 4.18V without. The discharging current is 3mA

[Skippic]

Sorry, forgot to measure after 48h, but this morning 60h after the start the first cells are 4.14V with BMS and 4.18V without. The discharging current on the BMS is 2.2mA.

The rest of the cells on the BMS are at 4.09V and discharging at 0.3mA.

In other words it's working just as expected.

[Jeremy Harris]

Have you tried testing it while charging a few times to see what happens when you get a tiny imbalance in SOC between cells?
This is the critical case, leaving it hooked up with cells being stored won't demonstrate anything very useful, unfortunately, as the circuit isn't doing anything at all other than presenting a tiny drain on the pack.

[dak664]

Or make a 10 ohm jumper cable with alligator clips that will pull 350 ma from individual cells. Discharge one by an amp hour and then see what the BMS does on the next charge.

A 50 milliamp limit on charging current would take 20 hours to bring it up again. That jumper cable then comes in handy (and another 10 just like it) to shunt all the full cells while the low one comes back up.

In ordinary operation there are often cells that lose a little each cycle, possibly because of temperature differences do to position, internal resistance, or connection resistance (there should be more attention paid to connection resistance, it is easy to measure). You can bring it back up each charge cycle with a few minutes extra, or you can bring it up every 100 charge cycles with a few hundred minutes extra. In either case the slight overvoltage on the other cells is probably not doing them any good.

[Skippic]

10 ohm jumper cable with alligator clips

My hope was to have something automatic. It might take time, but as long as it isn't mine...

[Alan B]

It is good to see folks trying different approaches. Occasionally it leads to a new solution, but it always leads to learning.

A gentle balancing force should be adequate if the unbalancing forces are also gentle. It fails when the unbalancing forces exceed the weak balancing.

For well balanced packs it should work okay.

When a more serious balance problem develops it will quickly overwhelm the balancer and the weak cells, and if not detected the weak cells will be destroyed. You might say this is okay since the cells are bad, but this is not the whole story. For example, in a 4P pack, one of the three cells can go bad, resulting in the destruction of the three good parallel cells. A better system will catch this condition and allow the user to replace the bad cell and not damage the others.

Combining weak balancing with a cell level high voltage cutoff for the charger would solve this issue and provide real protection.

The question is, when used with a pack level charge cutoff, does weak balancing provide more pack life or protection? Since it does not detect the problem, it would not provide any real protection. It might provide a little extra life by gently "steering" toward balance. The same thing is accomplished by periodically charging with an RC balance charger, "resetting" the pack balance, with the opportunity there to "learn" about the balance of the pack. I suppose the monitoring can occur here, more or less, if the pack balance is checked periodically with the cell meters. So the user becomes part of the BMS.

If one of the weak balancer circuits fail and get stuck on there is no indication, and it will kill the pack. Unless you happen to monitor it frequently enough to see the problem.

In some sense, using a weak balancer is like what lithium manganese chemistry does, providing a weak restoring force toward balance. This does seem to work, but not as well as a true balancer (I heard Makita recently added balancing to their LiMn tool system and bad pack incidence went way down).

In summary, this technique should produce some improvement over no balancing, and some risk of circuit failure without detection, and not nearly as much improvement as full balancing.

Very interesting work!

[Skippic]

I wish there was a simple way of adding cell monitoring to this. My hope for now is to just skip balance charging. I didn't feel like taking apart the 24s2p pack every time I needed to balance cells.

I will have a look at the pack tonight when it is fully charged to see if there are any cells of balance (some broken components/bad connections).

[Alan B]

You can balance charge a 24S pack without rewiring using one Hyperion 14S charger set up for 12S by plugging it into the pack twice, once for the lower, once for the upper half; or you can get a pair of Hyperions and do both halves at once. 1100 watts of balance charging.

Or you can use a 6S charger plugged in four times, or four 6S chargers, each fed from a separate power supply.

There are lots of ways to avoid rewiring a pack to charge it.

[Skippic]

Two weeks later, the cells are all balanced at 4.09V. The first cell, that was originally 4.18V is now discharging at 280uA, the rest of them at 49uA, so the voltage is probably still a little higher. The first cell on the unbalanced pack is still at 4.18V.

[sn0wchyld]

Two weeks later, the cells are all balanced at 4.09V. The first cell, that was originally 4.18V is now discharging at 280uA, the rest of them at 49uA, so the voltage is probably still a little higher. The first cell on the unbalanced pack is still at 4.18V.

sounds promising mate. where did you buy them form?

[Skippic]

I made them myself, very simple to do, maybe 2 hours of work. The cost of the components is about $21 for a 10Ah 24s pack. If you would be interested I can post links where I bought all my components.

[Jeremy Harris]

When you charge the pack, how are the shunts coping? Are they holding the cell voltages down OK as the pack reaches full charge?

This is what bothers me, as you know, as once the pack is charged then the shunts don't need to do anything at all, in fact the pack would be better off if they were disconnected and not causing even the tiny drain they take. Balancing is only a charge limitation process, not something you want to continue doing as the pack discharges. Once the pack is fully charge, with all cells at the cut-off voltage, there's no need for the balancing shunts to do anything at all.

[317537]

http://www.fairchildsemi.com/ds/PZ/PZTA14.pdf


I played with these darlingtons over a year ago, I had no clue what I was doing at the time trying to get them to draw over 800 ma @ 3.75v to balance a 15ah pack.

From memory they seemed a little sloppy drawing to much current below the threshold, but again, I didnt know what I was doing at the time.

These darlingtons have a very good beta. the charts are not that detailed for under a milliamp. But they can draw over 1.2 amps on their own. and dissipate 626mw. If it was .25v you wanted to burn, thats 2.5 amps well over what you need and what they can handle. I use them for swtich leds and I can get a 10ma led to switch on with the current from my fingers or a 1M0 resistor, off a 4.5v supply with about 780 ohm load. They would perform better if you heat paste glued them to a heat sink. I tried them on a cell and was tryng to get one amp out of them. If the curves arnt sexy enough they could be used in place of resistors. But the charts are only from 1ma to 1 amp , which is not very detailed to help understand its complete vesatility. I imagine the currnet curve almost goes off the screen under b.2v @5v Vce, and @ 1v can draw 200 ma. If you turn the chart side on so the current is going upwards intead of across, up then you get a better idea

Give me some feedback. I love these litle darlings just the same.

[Skippic]

When you charge the pack, how are the shunts coping? Are they holding the cell voltages down OK as the pack reaches full charge?

This is what bothers me, as you know, as once the pack is charged then the shunts don't need to do anything at all, in fact the pack would be better off if they were disconnected and not causing even the tiny drain they take. Balancing is only a charge limitation process, not something you want to continue doing as the pack discharges. Once the pack is fully charge, with all cells at the cut-off voltage, there's no need for the balancing shunts to do anything at all.

I didn't monitor them while charging, but my charger has it's HVC set to 4.1V/cell. So unless one cell is totally out of whack it just stops at 24x4.1V and the BMS does it's magic. The BMS will not make any significant voltage difference while charging, it will just slowly get the pack to be balanced at maximum charge. So I would expect a pack with +/-0.1V cells to be balanced after one week.

As for the self discharging when left alone below 4.1V, it's about 50uA, so according to my calculations it would take 2 years to discharge 1Ah or 10% of my pack. I too would prefer it to just turn off below 4.1V, but for me it's not worth adding more components.

[sn0wchyld]

I made them myself, very simple to do, maybe 2 hours of work. The cost of the components is about $21 for a 10Ah 24s pack. If you would be interested I can post links where I bought all my components.

yea sorry mate thats what i meant . cheapest i found the 4040's was like $5 each, and no mention to their accuracy.

[Jeremy Harris]

When you charge the pack, how are the shunts coping? Are they holding the cell voltages down OK as the pack reaches full charge?

This is what bothers me, as you know, as once the pack is charged then the shunts don't need to do anything at all, in fact the pack would be better off if they were disconnected and not causing even the tiny drain they take. Balancing is only a charge limitation process, not something you want to continue doing as the pack discharges. Once the pack is fully charge, with all cells at the cut-off voltage, there's no need for the balancing shunts to do anything at all.

I didn't monitor them while charging, but my charger has it's HVC set to 4.1V/cell. So unless one cell is totally out of whack it just stops at 24x4.1V and the BMS does it's magic. The BMS will not make any significant voltage difference while charging, it will just slowly get the pack to be balanced at maximum charge. So I would expect a pack with +/-0.1V cells to be balanced after one week.

As for the self discharging when left alone below 4.1V, it's about 50uA, so according to my calculations it would take 2 years to discharge 1Ah or 10% of my pack. I too would prefer it to just turn off below 4.1V, but for me it's not worth adding more components.

Maybe I'm being a bit dumb here, but what exactly are they doing then?

Shunt voltage limiting like this is only useful as safety protection for ensuring that a cell doesn't exceed the maximum allowable voltage during charge. In an ideal world you'd disconnect the shunt protection as soon as you'd finished charging, as it does nothing but create a small current drain from then on. Certainly adding shunts across a pack when it's just sitting or being discharged does nothing useful at all.

Even if you are limiting the overall charging voltage you can still occasionally get a cell imbalance that pushes a cell into an over-voltage situation, as we discussed earlier in this thread. The critical, and only useful, test for a cell shunt protection circuit like this is to see how it copes when charging, particularly when you get a cell out of whack.

[317537]

Yeah I would agree, dedicated charged based BMS's on seires cells wont do the trick unless there is a shunt trip anyway. So in the end why bother as it makes no difference. its just the two furthest ends of the BMS will charge say on a 48v pack to 60v. IMO any cell can go over if a one or worse or if a few cells are under. I thought about Isolated DC to DC converters,even isolated converters outputs become unilsolated when you stack them in series. I even tried four LM317Ts on a 3S 5ah puffy cells I picked up, just for the heck of it. The currnet just goes through still to the highest cell and no balance occurs. Although I did observed that the LM317t can burn energy and get hot if the outout exeeds the ADJ voltage set, if you set the load and control resistor low right for the job. They are not practical for obvious reasons, being that I needed more than 17v volt charge supply for 3s to make them effective. But the experiment was fun and good learning.