Unbalanced Pack Help

[slippery410]

Hi guys.
My first post on this forum and Id like to start by saying thanks in advance to reading as well as posting to me thread.
Ive been wanting to build an bike for some time now but with a family and one on the way, its a luxury I could only hope to one day have for my own. A few of my friends have bikes and every now again I get to enjoy theirs.
Good friends indeed, as Joe Cocker once said, "Anything is possible with a little help from your friends. Well my buddy looked out for me big time. He was rebuilding one of his bikes and upgraded the motor and controller. He could have sold the pair without a problem and instead opted to gifting me with them. All my other friends with bikes were there that managed to scrape up all the other parts to nearly complete a bike. So if you guys are reading, Thanks Again! Love you guys.
My final bud gifted me with a battery. Big one. But he says theirs a problem with it. Says he shorted out the bms to the pack a few months back. Rather then take the time it would take to repair, he purchased an even bigger pack which is what he wanted to do anyway. He never got around to having it repaired so he gave it to me in hopes I could use it for my bike.
He removed the bms and sat it on a shelf till now.
He wanted to see how balanced the cells were so we pulled out the Multimeter and took a few reading for you guys to judge
23s pack 6 cell group. Voltage taken from each cell group:
1- 4.207v 2- 8.401v 3- 12.59v. 4- 16.81v. 5- 21.01v. 6- 25.21v. 7- 29.41v 8- 33.62v. 9- 37.83v. 10- 42.03v. 11- 46.23v. 12- 49.91v. 13- 53.29v
14- 57.5v. 15- 61.72v. 16- 65.93v. 17- 70.2v. 18- 74.4v. 19- 78.6v. 20- 82.8v. 21- 87.v. 22- 91.2v 23- 95.4v
The cells are INR18650-30Q Samsung SDI 136

What id like to know if these are safe ranges from one cell to the next to adapt a bms to this pack as well as any recommendations as far as a bms.
Im not too keen on handling such a powerful pack so if anyone could recommend some place that could service this pack would be a great help as well...
Thanks again guys and with a little luck Ill be E grinning all the time.

 

[flat tire]

You want the difference between the voltage readings after the first battery, which would give you the voltage of each parallel group, with the (small) resistance of everything hooked up in between added since you're measureing everything from common. So to do a little better, you can also measure the individual cell groups by moving each probe up a wire after you measure the first pair. The way the data is presented now whoever reads it has to do a lot of extra work to figure out the voltage of each parallel group.

That said, from a glance it looks like your first few cells are reasonably balanced but more importantly your pack is way charged and some groups appear to be over 4.2 volts. Personally speaking, I wouldn't charge the pack higher than 4.0 volts as you will gain nearly no additional capacity and the extra .2 charge voltage will disproportionately reduce the battery's life / capacity over time.

 

[Jon NCal]

Hi, can you calculate the voltage of each cell group, ie 1 is 4.20 volts, group 2 is 4.20 volts. group 3 is 4.19, 4 is etc. Then we can compare more easily.

 

[slippery410]

Thanks for getting back to me guys...
And yes, I see what you mean about parallel group voltage.
1st group- 8.39v. 2- 8.42. 3- 8.40. 4- 8.42. 5- 8.40. 6- 8.42. 7- 8.42. 8- 8.43. 9- 8.42. 10- 8.43. 11- 8.43
Hopefully this would assist you better....
Thanks Flat & Jon

 

[Jon NCal]

Hi guys.
My first post on this forum and Id like to start by saying thanks in advance to reading as well as posting to me thread.
Ive been wanting to build an bike for some time now but with a family and one on the way, its a luxury I could only hope to one day have for my own. A few of my friends have bikes and every now again I get to enjoy theirs.
Good friends indeed, as Joe Cocker once said, "Anything is possible with a little help from your friends. Well my buddy looked out for me big time. He was rebuilding one of his bikes and upgraded the motor and controller. He could have sold the pair without a problem and instead opted to gifting me with them. All my other friends with bikes were there that managed to scrape up all the other parts to nearly complete a bike. So if you guys are reading, Thanks Again! Love you guys.
My final bud gifted me with a battery. Big one. But he says theirs a problem with it. Says he shorted out the bms to the pack a few months back. Rather then take the time it would take to repair, he purchased an even bigger pack which is what he wanted to do anyway. He never got around to having it repaired so he gave it to me in hopes I could use it for my bike.
He removed the bms and sat it on a shelf till now.
He wanted to see how balanced the cells were so we pulled out the Multimeter and took a few reading for you guys to judge
23s pack 6 cell group. Voltage taken from each cell group:
1- 4.207v 2- 8.401v 3- 12.59v. 4- 16.81v. 5- 21.01v. 6- 25.21v. 7- 29.41v 8- 33.62v. 9- 37.83v. 10- 42.03v. 11- 46.23v. 12- 49.91v. 13- 53.29v
14- 57.5v. 15- 61.72v. 16- 65.93v. 17- 70.2v. 18- 74.4v. 19- 78.6v. 20- 82.8v. 21- 87.v. 22- 91.2v 23- 95.4v
The cells are INR18650-30Q Samsung SDI 136

What id like to know if these are safe ranges from one cell to the next to adapt a bms to this pack as well as any recommendations as far as a bms.
Im not too keen on handling such a powerful pack so if anyone could recommend some place that could service this pack would be a great help as well...
Thanks again guys and with a little luck Ill be E grinning all the time.

You need the individual cell group voltages. group 23 is 95.4-91.2=4.20, group 22 is 91.2-87=4.2v etc. etc,

 

[Dauntless]

If you REALLY want to help with an unbalanced pack, go here. Really.

https://endless-sphere.com/forums/viewtopic.php?f=3&t=87000

 

[Matador]

See the Excel calculation file I made for you (recommendations included within ) : View attachment Your cells.xlsx

Matador.

 

[Sunder]

Matador beat me to it, but here it is anyway.

Cell 12 and 13 is low. If you have a single cell charger, use that to bring it up to 4.20v. If you don't, cut the head off a USB charger, and use that as a 5V source. However, KEEP A MULTIMETER ON IT AND WATCH IT AT ALL TIMES. It will take a long time to get from 3.38/3.68 to 4.0, but then go from 4.0v to 4.2v very quickly. At approximately 4.35v, you will start having a risk of fire. Over 4.20 volt shortens the life of the battery, but is only marginally an increased fire risk

Cell 17 is a tad high, but not dangerously so. You might want to consider draining it using a 5V lightbulb, or another similar load.

Unfortunately 23S BMSes are not cheap or easy to come by. I would recommend once you have it balanced:

1. Charge to 4.1v per cell - That is, terminate the charge at 94.3
2. Do not discharge below 3.5v per cell (80.5v) under load.
3. Check every few charges, that no cell string is going over 4.2v. If so, it's time to manually rebalance again.

 

[Sunder]

Im not too keen on handling such a powerful pack so if anyone could recommend some place that could service this pack would be a great help as well...

This is also a good point. There is a risk of death over 60VDC, so most jurisdictions make working on 50v or higher legal only for qualified electricians. However, with a bit of due care and respect for the pack, you'd have to be really unlucky to die from under 100vdc. You definitely do not want two points of your body in contact with that pack!

And I'm surprised nobody has pointed out or asked you "Where are you located" so we can recommend someone to service the pack. I'm in Sydney, Australia. From your language, it doesn't sound like you're here, but if you are, I might be able to give you some recommendations, or have a look myself.

 

[Matador]

Matador beat me to it, but here it is anyway.

Glad we could both help !

Now that you mention, I think you are completely right about risk fire..... No appreciable risk below 4.35V/cell... Just not good for the cell's life ... I guess I over-reacted to the 4.27V cells 6P group
After all, some cheap chargers allows for 0.05V error margin, so that'd be 4.25V after charge... So 4.25 V would the still okay.

I just never even tried to charge higher that 4.20/cell..... Though I recieved a new charger by mail today to add to my other chargers (3 x Foxnovo 4S digital chargers, one 14S 58.8V 5 amp from em3eV and one BC168 I'm still waiting for in the mail, one BST900W) It's the genuine SkyRC ImaxB6 mini.... There's that LiHV mode for those 4.35V rated cells..... I think I have some in my leftover, soon to be abandonned, laptop cells box. Even though it's not a very powerfull charger, and not a real balancer, the Imax B6 mini is a good charger. Another cool thing about the new B6 is the internal resistance mesurement fontion (of course I have no idea how accurate/inacurate it really is, but good for orders of magnitudes I guess) I yet have to find power supplys to feed the ImaxB6 mini (10-18V) : currently modding a 18.5V 4.9A laptop adaptor to step it down by adding trim posts to IC number 3 on it's PCB...
Also, the BC168 will pump a whole lot of juice... I'm thinking server PSU unit for this one. Also order a new chinease V-A-Ah-Wh meter to add to the 4 that I already own. This stuff is addictive !

slippery410, your battery was is a great gift because it's made of one of the highest end 18650 cells. IMHO, that pack is worth some troubleshooting and investment in tools for that matter.... As you can see with the numbers of chargers I have, I like to invest in tools rather than just rebying stuff when it does not work anymore... I think that in the long run, I will save a bit of money, but I will surely acquire knwledge and experience....

 

[Matador]

Cell 12 and 13 is low. If you have a single cell charger, use that to bring it up to 4.20v. If you don't, cut the head off a USB charger, and use that as a 5V source. However, KEEP A MULTIMETER ON IT AND WATCH IT AT ALL TIMES. It will take a long time to get from 3.38/3.68 to 4.0, but then go from 4.0v to 4.2v very quickly. At approximately 4.35v, you will start having a risk of fire. Over 4.20 volt shortens the life of the battery, but is only marginally an increased fire risk

Cell 17 is a tad high, but not dangerously so. You might want to consider draining it using a 5V lightbulb, or another similar load.

Unfortunately 23S BMSes are not cheap or easy to come by. I would recommend once you have it balanced:

1. Charge to 4.1v per cell - That is, terminate the charge at 94.3
2. Do not discharge below 3.5v per cell (80.5v) under load.
3. Check every few charges, that no cell string is going over 4.2v. If so, it's time to manually rebalance again.

Here's the new device I was talking about(the new V-A-Ah-W-Whr) that I order cheap on eBay (25$) : Ming He 120V 30A Wireless Power Monitor. The cool thing is that it has a 30A relay linked to it (there are also other version of it but with a big ammeter shunt on it instead).... So for instance, you could use it for a LVC or HVC.... But it's still preferable to keep trak of the voltage and watch it all the time. Anyways, great monitoring device for cheap.
Here a good review...
[youtube]YGp2wDAfC1k[/youtube]

As with most chinese cheap stuff, I still would recommend verifying calibration (especially for voltage readings) with a more precise volmeter (ideally Fluke 189 if you have one on hand or if you can borrow it from a friend for a day)...

Matador

 

[slippery410]

/quote]
You need the individual cell group voltages. group 23 is 95.4-91.2=4.20, group 22 is 91.2-87=4.2v etc. etc,[/quote]
Sorry if I made this confusing for you at all Jon. This is all new to me and you guys are teaching me with every post. That along with the wealth of information that could be found rite here on the Sphere. Guess I should have asked how to properly gauge one of these things. In all honesty I was shock when I first saw this thing and didn't even believe its was possible to have that many cells connected together. I really had no idea what it was that powered one of these bikes and simply because of the cost for some of the better packsI that they were made of some blue glowing rocks that are found inside of fallen meteors.
The creativity you guys posses is staggering.

See the Excel calculation file I made for you (recommendations included within ) :

Matador.

Thanks Matador for translating my numbers for me as well as all the fine suggestions you and Sunder had recommended.
My friends agree that the pack was a bit overcharged. Says this is the first time that pack has ever been that high and blamed it on the failing bms and quite possibly the charger as well and strongly pointed out that I take your advise and invest in a few case sensitive tools.
To the words suspect charger sprung yet another of my heroes into action that dropped an Eaton APR48-3G charger on the table. Says he has been using it to charge his Addapto Battery Pack and that it works flawlessly but that I should ask about its compatibility with the pack I was gifted. Says its better to be safe then to be running for the extinguisher.
Hopefully someone has some knowledge of this charger and if its applicable in this situation.

If you REALLY want to help with an unbalanced pack, go here. Really.

https://endless-sphere.com/forums/viewtopic.php?f=3&t=87000

Thanks Dauntless for the link. Made for an interesting read and was very helpful.

Im not too keen on handling such a powerful pack so if anyone could recommend some place that could service this pack would be a great help as well...

This is also a good point. There is a risk of death over 60VDC, so most jurisdictions make working on 50v or higher legal only for qualified electricians. However, with a bit of due care and respect for the pack, you'd have to be really unlucky to die from under 100vdc. You definitely do not want two points of your body in contact with that pack!

And I'm surprised nobody has pointed out or asked you "Where are you located" so we can recommend someone to service the pack. I'm in Sydney, Australia. From your language, it doesn't sound like you're here, but if you are, I might be able to give you some recommendations, or have a look myself.

Im in the US. Small state to the east called New Jersey.
I sat there looking at that pack and it simply looks frightening as if a thunderbolt is going to jump out and zapp me in the ass if I get too close. I couldn't work on this thing. Id be too afraid its going to send me for a ride of its own across the room. Ide be like a long tailed cat in a room full of rocking chairs probing this thing. Not even my friends like screwing around with their packs but say if you maintain them properly, you will never have to.
I would really feel more at ease to simply investing in this pack as you also mentioned. My buddy paid nearly 3 grand for this pack new. Says its not even six months old. Says it can't have more then a couple dozen charges if that. But that the warranty expired and the look on my face was worth more then the 3 grand he paid for it. Thanks Tito. You have always been a true friend.
I was given 2 different controller options but that I should first confirm the batteries condition if I wanted to go the an Adaptto controller. As this controller put a greater demand on the battery but that it is also a more powerful system to the Crytalite. Which would you recommend?

 

[Sunder]

Wow, $3000. I know Matador identified the cells at premium quality, but that's one hell of a battery and definitely worth saving. That's roughly what I've spent in total on all my eBikes over the last 5 years. (Maybe 6 by now).

Unfortunately, I don't know anyone in your patch of the world, but hopefully someone else here can suggest someone to help. If you must do it on your own, here's a few safety tips:

1. Flatten the battery to around 75v if you can do it on a per cell string basis. If not, drain the whole pack as much as possible, while not letting any single cell string go below 3.2v. The further you are away from that 100v mark, the better.
2. Always have dry hands. Dry skin is a great insulator. Wet skin is a great conductor - so is blood. So wet skin, blood, through heart to wet skin on your other hand is your highest chance of shock, and lowest chance of survival.
3. Wear safety glasses. At that voltage, a Lithium battery and blow molten metal a significant distance:

4. Always bear in mind that your tools are made of metal, and WILL cause a short.
5. Mask off anything you don't need to work on - preferably with gaffer's tape, but even masking tape and paper is fine. This reduces the chance of short circuit
6. Always do it with someone else away from you but present. If the worst happens, someone to immediately call 911 could be the difference between a scare and death. Electrocution isn't always instant death. You could enter fibrillation (irregular but not stopped heart beats), or embolism (believe it or not, the electricity can cause your blood to electrolyse - remember separating water into oxygen and hydrogen in high school?). These can both be recovered with prompt medical care.

I agree with your friend though - Once in good condition, and well looked after, you should never have to take that apart again. If you think 23S @ 96.6v is high though, I built a 36S, 151.2v battery in October last year. I pulled it out after 3 rides back then, and it was still perfectly balanced. Every ... Last ... Damn ... Single ... Cell had zero variance. I pulled it out again 3 weeks ago (What's that, about 5 months of nearly daily use?) and 2 cells were 0.02v low. At that rate, the pack will be dead from calendar age before it becomes significantly unbalanced.

Find someone who can re-balance it for you (I have seen a few people around here claim to be in New Jersey - maybe if you started a new thread in the General EBike forum asking "Can anyone help New Jersey Member?" you'll get more help - don't need to repeat your story, just link to this thread, then as I said above, lower the charging voltage to 4.1v per cell, and never run it below 3.5v per cell, and it will probably never need balancing again.

 

[slippery410]

Hey Sunder.
My buddies and I were just discussing the plan of attack on this battery laid out by you and Madator. After discharging the pack a bit we tried charging the weaker cells that began to charge rite up. Looks like we just may be able to brings those cells back to voltage.
Just then my buddy remembered something. Says shortly after getting the pack he was on his bike grabbing some really big air on a trail and drifted off line Into the trees. He took a nasty hit as did his bike. After checking over everything he noticed a gouge in the side of the battery that sit the two effete rows.
Upon inspection of the row of cells we found nothing out of the ordinary.
I also noticed that upon building this battery that the nickel strips were folded over in half at the center in order to stack the battery with a carbon fiber plate in the center.
Could any of things Ive described cause some kind of resistance that will not allow those two row to fully charge or become problematic in the future.

 

[999zip999]

So where do you live ?

 

[slippery410]

So where do you live ?

Elizabeth New Jersey

 

[Sunder]

Could any of things Ive described cause some kind of resistance that will not allow those two row to fully charge or become problematic in the future.

If those are row 12 and 13, it sounds like several of the parallel connections within that string broke, or semi-broke. If a string loses a cell, it will still keep supplying voltage to keep the pack voltage up. However, as it gets drained, it will drain faster that the other strings around it.

When it gets charged back up, in theory it should do the reverse, but because of the intricacies of how batteries work (different internal resistance at different voltages, if you really wanted to know), that doesn't happen.

You'll need to find any loose or gone spot welds in the parallel connections within those strings. This could be hard, as it might make spring force contact while sitting flat on a table, but break that contact when the battery is moved.

Good luck.

 

[slippery410]

Thanks again guys and for all your help and Sunder, I figured Id give you an update to my battery dilemma.

A friend recommended that I take your advise and slowly recharge the weaker sting of cells. I found a flashlight charger in my junk draw that was 4.5v at 500ma and a couple alligator clips. It took some time to get those weaker cells back up to charge. Mostly because I didn't want to leave this project unattended in any way. Safety first!
But being I have quite a bit to do with motor and controller installation, I had it running in the background always checking on it with my multi meter as the charge slowly continued to rise.
After debating which controller I was going to use my friends convinced me on going with the Adaptto Mid-E system gifted to me by my buddy Tito. The reason I was sold on this system was because not only was it a very powerful system but that it also incorporated its own BMS. Now, if you can recall the BMS on this battery was shorted out, and as you said kind of. a hard item to find as well as expensive. So I contacted a man over at Envolt by the name of Alex that was kind enough to send me a BMS as well as the balancing leads to install.
A bit overwhelming at first but this was something I was determined to do backed by my buddies who wanted to see it happen for me, or should I say helped to make it all happen for me.
You Guys Are The Best...
I didn't really feel too comfortable soldering all new leads to this battery and figured why solder new leads at all when there are leads already soldered.
With a terminal removing tool I popped out each pin one by one from their original terminal connection and placed them in their proper prospective positions on the Adaptto BMS. Turns out it was a lot simpler than I had first anticipated and worked out perfectly.
Upon initial bench test you found that the battery was still off as far as balancing so we began the whole setup and tuning, all the while discharging the battery further to be able to give it a good charge through the system and let the BMS do its job.
After a few good hours on the Eaton 3G charger we were able to get the battery to this point that my friend says is more then executable.
As for whether or not rows 12-13 are going to preform as expected has yet to be determined but at least I have a way of closely monitoring its performance. Hats off to you guys over at Adaptto.
Ill keep you posted on its outcome back to health. If for whatever reason the pack is going to need to be rebuilt I was hoping you could throw some advise my way on the proper assembly of a high amp pack.
Thanks again to all you guys that helped out as well as to those who run this forum.

 

[Matador]

Thanks again guys and for all your help and Sunder, I figured Id give you an update to my battery dilemma.

A friend recommended that I take your advise and slowly recharge the weaker sting of cells. I found a flashlight charger in my junk draw that was 4.5v at 500ma and a couple alligator clips. It took some time to get those weaker cells back up to charge. Mostly because I didn't want to leave this project unattended in any way. Safety first!
But being I have quite a bit to do with motor and controller installation, I had it running in the background always checking on it with my multi meter as the charge slowly continued to rise.
After debating which controller I was going to use my friends convinced me on going with the Adaptto Mid-E system gifted to me by my buddy Tito. The reason I was sold on this system was because not only was it a very powerful system but that it also incorporated its own BMS. Now, if you can recall the BMS on this battery was shorted out, and as you said kind of. a hard item to find as well as expensive. So I contacted a man over at Envolt by the name of Alex that was kind enough to send me a BMS as well as the balancing leads to install.
A bit overwhelming at first but this was something I was determined to do backed by my buddies who wanted to see it happen for me, or should I say helped to make it all happen for me.
You Guys Are The Best...
I didn't really feel too comfortable soldering all new leads to this battery and figured why solder new leads at all when there are leads already soldered.
With a terminal removing tool I popped out each pin one by one from their original terminal connection and placed them in their proper prospective positions on the Adaptto BMS. Turns out it was a lot simpler than I had first anticipated and worked out perfectly.
Upon initial bench test you found that the battery was still off as far as balancing so we began the whole setup and tuning, all the while discharging the battery further to be able to give it a good charge through the system and let the BMS do its job.
After a few good hours on the Eaton 3G charger we were able to get the battery to this point that my friend says is more then executable.
As for whether or not rows 12-13 are going to preform as expected has yet to be determined but at least I have a way of closely monitoring its performance. Hats off to you guys over at Adaptto.
Ill keep you posted on its outcome back to health. If for whatever reason the pack is going to need to be rebuilt I was hoping you could throw some advise my way on the proper assembly of a high amp pack.
Thanks again to all you guys that helped out as well as to those who run this forum.

Wow, seems like it's back in balance, all cells betweem 4.10 and 4.11 volts.

 

[slippery410]

[/quote]Wow, seems like it's back in balance, all cells betweem 4.10 and 4.11 volts.[/quote]
Couldn't have done it without you.... My sincerest thanks Matador

 

[Matador]

Glad we could help! Ride safe my friend

 

[Joost]

Bottom balancing would maybe help.
With bottom balancing will all groups bring too the same voltage level, and not in capacity
And top balancing will only leveling the capacity

 

[bobc]

Just a thought -
you mentioned a carbon fibre separator plate in the vicinity of the 2 cells that were slightly discharged. Carbon fibre forms a resistor - if the plates is allowed to touch terminals (or connectors and battery body sides) it will conduct and discharge (incidentally at a slight risk of fire)
I've seen a 12V SLA ruin a carbon fibre body panel when carelessly piled up by someone who didn't know the risk - plenty smoke on that occasion but no actual fire.
I'd 2nd the bottom balancing notion; if you can't get a 23s BMS, at least a bottom balanced pack will allow safe operation protected only by the speed controller LVC (essentially makes running "no BMS" safer)

 

[slippery410]

Bottom balancing would maybe help.
With bottom balancing will all groups bring too the same voltage level, and not in capacity
And top balancing will only leveling the capacity

Thanks Joost...I had no idea that the pack could be bottom balanced and from what Ive read this far concerning your suggestion would be bringing the voltage of all the cells down to their minimum voltages then recharge them all back up together. They say its actually better to bottom balance then top. But that would mean bringing the voltages of each cell to a rather dangerous lever. Matador explained how I should never bring the cells down past 3.5. 2.7 seems like a deep drain. Theoretically, sounds like a great way to correct the capacity of the cells.
Now, is this something I could do using the BMS of the Adaptto to balance them all down to 2.7 then charge them back up to their full capacity?

Ive read how juggling the voltage of the cells from 3.5 down to 2.7 and back up to 3.5 and down ect ect ect. That doing this before bringing them all up to full charge is even better way to bottom balance. Is this correct?
How would you go about undertaking such a task?

Just a thought -
you mentioned a carbon fibre separator plate in the vicinity of the 2 cells that were slightly discharged. Carbon fibre forms a resistor - if the plates is allowed to touch terminals (or connectors and battery body sides) it will conduct and discharge (incidentally at a slight risk of fire)
I've seen a 12V SLA ruin a carbon fibre body panel when carelessly piled up by someone who didn't know the risk - plenty smoke on that occasion but no actual fire.
I'd 2nd the bottom balancing notion; if you can't get a 23s BMS, at least a bottom balanced pack will allow safe operation protected only by the speed controller LVC (essentially makes running "no BMS" safer)

Thanks for pointing this out as it as a second look was required.
Its appears to be a large blank pcb board sandwiched between two sheets of redish cardboard. I don't know how I came up with carbon fiber. I never really studied it to well and at first glance its what came to mind but thanks all the same. Any suggestions you could add to bottom balancing would be great as well.

 

[bobc]

A manual bottom balance is easy, you just need a resistor, a meter and lots of time......
If you used a 1ohm resistor your balancing current will be 2.5 to 3A, all quite benign. It is blowing off up to 10 watts so one of those little resistors won't do - and whatever you use might get too hot to touch.
One day I'll make a simple circuit to do a single cell discharge to 2.7V or so.