How does a BMS board ACTUALLY work?

Well. Perhaps my mood was not so great this week, and I took it out on you. I've been watching one of my dogs slowly die for a year, and we finally put him down last night. Yeah, only a dog I know. But I've been bitchy all week knowing it was coming.

But it does sound like at least two of your cells aren't so great.

Anybody out there want to back me up? Is immediate self discharge to 3.3v of lifepo4 charged to 3.65v normal? Hasn't been for any I've owned. Sure, they drop to 3.5v, but 3.3?

I don't recall any real battery experts ever saying 3.3v is fully charged lifepo4. I think your sales man might be misinformed.

Also, don't edit when you quote me. REALLY good cells is what I said. I never said the cells were unserviceable. I just wouldn't call that the behaiviour of REALLY good cells.

At this point, the ping lifepo4 I own self discharges down to 3.4v after some abuse. But they sure didn't do that till I frocked em up racing.

Hopefully, your low cells will balance and all will be fine. It may take weeks on your charger, or manual charging of those two cells.
 
Re reading the entire thread, You linked to some 15 ah cells.

And two of em put out a whopping 8 ah so far. If I'm out a line give me a whack. But that fits my defenition of crap.

You say you have an accucell 8. It would be interesting indeed to see what those two individual cells do when charged with that charger using the lifepo4 setting.

Do they get to 3.65? Do they hold it then?
 
Fail... Flat tire after 1,5Ah...

Sorry about your dog, I totally understand, I feel the same way about my cats :)

The two lower cells I charged on the Turnigy Accucell, but on Liion to 4,1V more or less according to the technicians (spelling) specifications. Then I got more or less 15Ah in them, but as I said, a flat tire prevented me from testing...

However, after coming off the charger at 4,1V, they quickly dropt down to 3,8V ish, but after 1,5Ah drawn they were down to 3,33V. The others were at 3,31-3,34V.

I dont know what to think. When I researched the batteries before buying, they looked very promising. Selling to EV-car producers and solar power plant. Quick replys, not like china when you have to wait 2 days for a email.

I´ll try jumpstarting the two bad cells, give them a couple of charges and see what happens. I have spares so no bigger problems but I don´t want to change if not neccessary.
 
nothing wrong with them if they go to 3.34V. the only way you know if there is something wrong is to do a capacity test. voltage don't mean anything. they all go there and stay there for years. i find them at 3.3V after sitting unused for 3 years. and after one day off the charger.
 
I´m thinking about making some sort of rig for measuring capacity (Ah), but I dont really know how to go about it.
I have a couple of watt meters, but I don´t know how to burn the amps in a reasonable fashion.

Some power resistors could do the job, if I had any...
 
i use two oil bath space heaters. i have some andersons soldered onto the end of an extension cord and i can plug in the heaters to that. but i can only pull max of 13A into the heaters with a 48V battery.

but when i do capacity tests, i follow the book and do it at .1C to make sure the numbers are consistent with factory specs. so for a 20Ah pack i use the low setting on one heater for 2A.

you still have not measured the voltages while it was charging, and there is nothing more important to know. you measure when it is off the charger, so you don't know if the BMS is balancing it or if one is far outa balance when on the charger. which it appears to be.
 
Electric toasters or 1.5 KW mains-powered space heaters will discharge at a few amps @ 36 or 48 volts DC. Parallel them on a plugstrip if you really need the extra current.
 
The thing is I need something to draw amp at 3-4V, since I´m testing each individual cell, and not the whole pack.

The flat is fixed now at least, if the rain goes away I can go out and test the capacity with the two low cells boosted...
 
Well that's definitely progress. Maybe those two cells were just being stubborn in the break in process.

Once you ride about a mile, lifepo4 cells should be down in the 3.3-3.4v area. That's the meat of the flat spot in the discharge curve. They'll sit there for a long time discharging away without dropping much more till the very end of the discharge. Then they'll start dropping faster as you ride, finally going over the cliff and dropping like crazy just before the bms or controller lvc pops.

So at this point, they are looking more normal than I thought. They didn't just self discharge to 3.3 right away.

DONT charge them on lilo. Put the charger on lifepo4. Some manufacturers have claimed that charging lifepo4 to 3.7 does irreperable damage. I don't think that's quite true, but ping pouch cells come set to charge to 3.75, and that is higher than most. Most bike packs charge to about 3.625v.

Again, normal, to my knowledge for a 16 cell pack is to charge to 58-60v. Then surface charge dissipates to 56v for the whole pack. So if you can ride 100 feet, and still have 55.5-60v your pack is fine. On a good new pack, surface charge may hold overnight, but fairly soon it will start to dissipate faster. One touch of the throttle will also knock out the surface charge, that is a slight overcharge that helps the battery balance completely.

Keep leaving it on the charger ovenight for awhile, if not from now on.
After you put some wear on it, you might only have 54v like I do now with my pingbattery.
 
I´m at my friends house now, he bought some cells as well and has built a 24S pack (crazy bastard). He has the same type of cells, same type of BMS (but for 24S of course), and same type of charger.

He left his battery on the charger for like 24H or even more, and his cells all had around 3,6V in them. 87,xV in total. More or less perfect.
Maybe I have been too impatient with my charging and should leave the pack on for like a couple of days.
His controller burned up on the first ride, so we haven´t tested much, but battery wise everything seems fine for him.
 
you don't test individual cells when you do a capacity test. you discharge the entire battery through the load.

you can keep track of how much capacity each cell has by recording how many Ah the pack produces before it reaches the 2V endpoint.

when the first cell reacheds 2V or 2.1V then you have to stop the discharge by disconnecting the load but keeping the wattmeter connected so it maintains the data.

then you put a single cell charger on the lowest cell and charge it up to match the others so you can continue the discharge.

when that cell has enuff charge that it will not trigger endpoint, then you reconnect the load and measure how long it is to the next cell to reach endpoint, then shut off the load, charge up that low cell, then continue the discharge until you reach the endpoint of the third lowest cell, and so on.. until you have a grasp of how much capacity each cell has.

you don't actually have to discharge all the way to 2V since it usually is just tenth of an Ah away at 2.2V anyway so you have a good idea and can stop in that range, 2.2-2V.

we still do not know if your BMS balances so it is pointless to test capacity. you have never established that the cells are able to get fully charged to begin with. without that you cannot do capacity tests anyway. all the cells have to be charged to at least 3.65V or more to start the capacity test. you have to list the cell voltages while on the charger for us to give you accurate advice, otherwise it is just more guessing. i think your pack is still unbalanced because it is not getting fully charged from the beginning.
 
Thanks for your very detailed explanation!

I have gotten home from my friend now and will let the battery charge for at least 24h.
When his battery was on the charger the MOSFETs on the BMS got warm, this I have never felt on my BMS.
So I have problably not given it a chance to do its job properly since I have wanted to go out ridning all the time, it´s so much fun :mrgreen:

In 24h I will get back with voltage readings on every cell, while it´s connected to the charger.
 
the charging mosfets should not get warm. but the shunt resistors do get warm when the shunt transistor diverts the current around the cell and through the shunt resistor. when that current goes through the shunt resistor and it gets hot you know that the cell has reached full charge. so you can tell how balanced the pack is by running your finger over the shunt resistor while it is charging, except usually the balancing is for a short time at the very end of charging, when the pack is balanced that is.
 
Yeah, keep it on the charger anytime you aren't riding it for awhile, if not forever.

Your buddy got some good cells. holding 3.6v till discharged. How new ones ought to act for awhile.
 
I see no reason why his 24 cells whould be any different from my 16+16 cells. Ordered at the same time, probably same batch and everything. Only thing is that he was more patient and let them charge for longer.
I´m gonna leave them over night and all day tomorrow while I´m at work and see what happens...
 
It's looking better and better. But a bad cell can be there in any group. For sure we now know you don't have a whole pack made from bad cells.

I got the wrong impression for sure. I was thinking, if the guy thinks self discharge to 3.3v is normal, he must normaly have some bogus stuff. He meant normal to drop to 3.3v once you ride a bit I think. Most of a 48v lifepo4 ride is done between 52 and 56v.
 
So I´ve had the battery on charge for over 48h now.
All cells came up to around 3,7V while on the charger.
And now after 16h all cells are around 3,57V, except one that was only 3,4V.
Worth mentioning is that I accidently short circuited that cell for a couple of seconds with the wires to the multimeter.
Hard to believe that the short would drop the voltage that much, so I think I have at least one bad cell in my pack.

Hopefully I can go out riding after work in the morning and pull some Ah from the pack and see how it behaves...
 
Peter Sternersson said:
So I´ve had the battery on charge for over 48h now.
All cells came up to around 3,7V while on the charger.
And now after 16h all cells are around 3,57V, except one that was only 3,4V.
Worth mentioning is that I accidently short circuited that cell for a couple of seconds with the wires to the multimeter.
Hard to believe that the short would drop the voltage that much, so I think I have at least one bad cell in my pack.

Hopefully I can go out riding after work in the morning and pull some Ah from the pack and see how it behaves...
Click to expand...

Of course the 3.57v resting once hit with a short would chill down 3.4x

If you put a LiFePO4 cell under it's normal load, it will have an operating voltage of 3 to 3.2v.

A HOC (hot off the charger LiFePO4 at 3.57v) will not maintain 3.57v with a load. It's not designed to. Under load it will drop, hence,
shorting a cell is to put the cell under load and the upper residual voltage will dissipate off quickly.

When building a pack with LiFePO4 (headway for example) a 2C discharge will be a 3.2v per cell and a 4C discharge will be a 3.0v per cell until
the capacity is used.
When the cell capacity is used to 80% DOD and a LiFePO4 Charger is applied the cells may see 3 to 3.3xv with maximum current, then when the
Cells are reaching near capacity, the charger starts to climb to higher voltage (3.65v per cell) as the amperes start to drop.

When sizing to use LiFePO4, one should consider the 3.0-3.2v nominal working voltage during load for their design, not 3.5x - 3.7x volts.

How do I know this to be true?..... Experiment - Observe - Take Notes Read and understand different battery chemistry and their attributed.
I hope I was helpful. :)

Tommy L sends.... \\m//
 
I apologize for making you more worried that need be, but you can stop fretting now.

Yer good. Even that one cell is not that bad, and 80% of that drop is just the surface charge going away. You just needed a bit more break in time on the cells, more time on the charger to balance them. You may find that one cell not quite as nice as the others, but it's not that bad at all.

Likely in the future, you will see a full 56v when you start to ride, which will drop to 54-55v fairly quick, then hold very well betwen 52v and 54v not under load for most of your ride. Basicly, you'll be doing 90% of your ride at about 3.5-3.4v per cell.

In time, with some use, you might be starting the ride closer to 54v, and most of the ride will be more like 52-53v. That's how my ping behaves now, 3 years and 230 cycles later, some of those cycles being quite harsh.

I just put balance taps on the ping, so now I can have a cellog on it, and watch it charge, and dissipate the surface charge at cell level. I am only able to hold a charge of 3.35v per cell overnight. The difference between hot off the charger at 3.65v and 3.35v is now surface charge for this pack. I'm slowly losing capacity, down about 10% now. But it still charges fully and balances on all cells. It was quite out of balance when I put the 8s balance plugs on it, and two cells were waaaay off.

Single cell charging, now that Iv'e made that easy brought them up closer. Leaving it on the charger for about 16 hours did the rest. Those two cells still drop off first, but all cells still equalize close to 3.35v after 12 hours.
 
My goal has been reached! :D
Just came back from a ice cold 30 km ride.
Pulled 12,1Ah out of my 15Ah battery, which was my goal (80%).

When measuring all the cells had a voltage of 3,22V, except the one that I shorted earlier (which was also one of the previous low cells) that had only 2,97V.
I did put the battery on charge for a couple of minutes after I shorted that cell, but perhaps a longer charge whould have made the cell more even with the others.

However, I am really pleased how this turned out in the end.
No I can start build number two for my girlfriend :mrgreen:
 
i doubt if shorting it made any difference. there is something wrong if you only get 12Ah. it should be 15 if that is what the label says.

i have asked several times if you would post up the voltages of the cells while it is charging so we could tell if the pack ever balances. i suspect that the cell you said you shorted just never gets fully charged and so it ends up shutting off the BMS for LVC.

so you can call it a dead cell or a runt cell or whatever but if it never is fully charged there is no way it can deliver the same amount of charge as a fully charged cell.
 
I agree with Dogman.

I've been using these Headways for over a year and from what I've experienced and from what I've read
here on ES, LiFePO4's need some time to break in, but it is also imperative, if possible, to tap all the
cells to watch them. If one is low after charge, then it might need more charging on that particular low cell.

SCENARIO.... 15ah pack 48v 16s1p
Just think of all the cells filled with water and one is only 3/4 full. The problem with this scenario is that the
3/4 full cell, if full, is 15 ah. At 3/4 it's only 11.25ah. And you only want to use 80% of 15ah which is 12ah and the
cell that started out at 11.25ah is being discharged over 100% while the others are discharged to 20%.

This is why SERIES packs need to be watched to keep them in good health and not to damage them :)

I do not use a BMS, but it is to my understanding that the LVC would cut in to protect the low cell. If it does
you are only getting 11ah out of 15ah. This will limit your range or use by 30%.

Want to go as far as you can with your pack? Then fill up all the cells. Watch the cells, know the cells, be one with the cells :)

Tommy L sends.... \\m//
 
You must log in or register to reply here.

Similar threads

F
5S 20A BMS boards for Makita batteries clones. Worth installing?
Replies
0
Views
1,153
Firedog
F
J
hacking into chinese bms board to get overcharge signal
Replies
7
Views
1,587
JackB
J
electroneg
New to LiFePo4 and New Member - Battery, charging and BMS Questions.
Replies
24
Views
4,883
john61ct
J
A
How does ANT BMS balance the battery pack and how long does it take?
Replies
7
Views
3,666
Archocell
A
Jim Subzero
New battery build: one parallel group behaves differently
Replies
11
Views
983
Jim Subzero
Jim Subzero
Back
Top