BMS's, how do they work please?

[T i m]

Hi All,

It turns out I signed up here a while ago (probably when we picked up a PowaCycle Windsor in kit form s/h) but think this could be my first post. ;-)

<Edit>, Just checked and it's not but they were in 2009. ;-)

So, generic and probably basic starter question (and possibly answered elsewhere but not that I could find with a quich search), 'how do BMS's work'?

I'll expand on that a bit (sort of off (cycle) topic to start with but please bear with me) ... ;-)

I have been asked to look at a battery pack for a mates electric scooter and it's a 10s / 9aH Li-Ion unit (as may be fitted on some eBikes I'm guessing)? Whilst I'm reasonably good with electrical, electronics and older rechargable batteries, I've not really played too much with these more exotic battery forms and hence why I'm a little in the dark.

If I understand it correctly (from what I've found from exploring this Li-Ion pack) is that the junctions of all the cells go back to the BMS where they are monitored, presumably during both charge and discharge? The charging side seems to work but there is no voltage on the output wires (although there is across the pack itself), and there are no fuses(?), suggesting the the output side of the BMS has failed somehow? Now, I don't think that's because the battery (/ any cells) are bad because I monitored the charging voltage across both the battery and individual cells and all looked good (I saw the charging current drop down from a fairly steady 2A as the voltage neared 42 or whatever it was).

There are 4 power FETs on the BMS that all test out ok on my electronic tester so I'm guessing the fault is further inside. Anyroadup, I've found what may be a replcement BMS on eBay and will try that soon.

Bringing this back on topic and reminded by this scooter battery issue, I dug out the PowaCycle battery, a 26V Li-Po, opened the case (to be able to keep an eye on it and the temp etc) and put it on charge. It seems to take charge ok and whilst there I noticed the little balance (is it called?) connector and I just happened to have a 7s battery monitor thingy (bought with the scooter in mind) and plugged it in to see what would hapen. Voila, it fired up and I was able to see the voltage on each cell, and the total voltage etc (more below).

Now, being as I don't have the bike ready to test the battery on, is there any load I can put on it to give it a discharge and then recharge please and would this be a good thing to do in general? I'm guessing every cycle counts against the total but is a complete cycle a 'good thing' to do eveny so often, especially if the battery has been in store for a while (indoors), or is it too late in any case?

I have a string of 3 x 21W 12V car lamps that I can use as a smallish load and I'm assuming the BMS will protect the battery from over discharge in any case?

And what sort of voltage variation should one expect between cells please? At the moment on the PowaCycle pack I've got, left overnight after a full charge - (1 - 7):

4.128, 4.103, 4.181, 4.185, 4.193, 4.173 and 4.130V (29.09 Total).

So, I can imagine the BMS would disconnect the output if any cell went below a certain voltage (protecting the weakest) but does it do the same during charge, disconnecting the entire pack if one cell reaches the maximum allowed voltage? Or is it able to do something clever re managing the charge (at least) on each cell?

Sorry for the long post and all the questions ... I'm just trying to see the bigger picture here in the hope that I don't make matters worse and damage anything. ;-(

Cheers, T i m

 

[Nealh]

You have answered your post your self, no output if its not a cracked solder joint or broken wire then a pcb component more than likely has failed. The total lipo voltage is good with an optimal 29.4 possible, cells 1,2 & 7 are a little out of balance 0.020v is a reasonable variation either try and bring each sell up to within the others or discharge the 4 higher cells to the lower value then set then for a discharge/charge and see if they balance out better.

 

[T i m]

You have answered your post your self, no output if its not a cracked solder joint or broken wire then a pcb component more than likely has failed.

Thanks for your reply ... yes, I can't see anything amiss on the old BMS so believe it's a failed component or somesuch.

https://dl.dropboxusercontent.com/u/5772409/BCM 1.jpg
https://dl.dropboxusercontent.com/u/5772409/BCM 2.jpg
https://dl.dropboxusercontent.com/u/5772409/BCM 3.jpg

The total lipo voltage is good with an optimal 29.4 possible, cells 1,2 & 7 are a little out of balance 0.020v is a reasonable variation either try and bring each sell up to within the others or discharge the 4 higher cells to the lower value then set then for a discharge/charge and see if they balance out better.

Ah, so the BMS doesn't do the balancing, just the protection?

So (and as I mentioned, I'm new to Li-xx stuff), do I just discharge the individual cells with a small load of some sort and / or is there some kit that would do this automatically please?

Similar question re charging of any low cells. Would it just be a matter of attaching say a bench PSU at a maximum voltage for each cell (/chemistry) and limiting the current to something reasonable (less than c presumably) and just watching it on the DMM?

Once done ('balanced') do we hope the cells should say in sync for a good few cycles?

Is there a good primer on all this you could point me to please?

All the best and thanks again.

T i m

 

[flecc]

Ah, so the BMS doesn't do the balancing, just the protection?

The BMS does do the balancing in the late stages of a charge but on results doesn't always do a perfect job. In fact if the charger is left plugged in when charge is complete, the charge can restart if any cells are detected dropping low.

Another problem is that the BMS is often powered from the first cell, so that cell can read low due to the small current drain.
.

 

[Deleted member 4366]

There's a photo missing. What's on the backside of the board with the FETs on? You need to show the B- and P- connections and the shunt if it has one.

I see R0 is missing and it goes to those wires in the corner that I assume are a switch. If it is a switch, it needs to be on, obviously. The switch would switch a transistor that opens a gate to provide the signal power to the output mosfets. Without R0, it appears to be isolated, but it's not clear in the photo where those tracks go.

I didn't get the bit about why you have a 10S BMS and a 7S battery. Can you explain that again. Is there a faulty 7S BMS somewhere? Are there two batteries?

 

[Nealh]

Bms on a cased lipo should manage cells well to a point however the chemistry involved is a lttle more temperamental, once you notice a balance issue you have to be a bit proactive in keeping an eye on them.
Their discharge voltage is different to lion and can go wildly out of balance below 3.5v, generally ebike lipo users will only discharge to
3.5v -3.7v min and charge to 4.1 - 41.5v max, with lipo at 3.5/3.6v their is very little ah to use below this to be of real use. Lion is ultra safe with a BMS with lipo you need to know the basics and be more careful with them, lipo when not in use don't like to be left fully charged to long due to the chemistry they are happiest at sitting at 37 - 38.5v and then charged prior to use. Like most lipo users I make up larger battery packs from smaller
packs to get the battery size I require and thus has no BMS protection but relies on the user to closely monitor all parimeters of charge, discharge and storing. For fast/ bulk charging I use PSU's and monitor them closely
with a volt meter for discharge and balancing I use BC168 lipo chargers. C rate is important generally not more than 1c., good cells should remain fairly well balanced and more so with a BMS some will balance once a month or maybe every 10 cycles or so really depends on how they are used and treated.

Have read of this tutorial on Endless Sphere as it explains a lot about how to use them but this is for the more advanced users who don't use BMS.
https://endless-sphere.com/forums/viewtopic.php?f=14&t=52240&sid=e5367be435100d3821526801e7f7c4f4

 

[T i m]

The BMS does do the balancing in the late stages of a charge but on results doesn't always do a perfect job. In fact if the charger is left plugged in when charge is complete, the charge can restart if any cells are detected dropping low.

Oh, ok, thanks.

Another problem is that the BMS is often powered from the first cell, so that cell can read low due to the small current drain.
.

Ah, I though it would have been powered by the whole battery but what you say is equally feasable (with loads of electonics running from 3.3v these days etc).

Chers, T i m

 

[T i m]

There's a photo missing. What's on the backside of the board with the FETs on?

Nothing much but in case there is something worth seeing. ;-)

https://dl.dropboxusercontent.com/u/5772409/BCM 6.jpg

You need to show the B- and P- connections and the shunt if it has one.

I think the B- is on the top right of the FET board (J2?) P- is on the top left marked -d .. which is confirmed as P- (upside down) when you look at the back. ;-) I don't think there are any other external components (other than the throtle etc) but they would still be on the scooter if they were.

I see R0 is missing and it goes to those wires in the corner that I assume are a switch. If it is a switch, it needs to be on, obviously. The switch would switch a transistor that opens a gate to provide the signal power to the output mosfets. Without R0, it appears to be isolated, but it's not clear in the photo where those tracks go.

Sorry, I shouldn't have cropped that picture so tightly, that goes to a thermal fuse (that measure 0 ohms) and I think only protects the charging from overtemp? Is one side of the daughter board for charge monitoring and the other for discharge?

I didn't get the bit about why you have a 10S BMS and a 7S battery. Can you explain that again. Is there a faulty 7S BMS somewhere? Are there two batteries?

Sorry, it was just all buzzing around in my head and so I sorta mixed the two things together.

Yes. There is 10s Li-Ion battery out of the scooter (no output) and a 7s Li-Po out of the PowaCycle that was potentially new when the PO bought it to build the PC out of leftover and new parts before it was bought by me in 2009, but never used since (too many cycles / hobbies, not enough time). I presume it's ok but I've never used it outside a short test in 2009.

Sorry for any confusion.

Cheers, T i m

 

[T i m]

Bms on a cased lipo should manage cells well to a point however the chemistry involved is a lttle more temperamental, once you notice a balance issue you have to be a bit proactive in keeping an eye on them.

Ok.

Their discharge voltage is different to lion and can go wildly out of balance below 3.5v, generally ebike lipo users will only discharge to
3.5v -3.7v min and charge to 4.1 - 41.5v max,

So assuming this is observed on the move, are they generally just monitoring the terminal voltage of the battery, rather than the individual cells?

with lipo at 3.5/3.6v their is very little ah to use below this to be of real use.

Ok.

Lion is ultra safe with a BMS with lipo you need to know the basics and be more careful with them, lipo when not in use don't like to be left fully charged to long due to the chemistry they are happiest at sitting at 37 - 38.5v and then charged prior to use.

Hmm, so as feared I may not have done them any good leaving them unattended and unused for a good few months at a time (between top-up charges). ;-(

Like most lipo users I make up larger battery packs from smaller
packs to get the battery size I require and thus has no BMS protection

OOI, Is there anything (technical I mean) stopping you adding a BMS?

but relies on the user to closely monitor all parimeters of charge, discharge and storing.

From bitter experience I know how easily one can be distracted with such things and hence why I don't mind investing in 'automatic' kit if it's not too expensive. ;-)

For fast/ bulk charging I use PSU's and monitor them closely
with a volt meter for discharge and balancing I use BC168 lipo chargers. C rate is important generally not more than 1c., good cells should remain fairly well balanced and more so with a BMS some will balance once a month or maybe every 10 cycles or so really depends on how they are used and treated.

Loads more good info. So, the multi connector is called the 'balance connector' and is used for doing said charge balancing and / or as a connection to a BMS etc?

Have read of this tutorial on Endless Sphere as it explains a lot about how to use them but this is for the more advanced users who don't use BMS.
https://endless-sphere.com/forums/viewtopic.php?f=14&t=52240&sid=e5367be435100d3821526801e7f7c4f4

Thanks for that, I'll check it out properly tonight but from a quick scan it looks just what I was looking for (as a more generic backgrounder etc).

I love a good graph as with me 'a picture often does speak 1000 words' and also 'you can manage what you can measure'. ;-)

Cheers, T i m

 

[Deleted member 4366]

I must be thick. I still don't get what problem you're trying to solve. I thought you said something about the 10S BMS not giving power, but there's no wires on P- (power to bike) and B- (power from cell-pack).

Where do lipos suddenly come into it? No ready-made ebikes would use lipos.

 

[T i m]

I must be thick. I still don't get what problem you're trying to solve. I thought you said something about the 10S BMS not giving power, but there's no wires on P- (power to bike) and B- (power from cell-pack).

Where do lipos suddenly come into it? No ready-made ebikes would use lipos.

Ok, I'll try to clear things up. ;-)

1) A mate has asked me to look into why his scooter isn't working. It has a 10s Li-Ion pack.

2) Whilst helping him with that I've dug out the 7s Li-Po pack that goes with our PowaCycle and been testing / comparing the two systems.

3) Because of both batteries being different to what I have experiened previously, I was looking to get a general overview about Li-?? cells in general and (hopefully) confirmation that what little I though I knew about all of it was correct.

So, I am currently juggling two balls in my head in an effort to try to understand the 'bigger picture' with these Li-?? batteries as they (Li-Po/Ion) seem to have more in common than they do with the Ni-?? or Lead Acid batteries I've been playing with for many years (my 4 seater EV ... and the electric 'motorbike' I designed, built and raced are both Lead Acid powered and loads of RC cars and boats).

https://dl.dropboxusercontent.com/u/5772409/EV1.jpg

So, the PowaCycle Li-Po battery has been in my possesion since 2009, charged now and again but not used and all I know about it are the cell voltages I posted previously. I believe it works but I have no idea how well.

The Li-Ion pack from the scooter seems to accept a charge (through the BMS) but doesn't seem offer any output (also via the BMS, fine when loaded via the balance connector) and so we removed the battery from the scooter and the BMS from the pack in the effort of faultfinding. I have also lifted the legs of the fets off the board and tested them on a component tester and they all appear to be ok.

eg: https://dl.dropboxusercontent.com/u/5772409/FET 1.jpg

We have also bought a new BMS from eBay that is supposed to be for a 10s Li-Ion system and rated at 45A which may be appropriate for the scooter. I've not tested it yet because I wanted to qualify some of my basic understanding of how all this is supposed to work before hooking it up.

And that's about where we are now. ;-)

So, my questions were (and some have already been answered):

Could someone point me to somewhere that may help me understand some of this from a real-world eBike / scooter POV (as it may differ from say RC flight).

Are the voltages I'm seeing on my unused / stored for a long time Li-Po pack indicitave of a good or bad battery.

If the differences in off-load voltages on the Li-Po cells are 'off', how would one typically resolve this?

Is there a piece of kit (that would be affordable to me and recommended by the panel) that I could use for 'balancing' these Li-?? cells in combinations up to 10s (so far). I have a couple of smal quadcopters but they only use single cells and come with their own basic (USB) chargers so might like to charge those more intellegently.

Without said kit, is there a safe / recommended way to charge / discharge / balance the pack(s) to check their functionality / capacity etc.

The main question was really 'how do these BMS work', what do they do and to (hopefully) confirm my basic understanding and expand on that to help me both understand the 'bigger picture' around the whole Li-?? battery scene and therefore help me to understand why the scooter stopped working in the first place (and either repair the old BMS if possible and / or confirm the potential replacement is appropriate).

Cheers, T i m

 

[Deleted member 4366]

It would have been better to just deal with one problem at a time. Tell me if I got any of this wrong.

Lets start with problem 1: 10S battery doesn't give any output, but charges OK.

When a battery doesn't give output, the first thing you do is measure the voltage at the charge socket after charging, or, if you want to open the battery, measure across the cell pack. It should be 41.5v - 42.0v

Assuming that the result is lower than 41.5v, you must measure each cell-group voltage, which you can do on the multi-pin connector. As yours has 11 pins, the individual cell voltages are between each adjacent pair. The BMS is sensing those voltages. If any are below 3v, it won't give output.

Do the above measurements and then report back.

Problem 2: You have a 7S 24v battery, but you don't know if it's any good. From the measurements that you've shown, it looks pretty healthy as long as it's not marked anywhere as a Phylion battery and has 7 aluminium block cells inside, in which case the answer is that it's probably bad.

If you have any specific questions about how a BMS works, just ask.

The lipo batteries used for RC work exactly the same as ebike batteries, except that R/C guys don't use a BMS. Instead, they use an intelligent charger. Those batteries used to be good for ebikes where you wanted a lot of power, but now modern ebike batteries can also give very high currents and are a lot safer. There's nothing to stop you fitting a BMS to R/C lipos, as their cell voltages are the same. You could then charge then with an ebike charger, which would be a lot more convenient. All lithium cells will catch fire if overcharged or if over-discharged and then charged, but r/c lipos are the most voltile, so, if you use a BMS, you need to test that it's working properly by monitoring the cell voltages until you have confidence in it.

Whatever lithium battery you have, the cell voltages must be kept in the range 3.0v - 4.2v, with two exceptions: Some modern 18640 cells can now go to 4.3v, and for LiFePO4 cells, the range is 2.5v to 3.65v

 

[T i m]

It would have been better to just deal with one problem at a time.

Well, yes possibly, but there was only one real problem (the Li-Ion pack with no output) and the rest was me asking about the Li-?? cells and their relationship with their BMS in general.

Tell me if I got any of this wrong.

Ok. ;-)

Lets start with problem 1: 10S battery doesn't give any output, but charges OK.

I believe so (charging, to my best understanding) yes.

When a battery doesn't give output, the first thing you do is measure the voltage at the charge socket after charging, or, if you want to open the battery, measure across the cell pack. It should be 41.5v - 42.0v

Ok, well because the BMS is now removed I don't have an easy way of measuring the post charge voltages (but can rig it back up if very important for true faultfinding etc) but the terminal voltage from the charge that was conducted about two weeks ago is now 39.21.

Assuming that the result is lower than 41.5v, you must measure each cell-group voltage, which you can do on the multi-pin connector. As yours has 11 pins, the individual cell voltages are between each adjacent pair. The BMS is sensing those voltages. If any are below 3v, it won't give output.

Do the above measurements and then report back.

Ok, well if the 'charged two weeks ago values' (when I observed the on charge terminal voltage just under 42V) are:

01 = 3.960
02 = 3.943
03 = 3.787
04 = 3.971
05 = 3.946
06 = 3.730
07 = 3.983
08 = 3.995
09 = 3.961
10 = 4.013

T = 39.21

Problem 2: You have a 7S 24v battery, but you don't know if it's any good. From the measurements that you've shown, it looks pretty healthy as long as it's not marked anywhere as a Phylion battery and has 7 aluminium block cells inside, in which case the answer is that it's probably bad.

Ok thanks. I'm not sure if this is any help?

https://dl.dropboxusercontent.com/u/5772409/Li-Po 1.jpg
https://dl.dropboxusercontent.com/u/5772409/Li-Po 2.jpg

If you have any specific questions about how a BMS works, just ask.

Well, I don't know how specific they are but one question I wanted to ask was:

'If the BMS manages the charging procees and if it can do so cell by cell, how does it do so please?'

The lipo batteries used for RC work exactly the same as ebike batteries, except that R/C guys don't use a BMS. Instead, they use an intelligent charger.

Ok. So are they the ones with both 'power' and the 'balance' connections?

Those batteries used to be good for ebikes where you wanted a lot of power, but now modern ebike batteries can also give very high currents and are a lot safer.

Understood.

There's nothing to stop you fitting a BMS to R/C lipos, as their cell voltages are the same. You could then charge then with an ebike charger, which would be a lot more convenient. All lithium cells will catch fire if overcharged or if over-discharged and then charged, but r/c lipos are the most voltile, so, if you use a BMS, you need to test that it's working properly by monitoring the cell voltages until you have confidence in it.

Thanks for that. That was the sort of 'general' information I am (also) interested in.

Whatever lithium battery you have,

I believe the. Li-Ions are these (they have the same number on the cell outer but they are grey not yellow):

http://www.ax-ps.com/pdf/datenblatt/LR1865_SI.pdf
https://dl.dropboxusercontent.com/u/5772409/Li-Ion 1.jpg

the cell voltages must be kept in the range 3.0v - 4.2v, with two exceptions: Some modern 18640 cells can now go to 4.3v,

Ok, thanks.

and for LiFePO4 cells, the range is 2.5v to 3.65v

Again, thanks. I'm going to put together a handy reference doc of all these goodies and the links I've been offered etc.

I might also get myself a 10s charger if you can recommend one please (thet would deal with these ~10A packs).

Cheers, T i m

 

[Deleted member 4366]

There's 10 channels on your BMS. Each one monitors the voltage on each cell-group. How they do it is complicated, but the end result is that one of two transistors are switched that interrupt the gate voltage to the mosfets. There's one charge mosfet and three discharge ones on your BMS, The three discharge ones are in parallel and share the gate line. The charge mosfet has its own. When any cell goes above about 4.25v, the charge mosfet is switched off by interrupting the gate voltage. When any cell goes below 3.0v, the discharge mosfets are switched off. Additionally, if any cell goes below 2.5v, the charging will be switched off too.

There's also normally a shunt somewhere. The voltage drop down it is proportional to current. If that voltage drop is too high, the discharge mosfets will also be switched off.

You also have a thermistor that is connected to a transistor that will switch off all mosfets if the temperature is too high. I can't see a shunt on your BMS, which is unusual, but not unheard of.

Modern BMSs use microprocessors to figure out what's going on. They have many more functions like totally closing down when you've recharged a certain amount of times.

Another function of the BMS is balancing. When any cell goes over about 4.15v, a transistor on that channel opens up a line through the bleed resistor (those 330 Ohm ones) to drain it down to 4.15v so that any lower cells will catch up. The draining is pretty slow, so if the cells are way out of balance, it can't balance them in one charge.

Those cyclindrical cells are described by their size. They're all 65 mm long and most are 18mm in diameter, so 18650 cells. 22mm ones will be 22650 cells. There's hundreds of different types. It's never clear what's in them because the technology is changing so fast. Each manufacturer adds a different ingredient X to them to make them last longer, cope with a higher temperate range, allow them to be cycled deeper, give them more capacity, etc. As far as we're concerned, at the moment there's three types of lithium battery. R/C lipos, Li-ion batteries and LiFePO4 ones. they're all lithium-ion batteries, but that's how they're differentiated.

R/C lipos have the balance leads to a multi-pin connector so that you can charge them with a balance charger.

The cell voltages on your 10S battery show two a bit lower than is healthy, but they shouldn't stop the BMS from giving discharge. It's possible that there's some faulty cells in the group, so that the voltage collapses as soon as you put a load on it. To test that, you'd have to measure again when you connect your three light bulbs between the positive and negative power wires from the cell-pack, or you could have a look at the tops to see if any have leaked.

 

[T i m]

There's 10 channels on your BMS. Each one monitors the voltage on each cell-group. How they do it is complicated, but the end result is that one of two transistors are switched that interrupt the gate voltage to the mosfets.

Understood.

There's one charge mosfet and three discharge ones on your BMS, The three discharge ones are in parallel and share the gate line. The charge mosfet has its own.

Ah, that makes sense re why there are 3 of one type and one of another.

When any cell goes above about 4.25v, the charge mosfet is switched off by interrupting the gate voltage.

So that affects the whole pack so when charged the way most eBikes and scooters are charged, via a smart charger with a single connection to the battery (well, + & - etc, not the balance wires) the sharge to the wole pack is stopped (completely?).

When any cell goes below 3.0v, the discharge mosfets are switched off.

So presumably this would also impact the pack when the voltage is even temporarily depressed due to a momentarily high load? Is the output automatically re-enabled if the voltages recovers or do you (typically possibly) need to reset the supply line etc?

Additionally, if any cell goes below 2.5v, the charging will be switched off too.

Because the chemistry doesn't recover well when charged from such a low voltage?

There's also normally a shunt somewhere. The voltage drop down it is proportional to current. If that voltage drop is too high, the discharge mosfets will also be switched off.

Understood. When I designed me electric motorcycle I used the longest battery lead as a shunt (to save introducing any further resistance in the powertrain) and measured the voltage drop across that to display the current. I could use the same display (via some togle switches) to display the voltage on battery 1, battey 2 or the total voltage, and, display speed as resolved from a magnetic sensor, detecting the teeth on an output gear in the gearbox and going though an F to V converter and then to my 'display' (which was just a voltmeter). Much to my surprise it won me an award from the IEEE. <blush> ;-)

You also have a thermistor that is connected to a transistor that will switch off all mosfets if the temperature is too high.

Would this be in addition to the thermal fuse?

I can't see a shunt on your BMS, which is unusual, but not unheard of.

I'm pretty syre there were no other connections to the battery or BMS or devices on the scooter outside the speed control etc but I may be wrong.

Modern BMSs use microprocessors to figure out what's going on. They have many more functions like totally closing down when you've recharged a certain amount of times.

Oh, so would that be 'you have reached 500 cycles on this pack so that's your lot' (like you get on some PC printer carts?).

Another function of the BMS is balancing. When any cell goes over about 4.15v, a transistor on that channel opens up a line through the bleed resistor (those 330 Ohm ones) to drain it down to 4.15v so that any lower cells will catch up. The draining is pretty slow, so if the cells are way out of balance, it can't balance them in one charge.

Ah, that was the bit I thought existed but couldn't see how would be possible over the thin balance wires and the size of the components on the BMS. On the one that came with the scooter there was a heatsink on the back of the daughter board with some thermal foam that went across the run of resistors. They must be whe resisters you are referring to.

Those cyclindrical cells are described by their size. They're all 65 mm long and most are 18mm in diameter, so 18650 cells.

Ah, no so reference to the potential capacity for any given size?

22mm ones will be 22650 cells. There's hundreds of different types. It's never clear what's in them because the technology is changing so fast. Each manufacturer adds a different ingredient X to them to make them last longer, cope with a higher temperate range, allow them to be cycled deeper, give them more capacity, etc.

Understood.

As far as we're concerned, at the moment there's three types of lithium battery. R/C lipos, Li-ion batteries and LiFePO4 ones. they're all lithium-ion batteries, but that's how they're differentiated.

Understood.

R/C lipos have the balance leads to a multi-pin connector so that you can charge them with a balance charger.

So do you generally apply the charge via the main power leads but monitor / manage the cells via the balance leads?

The cell voltages on your 10S battery show two a bit lower than is healthy, but they shouldn't stop the BMS from giving discharge. It's possible that there's some faulty cells in the group, so that the voltage collapses as soon as you put a load on it. To test that, you'd have to measure again when you connect your three light bulbs between the positive and negative power wires from the cell-pack, or you could have a look at the tops to see if any have leaked.

Understood. I'll try to do that later today.

Thanks very much for all the info. It's all going in, allbeit slowly. ;-)

Cheers, T i m

 

[Deleted member 4366]

1.So that affects the whole pack so when charged the way most eBikes and scooters are charged, via a smart charger with a single connection to the battery (well, + & - etc, not the balance wires) the sharge to the wole pack is stopped (completely?).

2. So presumably this would also impact the pack when the voltage is even temporarily depressed due to a momentarily high load? Is the output automatically re-enabled if the voltages recovers or do you (typically possibly) need to reset the supply line etc?

3. Because the chemistry doesn't recover well when charged from such a low voltage?

4 Would this be in addition to the thermal fuse?

5.I'm pretty syre there were no other connections to the battery or BMS or devices on the scooter outside the speed control etc but I may be wrong.
6. Ah, no so reference to the potential capacity for any given size?
7.So do you generally apply the charge via the main power leads but monitor / manage the cells via the balance leads?

1. The charger is set to 42.0v, so is normally the first to switch off.
2. Sometimes yes. Sometimes no. Some latch when they hit low voltage. For the microprocessor ones, it depends on the programming. Also the controller's LVC is set a bit higher, so is normally first to switch off.
3. Yes. Lipos will swell up and maybe catch fire.
4. Every BMS is different. Thermistors or thermal fuses - whatever the designer chooses.
5. The shunt is normally on the BMS between B- and the discharge mosfets.
6. You have to look up the data sheet for the capacity, There should be a designation on the cells, like XXX18650XX.
7. Depends on your charger. Some charge directly through the balance leads, so they have say 6 seperate charging chanels, each individually controlled, which means that the cells are charged seperately. Other chargers charge through the main leads and bleed the cells for balancing, which can take a long time if out of balance.

 

[anotherkiwi]

Does a 10S Lipo BMS which has a LVC at 3.5-3.6 V per cell exist? All the ones I have seen have LVC set to 2.5 V per cell which makes such a BMS pretty useless.

For the moment I ride with the voltage displayed on the LCD and Lipo alarms on the packs, I am just thinking ahead.

 

[Deleted member 4366]

The smart BMS from BMSBattery does. There's a thread on ES somewhere about how to hack it and program it.

 

[T i m]

Does a 10S Lipo BMS which has a LVC at 3.5-3.6 V per cell exist? All the ones I have seen have LVC set to 2.5 V per cell which makes such a BMS pretty useless.

Jumping in here (I've not gone off this battery issue, just been distracted for a few days) but this is the spec for the BMS that I've bought in replacement for the faulty one on this 10S Li_Ion pack:

"Applicable: 36V 10 cells lithium battery protection board
Batteries Type: lithium cobalt oxide / manganese lithium / ternary materials
Single overcharge protection voltage 4.25V ± 0.05V (4.20-4.35V/0.05V per upgrade)
Single overcharge recovery voltage: 4.10-4.00V
Single over-discharge protection voltage: 2.50V ± 0.1V (2.50-3.0V/0.05V per upgrade)
Single over-discharge recovery voltage: 2.80V ± 0.1V
Protection Current consumption: =300UA
Short circuit protection current: 40A ± 3A
Short circuit protection time: 500MS
Temperature protection: 55/65/75 degrees
Discharge current: 30A
The maximum instantaneous current: 40A
Single balanced voltage: 4.19V ± 0.02V
Single balanced current: = 55MA
Charge current: =10A"

Cheers, T i m

 

[T i m]

The smart BMS from BMSBattery does. There's a thread on ES somewhere about how to hack it and program it.

I've now got my charger and have been doing some tests and balancing on this 10s Li-Ion pack and it looks like it's ok.

So, I'm about to conect it up to the eBay BMS but like 'anotherkiwi' I'm not sure if the low voltage protection is 'good' for these cells?

However, might it matter if this low voltage cutoff is set such that it compensates for any voltage sag when under load (I'm not sure if that 2.5V or '2.5 to 3V' / cell, so maybe ok if nearer 3V?)?

I'd like to also add how impressed I am with this 'iCharger 1010B+' charger. I have tried it on all sorts of chemistrys and it seems to do what it says on the tin. It has also helped me understand how these Li-** batteries work a bit better (no connection to the company etc).

I'm also getting to grips with LogView and the fact that some of it is still in German doesn't help that. ;-)

Cheers, T i m