My DIY 48V 16Ah battery cuts out at 47 Volts

[spaceship]

As the title says, I built a battery pack 13s5p using LG MH1 cells salvaged from scooter batteries. I tested each cell on a Opus battery charger (on the 1A discharge setting) and they were all showing around the 3200mAh they are rated at, so it should be a 48V 16Ah battery. I built it to run a smaller bike with a 500watt geared front hub for a friend to ride and it worked well, although we never pushed it to its limit. I recently swapped my 1500w rear hub to a full suspension bike where there is no room to mount the big 52v20Ah battery I was using before, so I decided to give this battery a try.

What I am finding though is that I am getting far less range than with my bigger battery, like less than half the range, when this battery has 3/4 the Wh. (1040wh with the 52V triangle battery vs 768 Wh with this battery). The controller starts cutting out when the battery is still showing a resting voltage of 47v, according to this chart I found there should still be some capacity left.



From what I understand it is voltage sag when load is applied that is causing the battery voltage to momentarily drop below my controllers low voltage cut-off, and so it shuts down. The battery is built using a 35A BMS. The MH1 cells are rated to 10A max drain, so at 35A that would be no more than 7A per cell. I tested all the salvaged cells and they all showed 95-100% their rated capacity using 1A discharge. My question is, is it just a limitation of the cells or could it be something in how I built that battery that can be improved upon (bms, nickel strip, spot welds, etc). From what I've heard these are quality cells and suitable for ebike/scooter batteries, and I dont think I am trying to pull an unreasonable amount of power from them. So is there anything I can change to get the most out of this battery?

 

[jonyjoe303]

When you built the pack did you check the internal resistance of all the cells before you put them together. If you got some High IR cells in there that will cause problems.
When I first started building packs I never checked the IR of the cells and my packs usually had problems which I could never figured out. Once I bought a IR tester, I could see first hand how even a cell that checked good on capacity tests failed the IR test. Just 1 high IR cell will cause problems. I also use the opus but it has a terrible IR checker. I paid about 50 dollars for the tester, definitely worth the price when working with recycle cells.

You can also try a different BMS, or verified that the motor is actually pulling 35 amps and not something higher.

yr1030 internal resistance tester

 

[amberwolf]

If your BMS can show you individual cell voltages while you ride (via BT to a phone app, for instance), you can see which cells are dropping by how much under how much load.

Cells are affected by more than just capacity, there is also internal resistance, which causes the voltage sag under load, and voltage rise during charging--the higher current is in either case, the more of a problem it causes.

Capacity and voltage sag are affected by load, too, so you should test the cells at the load current you expect them to have to supply, if you really want to know how they will perform in your application. Testing them at a load 1/10th that of expected reality will not give you as useful a result.

Cell ratings don't necessarily reflect their real in-use capabilities, in that Cell Model A might have a 10A rating but sag by a whole volt at that current, while Cell Model B might have the same rating but only sag by half a volt, etc. Or Cell Model C might have only a 5A rating but sag by a whole volt at that. D might have a 5A rating but only sag by a tenth of a volt. So C is really less capable than any of the, with A a bit better, B better than that, and D the best, even though it was only rated for 5A. In a real-world higher current application, D would perform better than any of the others, though based on it's rating should be used in groups of twice as many parallel cells as A or B.

Used cells will have lower capacity and higher resistance than new (so sag more in voltage, and sooner, than the same cells new), and different cell models also vary from each other and age differently from each other.

Lgyte.info.dk has a bunch of cell testing graphs for different models at different current rates, and Pajda (and others?) here on ES have cycle life testing threads on some cells; these can show you directly how the above works.

 

[Eastwood]

definitely worth the price when working with recycle cells.

Does the internal resistance need to be checked with new cells?

 

[amberwolf]

Since it's unlikely that the sellers of the cells you buy will do any cell testing and matching (much less precise matching), it's a good idea to do so, if you really want a pack that will perform as perfectly as possible for as long a lifespan as possible.

Note that if the cells are not tested / matched at all by the seller, then to get enough cells to build a well-matched (or perfectly matched) pack from, you might end up buying way more cells than you need for one pack (like several packs' worth or more). If they sell by batch / box from the manufacturer, they may at least be close, but given results others have posted over the years here for some of those I wouldn't make bets.

 

[spaceship]

If your BMS can show you individual cell voltages while you ride (via BT to a phone app, for instance), you can see which cells are dropping by how much under how much load.

no, I wanted to order a BT one but it was a lot bigger wouldn't fit in the case that I got with all the cells i wanted to fit. It was hard enough getting it closed up with the bms that came with the case.

When you built the pack did you check the internal resistance of all the cells before you put them together. If you got some High IR cells in there that will cause problems.

it was almost a year ago that I built it, but I am pretty sure that I used the opus to capacity check and then check IR on all the cells for matching and made sure that they all had less than 100 milliohms

Capacity and voltage sag are affected by load, too, so you should test the cells at the load current you expect them to have to supply, if you really want to know how they will perform in your application. Testing them at a load 1/10th that of expected reality will not give you as useful a result.

I spent some more time looking at the datasheet, and looking at this chart I think that I'm starting to understand more what the cells are actually capable of. This chart shows various discharge rates, and 7A is the blue line near the bottom and there is about a 0.4V drop. When it starts cutting out at 47V resting thats roughly 3.6V per cell which should be close on this chart to the red line, a very low current draw, and if I trace that to the 7A line thats about 3.15V or just over 40V for the pack.



And this chart shows a sharp increase in discharge rate between 3.1V and 3.2V.



SO I am thinking now that the battery is fine and is pretty much the way the datasheet describes these cells. My expectation of the battery capacity was a little higher than what it actually is, but other than that it performs fine. Definitely some more to think about before selecting cells next time.

 

[docw009]

I have bought several used scooter packs using the MH1 cells. and have gotten consistent results either using them as 36V batteries, or taking out the cells and putting them into a Hailong case like the OP. These are nominally 3200 mah when new tested at 3000-3100, , and mine yield about 2600-2700 mah in an ebike application,

I use them with 20A controllers. pulling 5A per cell in 4P batteries. I think they're about 4V of voltage sag at 20A. I can usually run a battery down to 3.3V-3.4/cell and not have it shut off. Their IR, measured with the above YR1035+ runs around 30-32 milliohms, so they're old. Would probably be under 16 milliohms new.

The OP could have stale cells with high IR while still showing good capacity in a tester. I've got some cells 6-7 years old, that were 2400 mah when new, and are still at 2000-2200 mah, but the IR is 40-44 miilliohms. NOt wanting to throw them away, I built two 10S-5P's. The total IR of the battery is around 90 milliohns, but I get 6 volts of sag at 20A. I can only use these to about 50% capacity, which is 36V. Then pulling 20a will trigger LVC. Should have thrown them out. The battery would be still useful for ebiking, if I didn;t already have better units. So they sit around.

I would suggest to the OP, just as a sanity check, to run a poll of all 13 cell groups just to see that they are close. If one group is markedly lower in voltage, try to probe the individual cells. Yes, they are in parallel, but sometimes on our DIY packs, a weld comes loose and they're not in parallel.

 

[jonyjoe303]

The problem with the opus when checking IR is that if you check the same cell 5 times you get a different result everytime, plus they also always read higher than they actually are.
Also for a high discharge cell you should be aiming at under 40 milliohms (those would be high quality panasonic cells), but true high discharge cells would be in the 20 milliohms are less (sony cells). Anything in the 60 milliohm or higher are usually laptop low discharge cells.
With the yr1030 IR tester, if I check a cell 5 different times, I get the same result everytime. Since I got the yr1030, I check every cell new or old for IR, its a quick check 5 to 10 seconds per cell. If you have a batch of cells (same manufacture) the majority will be in a certain range but once in a while you will find 1 or 2 that are higher than the other. You want all the cells in a pack to be within the same range.

 

[amberwolf]

it was almost a year ago that I built it, but I am pretty sure that I used the opus to capacity check and then check IR on all the cells for matching and made sure that they all had less than 100 milliohms

An entire pack of cells in series might ought to be that kind of resistance...but any single cell shoudln't be anywhere near that high. I would guess the test equipment is not measuring them correctly, but if they are that high, that seems pretty bad.

FWIW, if you really want a pack to perform well, don't just measure that capacities are more than some amount, and resistances are less than some amount, and use all of those cells, but use only cells that are *the same* capacity and *the same* resistance. The narrower the range you allow to be built into a pack, the more uniformly the cells will perform, and the better the pack will stay in balance and do the job you need it to.

(if you can't get enough cells within a close enough matching range, you should at least try to match groups to each other, meaning build groups out of disparate cells in a way that creates groups that are as close as possible to each other in all properties.

I spent some more time looking at the datasheet, and looking at this chart I think that I'm starting to understand more what the cells are actually capable of. This chart shows various discharge rates, and 7A is the blue line near the bottom and there is about a 0.4V drop. When it starts cutting out at 47V resting thats roughly 3.6V per cell which should be close on this chart to the red line, a very low current draw, and if I trace that to the 7A line thats about 3.15V or just over 40V for the pack.

And remember--that is for *new* cells. Used ones will be worse by some amount, and typically less consistent.

 

[Wattman]

OP, want to preface and start with, there are some real smart battery people on this site. A lot of great advice provided in response. Well done on building your battery and of course you stumbled upon how sensitive weaker batteries of questionable balance and capacity are susceptible to hungry controllers and motors.

But, I didn't see an elementary step but perhaps I missed it in my reading.
It is 'possible' your battery cells are out of balance in terms of voltage. Yes, this is an elementary step but perhaps lost in your consideration. Telltale is when you charge your battery. Is it charging all the way to 54.6v? Or, do you never charge all the way to this voltage in an effort to preserve your cell life? If you don't charge your battery once in a while to full strength, then possible your BMS is not balance charging your parallel cell groups.

You mentioned you didn't have BT BMS and btw I feel your pain having built Hailong/Reention batteries with limited internal case volume.
So without knowing the voltages of your parallel groups within your battery, in spite of some capacity and IR differences which is pretty irrevocable with salvage cells, you may be able to achieve a lower voltage cut off with your battery by balancing your cells with a balance charger set to 1S and discharge/charge all parallel groups for voltage equivalency.

So first I would suggest, measure all 13 parallel groups and post these voltage values if you care to or not.

Second step, if you own a balance charger is...a great investment if you are a battery geek like many here...you can also do a collective capacity check of your parallel groups. Btw, I am doing this right now on a 52V 20 amp-hr battery built with 21700 cells due to a relatively high voltage cutout like you are experiencing. My cells are well out of balance.

You do this by discharging each of 13 x 5p groups down to 3.0 V baseline say at 1-2 amp discharge depending on balance charger. Then charge the parallel group to 4.0 V and record the capacity in mAh....displayed on the balance charger in the right lower corner, time just to the left of this value. Most balance chargers or many have the same display.
You don't have to charge all the way to 4.2v. What matters is 'relative' capacity, parallel group to parallel group. If you have a low side outlier for capacity in one or more of the parallel groups, time to get out your snips and pull the nickel strips off those cells and replace. Once apart you can much easier evaluate each cell of course and a single bad apple can spoil the barrel which you may be able to ID when removed from the nickel strips.

This is a great thread and input from members here is terrific. Good luck OP.

 

[spaceship]

OP, want to preface and start with, there are some real smart battery people on this site. A lot of great advice provided in response. Well done on building your battery and of course you stumbled upon how sensitive weaker batteries of questionable balance and capacity are susceptible to hungry controllers and motors.

But, I didn't see an elementary step but perhaps I missed it in my reading.
It is 'possible' your battery cells are out of balance in terms of voltage. Yes, this is an elementary step but perhaps lost in your consideration. Telltale is when you charge your battery. Is it charging all the way to 54.6v? Or, do you never charge all the way to this voltage in an effort to preserve your cell life? If you don't charge your battery once in a while to full strength, then possible your BMS is not balance charging your parallel cell groups.

You mentioned you didn't have BT BMS and btw I feel your pain having built Hailong/Reention batteries with limited internal case volume.
So without knowing the voltages of your parallel groups within your battery, in spite of some capacity and IR differences which is pretty irrevocable with salvage cells, you may be able to achieve a lower voltage cut off with your battery by balancing your cells with a balance charger set to 1S and discharge/charge all parallel groups for voltage equivalency.

So first I would suggest, measure all 13 parallel groups and post these voltage values if you care to or not.

Second step, if you own a balance charger is...a great investment if you are a battery geek like many here...you can also do a collective capacity check of your parallel groups. Btw, I am doing this right now on a 52V 20 amp-hr battery built with 21700 cells due to a relatively high voltage cutout like you are experiencing. My cells are well out of balance.

You do this by discharging each of 13 x 5p groups down to 3.0 V baseline say at 1-2 amp discharge depending on balance charger. Then charge the parallel group to 4.0 V and record the capacity in mAh....displayed on the balance charger in the right lower corner, time just to the left of this value. Most balance chargers or many have the same display.
You don't have to charge all the way to 4.2v. What matters is 'relative' capacity, parallel group to parallel group. If you have a low side outlier for capacity in one or more of the parallel groups, time to get out your snips and pull the nickel strips off those cells and replace. Once apart you can much easier evaluate each cell of course and a single bad apple can spoil the barrel which you may be able to ID when removed from the nickel strips.

This is a great thread and input from members here is terrific. Good luck OP.

I do charge the battery fully, although it hasn't been used a lot or fully cycled many times since it was built. I will open it up and check how the cell groups are balanced later this week.

lots of great info on this thread. Building this battery was meant to be a lunching point and a learning experience, and it continues to be one. I'll post what I find when I open it up.

Thanks

 

[Wattman]

I do charge the battery fully, although it hasn't been used a lot or fully cycled many times since it was built. I will open it up and check how the cell groups are balanced later this week.

lots of great info on this thread. Building this battery was meant to be a lunching point and a learning experience, and it continues to be one. I'll post what I find when I open it up.

Thanks

We all learn about this stuff together and from one another. Yes, the telltale will to first measure the voltages of each parallel group. You can do this either through the BMS serial connector or my preference which is on the sides of the pack with multimeter with probes adjacent to each other.
If you care to share these values that would be great.
Next step is, to balance the battery. I am in fact going through this exercise as mentioned right now on one of my batteries.
After you balance charge your battery, you can do with a balance charger set to 1S and individually charge/discharge the cell groups to say 3.6 volts, then you can plug in your big charger and try to charge to capacity and see if the battery balances with the BMS. Most BMS's balance charge near peak voltage for the pack.

Then ride the bike a few times and see if the battery will still fully charge aka the capacity difference between the parallel groups isn't too great. If the battery falls out of balance quickly you know you have a broad disparity in capacity between one or more of the parallel groups.

Then, if you care to, you can perform a capacity check on the low voltage parallel group or groups you suspect is bad by another quick voltage check comparison and compare capacity of discrepant group to one of the more uniform groups. You can do this all with a $30 dollar balance charger. If you find out a parallel group or two or three have poor capacity compared to the majority of parallel groups, you get out your flush cut snipers and then remove the cells. Once you peel away all the nickel strips, if you own one, you do a capacity and IR check of each cell with a 4 slot charger to ID the bad cell or cells. Then, the goal it to rebuild the battery with equivalent cells which you can even test in aggregate capacity before you patch in a new parallel group. You can do this by parallel wiring whatever your battery parallel group no. is together...say 4 or 5 new or replaced cells in parallel and do a charge/discharge and measure the capacity and compare it to more uniform single parallel group capacity within the larger battery. You can buy little cell receptacles on Amazon. Btw, a 21700 receptacle will hold a 18650 with two pennies at the cathode aka positive little buttton side...only slight difference in battery length.

https://www.amazon.com/dp/B0BSDC7L9T?psc=1&ref=ppx_yo2ov_dt_b_product_details

That is a general guideline and hope it makes sense. Please share your journey with us if you care to.

 

[harrisonpatm]

The controller starts cutting out when the battery is still showing a resting voltage of 47v,

So it's the controller cutting out, not the BMS, is that correct?

Sorry to say, but in my opinion, if you want the 48v range of most controllers, you should have built a 14s, not 13s.

You're right that at 47v resting, the batteries should still have capacity left. It's not the battery's problem, its the controller that's using the battery. If you went with 14s, 47v would be at the bottom 20% or so of the pack's capacity.

Can you link the controller you are using so we can see if it's cutoff voltage is indeed 48V? Or 47V? If so, then there's likely nothing wrong with your battery, but rather a mismatch between the battery and controller. That's why your 52V battery (likely 14s) wouldn't hit the cutoff, while this 13s battery does.

I tested all the salvaged cells and they all showed 95-100% their rated capacity using 1A discharge.

When you discharge at a higher rate, you get less usable capacity. This is true with all cells. LG MH1 says it is indeed "rated" for 10A max, meaning you shouldn't even consider trying to pull more from it. 7A discharge is close to that max rate, and you were testing at 1A. Rule of thumb is that testing conditions should reflect expected use. That's why you're seeing more voltage sag than expected.

Edit: I reread Amberwolf's earlier response, he already said this better than me, didn't mean to beat a dead horse.

 

[Wattman]

Sorry harrison but what you wrote is incorrect based upon my experience. 95% probability it is a battery problem and not a controller problem. Also, nothing wrong with the battery configuration of 13s.
Spaceship likely has a low parallel group or two of cells within his battery. The BMS cuts out and not the bike controller when a voltage drop occurs for the lowest parallel group which drops it below 3.0v.
Again, this should be confirmed shortly by Spaceship measuring all of his parallel cells. I am going through this right now. I have two low cell groups in my 14s 4p battery and my BMS cut off when riding in a simliar fashion which btw, is extremely common if cells aren't really well matched when the battery is built which is also sadly very common and why this dynamic occurs too frequently.

 

[harrisonpatm]

Sorry harrison but what you wrote is incorrect based upon my experience

It's not about experience. If the controller cuts off at 47v, then it cuts off at 47 volts regardless of the SoC of the battery in question. Doesn't matter if the battery has more capacity left, if it drops below a voltage that the controller can use.

95% probability it is a battery problem and not a controller problem

I didn't say it was a controller problem, I said it was a controller/battery mismatch. If you put a 12v battery on a 48v controller, you wouldn't look for what's wrong with the battery.

Two ways to confirm whether i'm right. First, info on the controller. We can lookup its cutoff voltage.

Second, run the bike and 13s battery until the controller cuts off. Then plug the same battery into a load tester. Does it still have more energy? Then the battery isn't the issue, its the equipment using the battery.

 

[harrisonpatm]

The BMS cuts out and not the bike controller

OP didn't confirm this yet. If the BMS is the one that's cutting off at 47v, under load, then yes, there's an issue with the battery, and I'm wrong. If it's the controller cutting off, then the battery is just going through it's normal discharge curve, which eventually falls below the working voltage of the controller.

 

[E-HP]

OP didn't confirm this yet. If the BMS is the one that's cutting off at 47v, under load, then yes, there's an issue with the battery, and I'm wrong. If it's the controller cutting off, then the battery is just going through it's normal discharge curve, which eventually falls below the working voltage of the controller.

The terminology being used, cutting out, shutting down, etc. is vague, but both are used in the original post. Cutting out (which may or may not assume cutting back in) is usually related to the controller, which doesn't usually shut down the system when hitting LVC. Shutting down is usually more indicative of a battery problem, since shutting down doesn't usually imply that the power automatically resumes. But in either case, I see posters using the terms interchangeably so there's no way to assume, without clarification. Most of the post seems to be indicate it's cutting out (and back in), but this assumption brings that back into question
"From what I understand it is voltage sag when load is applied that is causing the battery voltage to momentarily drop below my controllers low voltage cut-off, and so it shuts down."

 

[Wattman]

Harrison, will all be confirmed if the OP opens his battery and measures the parallel voltages.
I had this exact issue and in fact am charging/discharging parallel groups as I write this.
Exceedingly rare for any controller to cut out at 47 volts even with voltage sage as you conjecture. By stark contrast, quite common for a higher charged battery to have one or two weak parallel cells and the battery shuts off under load to protect the low voltage cells via the BMS. Could be merely an out of balance condition or more problematic failed cell or cells which can only be proven with capacity testing either in a parallel group or take the pack apart and individually load test each cell.

I had two low cell groups and am balancing my battery presently to keep the battery aka BMS from shutting the battery off to protect the lowest voltage cells. I am bringing high volt cells down aka discharge and low side cells up aka 1S charge. Shooting for a nominal of 3.6 volts per group X 14 = 50.4v. Then, next step with battery still apart will be to 'balance charge' my entire pack with my 'dumb' generic 5 amp 52 volt charger which will charge the pack to 58.8 volts or try to. This experiment with the pack still apart will divulge whether the markedly weaker cells will charge at the same rate as the more robust cells which is a glimpse into their respective capacity capability. A pretty high probability that the weaker cells will not charge as quickly to the highest parallel cell voltage cells which will discontinue charging via the BMS to protect the highest charged parallel cells at 4.2v and lower charged cells as a result should follow a bit under this value. This is confirmed with a multimeter check of each parallel group.
After likely reaching a bit under 58.8 v, I will leave the pack on the generic aka dumb charger to see if the BMS will top balance all the cells in spite of a capacity difference. If it does, there is hope I don't have to take apart the pack and replace weaker cells.
Generally however there is a reason why cells fall out of balance irrespective of top balancing. Its because cell capacity isn't equivalent. A matter of degree. This is either manageable aka the BMS will do a bit of top balancing clean up...or...the pack will fall quickly out of balance because the capacity variance is too great among parallel groups and if the battery is to be salvaged, cells will need to be replaced.

 

[spaceship]

The terminology being used, cutting out, shutting down, etc. is vague, but both are used in the original post. Cutting out (which may or may not assume cutting back in) is usually related to the controller, which doesn't usually shut down the system when hitting LVC. Shutting down is usually more indicative of a battery problem, since shutting down doesn't usually imply that the power automatically resumes. But in either case, I see posters using the terms interchangeably so there's no way to assume, without clarification. Most of the post seems to be indicate it's cutting out (and back in), but this assumption brings that back into question
"From what I understand it is voltage sag when load is applied that is causing the battery voltage to momentarily drop below my controllers low voltage cut-off, and so it shuts down."

OK let me clarify, when riding it gets to a point where when I apply throttle, the display turns off and the motor no longer receives any power. It does not automatically turn back on, I need to turn off and back on the power switch on the battery, then I can turn the controller back on. If I lift the wheel and just apply low load (100 watts or so) it doesnt power off. Its only under heavy load, 700 watts or more, that the power goes out. At the point where this occurs if I remove the battery and check voltage it is resting at about 47v.

I fully charged the battery and I opened up the battery and measured voltage of all the groups:

Full Pack: 54.2v

1 - 4.194
2 - 4.196
3 - 4.20
4 - 4.20
5 - 3.930
6 - 4.200
7 - 4.192
8 - 4.194
9 - 4.186
10 - 4.181
11 - 4.194
12 - 4.189
13 - 4.196

So it appears that I have an issue with cell group #5. I do not own a balance charger, how would I use it to try and balance this group without remving it from the pack? If you recommend buying a balance charger are the cheap 1s-6s ones on Amazon suitable? Is it useful for other purposes? Or should I just cut the group out and test individual cells again with my Opus charger?

Thanks

 

[Wattman]

OK let me clarify, when riding it gets to a point where when I apply throttle, the display turns off and the motor no longer receives any power. It does not automatically turn back on, I need to turn off and back on the power switch on the battery, then I can turn the controller back on. If I lift the wheel and just apply low load (100 watts or so) it doesnt power off. Its only under heavy load, 700 watts or more, that the power goes out. At the point where this occurs if I remove the battery and check voltage it is resting at about 47v.

I fully charged the battery and I opened up the battery and measured voltage of all the groups:

Full Pack: 54.2v

1 - 4.194
2 - 4.196
3 - 4.020
4 - 4.020
5 - 3.930
6 - 4.200
7 - 4.192
8 - 4.194
9 - 4.186
10 - 4.181
11 - 4.194
12 - 4.189
13 - 4.196

So it appears that I have an issue with cell group #5. I do not own a balance charger, how would I use it to try and balance this group without remving it from the pack? If you recommend buying a balance charger are the cheap 1s-6s ones on Amazon suitable? Is it useful for other purposes? Or should I just cut the group out and test individual cells again with my Opus charger?

Thanks

First, well done. Second, I don't see a major battery issue here. That is not a gross unbalance of cells in the least. A BMS cut out typically occurs when one or more parallel cell groups drop below '3.0 volts' even though aggregate battery voltage is much higher. That said, you show parallel group no. 5 in its charged state. It may discharge much faster than the other cell groups.
Ideally you want to compare the capacity of this cell group to one of your other strong voltage parallel groups or a couple for reference.

Since you are a battery builder I do suggest you buy a balance charger. You also will need to purchase some cables that generally do not come with the charger to 1S balance charge parallel groups...or discharge them. These cables have a 4mm banana clip on one end and alligator clip on the other.


This is the stuff I have. You can find a cheaper balance charger. Most use the same architecture and screen. They vary in whether they have a fan and their watt potential mostly.

Balance Charger:

https://www.amazon.com/HTRC-Charger-Battery-Balance-Discharger/dp/B07RGK6P49/ref=sr_1_14_sspa?keywords=6s+lipo+battery+charger&qid=1679102727&sr=8-14-spons&ufe=app_do%3Aamzn1.fos.006c50ae-5d4c-4777-9bc0-4513d670b6bc&psc=1&smid=A32LIMT42X7BEN&spLa=ZW5jcnlwdGVkUXVhbGlmaWVyPUE0VkxSRTBJVVVDOEEmZW5jcnlwdGVkSWQ9QTAzNDM4MTUyNUZZT1lNUVFYTkhJJmVuY3J5cHRlZEFkSWQ9QTAxMzI5MDUyVkRaVjBUR1o5Q0hVJndpZGdldE5hbWU9c3BfbXRmJmFjdGlvbj1jbGlja1JlZGlyZWN0JmRvTm90TG9nQ2xpY2s9dHJ1ZQ==#customerReviews

Alligator cables for 1S charging:

https://www.amazon.com/dp/B09L7JHLFM?psc=1&ref=ppx_yo2ov_dt_b_product_details

The purpose of the balance charger is to equalize parallel group voltages. That is a starting point.

But what I suggest is use the balance charger for a higher purpose which is a capacity assessment of your weakest parallel group compared to your strongest voltage parallel group.

If you decide to buy this stuff and have a go, I can walk you through it.

I strongly suggest you do not tear your pack apart...yet. You can of course later and then perform a battery load/capacity check with the following charger but I would use your balance charger as a tool to analyze cell condition without cutting your nickel strips.

https://www.amazon.com/Battery-Charger-Lithium-Analyzer-BTC3400V31/dp/B0B2M1G22K/ref=sr_1_1?crid=2BYH7WLGFZ9T2&keywords=BTC3400V31&qid=1679160694&s=electronics&sprefix=btc3400v31%2Celectronics%2C148&sr=1-1

Let me know if above makes sense or if any concerns.

 

[jonyjoe303]

You have an issue with 3,4 and 5 those 3 are at about 4.00 volts, all the other cells are close to 4.20 volts. If you isolate those 3 cells you can probably charge them to 4.20 with the opus by running wires from the opus to the cells. That's how I charge large cells that don't fit on the opus.

I suspect you have high IR cells in the pack. Even if you balance it out, it will go out of balance again. Batterypacks that don't stay in balance will always give you problems. If you do take the pack apart to test each cell, you will need an IR tester besides the opus. Without an IR tester, there are ways to test the IR using a multimeter, there are videos on youtube that shows how to do it.

 

[spaceship]

You have an issue with 3,4 and 5 those 3 are at about 4.00 volts, all the other cells are close to 4.20 volts. If you isolate those 3 cells you can probably charge them to 4.20 with the opus by running wires from the opus to the cells. That's how I charge large cells that don't fit on the opus.

I suspect you have high IR cells in the pack. Even if you balance it out, it will go out of balance again. Batterypacks that don't stay in balance will always give you problems. If you do take the pack apart to test each cell, you will need an IR tester besides the opus. Without an IR tester, there are ways to test the IR using a multimeter, there are videos on youtube that shows how to do it.

Thanks for pointing out 3 and 4, I actually just typed those out wrong and did not realize. Double checked my hand written notes from when I was measuring and they were actually 4.20 not 4.02. I fixed the post.

I had thought about running wires from the Opus charger, but decided to just order a cheap balance charger discharger, it has some additional features that might come in handy at some point. This is the one I ordered because of quick shipping

https://www.amazon.ca/gp/product/B07NW99VLM/ref=ppx_yo_dt_b_asin_title_o01_s00?ie=UTF8&psc=1

First, well done. Second, I don't see a major battery issue here. That is not a gross unbalance of cells in the least. A BMS cut out typically occurs when one or more parallel cell groups drop below '3.0 volts' even though aggregate battery voltage is much higher. That said, you show parallel group no. 5 in its charged state. It may discharge much faster than the other cell groups.
Ideally you want to compare the capacity of this cell group to one of your other strong voltage parallel groups or a couple for reference.

Since you are a battery builder I do suggest you buy a balance charger. You also will need to purchase some cables that generally do not come with the charger to 1S balance charge parallel groups...or discharge them. These cables have a 4mm banana clip on one end and alligator clip on the other.


This is the stuff I have. You can find a cheaper balance charger. Most use the same architecture and screen. They vary in whether they have a fan and their watt potential mostly.

Balance Charger:

https://www.amazon.com/HTRC-Charger-Battery-Balance-Discharger/dp/B07RGK6P49/ref=sr_1_14_sspa?keywords=6s+lipo+battery+charger&qid=1679102727&sr=8-14-spons&ufe=app_do%3Aamzn1.fos.006c50ae-5d4c-4777-9bc0-4513d670b6bc&psc=1&smid=A32LIMT42X7BEN&spLa=ZW5jcnlwdGVkUXVhbGlmaWVyPUE0VkxSRTBJVVVDOEEmZW5jcnlwdGVkSWQ9QTAzNDM4MTUyNUZZT1lNUVFYTkhJJmVuY3J5cHRlZEFkSWQ9QTAxMzI5MDUyVkRaVjBUR1o5Q0hVJndpZGdldE5hbWU9c3BfbXRmJmFjdGlvbj1jbGlja1JlZGlyZWN0JmRvTm90TG9nQ2xpY2s9dHJ1ZQ==#customerReviews

Alligator cables for 1S charging:

https://www.amazon.com/dp/B09L7JHLFM?psc=1&ref=ppx_yo2ov_dt_b_product_details

The purpose of the balance charger is to equalize parallel group voltages. That is a starting point.

But what I suggest is use the balance charger for a higher purpose which is a capacity assessment of your weakest parallel group compared to your strongest voltage parallel group.

If you decide to buy this stuff and have a go, I can walk you through it.

I strongly suggest you do not tear your pack apart...yet. You can of course later and then perform a battery load/capacity check with the following charger but I would use your balance charger as a tool to analyze cell condition without cutting your nickel strips.

https://www.amazon.com/Battery-Charger-Lithium-Analyzer-BTC3400V31/dp/B0B2M1G22K/ref=sr_1_1?crid=2BYH7WLGFZ9T2&keywords=BTC3400V31&qid=1679160694&s=electronics&sprefix=btc3400v31%2Celectronics%2C148&sr=1-1

Let me know if above makes sense or if any concerns.

Thanks, I think I understand. I'll attach the clips of the balance charger to the low battery group and see if I can charge it up to 4.2v, and then run a capacity test on the questionable group and one or two of the other "good" groups to see how they compare.

I'll post my results in a couple days

 

[amberwolf]

OK let me clarify, when riding it gets to a point where when I apply throttle, the display turns off and the motor no longer receives any power. It does not automatically turn back on, I need to turn off and back on the power switch on the battery, then I can turn the controller back on.

That means the BMS itself is shutting off output to protect the cells against overdischarge and/or overcurrent.


Is your BMS a balancing BMS? If it has a balancing function, you can leave the pack on the charger and it will eventually rebalance the cells, making them all the same voltage at full charge.

This does NOT fix the problem, however (you would need to replace (all) the cells with (preferably new) matched-characteristics cells to do that). It just enables you to use more of the limited capacity the pack has under the load it is seeing on your system.

A BMS has a very small capability for rebalancing, so it may take hours or even days to rebalance the difference your cells have. (if you have a BMS with 50mA balance current, and there is a 1Ah difference between the lowest and highest cells, it will take 1Ah = 1000mAh / 50mA = 20 hours to correct, or more. (you can guesstimate the capacity difference using the voltage difference, and the chart you have for the cells showing capacity vs voltage, times the number of parallel cells).

 

[bobbill]

Same for ebikes, I must presume, 10s4p pack. Have no clue if balanced, but bike quit[ so replaced 2 cells later. Pack cells was way out of balance.....I switched cells and same thing happened,,,,? Bike quit. No clue,til now,,,,, and,
FWIW BMS Balance Wire Connections

 

[docw009]

I fully charged the battery and I opened up the battery and measured voltage of all the groups:

Full Pack: 54.2v

1 - 4.194
2 - 4.196
3 - 4.20
4 - 4.20
5 - 3.930
6 - 4.200
7 - 4.192
8 - 4.194
9 - 4.186
10 - 4.181
11 - 4.194
12 - 4.189
13 - 4.196

So it appears that I have an issue with cell group #5.

I agree. Now put the battery on the bike and run it til it drops dead. Measure those voltages again. You'll probably find #5 quite close to 3.0 volts while the others are around 3.7 Betcha a weld popped off and you've got 3P or 4P on #5.

Could also be you have a bad cell in that P-group, but it's more likely on a self built pack that the builder made an error, It's always nice when someone finally posts the group voltages. You can see a lot.