New LiFePO4-battery pack is underperforming

Batteries, Chargers, and Battery Management Systems.
User avatar
fivari   10 W

10 W
Posts: 88
Joined: Jun 17 2011 3:00am
Location: Leuven, Flanders, Belgium

Re: New LiFePO4-battery pack is underperforming

Post by fivari » Jul 06 2012 1:57am

dnmun wrote:nope. you have to test capacity by doing a bench test. using a wattmeter or ammeter.
I am not sure what you mean. Do you mean to test each cell of the first string individually? Anyway, I did a capacity test of the pack as a whole as described above: the BMS cuts off after much less than the nominal capacity of 40 Ah because the first string has low voltage. This way I was able to locate the problem and therefore, I plan to open the pack and locate the cell that causes the first string to leak energy. I just hope that I don't have to isolate each and every cell of that first string.

This morning, I found the first string still at 3.33V. The others were at 3.6V. To me it is obvious that the first string has a short cell that slowly brings down the voltage and the capacity of the whole pack. I guess that after I replace that one cell, the problem is solved.
Anyway, I will ask the chinese seller to send me a few cells as a compensation.

User avatar
Jeremy Harris   10 GW

10 GW
Posts: 4208
Joined: Oct 23 2007 4:15pm
Location: Salisbury, UK

Re: New LiFePO4-battery pack is underperforming

Post by Jeremy Harris » Jul 06 2012 2:22am

fivari wrote: I am not sure what you mean. Do you mean to test each cell of the first string individually? Anyway, I did a capacity test of the pack as a whole as described above: the BMS cuts off after much less than the nominal capacity of 40 Ah because the first string has low voltage. This way I was able to locate the problem and therefore, I plan to open the pack and locate the cell that causes the first string to leak energy. I just hope that I don't have to isolate each and every cell of that first string.

This morning, I found the first string still at 3.33V. The others were at 3.6V. To me it is obvious that the first string has a short cell that slowly brings down the voltage and the capacity of the whole pack. I guess that after I replace that one cell, the problem is solved.
Anyway, I will ask the chinese seller to send me a few cells as a compensation.
I doubt you have a shorted cell. It's much, much, more likely to just be crappy quality cells, probably with a few cells in the pack that are way under capacity, or maybe even open circuit. A voltage test isn't going to tell you much, if anything, about the state of charge of each cell group. You need to do some testing under load, plus ideally doing a capacity test on each cell group. My guess is that you'll find that the pack capacity isn't 40 Ah (it's pretty common for the vendors of these things to just openly lie about capacity) and you'll probably also find either some dead (open circuit) cells or some crappy internal connections (probably both) if you take the pack apart.

Unfortunately there are a LOT of nasty cylindrical cell packs around, more often than not of poor internal build quality. Dogman was spot on earlier with the comment earlier about not recommending those packs.

Good luck with getting recompense from the vendor, many of these Chinese ebay sellers are just buying up unknown quality surplus cells in the big electronic street markets in Guangzhou or Shenzen, assembling them into packs, selling them on ebay until they get caught out, then starting the process again with a new ebay ID. There are a few Chinese based battery sellers who have built a solid reputation, people like Li Ping and Cellman, for example, but there are also hoards of rip-offs to catch the unwary. The unscrupulous vendors rely on the high cost of postage and the absence of any consumer protection laws for international sales, so can get away with what we'd call fraud. I doubt the photos on their ebay advert have anything to do with their batteries, either, they were probably just copied from a reputable manufacturers site.
Please ask questions on the forum, rather than by PM, as it helps others and you'll get a better range of answers.

User avatar
fivari   10 W

10 W
Posts: 88
Joined: Jun 17 2011 3:00am
Location: Leuven, Flanders, Belgium

Re: New LiFePO4-battery pack is underperforming

Post by fivari » Jul 06 2012 3:00am

I have to agree that what seemed to be a good deal, turns out to cause some troubles.
I am just trying to work me out of this and I recognize that it will require quite some energy to overcome this.

But apart from that, is there anything wrong about the deduction I made?
Because the pack voltage drops soon after charge from 58.4V down to 54V
Because this one cell group shows low voltage after cut off,
because I was able to charge this one cell group an additional 22 Ah after fully charging the pack (a lot of which was probably lost right away because of the leaking)
because the voltage came down soon after fuly charing this first cell group,

==> i conclude that this one group has at least one cell shorted. What else could cause these three phenomena to happen simultaneously?

User avatar
Jeremy Harris   10 GW

10 GW
Posts: 4208
Joined: Oct 23 2007 4:15pm
Location: Salisbury, UK

Re: New LiFePO4-battery pack is underperforming

Post by Jeremy Harris » Jul 06 2012 3:28am

fivari wrote:I have to agree that what seemed to be a good deal, turns out to cause some troubles.
I am just trying to work me out of this and I recognize that it will require quite some energy to overcome this.

But apart from that, is there anything wrong about the deduction I made?
Because the pack voltage drops soon after charge from 58.4V down to 54V
Because this one cell group shows low voltage after cut off,
because I was able to charge this one cell group an additional 22 Ah after fully charging the pack (a lot of which was probably lost right away because of the leaking)
because the voltage came down soon after fuly charing this first cell group,

==> i conclude that this one group has at least one cell shorted. What else could cause these three phenomena to happen simultaneously?
Unfortunately. many of the cylindrical cells that find their way on to the Chinese electronic street markets are cells that have failed quality testing at the manufacturer. There are also some ebay vendors who aren't very good at assembling these cells into packs; we've seen many instances here of people with packs like this where the inter-cell connections are poor, where balance leads have fallen off and where the pack has been poorly secured inside the duct tape, allowing the cells to move and break internal connections.

It won't be a shorted cell causing the problem. If it was, then the pack would probably have gone up in flames as the other cells in that group discharged through it. It's very rare to get a shorted cell without it having been mechanically damaged, cells almost always fail by going open circuit, or by just not having the rated capacity. It's also quite possible that you have a few cells in the pack with a high self-discharge rate. You may also have a problem with the BMS, perhaps it isn't limiting cell voltage properly for each cell group.

If you have a few low capacity cells in the pack, then they will discharge more rapidly, and charge more rapidly, than the others and so will show voltage differences to the others in the pack. In all probability you'll find a few open circuit cells, together with perhaps a few that have a lower capacity than the others. You may be able to rescue the pack with new cells, but it might be a challenge to take the thing apart and find the dead cells, then fit new cells and get it balanced up again.
Please ask questions on the forum, rather than by PM, as it helps others and you'll get a better range of answers.

User avatar
fivari   10 W

10 W
Posts: 88
Joined: Jun 17 2011 3:00am
Location: Leuven, Flanders, Belgium

Re: New LiFePO4-battery pack is underperforming

Post by fivari » Jul 06 2012 4:04am

Thanks Jeremy.
Reading your answer, I realise that I used the wrong terms: shorted cell leaking energy. You are right, I should have called it a cell with high self discharge rate. Sorry for that. I am only beginning to learn about these topics.

I will keep this forum informed about the outcome of this problem.

User avatar
Jeremy Harris   10 GW

10 GW
Posts: 4208
Joined: Oct 23 2007 4:15pm
Location: Salisbury, UK

Re: New LiFePO4-battery pack is underperforming

Post by Jeremy Harris » Jul 06 2012 4:34am

fivari wrote:Thanks Jeremy.
Reading your answer, I realise that I used the wrong terms: shorted cell leaking energy. You are right, I should have called it a cell with high self discharge rate. Sorry for that. I am only beginning to learn about these topics.

I will keep this forum informed about the outcome of this problem.
Although cells with a higher than normal self discharge rate are a possibility, they won't be drawing much power from the pack, if they were then those cells would get hot. In practice you will get cells that discharge their "surface charge" (the bit from the hot of charge voltage to the working voltage) more quickly than others, but this doesn't have much energy and so isn't normally a major problem. It's rare to get a cell that self-discharges below the normal cell working voltage, but common to get cells with a low capacity, open circuit or high internal resistance.

If you take the duct tape off your pack we can probably advise better on the way forward. If it's a typical multicell cylindrical pack then what you'll probably find under the tape is a big stack of cells with spot welded tabs interconnecting them into cell groups. Repair will mean isolating the cell group with the low capacity (not necessarily the lowest voltage group) taking it out of the pack, splitting the cells in that group and checking for which ones may be open circuit, then rebuilding the pack by spot welding the new cells back together. You may be able to make a soldered repair, but this needs care, as it's possible to damage the cells be directly soldering to them and letting the cells get too hot. It is possible that the pack you have has been soldered together, in which case this is a possible reason for some cells to fail.
Please ask questions on the forum, rather than by PM, as it helps others and you'll get a better range of answers.

User avatar
dogman dan   100 GW

100 GW
Posts: 34691
Joined: May 17 2008 12:53pm
Location: Las Cruces New Mexico USA

Re: New LiFePO4-battery pack is underperforming

Post by dogman dan » Jul 06 2012 6:39am

It doesn't sound like simply popped spot welds this time. If it was, would that first cell group charge up faster with less cells to charge?

Often, one cell group runs the bms. Stored lifepo4 packs often destroy the first cell group by overdischarging that group. It can also happen if the shipping is slow enough, taking a month or more. So that seems like a plausible possibility.

Did you have to plug in the bms when it arrived? If so, then that theory is also out the window. Once you dissect out the bad cell group, replacing it should help the pack. It may still tend to self discharge to about 54-55v. That's normal once a pack has some wear on it.

But it may never put out the rated capacity even when you get it fixed. Pics of cell testers do not prove your cells are the ones in the good batch.

User avatar
fivari   10 W

10 W
Posts: 88
Joined: Jun 17 2011 3:00am
Location: Leuven, Flanders, Belgium

Re: New LiFePO4-battery pack is underperforming

Post by fivari » Jul 06 2012 8:37am

Thanks Dogman.
The shipping did last only one week. The BMS was connected to the batterypack.
The voltage drop yesterday, happened without the BMS being connected.
Opening the pack this evening will hopefully reveal the reason for its behaviour.

User avatar
dogman dan   100 GW

100 GW
Posts: 34691
Joined: May 17 2008 12:53pm
Location: Las Cruces New Mexico USA

Re: New LiFePO4-battery pack is underperforming

Post by dogman dan » Jul 06 2012 9:10am

Well so much for that theory. It could be the bms draws more than it should be from the cells that run it, or perhaps they had it stitting a month before they shipped. Pretty sad they still haven't learned to unplug the bms after the battery is tested. This problem has been known about for about 4 years now.

Unfortunately the bargian lifepo4 packs reputation for being a hairball of problems continues. It's been pretty much standard for a V power hong kong pack to put out 15 ah if it's a 20 ah pack in the past. Not that others we trust haven't had bms issues, or cells tested good spontaneously go bad fairly soon. For example, when odering batteries from Hobby King, we just expect and order enough to cover a 10% fail rate on the RC lipo packs. So you want 8 you buy 9. Much harder to deal with in an assembled pack. One reason we recomend pingbattery is that he will send replacement chargers, bms, and cells when needed without a paypall fight.

User avatar
fivari   10 W

10 W
Posts: 88
Joined: Jun 17 2011 3:00am
Location: Leuven, Flanders, Belgium

Re: New LiFePO4-battery pack is underperforming

Post by fivari » Jul 09 2012 1:18am

As promised, I keep you informed about the opening of the pack.
First comment: apperently the pack was built to never be opened again: a lot of hot glew, dubble duck taped, hard to open. But I succeeded without any damage to the wiring nor the cells.

Whereas the vendor told me that the pack consisted of 16s13p of 3100 mAh cells, the truth was 16s12p of LFP26650E which are according to internet datasheet 3000 mAh cells. I opened a paypal dispute in order to be paid back 10%. He refused using the argument that he selected cells that were tested 3300 mAh. Since the capacity test I did showed 35.7 Ah, I don't believe him.

Image

Second mistake was the connection between the different cell groups: only two thin tabs of 0,15mm x 10mm have to carry up to 60A. This is, in my opinion, not enough. Therefore I added tabs on every cell by soldering as shown on the picture below.

Image

Since I want to be able to measure each cell group individually I connected another set of wire to the batteries. This give me the possibility to charge a group seperately if necessary. I could have used the BMS wires, but in order to use these, I have to open the BMS, deconnect the connector. After doing this 20 times, the BMS-box would be broken.

Image

dnmun   100 GW

100 GW
Posts: 16190
Joined: Jun 09 2008 1:32pm
Location: portland, or and loveland, co

Re: New LiFePO4-battery pack is underperforming

Post by dnmun » Jul 09 2012 1:31am

if the spot welds did not come apart and you actually got almost 36Ah, you are way ahead of what we expected.

when i pulled the duct tape off the Vpower pack all the spot welds were pulling off the top of the cells with the tape.

remember that rubber mat is all that separates the two ends of the pack. if it dislodges you could end up shorting the entire pack.

he really said he had selected them for an extra 10% capacity? no way.

User avatar
Jeremy Harris   10 GW

10 GW
Posts: 4208
Joined: Oct 23 2007 4:15pm
Location: Salisbury, UK

Re: New LiFePO4-battery pack is underperforming

Post by Jeremy Harris » Jul 09 2012 1:56am

Sorry to hear that the seller misrepresented these batteries, but I am afraid to say that I'm not at all surprised. Misrepresentation is so common amongst some battery suppliers, particularly many of the Ebay sellers in the Far East, as to be almost normal.

I'd agree with dnmum and say that the measured capacity you've got is actually pretty good for the cells you have, even if it is less than the seller said in his advert. The construction is pretty typical for these cylindrical cell packs, too, and OK as long as the cells are held together so that they can't move relative to each other and break the tab welds.
Please ask questions on the forum, rather than by PM, as it helps others and you'll get a better range of answers.

User avatar
fivari   10 W

10 W
Posts: 88
Joined: Jun 17 2011 3:00am
Location: Leuven, Flanders, Belgium

Re: New LiFePO4-battery pack is underperforming

Post by fivari » Jul 09 2012 2:40am

I agree that the building quality is good. The tabs were point welded with 4 welds on every cell. The capacity corresponds to the cells used and this is a relief, but does not correspond to what I ordered and paid for. I can't let this happen. I believe I am rigth to start a paypal dispute, no? Does anyone has experience with paypal disputes? There is the option to "escalete to paypal claim", which I think I should do in case he does not agree.

Still, I don't understand that for parallelling the cells, the same tabs were used (0.15mmx10mm) as for the two, only two, serial tabs. The parallel tabs should not carry any current, wherease the serial tabs are taking all the current. Only for the first and the last cells group, the parallel tabs are doubled. I noticed that the voltage sag is less. Now it is down to 50V, before it was down to 48,5V.
The pack fits just in the battery space of my scooter, it is tightly duck taped. So, I don't expect the cells to move relative to each other. I replaced the rubber by a more demping material: 8 mm thick, used for accoustic insulation:

Image

icecube57   1 GW

1 GW
Posts: 3072
Joined: Apr 25 2008 8:42am
Location: Austell GA
Contact:

Re: New LiFePO4-battery pack is underperforming

Post by icecube57 » Jul 09 2012 2:59am

Be careful about certain foams.. ive found some to be conductive...

User avatar
dogman dan   100 GW

100 GW
Posts: 34691
Joined: May 17 2008 12:53pm
Location: Las Cruces New Mexico USA

Re: New LiFePO4-battery pack is underperforming

Post by dogman dan » Jul 09 2012 6:26am

Gee, I wonder how much more damage you just did with the soldering iron? Hope you did it hot and quick.

User avatar
fivari   10 W

10 W
Posts: 88
Joined: Jun 17 2011 3:00am
Location: Leuven, Flanders, Belgium

Re: New LiFePO4-battery pack is underperforming

Post by fivari » Jul 09 2012 8:11am

Hot and fast it was!! 80W iron, some added flux to speed up things even more, done in less than two seconds. A ventilator blowing fresh air over the cells to cool them down as fast as possible.

999zip999   100 GW

100 GW
Posts: 8598
Joined: Jan 10 2010 1:40pm
Location: Dana Point So. Cal

Re: New LiFePO4-battery pack is underperforming

Post by 999zip999 » Jul 10 2012 12:23am

I heard of adding a cell to the weak string to even up it's ah, but it may have other stuff happen. Hopping for a Jeremy Harris answer. Thanks for your insight.

Post Reply