What is going on with this cell?

madsci1016

New member
Joined
Dec 13, 2019
Messages
26
I'm geeking out (EE here) while doing some precision testing on some used Headways and I have noticed one cell is losing about 2-4% capacity every cycle, and it's nominal voltage during discharge was lower. So I captured it's discharge curve at 2C and compared it with another cell also at 2C. Both cells originally tested the same capacity, but you can see how this bad one, in orange, is now lower capacity, and you see how it's curve is 0.1V lower than the other. What phenomenon can cause such results in a LiFEPO4? Electrolyte oxidation from over charge? Design defect? I'm assuming this cell isn't going to last many more cycles, and it has a high internal resistance (explaining the 0.1V lower voltage) but I do not have a way to directly measure that.

 

Redpacket

Well-known member
Joined
Feb 28, 2018
Messages
1,280
Definitely something going on, obviously not for use in your battery pack...
Some sort of contamination or maybe air leak (bad seals?)
Does it have any obvious physical damage?
 

Overmind

Active member
Joined
Jan 16, 2019
Messages
422
I encountered something I called destabilized lattice in some liitokala Li-ion 26650s.
Basically, they had full capacity, good sustained current, but every discharge they would lose a small % of the capacity. I got from 5200 to 4200mAh after just a few full discharges.

Probably such cells are cells that failed QC testing and got dumped and re-labeled.

In LiFEPO4 it's even more strange because they suppose to last significantly more cycles than Li-ion.

Can you test a large number of cells to see if this is an isolated case or something that is widely spread among your cells ?
 

madsci1016

New member
Joined
Dec 13, 2019
Messages
26
Overmind said:
Can you test a large number of cells to see if this is an isolated case or something that is widely spread among your cells ?

I bought 48 from BH and I have 40 I'm testing. 4 show self discharge, wide range of capacities on the rest. But this is the only one with high resistance and diminishing returns per cycle.
 

OffGridInTheCity

Well-known member
Joined
Dec 15, 2018
Messages
1,577
I started with 18650 lithium-ion (many small cells) and am very far down that road buthave 'grass is always greener' for those doinglarger cells (fewerto process) for similar power. However, this info illustrates that you can't just buythe exact number as some may not test out - same as 18650 but with larger consequences. I'm sure its a bit discouraging to find suspect cells.Just curious- did youbuy a couple extrawith this possibility in mind? I wonder if there is ageneral consensus strategy when buying larger LifePo4cells in 48v context such as 'buy 1 extra per 16'?
 

madsci1016

New member
Joined
Dec 13, 2019
Messages
26
I didn't have any plans for most of the batteries I buy mostly, I'm just buying them because I'm a geek engineer and I don't have experience with all these battery chemistry so I buy them to play with them to gain that experience. One reason Ibought these to experiment with supplementing the weak cells in my BYD packs with these to try an balance my BYD packs a bit better. But that is proving to have diminishing results anyway.

I recently became a boat owner so now some of these cells plus some of the LTO BMW batteries I bought are going into the boat. LTO will be the starter battery and these headways will be a hotel battery.

I mean fair rule of thumb for most projects is buy 10% extra. Where I'm lost is what the expectations should be when you buy from places like Battery Hookup. Tom is a great guy, and if you do ask about issues he takes care of it. I wish he would be a little less passionate in his item descriptions and focus more on what testing he does and will do, what capacities he guarantees, where these batteries came from and how they were used(if he knows it), etc. He has most recently gotten better at this on the newest batteries. When I bought these headways, all that was said is they were "Engineering samples testing near full capacity" so I have no idea what that means for expected age / life / quality / % bad cells. And if I ended up with 10% bad batteries from my order, should I even talk to them about it (I know he will refund / replace if i do) or is that the nature of buying used batteries? I want Tom to stay in business after all.

Anyway, sorry the tangent.
 

OffGridInTheCity

Well-known member
Joined
Dec 15, 2018
Messages
1,577
madsci1016 said:
I didn't have any plans for most of the batteries I buy mostly, I'm just buying them because I'm a geek engineer and I don't have experience with all these battery chemistry so I buy them to play with them to gain that experience. One reason Ibought these to experiment with supplementing the weak cells in my BYD packs with these to try an balance my BYD packs a bit better. But that is proving to have diminishing results anyway.

I recently became a boat owner so now some of these cells plus some of the LTO BMW batteries I bought are going into the boat. LTO will be the starter battery and these headways will be a hotel battery.

I mean fair rule of thumb for most projects is buy 10% extra. Where I'm lost is what the expectations should be when you buy from places like Battery Hookup. Tom is a great guy, and if you do ask about issues he takes care of it. I wish he would be a little less passionate in his item descriptions and focus more on what testing he does and will do, what capacities he guarantees, where these batteries came from and how they were used(if he knows it), etc. He has most recently gotten better at this on the newest batteries. When I bought these headways, all that was said is they were "Engineering samples testing near full capacity" so I have no idea what that means for expected age / life / quality / % bad cells. And if I ended up with 10% bad batteries from my order, should I even talk to them about it (I know he will refund / replace if i do) or is that the nature of buying used batteries? I want Tom to stay in business after all.

Anyway, sorry the tangent.
I'm with you. I've purchased well over 6,000 cells from BatteryHookup - and agree that they have been responsive. For example, I bought a batch of 2,000 modem cells but 10% of them were dented inside the case!! - nothing I did or BatteryHookup did or visible till the case was opened. BatteryHookup made good on them.
However - I agree that you have to take off the rosy glasses and really think about what the descriptions 'don't say' :D

Another recent example is that I made 4 seperate purchases fromBatteryClearing hosue for RING packs. They come 100packs or 150packsin a box depending on when I ordered them. Overall I had 11 seperate boxes with 11 seperate FedEx trackings. One of them got lost in Kentucky by FedEx but FedEx says Battery Clearning House's problem to make a claim. I can't. Battery clearing house won't respond by email or phone soI'm stuck missing 10% of my overall orders. On the other hand, the cells are pristine / new / only 2 bad out of 500 so far.... far exceeding my expectations from other DIY packs.

1.5 years agoPower2Spare had a super sale on raw packs - with 8 NCR18650As per pack. They're description was pretty rosy and the good cells were indeed 90% or better. However, 30% of the cells were no good / 0v / heaters - huge loss % wise and much worse than I expected.

So it seems like you cannot reach 100% but relative to flat out false advertisting... I would reorder from any of these 3 places and just try to be more critical when I read the 'glowing descriptions' :)
 

madsci1016

New member
Joined
Dec 13, 2019
Messages
26
I think I've gotten to the point where I'm happy with the tested capacities, but really bummed that I'm up to 8 cells so far (8 left to test) showing self discharge. IE, they lose .3V in about a day after charging up to 3.65V. So I may ask for a refund on those.

Any advice on how I could use these cells otherwise? Or if I'm out of line to expect a refund for cells that demonstrate self discharge?
 

OffGridInTheCity

Well-known member
Joined
Dec 15, 2018
Messages
1,577
>One of them got lost in Kentucky by FedEx but FedEx says Battery Clearning House's problem to make a claim. I can't. Battery clearing house won't respond by email or phone soI'm stuck missing 10% of my overall orders.

Update - Battery Clearing House got me my missing cells. @floydR provided a goodemail to Battery Clearing House and they responded and made good on the missing box of cells. Just wanted to make a note since I complained about it above. Thank you @floydR and Battery Clearing House.
 

madsci1016

New member
Joined
Dec 13, 2019
Messages
26
gauss163 said:
^^^ Is the IR on those much higher than others?
Not sure I have a good way to test that. The one in my OP definitely does, as you can see the extreme difference under a 2C load. The rest I haven't paid attention to voltage drop.

Is a 0.3V drop in two days not enough of a tell? The rest settled at a 0.1V drop under the same conditions.
 

Korishan

Moderator
Staff member
Joined
Jan 7, 2017
Messages
6,445
0.3V drop in a few days is quite a bit, yeah.

Reading resistance in this case could be just as easy as doing a simple comparison. Let's say you have cheap DMM that doesn't do great on accurate Ohm readings. But by comparing the readings to known good cells, we can infer if the one cell is bad.
Example:
Known good cells are reading with cheap DMM in Ohm: 100-Ohm - We know that generally 100-Ohm is pretty high.
Known good 2nd cell reading: 90-Ohm - Still seems high
Known good 3rd cell reading: 108-Ohm - Again, high.
Ok, now we have a base line. The DMM is possibly reading about 60-Ohms too high; this is a guess, but we are comparing with known good cells, so we can infer that the values are much lower

Possible bad cell: 160-Ohm
We see here that this one is reading MUCH higher than the others. We don't know the exact amount, but we can get a good idea it's worse than the others by a lot.
 

madsci1016

New member
Joined
Dec 13, 2019
Messages
26
I may be missing the obvious. I though in terms of battery testing when we mention internal resistance measurements it's by a fancy device that does AC impedance testing on the battery. If it's as simple as a DMM on resistance mode, I have several expensive ones. What should I expect to read on a good cell vs bad?


NVM, just tried with all my DMMs, none will give a ohm reading on any of these headways. So I guess I'm confused by your instruction.
 

Korishan

Moderator
Staff member
Joined
Jan 7, 2017
Messages
6,445
A DMM Ohm reading is a general reading. It's not 100% accurate. But it does give you a good idea of what's going on. Kinda like putting your ear to someones chest to hear their heartbeat as opposed to using a stethoscope.

A standard DMM should be able to do an IR/Ohm test on any battery, as long as the voltage isn't too high (probably don't want to test it on a 48V system, for example).


Here's 2 examples of a DMM. You would use one of the settings circled in Yellow. For cells, on the meter on the left, you'd use 200 setting, this is for reading up to about 900-Ohm. If the meter doesn't register, drop down to the 20 reading and see if it reads then.
The one on the right is auto-ranging.
Also note that the Red probe must be inserted in the correct sensor port. The port will be labeled V?mA

After setting the DMM into this setting, take the probes and touch to the cells terminals. Orientation doesn't really matter for this test.

[retracted]
Use the following two posts as examples.
 

gauss163

Member
Joined
Jun 28, 2020
Messages
284
Unfortunately the above suggestion will not work, i.e. the ohms function on a multimeter cannot be used to measure the internal resistance (IR) of a battery (attempting to do so might even damage some multimeters).

Instead, one simple method is toemploya resistorand the voltage function, e.g. as described hereor here.

To learn more about thevarious methods of measuring internal resistance see the papers I cite here.
 

Redpacket

Well-known member
Joined
Feb 28, 2018
Messages
1,280
Yeah have to agree, a typical DMM ohms range won't measure a battery cells IR.
Typical DMM are only designed to measure resistance of unpowered passive devices, not cells (or circuits, etc) with voltage on them.

In addition to the links gauss163 provided, the OP could also try this fairly simple method I posted which is step load DC IR using two resistors.
https://secondlifestorage.com/showthread.php?tid=9694&pid=66405#pid66405
...beware: a bit of math is needed :)
 

madsci1016

New member
Joined
Dec 13, 2019
Messages
26
Thats what i thought, thanks guys. I have a digital load with a step program function, so I may try to do that.

But not really sure anyone answered my question. DO I need to do that to know for sure these are bad cells? Is 0.3V drop in 2 days not enough of an indication they have a self discharge issue?
 

gauss163

Member
Joined
Jun 28, 2020
Messages
284
Yes, there is likely high self-discharge, but similar big drops can occur when they have very high IR and the charge termination current is not lowered enough in compensation, e.g. see the graphs here.

Your cell may be suffering from both. What was your charge termination current?
 

madsci1016

New member
Joined
Dec 13, 2019
Messages
26
Ahh, I see why that was asked. All the cells were charged using a bench top power supply, and I let them sit at CV of 3.65V for about 15 extra minutes once current fell below 5-10mA before taking them off to make sure they absorbed everything they could.
 

gauss163

Member
Joined
Jun 28, 2020
Messages
284
Ah, with such a low termination current you charged them more fully than most CC/CV chargers, so that likely means that self-discharge is the primary culprit, probably due to an internal short.
 
Top