"Flash-amp" battery testing questions.

[eebowler]

Ok. This is something I believe in but am a bit uninformed about the hows and whys involved.

( You don't really need to read this paragraph.) I bought a ZTS tester a while ago and gave it away to someone simply because it didn't give me all the information I thought it would. I had some old AA and AAA rechargeable cells which wouldn't be charged by two of three different chargers. Besides the usual very high self discharge rate and low % of actual capacity, the AAA cells didn't work in a regulated headlamp of mine and the AA cells simply couldn't dish out enough current to run some (most) lights properly. In all cases, flash amp tests performed on these cells gave unusually low readings even when fresh off the charger. The ZTS tester however, still said that these cells had 100% capacity when tested... :shakehead My belief in the flash amp test was set in stone.

My ROP/1274/1164 light is run by 6 Titanium high discharge cells which are proclaimed to be able to handle a 10C discharge rate. Silverfox proved this to be true in his NiMH battery shootout tests. The cells were obtained early this year and have been cycled (completely discharged) no more than 15 times and recharged about the same number of times. I have 12 cells in all.

Got a new (though cheapish) DMM (Extech MN36) recently and have recently runflash amp tests on these cells. All except three cells gave me readings of over 14 amps during the 1-2 seconds they were tested. Two of the 'bad' cells gave readings of around 8 amps while one cell gave me a reading of just over 3A. I recharged the cells, ran the tests again and got approximately the same results.

:thinking:
Were these cells always bad or did they degraded over time? (The previous DMM gave <6A for all cells so there's no history.)

What could I have done wrong to cause this to happen?

What exactly goes on in a cell (chemically, physically) for it's internal resistance to increase? Can something be done about it or are the cells goners?

Do these low flash amp readings also mean that the 1800mAh capacity decreased as well? (I normally measure open circuit voltages of the cells before recharging and they are all approximately the same value ie, no one cell is discharged significantly more than another.

I have a crappy feeling that they'll eventually degrade like the Sanyo HR-3 cells did.

Please help if you can.

eebowler

 

[eebowler]

:bump:

 

[SilverFox]

Hello Eebowler,

I would guess that the cells were "different" from the beginning...

If you are matching cells on flash amps, you need to group them according to your tests, then continue on to match them by capacity and mid point voltage. Once you have done all of this matching, you should have a very good set of cells.

The internal resistance of a cell increases because of changes in the separator during use, or storage. Self discharge can produce large crystal growth, which will also contribute to higher internal resistance until the large crystal growth has been broken up.

Another thing that may help your under performing cells is to have them zapped. I am not sure of the construction of these cells, but they may respond to zapping and that may be enough to get them back into shape.

Tom

 

[mdocod]

consider the fact that flash amping a single cell (only 1.2V) can generate pretty low current readings inherently depending on how much resistant you are dealing with to make the loop through the DMM and back. If you took a flash amp reading of 2 cells in series, your number would likely shoot up quite a ways. (possibly high enough to kill a 20 amp unfused circuit on a meter)

I have also used flash amp testing to weed out problem cells. Seems like cells that are behaving badly DO seem to give poor flash amp results, I had some Kodak brand cells, i think they were like 2200mAH or something like that, but within a few cycles they started not working in anything worth a darn, but open circuit voltage looked great (1.45+V fresh off charger). When I took dead short flash amp readings they wouldn't even get above an amp. I tried cycling them repeatedly on the charger, tried discharging them into a near dead short a few times to warm em up and try to bust the crusty's loose, nothing I tried helped. They were junk and I have just accepted that. Sanyo Eneloops all the way now!

 

[TorchBoy]

Another thing that may help your under performing cells is to have them zapped.

So that's not just for NiCd?

 

[jayflash]

I've found that as the short-circuit (flash amps) current reading goes down, so does the cell's capacity.

 

[MikeLip]

OK, I may be dumb here, but "flashamp" seems a like very poor idea for at least two reasons.

It's inaccurate - or more correctly, inconsistent. The current will vary a lot depending on how good your connection is, since at these currents very small changes in contact resistance will make a big difference. You don't have time to mess around getting a good connection, because most cells are going to be seriously unhappy driving into a short load. Even a humble AA can get pretty hot driving into a short.

It's unnecessary. What you are doing is measuring the internal resistance of the cells. There is no reason why you can't do this into a reasonable load. If you have a (say) 3V terminal voltage battery and put a 1 ohm load across it, you have a maximum 3 amp load, which I personally would feel much more comfortable with. Measuring the loaded terminal voltage will give you a very exact number for the internal resistance of the cell is without abusing it. If you do this with a known good or brand new cell, you now have a baseline to compare questionable cells to.

 

[VidPro]

agrees kinda with mike, "flashamp" the cells your not going to use, do it just for testing the ones you LIKE and would like to keep working ok, just dont do that
depending on the cell type and the flash amp current, and the shunt and the leads on the DMM, you can damage a cell, course the massive hotwires themselves are probably doing an ok job of shorting the cell out as it is but why make it worse.

 

[SilverFox]

Hello Eebowler,

Energizer refers to flash amp testing in order to get an approximation of a cells internal resistance. They suggest using a 0.01 ohm resistor and flashing the cell for 0.2 seconds.

Here is the article.

Shorting a cell can be a safety issue. I would rather see testing done at the loads you are going to use the cell at, rather than relying on flash amp results. It would be interesting to see if your cells that scored low on your flash amp testing also score low on mid point voltage and capacity.

Tom

 

[Turak]

I know it is an accepted practice, but I see 'Flash-Amp' testing as an accident just waiting to happen.

There are MUCH safer ways to figure out the internal resistance, if that's all your after.

Talk about a surprise....just wait till you accidently 'flash-amp' test a non-protected lithium cell...

 

[eebowler]

Thank you for the responses guys!

Mike: I'll try your method to figure out the internal resistance of the cells. (high accuracy wasn't a necessity in the first place since I disconnected the probes just after the meter beeped after current passed the 10A mark.)

SilverFox: I need to get myself a battery capacity tester thingy! The only thing I can afford is the Triton Jr (refuse to afford a tester only). Do you think that's acceptable?

 

[SilverFox]

Hello Eebowler,

The Triton Jr is a great product, but it is limited to a maximum of 1 amp discharging. If that is enough for you, go for it. I understand that the newer units have been modified to allow you to do a 0.1C 16 hour charge. The first units out had a 90 minute timer that prevented this. Make sure you pick up one of the later units.

Tom

 

[Handlobraesing]

If you don't mind reduction in resolution, you can first figure out the internal resistance of your DMM + leads by measuring a known current and measuring the voltage present across the probe tips.

Let's say this is 0.100 ohms. I just used simply even values so I don't have to break out my calculator.

Now, you've got OCV of 1.200v and the flash current was 10.0A.

you can then OCV/(k1+k2+n)=I, and solve for n

Set the k as 0.1, k2=0.00(optional series resistor)
OCV 1.20
I=10.0A

Do the math, and n = 0.02 ohm.

The important thing is that you have to know exactly what k is(which is easy to figure out with ohm's law using two DMMs).

Also, the greater the ratio of k to n, lower the resolution, but for battery packs that would put too much current through the DMM you could just use an additional known value resistor, shown as k2.

For my test, I use a 50mV/50A calibrated shunt, so I can directly read out 1mV=1A, then leads to connect to battery.
I know that shunt is 1.00m-ohm and that's where the measurement is taken. Leads will add 10 or 20 m-ohms of resistance, but since flash amp reading isn't that accurate anyways, I will ignore it and do the math assuming instrumentation shunt and its associated leads have no resistance.

So, if I get a flash current of 45A (which I did on a Duracell 2650mAh cell in the past) and it had a pre-measurement voltage of 1.25v, so I know that its internal resistance is ~28m-ohms

 

[eebowler]

I got a Triton 2 charger a couple weeks ago and have been testing some of my cells.

The data isn't with me tonight but, I must ask a question.

I've noticed that for every NiMH battery I cycle, the discharge capacity is ALWAYS lower than the charge capacity. It's true for a 1.66 C(3A) discarge rate for 1800 mAh AA cells and true for a 0.3C(1.5A) discharge rate for 5000 mAh cells. The second discharge capacity is never as high as the first charge capacity and usually within +/- 10% of any other discharge capacity.

If heat generation is the answer, then why does the C size cell discharged at 0.3C exhibit the same behavior?

 

[Handlobraesing]

I got a Triton 2 charger a couple weeks ago and have been testing some of my cells.

The data isn't with me tonight but, I must ask a question.

I've noticed that for every NiMH battery I cycle, the discharge capacity is ALWAYS lower than the charge capacity. It's true for a 1.66 C(3A) discarge rate for 1800 mAh AA cells and true for a 0.3C(1.5A) discharge rate for 5000 mAh cells. The second discharge capacity is never as high as the first charge capacity and usually within +/- 10% of any other discharge capacity.

If heat generation is the answer, then why does the C size cell discharged at 0.3C exhibit the same behavior?

NiMHs don't quite act like a regular bucket. What you're describing is less than perfect coulombic efficiency. No NiMH have a perfect efficiency.

Lithium ion comes pretty close to 100% coulombic efficiency, but the charge voltage curve is higher than the discharge curve, so the charge-energy efficiency is still not 100%. There is no such thing as a battery with 100% charge efficiency.

 

[Mr Happy]

There is no such thing as a battery with 100% charge efficiency.

Someone should invent one. A battery where you get more electricity out than you put in would be really useful.