Measuring cell voltage - under load?

[fireguy]

Ok, I'm thinking aloud here, perhaps you can tell me if this will work ... :thinking:

If you measure a cell's voltage with a DMM, you'll get a higher reading because it's not under load. Using a tester that puts a load on it, you get a more accurate reading.

If I want to determine a cell's true voltage, can I just slip it into a slot on the C9000, select discharge and a low rate such as 200 mAh, then wait for it to cycle to the voltage? Wouldn't this be the voltage under load, and mean that I wouldn't have to buy a tester to do the same function?

 

[Mr Happy]

Yes, I think that can work very well to determine the condition of a cell.

I don't have any charged cells to hand right now, but let me just charge a couple up and I will take some measurements and expand on the concept in a short while...

 

[fireguy]

Yes, I think that can work very well to determine the condition of a cell.

I don't have any charged cells to hand right now, but let me just charge a couple up and I will take some measurements and expand on the concept in a short while...

Thank you! :bow:

 

[Mr Happy]

OK. There is something I wanted to confirm, which is whether you can get different voltages on the C9000 by selecting different discharge currents. I suspected that you cannot.

So I took a couple of fully charged cells. Each measured 1.40 V on a voltmeter. On the C9000 at different discharge rates, these were the initial voltages indicated:

100 mA: 1.30 V
500 mA: 1.30 V
1000 mA: 1.29 V

Why no difference between discharge rates, you may be wondering?

The answer is that the C9000 always uses a 1000 mA discharge current, and measures the voltage under that load. If you ask it to discharge at 100 mA it will use a 1000 mA discharge current with a 10% duty cycle, e.g. 0.2 seconds of 1000 mA followed by 1.8 seconds of 0 mA.

With that out of the way, all is not lost. In fact a 1000 mA load is a piece of cake for an NiMH cell, so doing a short discharge on the C9000 at a requested 100 mA is a good battery test. The C9000 will apply short load pulses of 1000 mA and measure the voltage during those pulses.

Do the test with a cell that has been freshly charged and allowed to rest for two hours. If the cell is in good condition, you should expect to see an indicated voltage in the 1.25 - 1.30 V range, as I saw above. If you see a voltage in the 1.20 - 1.25 V range, the cell is probably in need of refreshing. If you see a voltage below 1.20 V the cell is not in good shape and may benefit from a break-in cycle. If after refreshing and conditioning you cannot get the fresh discharge voltage above 1.25 V the cell is basically no good and should be recycled.

 

[bob_ninja]

Wow Mr Happy that is great idea, thanks

Nice scale, 1.3-1.25-1.2 (will have to label them on my C9K)

You said "allowed to rest for two hours"
What do you think about allowing 1-2 day rest before doing the test? That might reveal which cells developed a high SD, right? Couple of hours would still leave a high SD cell looking good.

I just wonder if I am not in a hurry, how would much longer rest time (days) impact the test, if at all?

 

[PeAK]

In another thread, the determination of internal cell resistance was discussed:

.
.
.
We used -

Internal resistance = (V1 - V2) / (I2 - I1)

and for the special case of using a single resistor and the Open-Circuit voltage

Rb = (V1 - V2)/(I2 - I1)
= (Vs - V) / (I - 0)
​

For your case Vs=1.4, V=1.29 and I=1 amp and the results is that the resistance works out to be:

Rb=(1.4-1.29)/1 = .11 ohms
​

...or 110 milli-ohms. Similarly, a cell with a 1.20V reading would have a corresponding value of resistance of 200 milli-ohms.

I have found this to be a quick and invaluable way grouping batteries which are used in pairs in electronics and the general trend is that larger/longer discharge capacity tends to correlate with lower intenal source resistance. The ones that measure high are used for powering my low powered wireless mouse for about a month at a time between charges....some would say they ready for the recycle bin...for me it's the mouse bin

Seems like you've discovered a hidden feature of the Maha.

It would be better practice to measured the open circuit voltage after making the "under load" voltage measurement.


PeAK

 

[Mr Happy]

I did measure the open circuit voltage after making the loaded measurements and it had returned to within a few millivolts of 1.40 V.

The cells I did the test with are the cheap Chicago Electric ones from Harbor Freight that get quite warm during charging. That is consistent with them having a highish internal resistance as measured here.

 

[Mr Happy]

Incidentally, in response to bob_ninja's question, I just measured the two cells again after an overnight rest. The OC voltage is now 1.375/1.375 and the loaded voltage is 1.27/1.28.

That illustrates how the voltage drops a bit after charging before it stabilizes.

 

[fireguy]

Well, that settles it for me, Mr. Happy. There's no need for me to buy a ZTS tester if I want to measure under-load voltage of an AA or AAA. I wonder what load the ZTS tester would put on the battery when it tests it?

I'm also going to note the voltage ranges you mentioned and put them near the C9000.

Thanks for your research and answer!

 

[Mr Happy]

It is also of significance to note that this test can tell you about the condition or "vibrancy" of a cell, but not about its state of charge. For instance in this thread I discovered an eneloop that measured 1.05 V under load, but this cell was in fact fully charged and sustained a voltage of 1.05 V for a whole 1800 mAh of discharge.

This strange behavior is why it is not practical to use battery testers like the ZTS on NiMH cells to determine their state of charge.

 

[Bones]

Well, that settles it for me, Mr. Happy. There's no need for me to buy a ZTS tester if I want to measure under-load voltage of an AA or AAA. I wonder what load the ZTS tester would put on the battery when it tests it?

I'm also going to note the voltage ranges you mentioned and put them near the C9000.

Thanks for your research and answer!

Or me!

Thanks, gentlemen, especially for working this into such tidy package.

Notes taken.

 

[bob_ninja]

Incidentally, in response to bob_ninja's question, I just measured the two cells again after an overnight rest. The OC voltage is now 1.375/1.375 and the loaded voltage is 1.27/1.28.

That illustrates how the voltage drops a bit after charging before it stabilizes.

Nice, thanks
So it still falls in the 1.3 - 1.25 range = green
That is exactly what I am interested in, cell health test.
Cool

 

[PeAK]

Well, that settles it for me, Mr. Happy. There's no need for me to buy a ZTS tester if I want to measure under-load voltage of an AA or AAA. I wonder what load the ZTS tester would put on the battery when it tests it?

.
.
.

There is probably some correlation between the voltage drop under load (i.e. internal resistance value) and the capacity that ZTS is basing their lab measuring instrument upon. They acknowledge this variation and average the characteristics from many batteries of the same make to come up with a characteristic curve. The more direct (and longer) way is the well know "discharge capacity" function available on a number of high end charger/analyzers.

The best use of the ZTS, in my opinioin, would be to give a grade on the batteries condition in terms of lifecyle...but as many of you have surmize, this can be simply done by measuring the voltage difference obtained from a battery under load and "no load". You've made a wise decision, my son, to forgo the expense of the ZTS.

PeAK

 

[Russel]

Nice, thanks
So it still falls in the 1.3 - 1.25 range = green
That is exactly what I am interested in, cell health test.
Cool

Hmm...

So, an eneloop should hold 1.25 to 1.3 volts under a 1000ma discharge? Or do these figures apply to testing with the MH-C9000 specifically?

You guys got me thinking (I know, that could be dangerous!) and I did a few tests with a CBA II:

I tested one of my Eneloop cells for 30 seconds under 1000ma discharge, then 250ma discharge, then 500ma, and finally 750ma. Each discharge is followed with 30 seconds no discharge to see the recovery voltage.

While testing eight of my Eneloop cells at 1000ma discharge I discovered every cell had aproximately the same voltage drop except cell 2. It didn't drop quit as much. Here is cell 2 compaired with cell 1, cells 3 to 8 were almost identical to cell 1.

Whoo Hooo! this is my first message as Enlightened! 20th post!

 

[Black Rose]

Ok, I'm thinking aloud here, perhaps you can tell me if this will work ... :thinking:

If you measure a cell's voltage with a DMM, you'll get a higher reading because it's not under load.

That depends on the DMM.

I have one of those Innova/iEquus 4320a DMMs that Canadian Tire has on sale regularly.
It has a battery test function that applies a 10 mA load when testing 1.5 & 9V batteries.

Not sure how useful a 10 mA load actually is, but it does produce a different reading than simply using the DCV setting on a DMM.

Nice work Mr. Happy.

Some very interesting info in this thread. I'm going to subscribe to it so I don't lose track of it.

 

[digitor]

You guys got me thinking (I know, that could be dangerous!) and I did a few tests with a CBA II:

The voltages you've got here are so low that I think something is not quite right with your measurement setup. Have a look at the graphs here http://www.candlepowerforums.com/vb/showthread.php?t=79302 for instance, the Eneloop is holding over 1.25 Volts at 1 Amp for most of its discharge.

Cheers

 

[Mr Happy]

So, an eneloop should hold 1.25 to 1.3 volts under a 1000ma discharge?

Yes, it certainly should. Note that this test is to be performed on a freshly charged cell though, and the measurement is to be taken for just a few seconds so that the cell does not get noticeably discharged.

Or do these figures apply to testing with the MH-C9000 specifically?

No, not specifically.

You guys got me thinking (I know, that could be dangerous!) and I did a few tests with a CBA II:

That is good thinking.

I tested one of my Eneloop cells for 30 seconds under 1000ma discharge, then 250ma discharge, then 500ma, and finally 750ma. Each discharge is followed with 30 seconds no discharge to see the recovery voltage...

Well, let's think about these results.

Firstly an eneloop should come fresh off the charger at about 1.45 V, so yours perhaps have been sitting on the shelf for a while after charging. An open circuit voltage of 1.38 V suggests several weeks, or even months, of resting.

Now an eneloop in good condition should have an internal resistance of about 50 milliohms, or 0.05 ohms. Therefore on a 1000 mA load we would expect the voltage to drop by about 0.05 V, meaning 1.45 V drops to 1.40 V, and 1.38 V drops to 1.33 V.

In your test the voltage apparently dropped to 1.05 V. What this says is that your eneloops are in poor condition and have exceptionally high internal resistance. (Edit: Or as digitor says, maybe you have something wrong with the way you are conducting the measurements...?) This is a natural change that happens after a long period of storage, and also may be induced by poor charging technique.

What you need to do here is refresh your eneloops and see if you can make them better. At first you should discharge them right down to 0.9 V at a low rate, then charge them rapidly using a charge current of 1000 mA or more, then repeat the 1000 mA load test. Try up to two more times until no more improvement is obtained. If you are still not getting good results, it is time to discharge down to 0.9 V again and then run a timed conditioning charge of 200 mA for 16 hours. Repeat a second time if necessary.

If all this refreshing does not bring your eneloops back we would have to say they are damaged. Then the question to ask is how did they get damaged and how can you avoid it happening to more eneloops?

 

[Russel]

The voltages you've got here are so low that I think something is not quite right with your measurement setup. Have a look at the graphs here http://www.candlepowerforums.com/vb/showthread.php?t=79302 for instance, the Eneloop is holding over 1.25 Volts at 1 Amp for most of its discharge.

Cheers

If I'm not doing something right with the test my Eneloops arn't in very good shape! [Edit: If my test is right my Eneloops arn't very good]

I need to make a low resistance AA cell holder and try a complete 1000ma discharge test to .9 volts with one of the cells.

I am using a multi cell AA holder with a power pole connector from the CBAII to a 9 volt battery connector to the cell holder with a #16 wire jumper with aligator clips so that I can use only one compartment of the AA cell holder. I guess that means that I have far to many connections, causing excessive voltage drop.

 

[Mr Happy]

If I'm not doing something right with the test my Eneloops arn't in very good shape!

I need to make a low resistance AA cell holder and try a complete 1000ma discharge test to .9 volts with one of the cells.

I am using a multi cell AA holder with a power pole connector from the CBAII to a 9 volt battery connector to the cell holder with a #16 wire jumper with aligator clips so that I can use only one compartment of the AA cell holder. I guess that means that I have far to many connections, causing excessive voltage drop.

Absolutely for sure your test setup is not sufficient. You need to make one of these: http://www.candlepowerforums.com/vb/showthread.php?t=89363

What would make the CBA II better is if it used a four point battery connection. I do not know if it has that facility?

 

[Russel]

Firstly an eneloop should come fresh off the charger at about 1.45 V, so yours perhaps have been sitting on the shelf for a while after charging. An open circuit voltage of 1.38 V suggests several weeks, or even months, of resting.

Yes, the cells have been sitting for about 6 or 7 weeks from the last time they were charged.

Now an eneloop in good condition should have an internal resistance of about 50 milliohms, or 0.05 ohms. Therefore on a 1000 mA load we would expect the voltage to drop by about 0.05 V, meaning 1.45 V drops to 1.40 V, and 1.38 V drops to 1.33 V.

In your test the voltage apparently dropped to 1.05 V. What this says is that your eneloops are in poor condition and have exceptionally high internal resistance. (Edit: Or as digitor says, maybe you have something wrong with the way you are conducting the measurements...?)

I would venture to say that I'm causing too much voltage drop with the number of connections in my setup. I count 8 connections from the CBA II through everything and back to the CBA II. That may be my problem or at least part of it.

If all this refreshing does not bring your eneloops back we would have to say they are damaged. Then the question to ask is how did they get damaged and how can you avoid it happening to more eneloops?

First I would like to try a better test setup with a maximum of 4 good connections. Two at the powerpole connector and two at the cell. If that doesn't make any significant differance, I'll refresh the batteries and see how they do then. In order to test my test setup I want to leave the batteries as is and see if the change to my test setup makes a differance.

Russ