Voltage = performance?

[entoptics]

Uber newb here. Tried the search, but my question may be too dumb (and use too many common terms) to get useful hits.

So here it is...When I pull my cells out of the charger, is it reasonable to assume the highest voltage cells are going to last the longest?

In other words, with all other things being equal, does voltage correlate to stored energy in the battery, or are the two independent of each other.

I have 30 billion NiMH cells of various ages and usage, and would like to go through them and cull out the ones that are likely to crap out quick, and separate the ones that will likely be the highest performers.

I only have a DMM, and I have no intention of buying more equipment. My DMM does have a setting labeled "Battery load test" which reads in DC Volts and has little green, yellow, and red LEDs which indicate good, bad, and ugly, but I don't really know what that is actually testing beyond voltage.

Can voltage alone tell me what I need to know?

 

[Ragiska]

for nimh cells, no, open circuit voltage doesn't tell you much of anything.

except for li-ion cells, there is no easy or reliable way to determine the instantaneous charge state or capacity of a cell.

 

[entoptics]

Well crap.

Any good methods for testing then? I'm doing a lot of beam shot time lapse videos lately, but I'd rather not do 30 billion of them.

Any other ways to "quickly and efficiently" test cells?

 

[Jack Reacher]

Voltage ‡ performance...

I'm a (relatively speaking) battery newbie too, and I'd probably go for something like this, particularly for a "quick and dirty" test of a large number of cells:

http://servaas.com.au/maha-powerex-aa-aaa-9v-battery-tester-p-120.html

Batteries can be more — or less — serviceable considering their current draw characteristics, over time, and as Ragiska has said, voltage alone doesn't tell you much about a cell's charge state or the efficiency of its internal chemistry — unless of course it's totally stuffed, and unservicable! Batteries (generally speaking) need to be loaded to see how well they're working — or not.

— Jack.

 

[march.brown]

Re: Voltage ‡ performance...

"Why do today what you can do tomorrow?"
— Jack.

Why do today or tomorrow if you couldn't be bothered to do it yesterday.

 

[InHisName]

Well crap.

Any good methods for testing then? I'm doing a lot of beam shot time lapse videos lately, but I'd rather not do 30 billion of them.

Any other ways to "quickly and efficiently" test cells?

Yes, go buy a 10 ohm resistor from Radio Shack or whom ever you want. Cost should be less than $1. Then connect it up between the lead of the voltmeter and measure the voltage across the resistor that the battery can deliver. OH and be sure batts are fully charged to have enough meaning.

Good 1.22 and up
medium 1.1 - 1.21
Ugly less than .9 for sure

Many others will want to use slightly different numbers. That's ok. The essence is biggest numbers are good, medium are ok, and smallest are UGLY.

 

[HKJ]

In other words, with all other things being equal, does voltage correlate to stored energy in the battery, or are the two independent of each other.

...

Can voltage alone tell me what I need to know?

The best way is to buy a MAHA MH-C9000 charger, it can analyse the batteries for capacity, but it will take a few hours for each set of 4 batteries. If the C9000 is only used for discharge and you use some other charges to fill the batteries, you can do it faster.

Another way, that will only tell if the battery works, not how much capacity it has, is to measure the voltage with a load on the battery. I have written some instructions here.

 

[TakeTheActive]

Uber newb here...

CLICK on my Sig Line LINK for LOTs of useful, informative "Newbie" LINKs.

...Tried the search, but my question may be too dumb (and use too many common terms) to get useful hits...

Reading FAQs will give you a 'Knowledge Foundation' and enable you to chose proper SEARCH terms to further build your understanding.

...When I pull my cells out of the charger, is it reasonable to assume the highest voltage cells are going to last the longest?

No. The cell's Internal Resistance plays a big part in determining its ability to deliver current under various loads.

...I have 30 billion NiMH cells of various ages and usage, and would like to go through them and cull out the ones that are likely to crap out quick, and separate the ones that will likely be the highest performers.

I only have a DMM, and I have no intention of buying more equipment. My DMM does have a setting labeled "Battery load test" which reads in DC Volts and has little green, yellow, and red LEDs which indicate good, bad, and ugly, but I don't really know what that is actually testing beyond voltage...

"30 billion NiMH cells of various ages and usage..." is quite a lot! Why did you buy so many?

To start off, you could probably use your DMM "Battery Load Test" (equivalent to the standalone 'Tester' suggested by Jack Reacher) to identify the REAL duds. If you could post the 'Amount of Load / Current Draw' (printed either on the DMM or in the manual), it would help. The 10 ohm resistor suggested by InHisName is low (i.e. won't tax *CRAP* cells). Using the formula V=I*R, at 1.2VDC a 10 ohm load will only draw 120mA.

Using the DMM instructions linked to by HKJ, duplicate the 'Experiment' performed by Mr Happy in Ni-Zn from UltraLast choosing the appropriate load resistors to calculate 0.5C and 1.0C for *YOUR* capacity cells. The results will give you a idea of your good performers without buying any additional equipment besides a few resistors.

 

[entoptics]

Excellent info guys.

The "Battery Load Test" setting on my DMM (INNOVA 3320) has a "Load Current" of ~10 mA according to the owner's manual. Using V=IR, am I correct in assuming that corresponds to a 150 ohm resistor for the 1.5 V setting?

I'm not sure what the "C" you are referring to is (0.5C and 1.0C). My googling got me to Peukert's Law, but there seems to be a constant (k or n depending on the equation source) that I don't know how to get at.

Also, doing the loaded discharge experiment appears to require me to sit and stare at my DMM for hours at a time, or am I missing the trick?

 

[TakeTheActive]

...The "Battery Load Test" setting on my DMM (INNOVA 3320) has a "Load Current" of ~10 mA according to the owner's manual. Using V=IR, am I correct in assuming that corresponds to a 150 ohm resistor for the 1.5 V setting?

Maybe. Then, at 1.2VDC a 150 ohm (Ω) load will draw 8mA - not very useful.

You want to use resistors like 2.2 Ω, 1.1 Ω and 0.6 Ω in the 'Experiment'.

...I'm not sure what the "C" you are referring to is (0.5C and 1.0C)...

"C" is C)APACITY, which is explained in the FAQ under Definitions.

...doing the loaded discharge experiment appears to require me to sit and stare at my DMM for hours at a time, or am I missing the trick?

And that's why folks buy Smart Charger/Analyzers or 'Data-Logging' DVMs.

But, if you put a 1.2 Ω load on an 'unknown condition' 1.2VDC 2000mAh cell, drawing 1000mA (0.5C), HOLDING 1.2VDC 'Under Load' will give you a quick indication. :thumbsup:

 

[InHisName]

You want to use resistors like 2.2 Ω, 1.1 Ω and 0.6 Ω in the 'Experiment'.

You/ve inspired me. Here's my idea....

1. Get a battery holder or build your own clamping arrangement. 
2. Get those three resistors: 2.2, 1.1 and 0.6 ohms, 3 momentary switches, some wire and some solder.
3. Wire up things as below 'ascii' diagram below:
4. Measure 4 values: open circuit, Low quality(2.2), medium q(1.1) and high q(0.6)
5. Look at diff oc-1 < 0.9, 1-2 < 0.02 2-3 < 0.3
6. Get a 'baseline' by measuring 4 good eneloops or duraloops.
7. Compare all your fully charged cells against the eneloop values.
8. junk cells fail all three with too big differences.
9. low cells pass 1st one
10. med cells pass 1st two
11. high cells pass them all
12. I use junk cells in 1 celled 10 lumen lights, clocks, and remotes.
13. Test which items you can get by with and mark them.
14. Diagram:  ________________________________
              |       |       |       |      |
              DMM    0.6     1.1     2.2    bat
              |       |       |       |    hldr
              |      sw       sw      sw     |
              --------------------------------
15. these are momentary switches: push to contact, not push to open circuit.

I like it so much, I think I'll build it too.

 

[TorchBoy]

I only have a DMM, and I have no intention of buying more equipment.

Ah, but think how much you could use something like... a DMM with a temperature probe! :tinfoil: (Unless your DMM already has one.)

 

[MrGman]

all you really need is one resistor in about the 3 to 5 ohm range. Testing all the NiMH across the same load and match up the ones with the highest voltage under load to each other in the set quantities you need to make whatever gizmo work. Any freshly charged battery that isn't holding at least 1.05 V under load get rid of, simply not worth the time. The load current doesn't have to be up to 1 amp in order to tell if the batteryis good. 240 mA to 500mA for 10 seconds is plenty enough to determine if they are going to be useable. Just put all the best ones together first and be amazed at how the total performance in whatever they go in will go up.

 

[HKJ]

all you really need is one resistor in about the 3 to 5 ohm range. Testing all the NiMH across the same load and match up the ones with the highest voltage under load to each other in the set quantities you need to make whatever gizmo work. Any freshly charged battery that isn't holding at least 1.05 V under load get rid of, simply not worth the time. The load current doesn't have to be up to 1 amp in order to tell if the batteryis good. 240 mA to 500mA for 10 seconds is plenty enough to determine if they are going to be useable. Just put all the best ones together first and be amazed at how the total performance in whatever they go in will go up.

But this test will not show if it is a good 2000mAh or good 2500mAh battery.

 

[MrGman]

But this test will not show if it is a good 2000mAh or good 2500mAh battery.

it should already been known what the ampere hour rating is on the battery and to keep like batteries the same, and simply match peak voltage under load for the same battery rating types. This is not meant as a sit there and watch for complete discharge of each battery and determine their ampere hour rating the hard way.

if the batteries are not labeled or some how that information is lost that you have various ampere hour rating batteries mixed together then you really need get the MAHA charger/discharger and let that test them. To sit there and watch batteries discharge from 1 to 2.5 hours each under a 1 amp load isn't worth the cost of the batteries in a person's time. If you don't trust the batteries are really the ampere hour rating that they say they are and want to test them for complete discharge ampere hour capacity then you should still get the MAHA unit.

I always keep my various brand and ampere hour rating batteries separated from each other and simply test each type under load and put the best of that type together in matched sets for the specific electronic device they go to. I do that with my Alkalines too for things like my RC controllers. never had a problem.

30 years ago I built a switchable load box to go from 270 ohms down to 0.25 ohms load over 16 steps. Testing, testing, and more testing. Wasn't really worth the time I wasted. Something around a 400mA load for AA batteries in the same category whill tell you which ones are holding the best charge and simply put them together in a group.

Otherwise if you really want to get picky about it then use the electronic device that will do all that for you in such a way that you don't have to sit there and baby sit it and waste your time and it will more than pay for itself over the long run unless your time is really only worth about $5.00 an hour or less.

 

[45/70]

There have been some good suggestions presented here. I think the Maha is probably the best and most practical solution, as you get a really good charger for NiMH and NiCd cells out of the deal, as well.

The simplest solution I think though, is to use the old fashioned, dirt simple, basic, pre digital era, "is it crap, or is it usable" test, "flash" Amps. It's not very scientific and it won't give you results usable for a studied comparison, but you will be able to weed out the really bad cells. I might add here, that this method should not be used for testing lithium, or lithium ion cells. Not only might this be potentially hazardous, but the results testing these chemistry cells in this way, is less accurate.

To test for flash Amps, you need a DMM, or ammeter capable of at least 10A, preferably 20A, for AA cells. Also it's really easier to use an analog meter, but a digital one will work.

To execute the test, make sure your test leads are hooked into the proper jacks for testing 10A, or 20A. On most multimeters this means the black lead into "COM" and the red lead into "10A", or "20A" (instead of the usual "+" jack) depending on your meter. Also, make sure any switches and/or dials are setup for testing this range.

Then, while watching the meter, you simply hold the test subject in one hand with the black lead held to the bottom (- side) of the cell with your finger(s) of the same hand, and touch the red lead to the top side (+) with your other hand. With a good cell, things will heat up here pretty fast, so you don't want to do this for very long, maybe 5 seconds max. Also, with a lot of modern cells such as LSD's or high discharge rate cells, even a 20A capable digital meter will immediately read "out of range" or whatever, or an analog meter's needle will "peg". With such cells, you want to disconnect the "+" lead immediately. For all practical purposes, these cells "pass" the test.

What you're looking for is "bad" cells. You will have to decide for yourself which cells these are by comparing your "results" with cells that peg the meter etc. A bad cell often, will initailly read fairly high on the scale, but almost immediately start dropping off to a much lower level. These are the cells that suffer from voltage depression etc. and are the ones suspect to needing to be replaced, or at the very least, an attempt to recover them is necessary.

Normally, your meter's internals won't get too hot with this method, as it takes long enough while your fiddling around when switching cells to be tested, that things cool off. You might give it a rest every few "tests" though to make sure you don't burn it out, just in case. Also, testing cells in this manner is hard on the cells, as they were not designed to be short circuited (which is pretty much what you're doing), but if you keep the "test" period short and don't make a habit of it (as I'm sure some here will anyway ), you should be OK.

Again, this is a crude way to weed out the worst cells in a group. You won't be able to categorize cells into twenty different levels of usefulness, or anything like that, but it is a good way to get a general idea of the condition of your cells, and pick out the really bad ones.

Dave

 

[MrGman]

There have been some good suggestions presented here. I think the Maha is probably the best and most practical solution, as you get a really good charger for NiMH and NiCd cells out of the deal, as well.

The simplest solution I think though, is to use the old fashioned, dirt simple, basic, pre digital era, "is it crap, or is it usable" test, "flash" Amps. It's not very scientific and it won't give you results usable for a studied comparison, but you will be able to weed out the really bad cells. I might add here, that this method should not be used for testing lithium, or lithium ion cells. Not only might this be potentially hazardous, but the results testing these chemistry cells in this way, is less accurate.

To test for flash Amps, you need a DMM, or ammeter capable of at least 10A, preferably 20A, for AA cells. Also it's really easier to use an analog meter, but a digital one will work.

To execute the test, make sure your test leads are hooked into the proper jacks for testing 10A, or 20A. On most multimeters this means the black lead into "COM" and the red lead into "10A", or "20A" (instead of the usual "+" jack) depending on your meter. Also, make sure any switches and/or dials are setup for testing this range.

Then, while watching the meter, you simply hold the test subject in one hand with the black lead held to the bottom (- side) of the cell with your finger(s) of the same hand, and touch the red lead to the top side (+) with your other hand. With a good cell, things will heat up here pretty fast, so you don't want to do this for very long, maybe 5 seconds max. Also, with a lot of modern cells such as LSD's or high discharge rate cells, even a 20A capable digital meter will immediately read "out of range" or whatever, or an analog meter's needle will "peg". With such cells, you want to disconnect the "+" lead immediately. For all practical purposes, these cells "pass" the test.

What you're looking for is "bad" cells. You will have to decide for yourself which cells these are by comparing your "results" with cells that peg the meter etc. A bad cell often, will initailly read fairly high on the scale, but almost immediately start dropping off to a much lower level. These are the cells that suffer from voltage depression etc. and are the ones suspect to needing to be replaced, or at the very least, an attempt to recover them is necessary.

Normally, your meter's internals won't get too hot with this method, as it takes long enough while your fiddling around when switching cells to be tested, that things cool off. You might give it a rest every few "tests" though to make sure you don't burn it out, just in case. Also, testing cells in this manner is hard on the cells, as they were not designed to be short circuited (which is pretty much what you're doing), but if you keep the "test" period short and don't make a habit of it (as I'm sure some here will anyway ), you should be OK.

Again, this is a crude way to weed out the worst cells in a group. You won't be able to categorize cells into twenty different levels of usefulness, or anything like that, but it is a good way to get a general idea of the condition of your cells, and pick out the really bad ones.

Dave

this is a good way to get people who don't know what they are doing all that well to damage their meters or batteries. Possibly have a battery explode in their face. This is a ridiculous test to recommend to inexperienced people. Even though you said not to do it on Lithium Cells the warning may be ignored or overlooked. I wouldn't even risk doing this type of test with my equipment. And I have been testing electronics and batteries now for over 35 years. Basically you are using the shunt resistor in ammeter itself and the wires as the only load resistance to the battery and see if it will deliver extremely high current into a short as a load. On good meters that can test to 10 or 20 amps the shunt sense resistor is only 10 milliohms resistance. The leads are more than that, but total loading is usually less than 0.040 ohms to the battery.

A far more sensible thing to do would be to build a 0.100 ohms 10 watt rated load resistor to put across one individual cell and test it while monitoring the voltage with the multimeter across the load resistor. No risk of damaging the meter or popping a fuse. But its still unnecessary to do such high current testing to find weak cells for AA batteries.
If they are really that bad then testing them with a 0.50 ohm load 5 watt resistor connected to a volt meter and the battery for a minute will tell you plenty, if they aren't holding a charge the battery will fade fast enough. these are for AA NiMH batteries.

 

[45/70]

this is a good way to get people who don't know what they are doing all that well to damage their meters or batteries. Possibly have a battery explode in their face.

Perhaps you are right, MrG. Although I don't see any consumer size NiMH cells exploding, in such a short duration. Many cells are shorted accidentally, for longer periods than this, I suppose it would be possible though. At the same time, a lot of people, including myself have done this admitedly, "crude" test many times over the years and never had a problem with cells, or meters. You're right though, someone could get into trouble not following the procedure properly, eg. testing very large cells, or testing for too long etc, but I was referring to consumer type AA NiMH cells, as an example.

I'll admit with better ways to test for under performing cells at hand, I don't do this much anymore. And, no I probably wouldn't use a Fluke 8808A to run the tests with. I was under the impression that entoptics was looking for an easy way to sort through many cells without buying, fabricating, or otherwise getting deeper into the problem he is faced with already, and I just thought I'd mention it.

Dave