# Home battery tester: measures voltage or current?

I am in the UK. I have a battery tester from years ago which is still available. It may also be sold in the US.
http://www.avenuesupplies.co.uk/getimage.php?id ˜&type=1&format=2
My multimeter shows that this battery tester puts a load of 500 mA on the 1.5v battery under test.
I have alkaline, NiCad, and NiMH batteries. I have AA and AAA.
(1) Almost all gets a steady reading of 'GOOD' in green. (2) No battery goes to 'REPLACE RECHARGE' in red. (3) One battery starts in GREEN, then slides into RED over 10 seconds.
Is this tester measuring: (a) the general "health" of the battery (b) the battery's current state of charge?
Presumably (a) could be done crudely by displaying current and (b) by displaying voltage? Is this correct?
âœ–
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>

Wont load here. http://www.avenuesupplies.co.uk/index.php?id `2&pid˜&sid=1 does tho.

Not ever, even when flat ?

What happens to the voltage of that battery when it does that ?

Not even possible with the NiCad and NiMH batterys with something that cheap.

Not feasible to work that out with alkaline, NiCad and NiMH batterys.

Nope, its actually displaying the voltage under load.
âœ–
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
On Thu, 6 Jul 2006 16:02:19 +1000, "Rod Speed"

They place a fixed load, and test for voltage. The problem is that with the batteries the world has been using, the charge/discharge curve is a long straight line right up until it avalanches with a pretty quick slew to near zero volts.

I'll bet that repeat tests will have the cycle occur more quickly each time. That battery is likely near discharge.

The meter is reading the voltage through a fixed load, so I'd say it's the voltage that is dropping. When removed from the load, certain batteries "bounce back" a bit. This is only by voltage, and it will again fall under loaded tests, eventually yielding no further rebounds of significance.

To be certain.

Loaded testing has always puzzled me with batteries, as they are limited fill storage devices. Upon a loaded test, I find myself wanting to "top off" the battery again. Also, as you stated, performing such tests doesn't really reveal charge level. At the rates modern batteries are slow charged, the best solution to one's unknown battery condition is to "top 'em off". In other words, put them back in the charger and let its built in detection routine decide the battery's fill level. Those "watchdog" chips are pretty cool stuff.

Yep.
For charge rate, and or fill level, one would have to know the physical characteristics of the battery under test.
It's fully charged internal resistance, and it's internal resistance right at the discharge avalanche point, and its very nearly fully discharged internal resistance. With this knowledge, one can test a battery while it is being charged at a known fixed current limited rate. It can be determined by knowing the at rest fully charged cell voltage, and comparing it to the voltage required to get the battery to take charge at any given point during a charge cycle (not including fully charged of course).
With these chargers being current limited, what takes place is that the voltage is just above that required to pump electrons into the battery. Any more and the charge rate current would be exceeded.
So, a dead battery impresses a very small voltage, or emf when near dead. Hook up a smart charger, and it will raise its voltage until it just starts to pump current into the battery, it will bring it up to the current limit rate and slowly raise the voltage as the battery charges up keeping said current rate steady as she goes. At some point near the end of the charge cycle the chip is programmed to provide, the voltage of the battery will no longer continue to rise and the current will begin to fall off. The chip will sense this and change the charge indicator to green and discontinue charging operations. Usually, in chargers, there is a chip for each battery. I have even seen batteries themselves with them built in. Particularly when there is an array, or true "battery" of cells arranged, and designers want to insure that all cells get charged evenly and fully. Memory effect got shot in da head.
âœ–
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>

What I said below.

Its nothing like near zero, and that is why I asked if it never shows red even when the battery is flat.

Which is why I asked if the others ever show red.

Yes, but its rather surprising that only one battery ever does that.

It shouldnt. Its pretty pointless just measuring the unloaded voltage with many battery types.

You dont normally leave it on the tester for long.

Sure, but that 'tester' is clearly a very crude approach to battery testing.

Not feasible with a simple cheap tester like that one tho.

Sure, but thats an entirely separate matter to what that cheap battery tester can do.
âœ–
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
On Thu, 6 Jul 2006 18:55:32 +1000, "Rod Speed"

The activity doesn't puzzle me, the logic does.
Even a loaded test doesn't tell one a battery is full. It can only tell one something if one witnesses the battery in its death throws.
So to me, it is pointless at any point during a battery's life as the thing is LIMITED, why take more away from a fixed amount?
A simple voltage test, and knowledge of the battery tells a lot. If it is a 1.2 or a 1.5 volt battery, and you are reading that or even a bit more, you know the dang thing is charged.
I have seen NiCads in particular discharge with an incremental drop in voltage through the cycle. I can tell when it is charged too. I was using an HP lab supply with a 4.5 digit meter on it and very precise constant current or constant voltage modes.
Really all one needs is a good meter and a lab supply. Read the battery voltage. Set the supply to just over that, say a half a tenth of a volt to start. Set the current limit on the supply to the battery maker's spec. Place the battery on the supply, and read voltage with the handheld, reading current on the supply meter. If the power supply is set to the battery spec charged voltage, and there is no current flow, it is charged. Turn up the voltage to see the current rise up to the limit point, and read the supply voltage. It should not be much more than the correctly charged battery voltage is.
Remove battery from charger, and let it sit for a few minutes to let its internal thermals re-homogenize. Read battery voltage with meter. Make a report that has each battery's serial number and fully charged and settled battery voltage (we should really be saying "cell" Replace the word "battery" with "cell" throughout this post). Do that with all of your CELLS. Any time you want to know a CELL'S condition, read its voltage, and refer to the chart. The closer it is to that value, the closer to fully charged it is. The increment is very small.
One could also find out the rate the battery is, and discharge what would be half that, and take readings on each battery, and log the value. Then one could extrapolate charge level from voltage reading fairly well. They avalanche late in their duty cycle, but the line from the beginning to the avalanche point is a smooth, slow, small decline over a couple tenths of a volt. Meter needs to read hundredths at least.
âœ–
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>

The logic is fine. Its essentially allowing for the internal resistance.

Yes, but it can certainly tell you when the internal resistance is high.

Nope, it also tells you when the internal resistance is high.

Basically you are taking bugger all of that fixed amount for the test.

Doesnt tell you anything about the internal resistance.

Yes, but there is more involved than just charged/not fully charged.

And that cheap tester is nothing like that.

That costs a tad more than that tester does.

That costs a tad more than that tester does.

That costs a tad more than that tester does.
âœ–
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
On Thu, 6 Jul 2006 19:31:44 +1000, "Rod Speed"

The only time you need to worry about that is with a defective cell.
âœ–
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Yes, but any decent battery tester should do that.
Particularly when it doesnt cost any more to test it under load.
âœ–
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
On Fri, 7 Jul 2006 03:06:31 +1000, "Rod Speed"

It does cost. It costs "juice". There is only so much in there.
âœ–
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Bugger all if you dont leave it on the tester longer than you need to.
âœ–
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
On Fri, 7 Jul 2006 03:06:31 +1000, "Rod Speed"

Wrong. Internal resistance is a property of ALL batteries, including brand new, fully charged, perfectly working ones. ALL OF THEM. The load current must be weighed against the particular cell's resistance to determine what the acceptible voltage drop will be. The cost is that cells may be deemed more drained than they actually are and discarded prematurely. In other words, as it applies to this thread, 500mA is a rather high current for smaller alkalines, but more appropriate for mid-sized. Likewise, it's way too low for a battery with even lower resistance like a car battery or typical laptop battery.
Anyone who's tried to use Alkaline batteries in a digital camera may have observed this, the batteries don't seem to last very long because their internal resistance is high relative to the current consumed by the camera.
âœ–
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
We'll see...

Duh. What matters is whether the internal resistance has increased substantially so that it can no longer deliver the voltage it should deliver with a decent load, particularly when the batterys are used to power the higher current devices.
Just measuring the unloaded voltage doesnt tell you anything about batterys that have gone higher resistance.

And thats not feasible with as simple a battery tester as the one being discussed. It certainly makes more sense to load the battery than not load it at all tho.

Yes, but thats better than claiming the battery is fine when its internal resistance has increased substantially and it will no longer supply the voltage its supposed to supply when loaded.

Not for the devices that have motors in them etc.

It clearly isnt intended to be used for either of those.

Separate matter entirely to whether its better for a very cheap battery tester to measure loaded or unloaded batterys.
âœ–
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
On Sat, 8 Jul 2006 04:37:58 +1000, "Rod Speed"

Pretty much splitting hairs there, fact is that on a perfectly working cell it's impedance is an issue when driving relatively high current loads, it is not a matter of defect alone, it is a factor in every use of brand new perfect cells too.

never suggested it did. The load has to be appropriate for the battery chemistry, design and size and the margin for error rapidly goes up as one deviates from that ideal.

never claimed otherwise, but only randomly thinking "load" vs "no load" is pretty irrelevant, the ACTUAL load has beconsidered (by either the tester or designer of the test equipment).

not better or worse overall, depends on the criteria. Some want most lifespan, or lowest cost, or longest runtime per outting, several different things can matter most.
The one thing that should always be avoided is unreliable indication of the battery state so one can't even make a reasonable attempt to choose their own strategy for replacement or recharge intervals.

Yes for devices that have motors. While the motor might load it more, that does not matter! To indicate the battery state it is necessary to load it appropriate to it's size, design and chemistry.
Smaller alkalines shouldn't be used in any high current motorized device to begin with, but if they're used, there will be a voltage depression from the higher current load (same as when tested), a voltage that will RISE AGAIN after the device is turned off, relative to what it was when last running. This completely invalidates a high (relative) current test on those cells, unless your only goal was not testing the battery at all, rather qualifying for particular device usage. Toward that end it is a fair test, a determination of alkaline (for example) is fit at all, or better to use something like NiCad.
âœ–
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>

Nope.
Nope, the device obviously has to be designed to allow for the internal resistance of a good cell.

Only for the designer, not for the individual deciding if a particular battery is good or not.
Who's mindlessly hair splitting now ?

Never suggested you did.

And that cheap battery tester clearly isnt even attempting to consider any of that.

Never claimed you did.

Nope, not with a cheap battery tester like that.

Not even possible with a cheap battery tester like that.
ALL the designer of a cheap battery tester like that gets to do is either measure it loaded or unloaded and to use what he considers to be a useful load that will indicate whether the battery can deliver that sort of load fine.

Gets sillier by the minute. The criteria with a cheap battery tester like that, there is just ONE criteria, is the battery still usable or isnt it.

Not WITH A CHEAP BATTERY TESTER LIKE THAT.

And measuring the battery loaded is the best way to do that.

Corse it does when testing to see if the battery can deliver the voltage it should do with that load.

Not even possible with very cheap battery testers like that.

Tell that to the designers who do just that.

Irrelevant to what is feasible with as cheap a battery tester as that.

No it doesnt, it shows that the battery can deliver that current fine. And so will handle a lighter load fine too.

Its clearly a battery tester.

That aint what that particular battery tester is about.
âœ–
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
On Sat, 8 Jul 2006 10:17:10 +1000, "Rod Speed"

Obviously you haven't a clue. The devices are designed for a target battery type. "Some" can be used with another type but it's not the target. Take a laptop, did you think you could chuck out the Li or NiMH and put some Alkaline in there and use it? Good luck, you'll need it. Too high a load for alkaline.

The individual putting a load on a cell has to use a load causing a current draw that is reaonably compatible with what the battery can provide. To do othewise will result in a misleading reading.
âœ–
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>

Never ever could bullshit its way out of a wet paper bag.

Not those very cheap battery testers.

Corse it is with those very cheap battery testers.

No thanks, those very cheap battery testers dont even claim to test those batterys.

Completely and utterly irrelevant to what those very cheap battery testers test battery technology wise.

And the load that particular battery tester puts on the individual battery is fine in that regard.
âœ–
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
On Thu, 6 Jul 2006 18:55:32 +1000, "Rod Speed"
Absolutely.
âœ–
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
On Thu, 6 Jul 2006 18:55:32 +1000, "Rod Speed"
Any amount is still less than it had.
âœ–
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
âœ–
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
On Thu, 6 Jul 2006 19:32:43 +1000, "Rod Speed"

I am aware of that.