Battery matching solution?

[Bruce W.1]

Please see previous thread.

I'm trying to match AA battery capacities so they can be used together
in battery packs and I think created a low-tech way to do this. I'd
like to hear more informed opinions.

This would apply to NiCads and NiMh but it should work with lithium-ions
too.

Take all your batteries of the same type and charge them all in parallel
to the same voltage. Then one at a time remove them from the float
voltage and attach them to a fixed size resistor (say 6 ohms) for a
fixed period of time (say 3 hours) then measure their voltage. Well
matched batteries should have similar voltages.

Does this wash?

The only thing I wonder about is the full charge state voltage. Say I'm
doing Ni-Cads which usually top-out around 1.65 volts (where they
experience a voltage drop), and I set the float voltage for the bunch at
1.5 volts. Would the maximum voltage of each cell make a difference
here?

Thanks for your help.

[Roy Lewallen]

What's really important for matching in a battery pack is that the cells
have very close to the same capacity. If one has lower capacity than the
others, the others will reverse charge it after it discharges. This
further reduces its capacity, making it go flat even sooner the next
time. The absolute cell voltage at any point in the discharge process is
unimportant.

Unfortunately, you can't discern the remaining capacity of a NiCd or
NiMH cell by looking at the cell voltage. The only way to accomplish the
matching is to discharge each cell until its energy is essentially
completely gone (i.e., to 1.0 volt at moderate drains), and select cells
on the basis of how long they take to get there. Ideally, the discharge
rate would be about the same as in the device you intend to use the
battery for. But in practice, the capacity of NiCd and NiMH cells isn't
a very strong function of the load current, so you've got quite a bit of
leeway.

Many years ago I made up a discharger for this purpose. It's basically
an adjustable constant current load with an adjustable voltage
comparator. When the voltage reaches the comparison voltage, it
disconnects the load. An old analog electric clock is used to record the
time -- the comparator also removes power from the clock. Initially set
at 12:00, it directly reads the time it took for the cell to discharge
to the specified voltage level. These days it would make a lot more
sense to use an A/D converter computer board and/or a microcomputer chip
of some sort.

Roy Lewallen, W7EL

Post by Bruce W.1
Please see previous thread.
I'm trying to match AA battery capacities so they can be used together
in battery packs and I think created a low-tech way to do this. I'd
like to hear more informed opinions.
This would apply to NiCads and NiMh but it should work with lithium-ions
too.
Take all your batteries of the same type and charge them all in parallel
to the same voltage. Then one at a time remove them from the float
voltage and attach them to a fixed size resistor (say 6 ohms) for a
fixed period of time (say 3 hours) then measure their voltage. Well
matched batteries should have similar voltages.
Does this wash?
The only thing I wonder about is the full charge state voltage. Say I'm
doing Ni-Cads which usually top-out around 1.65 volts (where they
experience a voltage drop), and I set the float voltage for the bunch at
1.5 volts. Would the maximum voltage of each cell make a difference
here?
Thanks for your help.

[Bruce W.1]

Post by Roy Lewallen
What's really important for matching in a battery pack is that the cells
have very close to the same capacity. If one has lower capacity than the
others, the others will reverse charge it after it discharges. This
further reduces its capacity, making it go flat even sooner the next
time. The absolute cell voltage at any point in the discharge process is
unimportant.
Unfortunately, you can't discern the remaining capacity of a NiCd or
NiMH cell by looking at the cell voltage. The only way to accomplish the
matching is to discharge each cell until its energy is essentially
completely gone (i.e., to 1.0 volt at moderate drains), and select cells
on the basis of how long they take to get there. Ideally, the discharge
rate would be about the same as in the device you intend to use the
battery for. But in practice, the capacity of NiCd and NiMH cells isn't
a very strong function of the load current, so you've got quite a bit of
leeway.
Many years ago I made up a discharger for this purpose. It's basically
an adjustable constant current load with an adjustable voltage
comparator. When the voltage reaches the comparison voltage, it
disconnects the load. An old analog electric clock is used to record the
time -- the comparator also removes power from the clock. Initially set
a_pag:00, it directly reads the time it took for the cell to discharge
to the specified voltage level. These days it would make a lot more
sense to use an A/D converter computer board and/or a microcomputer chip
of some sort.
Roy Lewallen, W7EL

========================================================

Seems the RC car community matches batteries based on a number of
criteria like capacity, internal resistance, and average voltage. The
websites I found on this don't give specifics.

Buying a $200 pieces of test equipment to measure all this is slight
overkill. I think I came up with a way. What to you think of this?

Charge all the batteries in parallel to the same voltage, say 1.5
volts. Then partially discharge them all in series. At this point
quickly measure the voltage of each cell while still under load.
Matched cells would have the same voltage.

[Tarmo Tammaru]

Why charge them in parallel? Having them all charge the same in the real
world should be part of the matching. In fact, if you start out with all
batteries in the same state, and charge them in series, you can quickly pick
out oddball by measuring the charging voltage across each cell.

Tam/WB2TT

Post by Bruce W.1
Please see previous thread.
I'm trying to match AA battery capacities so they can be used together
in battery packs and I think created a low-tech way to do this. I'd
like to hear more informed opinions.
This would apply to NiCads and NiMh but it should work with lithium-ions
too.
Take all your batteries of the same type and charge them all in parallel
to the same voltage. Then one at a time remove them from the float
voltage and attach them to a fixed size resistor (say 6 ohms) for a
fixed period of time (say 3 hours) then measure their voltage. Well
matched batteries should have similar voltages.
Does this wash?
The only thing I wonder about is the full charge state voltage. Say I'm
doing Ni-Cads which usually top-out around 1.65 volts (where they
experience a voltage drop), and I set the float voltage for the bunch at
1.5 volts. Would the maximum voltage of each cell make a difference
here?
Thanks for your help.

[Kavall]

Battery matching.

Here is how I would go about it. Charge the cells individually untill
their voltage starts to drop, aka. peak charge them. Measure the peak
voltage. Then hook them to a fixed load and measure their discharge
time to 0.9V. Let the cell rest. The next day repeat. Take the
measurements and compare close discharge times and close peak
voltages. If you could measure internal resistance it would be nice,
but I don't know of a simple way. In my opinion the important
qualities are.
1. Close discharge times
2. Close and low internal resistances
3. Close peak voltages

[Joshua K Drumeller]

The internal resistance of the battery changes as the battery is discharged.
Also things like temperature, age and other factors effect it.
I think the proper way to test your battery is to fully charge it to the
manufacturer specification and discharge it at a given amp hour rate.
Compare your discharge curve with the manufacturer. Choose the same
discharge rate that the manufacturer does btw. If your curve is a little
less than what the manufacturer says then your probably ok. If it is much
less then you need a new battery.

Josh

Post by Kavall
Battery matching.
Here is how I would go about it. Charge the cells individually untill
their voltage starts to drop, aka. peak charge them. Measure the peak
voltage. Then hook them to a fixed load and measure their discharge
time to 0.9V. Let the cell rest. The next day repeat. Take the
measurements and compare close discharge times and close peak
voltages. If you could measure internal resistance it would be nice,
but I don't know of a simple way. In my opinion the important
qualities are.
1. Close discharge times
2. Close and low internal resistances
3. Close peak voltages

[d***@gmail.com]

Post by Bruce W.1
Please see previous thread.
I'm trying to match AA battery capacities so they can be used together
in battery packs and I think created a low-tech way to do this. I'd
like to hear more informed opinions.
This would apply to NiCads and NiMh but it should work with lithium-ions
too.
Take all your batteries of the same type and charge them all in parallel
to the same voltage. Then one at a time remove them from the float
voltage and attach them to a fixed size resistor (say 6 ohms) for a
fixed period of time (say 3 hours) then measure their voltage. Well
matched batteries should have similar voltages.
Does this wash?
The only thing I wonder about is the full charge state voltage. Say I'm
doing Ni-Cads which usually top-out around 1.65 volts (where they
experience a voltage drop), and I set the float voltage for the bunch at
1.5 volts. Would the maximum voltage of each cell make a difference
here?
Thanks for your help.

[d***@hotmail.com]

Post by Bruce W.1
Please see previous thread.
I'm trying to match AA battery capacities so they can be used together
in battery packs and I think created a low-tech way to do this. I'd
like to hear more informed opinions.
This would apply to NiCads and NiMh but it should work with lithium-ions
too.
Take all your batteries of the same type and charge them all in parallel
to the same voltage. Then one at a time remove them from the float
voltage and attach them to a fixed size resistor (say 6 ohms) for a
fixed period of time (say 3 hours) then measure their voltage. Well
matched batteries should have similar voltages.
Does this wash?
The only thing I wonder about is the full charge state voltage. Say I'm
doing Ni-Cads which usually top-out around 1.65 volts (where they
experience a voltage drop), and I set the float voltage for the bunch at
1.5 volts. Would the maximum voltage of each cell make a difference
here?
Thanks for your help.

[KB]

Post by Bruce W.1
Please see previous thread.
I'm trying to match AA battery capacities so they can be used together
in battery packs and I think created a low-tech way to do this. I'd
like to hear more informed opinions.
This would apply to NiCads and NiMh but it should work with lithium-ions
too.
Take all your batteries of the same type and charge them all in parallel
to the same voltage. Then one at a time remove them from the float
voltage and attach them to a fixed size resistor (say 6 ohms) for a
fixed period of time (say 3 hours) then measure their voltage. Well
matched batteries should have similar voltages.
Does this wash?
The only thing I wonder about is the full charge state voltage. Say I'm
doing Ni-Cads which usually top-out around 1.65 volts (where they
experience a voltage drop), and I set the float voltage for the bunch at
1.5 volts. Would the maximum voltage of each cell make a difference
here?
Thanks for your help.

Hi,
I would measure the voltage of each battery after charging. Use only
those with similar or same voltages. Also, I think that 6 ohms is much
too low an ohmage for the resistor. Use a much larger size. Also
"gesstamate" a suitable wattage for the first trial.
Then perhaps a smaller size would be suitable.
Good luck,
Ken B / WB8KQJ

ps: I've done this several times with battery packs for small electric
drills.
kb