The discharge curves for NICAD and NiMH batteries are rather flat, ie the volage does not change much between fully charged and nearly completely discharged. A voltmeter is of limited use.
If you want get a better indication of the charge in the battery you will have to integrate the charge/discharge current with respect to time. This is getting away from the idea of "simple".
There are integrated circuits which implement a "gas gauge" function for use in laptop batteries etc, try searching the Maxim website, I think they do one or two of these things.
There are integrated circuits which implement a "gas gauge" function for use in laptop batteries etc, try searching the Maxim website, I think they do one or two of these things.
Testing a batteries state of charge is not that complicated. Open circuit voltage is as Nigel said relativly useless by itself. But you can get a very good idea of the state of charge of a battery by comparing a no load voltage with a loaded voltage. A good 'load' current to put a cell under is 1/10th of it's rated capacity. So for a 2000ma NiMh, put it under a 200ma load and compare the voltages. You'll get a pretty decent representation of the battery charge.
If you say compared 10ma 100ma and 500ma load voltages with the no load voltage I'm sure you could calibrate a 'gas gauge' style meter within 5% accuracy.
If you've ever seen those built in battery gauges on AA's (I think Energizer and Duracell both did this for a time) when you press the side of the battery it shorts the battery across a thermal strip and the relative current through it causes it to heat up and show a charge state, same general concept.
If you've ever seen those built in battery gauges on AA's (I think Energizer and Duracell both did this for a time) when you press the side of the battery it shorts the battery across a thermal strip and the relative current through it causes it to heat up and show a charge state, same general concept.
Yes, but a solid state digital circuit such as perhaps a PIC or other compareable micro controller with an ADC can read the loaded voltage (switched easily via a transistor) in less than a milisecond. At 1 reading a second that's 1ma average draw for a 1 second update on battery capacity with a 100ma draw... Check it once a minute and the wasted current is almost less than self discharge. I wouldn't recommend anything less than a 1ms 'short' to the load before sampling simply because internal surface charge of the batteries chemistry can artificially 'liven' the loaded voltage for a brief period of time.
No I have not, but that is the point I was trying to make by my earlier statement:-
JimB said:
If you want get a better indication of the charge in the battery you will have to integrate the charge/discharge current with respect to time. This is getting away from the idea of "simple".