Hi folks,
I am looking for some help and advice on how to build a circuit that can measure the internal resistance of a battery whilst it is still discharging into a load.
I believe that the method is called an ‘AC conductance’ test and involves applying a small AC signal to the battery in order to measure the impedance.

Obviously any loads connected would have to taken into account and would probably require chokes to avoid skewing the results too much.
I absolutely see how this should work in theory but really don’t have a clue where to start.

I would appreciate any thoughts at all and if possible a pointer to some practical circuits.

The bats in question are circa 400Ah AGM’s and I would like the choke to handle as high a current as possible. Peek current could potentially be 70 amps but my typical demand seldom exceeds 15A.

Why do I want to do this ...? :-

I have a plot of SOC (state of charge) vs OCV (open circuit voltage) for my speciffic cells but that is only useful if the batteries are rested, which mine never are.

I want to measure internal resistance, which is primerrally dependant on discharge current, SOC and temperature.
Once I have the internal resistance I can it to calculate the internal voltage drop.

My assumption being that the internal voltage drop added to the terminal voltage will be close to the voltage that I would see if the bats were disconnected and rested which is a reasonable indication of SOC when used to look up the value from my OCV vs SOC chart.

I amy find that other factors need to be considdered but getting the resistanve seems like a good first step to me.

Any and all thoughts welcome folks …
Thanks
Al

I just found something pertanant I think ....

The internal resistance of a battery is generally measured by noting the change in its output voltage under different load currents. Suppose that when the battery supplies one current I1, its output voltage is V1, and when it supplies a current I2 its output voltage is V2. The internal resistance is given by R = (V1-V2)/(I1-I2)

(Adjuster - 05-05-2011 11:01 PM )

It leaves me wondering about switching in aditional test loads for short periods just to get a comparison reading or possibly just comparing old values to new values after any significant change in load.

Still looking for comments and opinions
Thanks
Al

 

Using ‘AC conductance’ sounds simple but the actual usage of the data collected is a complex problem. If the measurement system is always connected to the battery under test then a standard Ah in/out Book–keeping model might work better

Background link on battery SoC testing history.
http:////web.cecs.pdx.edu/~strom/ev1/articles/state_of_art_battery_charge_determination.pdf

 

If you are trying to measure the internal resistance of a battery, then I suggest using battery testing products from Futek Instruments. (It also allows you to measure temperature, voltage, currents, pressure, etc...) It precisely collects data and graphs the data to your computer. You can either use their free software or use their drivers that are compatible with the software of your choice. Just browse through their battery testing products at www.futekinst.com They products are affordable and easy to use. They also build custom devices that meet customer specifications. I previously purchased one of their product for laboratory use, and I have to say it was a neat gadget.

 

Why don't you try an experiment to measure the battery impedance, on and off load. You may find that at a low enough frequency the load makes little difference: clearly the battery impedance is a minute fraction of the static V/I load, unless this is drawing many hundreds of Amps. The choke you are planning may not therefore be needed. I might be wrong, but maybe it's worth a try.

Achieving adequate rejection of all the load derived splats and hash may be another matter. Think lock-in amplifier, or at least a synchronous detector, possibly these days implemented with the dark art of DSP. (Which is not my subject.) Good luck!