# Measuring Batteries in Series

Discussion in 'Power Electronics' started by Arouse1973, May 18, 2014.

5,165
1,087
Dec 18, 2013
I was involved in a recent thread regarding measuring the individual battery voltage of series connected cells. The original OP didn't have the experience to design or construct anything too complex, so he was looking for an off the shelf solution.

Nothing could be found so he opted for taking the battery off load periodically to check the status of the cells. So I thought I would have ago and I remember years ago reading about using a transformer. So here it is. Fig 1 showing one battery out of a set of series cells. The extra earth connection was added to keep the simulator happy. If measuring any other cell other than the one connected to 0V then this connection is omitted.

The circuit pulses a 1:1 transformer, the voltage seen across the resistor R1 is clamped by the battery voltage minus a diode drop. The output is diode biased to take into account the diode in series with the battery. I get about 50mV error, probably due to the bias current for the diode being out a bit.

You may have seen something similar before, but if not I hope you find this interesting.
Thanks

Fig 1

Fig 2

Last edited: May 19, 2014
Harald Kapp likes this.
2. ### Harald KappModeratorModerator

11,439
2,627
Nov 17, 2011

5,165
1,087
Dec 18, 2013
Neat idea isn't it?

4. ### Harald KappModeratorModerator

11,439
2,627
Nov 17, 2011
@Arouse1973 :
Hi Adam, I revisited your little circuit. While trying to understand it in more detail I came across an issue with this part of your explanation:
I'd say the voltage across N1 is clamped in this way. The voltage across R1 is VG1-V(N1).
It is coincidence that with your choice of a 12 V step and a 6 V battery the output voltage is 6 V, too.
When you step e.g. V1 from 2 V ...10 V in steps of 2 V, you'lll notice that Vout=10 V for V1=2 V, Vout=8 V for V1 = 4 V etc.

5,165
1,087
Dec 18, 2013
Hi Harald
Why are you setting VG1 bellow the battery voltage? This is not how the circuit is supposed to work. Do you have the sim of your circuit? Have you tried to change the battery voltage whilst driving the circuit with 12 Volts? Give me some time to have another look, I may have made an error in my circuit.

6. ### Harald KappModeratorModerator

11,439
2,627
Nov 17, 2011
I didn't step VG1, I stepped V1 (6V in your sim), the battery voltage. The other voltages are as in your sim. I did an LTSPICE simulation and it looks quite o.k. (I had to set the inductance of the transformer windings to >10H, however, otherwise I couldn't get the stepwise waveforms.

No, I don't think so, just in the interpretation of the result. When you measure Vout-VG1 you get the expected voltages.

5,165
1,087
Dec 18, 2013
Ah yes ok. If I remember the original design had the resistor connected to the coil and the coil connected to 0V. This may be where I got confused. Like this I think. .

8. ### Harald KappModeratorModerator

11,439
2,627
Nov 17, 2011
this circuit matches my observations. Thanks for taking the time to check.

Harald