Battery Monitoring

[lorhve]

Hi,

I have an op-amp set up as depicted in the attachment file. I have put a physical wire to VCCSENSE and varied the input voltage with a DC power supply to determine the proportionality of the reading. And, I have observed something unexpected. First of all, the reading seems to be way off from the 'theoretical' value, which is the obvious straight line in the graph. The other two lines are readings from two different circuit boards. Also, there is a sudden slope near the 15V reading. Does anyone have an idea of why the reading is so different from the theoretical value only except at 11.5V and there is a slope change at a higher voltage level?

Thanks,
Lorhve

[dsp_redux]

If you want a follower for proportionnal reading, shouldn't you place R69 after the feedback loop instead of before? Just guessing, maybe more info on what you are trying to do would help?

[SgtWookie]

You don't show the supply voltages for the TL072, but I'll bet you have the negative supply terminal at ground instead of a negative voltage.

The TL072 is a good, quiet opamp, but it does not have rail-to-rail inputs and outputs; you need to have the negative rail supply 3v lower than the lowest expected input voltage.

You need to either add a negative rail of at least -3v, change your input voltage divider, or use an opamp who's common mode input includes ground.

[lorhve]

The op-amp has a single-power supply (5V and GND). I didn't particularly design this part of the circuit but do you know the importance of C53, in parallel with the voltage divider, and why the feedback is taken after R69?
And when it doesn't support rail-to-rail inputs and ouputs, what does that mean?

[SgtWookie]

5v is not enough for a TL072 to work properly. Either increase it's supply voltage to +5v/-5v, or select another opamp.

Resistors create noise; it's sort of like water going through a garden hose nozzle, you can hear the hiss. C53 is there to quiet the noise down, similar to placing the garden hose nozzle in a bucket of water.

R69 is there to slow the response time of the circuit; combined with C154 it's a low-pass filter. A higher frequency input signal would be dampened.

"rail to rail" refers to the ability of an analog device to use inputs and/or provide output signals that are in a voltage range from the negative power supply rail to the positive power supply rail.

The +/- supplies to a circuit are frequently called "rails", as if drawn in a schematic they are usually parallel, with devices connected in between them. There is a resemblance to railroad tracks.

If you wish to use your circuit as-is, I suggest that you change the opamp to an MC33172. It is pin-compatible with the TL072, has near-RRIO, (rail-to-rail inputs and outputs) and will operate on a single 5v supply with no problems.