Unknown behavior of differential amplifier for high-side current sensing

Unknown behaviour of differential amplifier for high side current sensing

Dear Experts,

To fulfill the objective of high side current sensing, I am evaluating a differential amplifier operating from a 5V single supply.

The current is being sensed for a performing MPPT for a Solar battery charger and has to be with a low target cost.
Due to low cost consideration, I have refrained from using readily available current sense amplifiers, and instead focusing on building a discrete differential amplifier.

I have chosen the LMV358 which is a low supply, rail-rail op amp.

However, simulating the differential amplifier shows that the op-amp is going beyond saturation and having a voltage higher than VDD at the output. I am unable to find the reason for this. Simulation attached below.



If I replace LMV358 with another rail-rail op amp OPA2340, which has a high unit cost with similar voltage ratings, but low offset voltages. I am able to achieve the desired and proportional result as shown below.

Vout = Rf/Rin*V(Rsense)

Vout ~= 10k/1k * 159.52mV = 1.67V



The voltages at the inputs of the op-amp is varying drastically and seems to be the cause for it. Probably due to input offset voltages?

I would really want to stick to developing a current sensing solution around LMV358. Could someone point me in the right direction to fix this issue? Thanks!

Re: Unknown behaviour of differential amplifier for high side current sensing

Hi,

However, simulating the differential amplifier shows that the op-amp is going beyond saturation and having a voltage higher than VDD at the output. I am unable to find the reason for this. Simulation attached below.

Click to expand...

Your absolute input values for IN+ and IN- referenced to GND are much too high for the OPAMP.
Read datasheet about
* absolute input voltage range
* common mode input voltage range.
The OPAMP behaviour you see is expectable.

If I replace LMV358 with another rail-rail op amp OPA2340, which has a high unit cost with similar voltage ratings, but low offset voltages. I am able to achieve the desired and proportional result as shown below.

Click to expand...

This is a simulation error.
In reality this won´t work either.

I recommend to use a current sense amplifier. They aren´t really expensive... compared to a good OPAMP and the essential high accuracy resistors.

Klaus

Re: Unknown behaviour of differential amplifier for high side current sensing

You should read datasheets thoroughly.
1. LMV358 is "rail-to-rail output" OP, not rail-to-rail input.
2. It's supply voltage range is restricted to 5.5V.

Re: Unknown behaviour of differential amplifier for high side current sensing

Common mode range is exceeded by about 10 V. No OP Amp can work with that common mode from a 5 V supply.

Re: Unknown behaviour of differential amplifier for high side current sensing

No OP Amp can work with that common mode from a 5 V supply.

Click to expand...

No regular OP actually. Special "over-the-top" current sense amplifiers can, by supplying the input stage from the shunt terminals.

Re: Unknown behaviour of differential amplifier for high side current sensing

Thank you everyone.

I have changed by design and implemented current sensing using INA180, that are high side current sense amplifiers able to tolerate common-mode voltages up to +26 V.

I still don't understand why the simulation using OPA2340 gave a suitable result?

Re: Unknown behaviour of differential amplifier for high side current sensing

I still don't understand why the simulation using OPA2340 gave a suitable result?

Click to expand...

I guess the model is either too simple (ignoring limited common mode range) or too optimistic (still working with clamp diodes in forward bias). Please notice that the actual output voltage doesn't exactly fit the expected value according to the observed input offset. In any case, it won't give reasonable results in real hardware.