µV signal amplification

Hi all,

I want to measure 50µV, 50/60Hz AC signal. with gain of say 100, in low cost way also I need to keep supply current of the amplifier below 100uA. I searched precision instrumentation amplifiers but either offset voltage is more or supply current is more. Can I do it with normal opamp having lower offset voltage? or I must go for instrumentation amps only? Can I nullify the offset without any special hardware using software if I am using ADC to measure o/p of amplifier? Can I use chopper stabilized op-amps which gives me low offset & low supply current for the same? e.g. ISL28133. I think analog devices also have similar opamps..

Thnx in advance.

If you are measuring an AC signal then offsets shouldn't matter - you AC couple after the amplifier. An instrumentation amplifier is for differential signals so it depends what you signal source is.

Keith

Thnx,
I need to read it using ADC. So I need to give DC shift of AVCC/2 to it. I am sensing very low magnitude of AC current using resistor whose one end is grounded.

In which case you simply AC couple after the amplifier so the offset is irrelevant. Bias the ADC with some resistors to the middle of its working range. You will want a gain of more than 100 I would have thought. The choice between an opamp and instrumentation amplifier depends on the location of the resistor. If it is close to the amplifier & ADC then an opamp would do the job. If there is a significant amount of wire between the amplifier and the resistor then an instrumentation amplifier would be best.

Keith.

Suryakant,

I suspect your bigest problem is going to be noise in it's various forms.

Thus I would sugest a differential reading across the resistor and ignore/mitigate the fact that one side is grounded. so unless you AC couple you will need to find some way of effectivly bringing the voltage across both sides of the resistor to aproximatly half supply for you sensing amp.

As usual it's difficult to be more speciffic than very general without knowing more about what it is you are actually trying to do.

I can represent the circuit on my mind as follows... plz let me know ur comments

keith1200rs said:

In which case you simply AC couple after the amplifier so the offset is irrelevant. Bias the ADC with some resistors to the middle of its working range. You will want a gain of more than 100 I would have thought. The choice between an opamp and instrumentation amplifier depends on the location of the resistor. If it is close to the amplifier & ADC then an opamp would do the job. If there is a significant amount of wire between the amplifier and the resistor then an instrumentation amplifier would be best.

Keith.

Click to expand...

Thnx Keith. all components will be on the same PCB.. distance will be within 1-2 inches max.

Yes, that will give you an amplified AC signal centred on 1.65V (for a 3.3V ADC). You can generate the mid rail without a voltage reference by using two 2M resistors instead of R11: one to 3.3V the other to 0V.

Keith.

keith1200rs said:

Yes, that will give you an amplified AC signal centred on 1.65V (for a 3.3V ADC). You can generate the mid rail without a voltage reference by using two 2M resistors instead of R11: one to 3.3V the other to 0V.

Keith.

Click to expand...

But my question is can i use normal opamp with less offset(10uV or so) instead of instrumentation amplifier to achieve my goal?

Yes, an opamp would work. The offset will get multiplied by the gain, but as it is an AC signal that shouldn't matter, unless you are assuming the signal will be exactly centred around 1.65V. Even so, you could capacitively couple from the opamp output to a simple potential divider on the ADC input.

Instrumentation amplifiers are important for rejecting common mode signals. As your signal is not differential it isn't critical.

Keith.