Hi,

Can anyone tell me for an AC source how is the frequency decided to be?

 

Source for what?

 

An AC excitation source for a strain gauge full-bridge. Am I right in thinking I should do a frequency response test for the output of my circuit to see the ranges where the gain doesnt decrease?

 

I'm not familiar with AC excitation for strain gauges. I'll need some context, like a schematic or a link to a page, to understand what you need. I'm a precision analog guy. I'm thinking strain gauges are for measuring something that moves rather slowly, so frequency doesn't even need to be considered.

or some other helper that knows about this might jump in about now...?

 

I think that the linearity of your circuit is more important. As strain gauges vary slowly, the return 1 KHZ will only have a very narrow bandwidth, like .1 HZ, so your circuits gain will not change very much across 1K/.1 = .01% of its bandwidth.
Frank

 

The circuit is based on phase sensitive detection I have a four strain gauge full-bridge going to a three-op-amp instrumentation amplifier then from here to a Phase sensitive detector that switches between two paths that inverts and non inverts the amplified signal. This signal is rectified for both positive and negative voltages and is passed through a LPF that smooths the signal to DC and then the DC signal is thorough a Howland Current Source as a 4-20mA current.

To be honest I've never come across DC excitation, ever book Ive looked at talks about AC

 

I do not know much about this but strain gauges are resistive so I would not expect any phase difference. Perhaps you are talking about a lock in amplifier which is a synchronous detection method and is roughly equivalent to a very sensitive filter/amplifier. The speed of response has pretty much nothing to do with the frequency of the ac, the speed relates to the time for a change in resistance.

Choosing a frequency -- high to stay out of dc effects and keep any caps and inductive effects low, the strain gauge will probably have an inductance. Also high frequency tends to keep caps and inductiors if used, small. You want to be well within the frequency range for you amplifies to have high gain with low distortion and noise.

If it is a lock in amplifier there should be some sort of spec with the circuit.

All said I would guess from 1k to 10k Hz. Just a guess.

 

It relates to this ...

http://forum.allaboutcircuits.com/showthread.php?t=69751