Hi All,

I am an electronics student doing a HNC course. I also work for a measurement company, we measure strain and a variety of material properties.

Our equipment can output a strain signal that is either digital (USB or simply a multi-bit digital signal).

However we need to be able to interface with material test machines. Some of these have a port into which an Extensometer is plugged into. These have Wheatsone bridges in them that generate varying voltage signals of a few mVs. This signal is always scaled by another factor: the exiciation voltage. This is a signal generated by the test machine.

My boss has tasked me with building a circuit that can replicate the function of the Wheatsone bridge, but using the signal from our equipment as the strain signal.

It is very important that it is a linear

I have a few ideas so far as to how this can be achived:

• Convert to digital and use a microprocessor/controller
• Using digital resistors in the form of a bridge but use some circuit to control them.
• Using FETs as voltage controlled resistors.
• Using analogue multipliers to multiply the signals.

Does anyone have any experince of similar circuits and are there any ideas which I might be missing?

Thanks for any help.

Neil

 

Sorry but I really don't understand what you are trying to do.
If you are studying to be an engineer than one of the most important things you need to learn to be precise and complete in detailing your requirements.
Many designs fail to do the required task because this information wasn't complete.
Think about all the information you would need if someone told you to do this task without knowing anything else about the system.
For example what does "My boss has tasked me with building a circuit that can replicate the function of the Wheatsone bridge, but using the signal from our equipment as the strain signal." mean?

What are the actual signals you are trying to "convert"?

Your blocks need more detail.

 

No, this isn't entirely clear. I'm taking it that the assignment is to design a "dummy extensometer" which would accept an excitation voltage from the outside, and produce programmable outputs, equal to a gauge factor times the excitation, which could be used to test the company's equipment. Typically strain gauges produce an output in the millivolt range (but centered on half the excitation voltage) when the excitation is 10V or 12V, so it would be a case of controlling a very small signal. But I'm not sure if this is what's really wanted.

Edited to say, if you wanted to do a complete job of simulating strain gauges, you'd need to add an adjustment to vary the common-mode voltage a little, and also a settable output impedance, equal to half the resistance of whatever gauge you were modeling. Not difficult stuff.