Simulating a Wheatstone Bridge/Extensometer

[Neil_Measurement]

Hi,

I would like to build a device which simulates the operation of a bridge circuit in an extensometer. Instead of using an extesnometer to measure strain, we are taking in a strain signal (this can be either a digital or analogue signal (+/- 10 volt), whichever is easiest). The device needs to produce about 3 mV at full scale and must be proportional to the excitation voltage, as an extensometer is.

We have three signals:
- Measured strain;
- Exciation voltage;
- Strain Output signal;

This device would therefore allow us to input a strain signal into a test machine without having to use an extensometer. It is important that it is as linear as possible.

Any ideas?

Thanks,


Neil

 

[CDRIVE]

Hello Neil and welcome to EP. Firstly, this appears to be a class assignment which should have been posted in our Electronics Homework section.
https://www.electronicspoint.com/forums/electronics-homework-help.106/

That said the replies you receive will only aid in pointing you in the right direction as we don't do a students homework for them. The more you contribute at your end the more our members will assist you in finding applicable solutions.

We speak "Schematic" here. So the more of them (and or block diagrams) you provide the more the likelihood of member participation.

Chris

 

[OBW0549]

Firstly, this appears to be a class assignment which should have been posted in our Electronics Homework section.

I'm a bit new here, I'll admit; but I'm puzzled as to why you inferred that this was a class assignment. I would have guessed that the fellow is an engineer or a researcher, perhaps in a Materials testing lab.

If this is indeed a class assignment, I won't post a solution; if it is not, I may be able to help.

 

[Neil_Measurement]

Hi,

Although I am an Electronics student at a local college, this is only a day release program. This task is related to my work as a trainee technical apprentice at a Video Measurement firm. We use our systems to measure strain induced by Universal Test Machines. My boss has asked me to generate ideas for how the converter device works and to also ask other people.

The circuit we are looking to replace is the Wheatstone bridge, found in extensometers. When this is under strain it produces a voltage as the resistors change resistance proportionally with strain. Please see the first diagram.

We would like to replace this with a device that takes in another signal as the strain. This will be complicated because it has to be highly linear and also proportional to the excitation voltage.

Let me know if you require more info. Thanks for any help.

Thanks,


Martin

 

[CDRIVE]

I'm a bit new here, I'll admit; but I'm puzzled as to why you inferred that this was a class assignment. I would have guessed that the fellow is an engineer or a researcher, perhaps in a Materials testing lab.

If this is indeed a class assignment, I won't post a solution; if it is not, I may be able to help.

Martin's post included the word "we" which is often a red flag. Since it appears that this isn't a class assignment after all and I've not worked with Extensometers, please feel free to assist.

Martin, have you researched Piezo elements in conjunction with strain measurement?

Chris

 

[OBW0549]

Martin's post included the word "we" which is often a red flag. Since it appears that this isn't a class assignment after all and I've not worked with Extensometers, please feel free to assist.

Thanks.

I'm not sure whether this will help you or not, Martin, but the attached .pdf shows a schematic for a strain gauge simulator that I've used successfully in the past for testing precision strain gauge amplifiers for linearity, scale factor and offset voltage.

I've modified it to produce a +/- 3 mV output range based on a 10 volt bridge excitation voltage. Control of the output is via a logic-level PWM (pulse-width modulation) signal from a microcontroller; 0% on the PWM gives -3 mV output, 100% gives +3 mV, and 50% gives zero output.

If you have any questions, let me know.

(EDIT: Here are links to the manufacturer's data sheets for U1 and U2. The TL072 is inexpensive but it doesn't have a terribly high open-loop gain; it's possible that linearity could be improved slightly by substituting a much better op amp in U2, such as a MAX44246.)

 

[CDRIVE]

I'm glad your jumping in here because I thought the question revolved around the strain sensing mechanism / sensor not the output circuitry of the bridge.

I'll be lurking and learning while y'all work it out.

Chris

 

[Neil_Measurement]

Hi OBW0549,

Thanks, that looks like a nice circuit. Using op amps does make sense for this project, thanks for the recommendation on ICs as well. I will have to think it through a bit more in my head though.

Hi Chris, I'm surprised you spotted the 'we' but missed I had changed names .

This problem can be boiled down to multiplying the excitation voltage by the strain measurement and then by a pre-set factor that is related to the bridge resistance.

The ideas I have had (well, me and mostly my boss) are the following:
- Convert everything using ADCs/DACs and do all the operations on a micro controller. This is my preferred idea.
- Use digital resistors and create a real Wheatstone bridge, but control it using the strain signal.
- Use the Op-Amp method to multiply the signals.

I will post some diagrams (not yet Schematics) up soon.

Thanks,


Neil

 

[CDRIVE]

Neil, if your looking for a virtually painless transition into microcontrollers check out http://www.picaxe.com/

Chris

 

[OBW0549]

The ideas I have had (well, me and mostly my boss) are the following:
- Convert everything using ADCs/DACs and do all the operations on a micro controller. This is my preferred idea.

I like that idea, and it actually simplifies things a bit:

EDIT: There's an error in note 5 on the schematic: 0x0400 should be 0x0800.