Ideas/pointers for measuring resistance changes in time

[Redhouse]

I find myself in need of measuring a varying resistance over a period of about 5 or 10 seconds. I've been rolling the idea around in my head maybe using a PIC, or Arduino, and logging to an SC card, but was wondering how to actually measure the changing resistance without influencing the sensor as it records the changes.

I'm not looking for anyone to "design" it for me, I'm just asking for ideas/pointers/tips on how one might implement this kind of thing. I have some experience with PIC's and I've been meaning to get into Arduino's so that's why I ask about both/either.

[zac]

I find myself in need of measuring a varying resistance over a period of about 5 or 10 seconds. I've been rolling the idea around in my head maybe using a PIC, or Arduino, and logging to an SC card, but was wondering how to actually measure the changing resistance without influencing the sensor as it records the changes.

I'm not looking for anyone to "design" it for me, I'm just asking for ideas/pointers/tips on how one might implement this kind of thing. I have some experience with PIC's and I've been meaning to get into Arduino's so that's why I ask about both/either.

The conventional method is to measure the voltage drop with a known current.  I would use a constant current source to put a known current through the resistance and then sample the voltage that develops across it.  If you need to measure a wide range of resistance then being able to control the current source would provide better resolution. 

[intel]

Redhouse, I think it would be the following links helpful.


http://www.microchip.com/stellent/idcplg?IdcService=SS_GET_PAGE&nodeId=1824&appnote=en011113

http://ww1.microchip.com/downloads/en/AppNotes/00611b.pdf

http://ww1.microchip.com/downloads/en/AppNotes/00611.zip

[Vineyards]

You could try the ratiometric method which gives you immunity from problems associated with a constant current source not being very constant at all. It should also provide (better) thermal stability.

[Gallymimu]

how fast are you needing to sample?  What accuracy, precision, resolution do you need?  What is the range of resistance you will be looking at.  This is another one of those projects where you need to provide a lot more information for anyone to be able to offer a quality solution for you.

[zac]

You could try the ratiometric method which gives you immunity from problems associated with a constant current source not being very constant at all. It should also provide (better) thermal stability.

Good article about that:

http://www.edn.com/design/test-and-measurement/4329289/Accurately-measure-resistance-with-less-than-perfect-components