Measure a resistance with IC

Hello guys,

I need to do something a little bit tricky (for me)..

I have several unknown elect. resistances that range from 1 to 400 Ohms.
I need to be able to measure them with an accuracy of 1Ohm, so that 2 Ohms is actually different than 3 Ohms even though the scale ranges from 0 to 400.

I though of Weathstone bridges but I dont think they have the necessary accuracy for the range I need. Another idea is to use the time constant of an RC filter but I am afraid that once again I'lll get an approximate value and not a accurate one.

Obviously I cannot use a multimeter.. the idea is to replace the use of a multimeter by creating a specific tool.. The tool however can use the mechanism that a multimeter uses (I honestly don't know what it is..)

Thank you in advance for all your input.
Must be easy for top notch electronics engineers such as you guys!

hi,
A common method is to use a constant current source to drive the resistor under test and measure the voltage across the resistor, use the PIC's adc for the conversion.

You beat me to it Eric, I was just doing the math to see if that idea would work!

gaspode42 said:

You beat me to it Eric, I was just doing the math to see if that idea would work!

Click to expand...

hi R.
Many modern DVM use that method.

I am suprised to hear him say that a bridge isnt accurate enough.?

Thx,

I also thought on that one, just didnt know if I could use the 5 volt to measure 400 Ohm as the adc would have to have a resolution of 1.25 mV. that was one of the problems that I had with the withstone bridge idea.

Can you think of a way that would be independant of the voltage level? I can only think of the time costant to charge the capacitor....

Anyway I wanna give that idea a go, what do you suggest for a reliable constant current source (yes, I am that rookie!)


Thanks to both of you guys!
Pedro

it may be accurate I didnt give it a go beacuse I was afraid of the accuracy of teh adc.. 1mv Accuracy I feel like any little noise could mess up the reading...

Hi

Just checked the maths on this.
If you use a 10mA constant current source that would give 0.02V for 2Ω and 4V for 400Ω.
If you use 5V as the A/D reference that would give you a 0.0048V per 'digit' resoultion so the above would work perfectly.

Hope that helps

hi,
As pointed out by gaspode42 10mA gives a good result.

If 10mA is to high a current, use 1mA and ampilfy the signal 10 or so.

You can make a CCS with a transistor, a couple of resistors and a zener.
Have a look here there may be a better way, but I am sorry I don't know.

p.s. Where did you get 1.25mV from?