Go Down

Topic: Best accuracy to measure resistor variation (Read 1 time)previous topic - next topic

Yolco

Jan 10, 2013, 03:19 pm
Hi,

I'm trying to develop a resistor measurer with high accuracy for arduino. I'm going to explain the best I can my project and my doubts.

- I have an arduino Mega2560, and a long wire mesh (aka. mesh).
* This mesh has a great conductivity also low resistance parameters.
- Connecting the mesh with a low value pull-up resistor (input voltage is 5V from Arduino), building a voltage divider with them (the mesh is connected to ground from arduino).
- I take measurements from the voltage divider with arduino, so I can know resistor mesh values.
- The target is measure the minimun variation of resistance, in case the mesh will be sliced.

Is there any way to increase arduino accuracy/resolution so this project will be succesful??

Regards!

retrolefty

#1
Jan 10, 2013, 04:47 pm
Well you are going to have to put real world numbers on the 'low-value' pull-up and mesh resistance variation. If they are too low they may draw too much current from the arduino 5V pin. There are ways to improve the ADC resolution (internal 1.1vdc bandgap or external Aref applied voltage) but until we can see the circuit resistance values so we can see the measurement range you will need it's hard to give you specific advice.

Lefty

Magician

#2
Jan 10, 2013, 06:13 pm
You better to measure resistance via indirect method, where RC time delay counted by Timer1.  There are projects to measure capacitance based on this approach, of course, it 'd work to measure resistance as well. Benefits - 16-bit over 10-bits of ADC.
http://interface.khm.de/index.php/lab/experiments/ (Clue, theremin)

liuzengqiang

#3
Jan 10, 2013, 06:46 pm

You better to measure resistance via indirect method, where RC time delay counted by Timer1.  There are projects to measure capacitance based on this approach, of course, it 'd work to measure resistance as well. Benefits - 16-bit over 10-bits of ADC.
http://interface.khm.de/index.php/lab/experiments/ (Clue, theremin)

Cool! I know you can measure very large resistance with RC circuits if you have a known capacitor, such as dissolved solid content in water or moisture in soil. For a very small resistor you would need a large capacitor. I wonder if the charging/discharging current would be too high for arduino power supply.
Serial LCD keypad panel,phi_prompt user interface library,SDI-12 USB Adapter

dhenry

#4
Jan 10, 2013, 08:51 pm
Quote
The target is measure the minimun variation of resistance

It depends on how minimum your "minimum variation" is. The typical solution here is to measure voltage under a constant current drive.

And you want to place multiple points along the way to help you pin-point where the cut is, and to obtain the maximum variation.

liuzengqiang

#5
Jan 10, 2013, 09:04 pm
What is the mesh made of and how large is it? How is it being cut and how quickly do you want to detect the cut? You may be able to use the mesh as a capacitor and use cap sensing. More details are needed.
Serial LCD keypad panel,phi_prompt user interface library,SDI-12 USB Adapter

Yolco

#6
Jan 11, 2013, 09:04 am
Hi,

thanks to all!!

- The pull-up resistor is low enough to measure the mesh resistance, if I have a high resolution measurer, about [1-20] Ohms.
- I'll study the RC circuit, but I think it will need a very large capacitor for this project, so it will be difficult. However I'm going to make a test with it.
- Arduino will have a lithium battery as power source. So, it will take measurements each 10 minutes. Not a constant measure will be implemented.
- The mesh resistance is under 1 KOhm, but I can assure you the real value because it depends on the mesh size.

Regards!!

Go Up