Hi,
I'm trying to measure some very high resistances using the CR1000 logger.
I was originally going to use the BrHalf4W command to measure relative to a precision 1MOhm resistor, but wasn't getting sensible answers. I checked and the input resistance is ~20GOhms for the CR1000. I then tried to write my own version with longer settling times thinking that the problem was due to stray capacitance from the wire (~5-6m) and the large resistances involved.
Has anyone managed to measure ~1-20MOhm resistances with the CR1000? Do you think I've diagnosed the problem correctly? The longest settling time that can be used is 50ms, even in the individual VoltDiff commands. Using 50ms certainly reduces the problem, but doesn't eliminate it.
Wire is 213pF/m 6 m long, Resistance ~10MOhms, RC ~13ms
With 1MOhm resistors mounted directly into the terminals on the logger I do get sensible readings-ie very close to unity for the voltage ratio.
Maybe I need to buy low capacitance wire and perhaps add another resistor in parallel with the one I'm trying to measure to reduce settling time...?
Is there any programming trick that would allow a longer settling time to be used?-measurements are taken slowly in this application.
Any ideas welcome, Thanks in advance,
A
You can have a longer delay by first using the ExciteV command to turn on the excitation but with the delay set to zero. Then use a delay instruction, delaying the measurement sequence for the time you choose, then follow on with the voltage measurement instruction required.
Having seen an email you sent into CS, explaining this is to measure the resistance of wood, can I also recommend that to prevent long term polarisation issues that you repeat the excitation, delay and voltage measurement twice with the opposite polarity. You can correct the polarity of the second measurment (use a negative multiplier) and use the average of the two measurements, to give a better measurement as well.
One thing to beware of though and that is if you delay too long you may see the measurement start to change due to polarisation effects within the measurement, so do not be tempted to use too long a delay.