# Quick question on permissible resistance

Hello, I am currently using Arduino Uno to read the voltage output of a sensor that I am working on.

When I use 10 Mega Ohms resistor with one end connected to AnalogInput and the other to the GND,

the voltage level when no output is given reads 0.0 on my serial monitor.

However, here is the problem.

When I use the 10 Tera Ohms... the voltage fluctuates like crazy.

It draws sinusoidal wave with +- 0.2 Volts at neutral voltage of 3.55V.

I am assuming this is the noise produced by the Arduino, which is then amplified by high

resistor to show such high voltage, but I was wondering if this actually is the case.

Any help would be appreciated.

Thanks.

Where the heck do you even get a 10tera-ohm resistor?

This is not a āproblemā, itās essentially a floating input, and if you are seeing a sine wave then itās picking up the mains in your house.

Where the heck do you even get a 10tera-ohm resistor?

Probably a zero Ohm resistor cut in half

Where the heck do you even get a 10tera-ohm resistor?

This is not a āproblemā, itās essentially a floating input, and if you are seeing a sine wave then itās picking up the mains in your house.

Thanks for your reply. You can get one at one of those retailers that deal high resistors ha ha.

It costs 100-200 dollars or something. (depends on accuracy obviously).

I was trying to make a simple electrometer with an Arduino to read open circuit voltage out of

0.5 Tera Ohms electrical source.

So you are saying this kind of fluctuation is ānormalā on 10 Tera Ohms?

Incidentally, even 10MĪ© is too high for general use connected to an analogue input. It's better to keep the resistance considerably lower, ā¤10K if possible.
From the ATMega328P datasheet:-

The ADC is optimized for analog signals with an output impedance of approximately 10 kĪ© or
less. If such a source is used, the sampling time will be negligible. If a source with higher impedance
is used, the sampling time will depend on how long time the source needs to charge the
S/H capacitor, with can vary widely. The user is recommended to only use low impedance
sources with slowly varying signals, since this minimizes the required charge transfer to the S/H
capacitor.

OldSteve:
Incidentally, even 10MĪ© is too high for general use connected to an analogue input. It's better to keep the resistance considerably lower, ā¤10K if possible.
From the ATMega328P datasheet:-

Wow thanks

Maybe I should look up for a different chip.

Jasombre:
Wow thanks

Maybe I should look up for a different chip.

Possibly, but I suspect that most AVR chips will have similar specs. The ATMega328P, ATMega2560, ATtinyx5, ATtinyx4 and ATMega1284P are all the same in that regard.
You might need something more exotic, or you might need to use an op-amp buffer (voltage follower) if youāre sampling a really high-impedance source.

You'd need a super low input current opamp / buffer to interface such high impedances - and have to be very careful with layout, shielding and leakage. Probably a FET device.

NatSemi, LT, Texas all make suitable devices - you'll need sub - picoamps level spec for such a gadget.

If you're not worried about absolute dc accuracy you could use a FET.

NatSemi LT, Texas all make good devices.

regards

Allan.

OldSteve:
Possibly, but I suspect that most AVR chips will have similar specs. The ATMega328P, ATMega2560, ATtinyx5, ATtinyx4 and ATMega1284P are all the same in that regard.
You might need something more exotic, or you might need to use an op-amp buffer (voltage follower) if you're sampling a really high-impedance source.

I am not an electrical engineer so I do not know how op-amp buffer circuit is set up, but I will give it a try.

Thanks again.

Any comments on the fluctuation issue?

Jasombre:
Any comments on the fluctuation issue?

Effectively, a 10TĪ© source impedance is an open/floating pin. A floating pin will not be stable, and you can expect it to fluctuate. (Or do whatever it wants - it'll be unpredictable.)

Edit: I should add that if you're not fairly experienced with electronics, you'll have a lot of trouble making what you want. It requires fairly specialised knowledge.

allanhurst:
Youād need a super low input current opamp / buffer to interface such high impedances - and have to be very careful with layout, shielding and leakage. Probably a FET device.

NatSemi, LT, Texas all make suitable devices - youāll need sub - picoamps level spec for such a gadget.

If youāre not worried about absolute dc accuracy you could use a FET.

NatSemi LT, Texas all make good devices.

regards

Allan.

I guess I will just have to use the 6514 Electrometer that I purchasedā¦ Thanks.

I wanted to use Arduino as a Master controller so that it can collect sensor data (ex. Force) and

actuate a motor (10N = 5 rev. 20N = 10 rev. something like this).

OldSteve:
Effectively, a 10TĪ© source impedance is an open/floating pin. A floating pin will not be stable, and you can expect it to fluctuate. (Or do whatever it wants - it'll be unpredictable.)

Edit: I should add that if you're not fairly experienced with electronics, you'll have a lot of trouble making what you want. It requires fairly specialised knowledge.

Judging by the stable sine wave, I thought maybe this was a noise generated by Arduino itself.

I will have to stick to the electrometer that I have purchased.

I just wanted to try whether or not Arduino can function as an electrometer.

Thank you so much for your help.

Jasombre:
Judging by the stable sine wave, I thought maybe this was a noise generated by Arduino itself.

The sine wave youāre seeing is most likely at the same frequency as your mains supply, being picked up inductively. (I get the same result, a nice neat sine wave, if I use āanalogRead()ā with nothing connected to the analogue pin, then print the results on the serial plotter. I also get that result using my oscilloscope with nothing connected to the probe.)

Hi,
Have you looked for an electrometer with PC connection?

That model you have has computer connection capability, RS232 among others.

Tom...

1 Like