Redpitaya Forum

I recently bought a red pitaya and just quickly hooked it up in order to see what noise levels I get from it.
As I can see in the specs the ADC has a resolution of 14 bits. With a range of -1V to +1V that gives a quantization of approx. 122 uV.

So far so good. Now I hooked up a probe to one of the pitayas inputs and let it float (unconnected to anything) expecting to measure 50 Hz disturbances from the environment. The graph showed this:



I get noise with a standard deviation of around 1.4 mV.


Then I tried to short the probe to ground



which suddenly gives me a noisier signal and even some spikes (see the picture). The standard deviation is now around 2.3 mV.
Of course I can use the moving average filter but then high-frequency parts of a signal will get filtered out which I do not desire.

Are these normal noise levels for a Red Pitaya? If not what could be the cause?

So I tried shielding, using different power sources for the RP, removing any other electrical devices from the vincinity pf the RP. As far as my tests go I can say that about 4 bits of the ADC are lost due to noise. Is this normal or is my RP broken?

If you want to actually test the noise level of the RP, try simply connecting a 50 ohm load to the ADC.
Even with that, I have noticed about the same noise levels has you have. Maybe even more :S.

I suspect that 4 bits of noise is pretty good. Honestly, most scopes have 8 bit ADC because anymore more than that requires serious calibration.

I think the parts themselves quote 80 dB SNR, but that is under ideal power supply situations.