Sampling with two channels

[AlessioCarnemolla]

I tried to sample two different signals with two channels (GPIO26, GPIO27) using example_ADC_async.py. One of them was linked to a Function Generator and the other was linked with a Jumper to GP2. Processing the two signals it came out they are the same signal and, doing the FFT, it shows they have same peak frequency. I didn't modify any line of the code. Is there something I have to add? How can i record two different signals precisely?

[FilipDominec]

Hi, thanks for reporting. I will check it on my side. But first I must ask:

1. Did you use the stable release 0.2, or the latest github version of rp2daq?
2. What script did you use, was it example_ADC_async.py or example_ADC_sync_plot.py or just the short snippet from the README?
3. Are the results really 1-to-1 numerically the same (i.e. doubled values), or is there some little random difference (i.e. ADC properly sampling the same analog signal)?
4. Is there anything else that seems weird to you, e.g. 1.0 V input voltage possibly not mapping somewhere close to values 1220-1250 at ADC?

[AlessioCarnemolla]

Hi, I used the latest release of rp2daq. I used the 'example_ADC_async.py' code. There are some difference between the values of the signal but the shape is the same, the mean value is almost the same and the peak frequency is precisely the same. Here the graphs and the post processing done by Matlab. Thanks a lot for your avaliability.

[AlessioCarnemolla]

The peak Frequency of the signal is 190Hz, the frequency of the wave generated on the GP2. The Function Generator produced a wave of 20Hz. The GPIO 26 is connected with GP2, the GPIO27 is connected with the FG.

[FilipDominec]

Puzzling indeed.

But what you show is some 20mV pk-pk signal atop of some 80mV background - this is fairly weak. What we see may even be some voltage noise picked by disconnected high-impedance pins.

What do you get when you connect e.g. 1 V DC signal to one of the pins?

[AlessioCarnemolla]

Now the FG produces a wave of 17Hz Peak and 3V pk-pk on GPIO27. GPIO26 is still connected to GP2. This is what I get (peak frequency in the graphs corresponds to 17Hz).

[AlessioCarnemolla]

It seems that the code detects only one frequency from the two channel, printing one of them for both, casually. May it be a hardware problem? I tried with different jumpers.

[FilipDominec]

I am very busy now, but I will try to replicate & fix it until Friday. Thanks for understanding.

[AlessioCarnemolla]

Thank you a lot, for your avaliability.

[FilipDominec]

Hi, so I confirmed example_ADC_async.py works exactly as expected with the fresh firmware from git. Here is the tested device with a simple low-pass filter for making a nice example waveform (2kΩ resistor from GPIO2 to GPIO26, and 1μF capacitor from GPIO26 to ground). GPIO2 has a ~160Hz PWM output just for demonstration.

[FilipDominec]

Since you are using the previous release 0.2.0, I downgraded the firmware and run a similar script from the release's directory. And I got basically the same result. I guess there might be some hardware trouble with your device?

[FilipDominec]

Hi, having switched to a new board, I can confirm this erroneous behaviour. So thank you again. Now I can take a look at it and fix it!

The board at my desk is a new one with a RP2350 chip, not sure if this fact is related to the problem. But you probably had RP2040, as you didn't have re-compiled firmware for the RP2350 available. 99.9% it is my firmware issue.

[FilipDominec]

@AlessioCarnemolla I can confirm the signal on both channels comes from the first user-specified channel. The channels are slightly shifted, so it's coming from properly sampled values being wrongly interpreted as two interlaced channels.

I will try to fix this over the weekend. Then I will make a new 0.3.0 release, if there are no other troubles.

[FilipDominec]

@AlessioCarnemolla I will be releasing the new firmware soon and it seems fixed; will you be available to test it in your setup?

[AlessioCarnemolla]

Yes, I will try to sample with two microphone capsules. Thank You

[AlessioCarnemolla]

When will you release the new firmware? I will have the possibility to try it untill the 3rd of april.

Thank you for your avaliability.