• Interactivity

  • BBG WiFi support??[WAS: Noisy analog inputs with Pd]

Hi,

I've been working with the analog inputs in Pd, just using a pot via the adc~. It seems like the input is super noisy. When the pot is standing still, I'm getting values that range from .99 - .005. Has anyone else had these issues, or is there a way to calm down the inputs that I'm using, or could I be missing something simple? I've used various pots connected to 3.3v/GND/various analog inputs.

Thanks,
Colin

Hi Colin, I was having a similar problem trying to get the "theremin" sample code to work. I found that I needed to connect the ground side of my potentiometers to the gnd pin in the same header row with the other analog pins, rather than the gnd pin on the beaglebone expansion pins. HTH.

Thanks for the quick reply. Unfortunately that doesn't seem to help. Still getting values all over the place. I'm assuming that the 3.3v supply on P_9 is the one I should be using for power....?

    Make sure you are addressing the correct input pin. When nothing is connected to an analog input channel, it will be floating, because they have a very high impedance. They may also have some crosstalk from neighbouring channels when floating.
    In Pd, [adc~ 3] is analog input 0 on the board, [adc~ 4] is analog input 1 and so on until [adc~ 10], which is analog input 7.

    To check whether you are addressing the correct pin, try connecting it directly to P9_03 and P9_04 (which are 3.3V) and you should read about 0.81.

    czyskows you might want to try using the 5V supply from Bela's analog out header, as the voltage range of the analog ins is 0-4.096V, though you should be able to use any of the power and ground signals from the BeagleBone.

    Try using the scope-analog example from the IDE to visualise the signal from the potentiometer and see if that gives any hints as to what is causing the noise.

      The signal is a tiny bit better when using the 5v supply. It's at least stable at the top and bottom of the pot. Anywhere else is still incredibly noisy. I've now tried this with a couple different Belas. I'm positive I'm using the correct adc~/pins. I got an image from the scope (it would be awesome to be able to attach files in the forum, wink wink, nudge nudge) and that signal displays the noise - is there something else I should be looking for with the scope? I've tested the power with my real-life scope, and the signal is stable, which makes me think it's not a power issue.

        czyskows you're right about being able to upload images, we'll have a look into that. Meanwhile, could you email the image to info@bela.io?

        If you look at the potentiometer's output with your bench-top oscilloscope, does it display less noise than Bela's scope?

          czyskows you mentioned that the signal goes all the way between 0.005 and 0.99, is it still the case?
          can you confirm your pot is connected to the analog inputs and not to the analog outputs?

          Could you send a picture of your setup along with a pic of the scope?
          Thanks

            Images sent. By the way, and I don't know if this is important, but this is all happening on the BB Green.

            Are the analog inputs 0-7 DC coupled?

            I did - the c++ example does the same thing.

            Can you confirm the pot is pushed down firmly on the breadboard? Also, can you test by connecting the 3.3V line from P9_03 straight to the analog input ?

            Yeah, I've done both of those. The values right from the 3v3 supply do the same thing. And that supply is pretty stable when I test it on my hardware scope.

            Hi,

            I just traded out the BB Green for a BB Black and everything seems fine. No more noise.

            Unfortunately, I'm using the Belas for a summer class I'm teaching and we just bought 10 BB Greens. Oops.

            If you might happen to know of a workaround for the Green, that would be awesome....

            Anyway, thanks for the assistance. I'll keep you posted as to any more Green/Black issues I might run into.

            Cheers,
            Colin

            Hi Colin,

            Strange problem. We'll look into what's going on with the BBG -- it should be the same in every practical respect so hopefully there's a simple workaround. Something you might check in the meantime, since you have a hardware scope, is the quality of the analog outputs (use one of the demos which generates a signal on them). If that looks okay then the problem is specifically in the ADC, otherwise it might have something to do with the SPI bus that drives both the converters.

            Also, when you mention you tried this with two Belas, did you also try with two different BeagleBone Green?

            Yeah, I tried with two different Greens and at least a few Belas.

            alright thanks, I will test the BBG we have tomorrow.

            I have no troubles with the BBGs we have here. These are 1-year old now, we bought them as soon as it first came out. Manufactured by Seeed, serial numbers BBG 115080763, BBG 115081013.
            Is it a regular BBG or the BBG wi-fi you are using?

            How are you powering yours? I am powering mine from my MacBook via usb.
            If you are powering from a laptop via USB, do you observe the same behaviour when powering from different laptop and with/without it being connected to the power adapter?