Noise on input channels

Hello,
I have a micro maestro with 1 servo and three inputs. I had an older maestro which was so old it is 1.01 firmware. Once flashed, it worked flawlessly. I bought to more. However, these newer maestros seem to have noise in the line when I configure them with three inputs. This is before I ever attach any external devices. Once I configure the channel as input in the Channel Settings, if I go back to status, there is noise between 0 and the first 25% tick mark. This is for both maestros, but again, the older one does not behave this way. I also wired up the the maestro to see if the behavior would exist after a common ground and individual +5 voltages were applied and the noise seemed to get worse. Firing one channel high would bring up the other channels to 60% to 80%.
Are the newer Maestros expected to behave like that? Did I somehow miss some setting with the second and third maestro?
Thank you

Hi.

In general, it is not unusual for an input with nothing attached to it to read random noise. In your case, it seems like you are changing the output of other nearby pins, so the noise you are seeing might be bleeding over from those, and not really random at all. Could you try connecting those inputs to some analog voltage or ground and seeing if you get any noise on the readings then? If you want those pins to have no noise even when nothing is connected, you could add a pull-up or pull-down resistor.

-Claire

Thank you. A pull down resistor did work. Since I didn’t need one for the other maestro, I resisted doing it here. Thank you for the push.

I have this issue.

I use 3 inputs to read voltage values off a 3 axis accelerometer as voltage.
It only outputs 3.3v to 5v range input but its enough.
It jitters quite a bit and that’s not at the accelerometer output, only when plugged into the maestro

I run 2 maestro boards, one for servos with separate power and one only for inputs to keep noise down on the input maestro board.
What I ended up doing was using the values coming off a 4th ch open pin as a reference voltage value to the 3 values coming off ch 1 2 and 3.
The open pin value jitters around a bit so figure if the ref voltage follows the jitters in the other 3 inputs then it will correct the issue, but improvement is only minor.
And the problem is I cant put a pull down resistor on the 3 inputs from the accelerometer because it just shorts the signals out so no way to stabilize them.
So it just jumps around a bit and had to add dead band code to software to reduce it.

I don’t know if the input floating issue can ever be fixed.
I concluded it needs to voltage follower circuit in between that could take the low current value off the 3 accelerometer axis and convert that to higher current (over 5v range too) with pull down at the maestro input to clamp it down.

Hi.

It does not sound like you are having the same issue as the original poster. Their issue was that some of the input channels on the Maestro that were not connected to anything were reading noise. In your case it sounds like you are getting more jitter than you want on readings you are doing on lines connected to the analog output of your accelerometer. You mentioned that you do not see the same jitter when you directly measure the output of the accelerometer. How are you measuring it? How are you seeing the jitter when using the Maestro? Can you quantify the amount of jitter you are getting in both cases? What accelerometer are you using?

-Claire

You might be able to replicate it.

This is the accelerometer

pololu.com/product/1246

I use ch 1 2 3 in Maestro for the 3 axis off it
I run the voltage from it to ch 4 as reference voltage to the other 3.

I was thinking perhaps the outputs from the accelerometer need to be buffered to decouple them from the Maestro inputs.
But for my purposes I use the output as a visual indicator of angular G on a round target so it just causes only a minor jitter of the indicator.
If I had to use the values for anything more then that Id need to get it more solid.

It is possible that the jitter you are seeing is within what we would expect from that accelerometer, but it is very hard for me to tell if you do not try to quantify how much you are getting and how you are measuring it. I think it is very unlikely that the jitter you are seeing is due to whatever ambient noise is causing your open pin to float, and I do not think that trying to reference the analog readings of the accelerometer to a floating pin will help you eliminate the jitter. Instead, you should consider implementing some kind of averaging of the accelerometer outputs.

-Claire

Thanks Claire

I figured it might just be the tolerance of it
Also that it outputs a range up to 3.3 and the input range is 5v / 1023 so the output looses some resolution at input because 3.3 is over a smaller resolution than 1023, thus the jitter may be amplified due to a reduced resolution.

If the Maestro could be switch the a range from 5v in to scale to 3.3 (like the Arduino does I believe) then it would have 3.3 / 1023

That said for my use the output is only used as a visual indicator so it’s good enough.