I see this in the article on adding pots:

>we suggest adding a .1uH inductor in place of the green jumper wire, which is right next to the usb port on the logic board. this will clean up the analog signal.

Is there an equivalent improvement which can be done to the kit logic
board?

edit

also, I can't yet get _40h_MIDI to work with the adc port (button 
presses are making it through fine..)

is there any way to calibrate the inputs in monomeserial? (the fader 
tops out at approx .875, so I can't generate up to 127)

monomeserial provides no calibration. 40h_midi does, however. did you 
turn on the ADC ports in monomeserial?

the kit has the inductor on the FTDI board already.

thanks for your reply tehn


OK, figured it out - it had to do with offset settings I had in 
monomeserial

how easy would it be to modify the _40h_midi patch to span multiple 
devices?

any ideas of what is causing the fader to give really jumpy signal? 
maybe it's crap - i'm going to test with a rotary pot.  but I have tried 
2 different faders and they act about the same.

another question:  do all three connections to each pot need to be 
discrete?  could the ground or the 5v be common to more than one?  just 
trying to see if I can reduce the # of wires here.

same result with a rotary pot

in the end, the jumpy signal means that turning the pot down can 
sometimes actually increase the value...  then turning it down a little 
more will make a big jump down, not smooth

hmm

I guess next test will be to try it on the other logic board

I'm experiencing a little jitter after I stop moving my pots. But 
nothing too drastic - basically I'm sometimes seeing X come in followed 
by X + 1, X, X + 1, X, etc.

I'm compensating by storing the last two adc vals that have arrived.  If 
the new value matches either of the previous two, I ignore the new one 
(this is done per adc port):

  if (mPrevAdcVals[0] == newAdcVal ||
      mPrevAdcVals[1] == newAdcVal)
    return;

It does mean that a change in direction isn't picked up immediately but 
that hasn't been an issue so far.

I'm finding that something is lagging when I go from full off to full on 
very quickly.  Either I'm not getting the adc changes as fast as I'd 
like or my processing of the events are causing a lag.  I'm talking a 
lag of less than 1/2 a second - probably even less than 250ms - but 
noticeable compared to my MIDI Solutions pedal controller (which is 
being swapped out for the monome kit).

@longjohns

are you using a linear pot, or an audio-fade? you must use a linear.

reducing wire count, yes, you can use the same ground/5+ for all pots.

@sean

filtering/smoothing is always an issue. we send the full 10 bits of 
data, and there's bound to be noise on in the conversion. there's 
smoothing in the firmware, but you'll probably want additional smoothing 
(like you mentioned) in your own app.

as for the full-on to full-off, try hooking up a hard switch and seeing 
what the delay time looks like. at that point, it's purely due to 
firmware smoothing, which could be reduced but would require reflashing 
the chip, etc.'

I tried adding a 10k resistor on the ground.  That made the signal much 
cleaner, but really limits the range.  Which makes me want calibration 
on monomeserial even more!!

;)
Also tried some larger valued resistors, which cut down the range more 
and more

Would it be correct to say that the greater the value of the resistor, 
the cleaner the signal?