Input polarity

[estregan]

Hi, I have trigger pads designed for use with Roland modules. As you may know, Roland pads have the head piezos wired negative while pads for Alesis modules are wired positively. Would the megadrum accept either kind of head input or just one kind.

Here's a picture of the wiring I'm talking about:

[BlueDragon]

Interesting Question

I was going to wire mine like the Alesis, but I am not sure that it matters

I think that a piezo will create an AC voltage that swings from + to - - to + more like -2v to +2v until it stabilizes (bigger swing harder you hit it)

not sure about the actual voltage produced

http://www.pc-control.co.uk/piezoelectric_effect.htm

I could be wrong though, they do seem to wire a piezo with the ceramic Red + and outer disk Black -

that would mean Roland solder the red wire to ground/earth/common on the head piezo

http://www.edrum.info/theory.html

Edit:
hmm been thinking about this and if both piezo are hit at same time on the Roland pad, the hit would be out of phase e.g rim in - and head in + or vis versa maybe that is why they do it

[dmitri]

The piezo signal does swing between - and +. MegaDrum samples continuously and will measure several positive swings during a sampling cycle so it won't matter how you connect them. Tried connecting piezos using either polarity and did not see any difference.

[aquawicket]

Interesting Question
hmm been thinking about this and if both piezo are hit at same time on the Roland pad, the hit would be out of phase e.g rim in - and head in + or vis versa maybe that is why they do it

I've always wondered why roland did this and also noticed that they have the smoothest position sensing drum modules. I'm not sure if the differences in polarity give them any kind of advantage when determining position or not. But next time you goto the music store, play one of those roland V drum kit snares and slowly move the stick from the middle to the rim and notice how accurate it is.

[BlueDragon]

Ok its based on what I know but unproven guesstimate

I see now it might be better for Roland, less interference / crosstalk maybe

Very Interesting, so I did a sketch of the how I see the signal

Roland_Output3.jpg

timescale will properly be in ms