All midi signals maxed out

[huit]

I finally (many years on) *finished* building my megadrum (Synthex kit v3.1 56i now loaded with firmware 20100516) and have a problem with the midi signals always being maxed out. I've adjusted the curve, compression, set all gains low and with a very low curve/high compression I can get values below 127, but they are always the same, i.e. no dynamic range.

My first assumption is that the simple piezo trigger taped under rubber pads I'm using for testing are just too sensitive, but wouldn't I at least be able to get some values below 127 with really gentle hits? I've tried isolating the triggers with foam under a DIY mesh pad setup with no improvement.

Any ideas? problem on the board anyone might be able to diagnose? Any help will be appreciated and I'll put up some photos of the build when it's all running

[dmitri]

You forgot to adjust HighLevel.

[huit]

I tried high level auto "YES", big hits, and high level is now on 985, changed back to high level auto "NO" to lock it in but all hits are still maxing out Something is working though because I just tested the same process with weaker hits and got a lower high level value (143).

P.S thx for the fast reply!

[dmitri]

Hitting(or even touching) bare piezos produces high voltage.

[huit]

I'm not hitting the piezos directly, i have two test drums. One is a rubber practice pad with mdf (glue and wood fibre) base that I have taped the piezo underneath, and the other is a mesh pad that has the piezo mounted between foam pads on a mounting bracket.

I am guessing that the fact I get variable values from HiLvlAuto reading into HighLevel suggests the voltage being received from the trigger source is variable and in a suitable range for megadrum. This makes me think modification of my triggers is unlikely to help. Is this a reasonable interpretation?

If so, it would seem the problem is in processing and/or output and I'm hoping this information is enough for someone with a little more knowledge about the signal processing to see a potential source of the problem. Any help will be appreciated, I'm so excited about the possibility of getting this thing up and running.

EDIT: On thorough examination of the board I had swapped R4 and R29 (i.e. a 47k on R4 and a 4.7K on R29) and R8 and R21 (1k on R8 and 10k on R21). I'll swap these now and hope that was my problem!

Well the LCD seems clearer now but the MD is still maxing out all midi signals

I just tried a new piezo that only yields HighLevel of 340 on HiLvlAuto, and this one still maxes out to 127 on all hits. This seems like further evidence that it's something in the signal processing rather than on the trigger end but once again I'm open to suggestion.

[huit]

I tried with a resistor (62K) in line on the piezo and despite requiring much bigger hits to trigger and HiLvlAuto (with very solid hits) writing a lower value (~300) to HiLvl I still only get 127 midi values. Does anyone have any other ideas?

[dmitri]

You actually have to raise HighLevel to make MIDI velocity lower with the same input signal level.

[huit]

Thanks again for your input, I really appreciate your time and I'm sorry for not being entirely clear.

With the 62K resistor in line (between the piezo and the "tip" on the TRS jack) and "all gains low" set to yes the highest value the HiLvlAuto inputs to HiLevel, even on very hard hits to the pad, is ~100. Despite this, even manually raising the HiLevel value to its highest setting (1023), I still get only 127 values in the velocity value of the midi signal on the lightest hits that trigger. This is true even for "ghost" hits triggered when the threshold is lowered to around 5.

[dmitri]

May be you set Curve to Max?

[huit]

On all of the pre-loaded curves, including "Max" I get the same value.