Bernard Mentink wrote:
> Hi Guys,
> 
> To go with Hydrogen I am developing an open hardware project of a 
> trigger->usb module that appears as a midi device
> to Hydrogen. I am aiming the cost build to be <$50.

You should check out edrum.info and megadrum.info to see if they might 
meet your needs.  Megadrum in particular sounds very similar to what 
you're trying to do.  With 12 inputs, I wouldn't think the price to 
build would be much more than $50.  I've built edrum, and although it's 
more complex a design than I'd like (complex for some good reasons, I 
should add), it does work very nicely.  It can be built for not much 
more than $50 as well, albeit without a USB interface.  I bought an E-mu 
X-midi 1x1 for $30.

Having said all that, I'm considering developing a device similar to 
yours; perhaps we could collaborate.

The reasons I think it might be useful to have another drum brain design 
are:

1) <$50 parts cost
2) open source firmware, hopefully resulting in faster development
3) minimal and modular hardware design, for an as easy as possible build

Does that sound compatible with what you're trying to do?

I have put together a parts list for what I think is everything 
required, and it comes out to just under $50, much of which is in the 
case, proto board and jacks.  I haven't considered the USB interface, 
however; I'm not sure what that does to the parts list.

> My aim is to provide 12 trigger inputs (from a piezo sensor) and several 
> switch inputs (for close/open high hat switch and jazz/rock snare 
> selection .. and probably some others)

What processor are you considering?  You can get 8 analog inputs from 
the atmega644; more than that will require a multiplexer.  You might get 
some good ideas from asking in arduino forums.

A $15 off the shelf keyboard sustain pedal will get you open/closed when 
connected to a digital input pin.  I need a little more control than 
just closed/open, at least closed/half-open/open/free, but that's easy 
to implement.  Admir has a nice optical foot controller design at:

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

For the switches, it'd be easy to implement a mux to give you pretty 
much as many switches as you wanted.  If you wanted to, though, you 
could probably get away without dedicated hardware support for changing 
samples (e.g., jazz snare/rock snare).  You could have your device or 
sequencer respond to a strike from one trigger by changing the note or 
sample for another trigger.

> So far I have my hardware enumerated correctly  as a midi/audio device 
> and I can send midi events to Hydrogen ... so the usb midi
> stack is working.

Cool.  That's a piece I was considering a nice-to-have.  I was just 
planning on simple MIDI jacks for output, but USB support is really 
nice, and saves people from having to buy a $30 external USB-MIDI interface.

> Next, is to do A2D work of scanning all trigger inputs, setting 
> thresholds, measuring the volume of the hit, and sending messages back 
> to the midi interface.

I have not settled on an algorithm to calculate velocity, but I was 
planning on using a very simple rise-time based algorithm for the 
initial implementation, similar to:

http://todbot.com/blog/2006/10/29/spooky-arduino-projects-4-and-musical-arduino/
http://todbot.com/arduino/sketches/midi_drum_kit/midi_drum_kit.pde

My plan is to get a reliable device working with limited sensitivity and 
refine the algorithm as it becomes apparent that it's not accurate enough.

There are some good discussions of ADC at:
http://www.arduino.cc/playground/CourseWare/AnalogInput
http://forums.makezine.com/comments.php?DiscussionID=232

As far as measuring the peak voltage, you have to be careful that you 
don't try to measure a voltage that will destroy the ADC.  My DIY 
triggers generate ~30v or more from a hard strike, so over-voltage 
protection is required.

Dave


> I believe there are others on the list that have done similar things. I 
> am wondering if anyone has experience with the signal-processing side of 
> reliably detecting a "hit" and measuring the peak/avg to generate the 
> volume parameter. Any help most appreciated.  I though maybe of 
> differentiating the signal to get the trigger and integrating to get the 
> volume ... any one have some algorithms?
> 
> Many Thanks,
> Bernie
> 
> 
> ------------------------------------------------------------------------
> 
> ------------------------------------------------------------------------------
> Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day 
> trial. Simplify your report design, integration and deployment - and focus on 
> what you do best, core application coding. Discover what's new with
> Crystal Reports now.  http://p.sf.net/sfu/bobj-july
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> Hydrogen-devel mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/hydrogen-devel


------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day 
trial. Simplify your report design, integration and deployment - and focus on 
what you do best, core application coding. Discover what's new with
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
_______________________________________________
Hydrogen-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/hydrogen-devel

Reply via email to