Can you suggest an easy and inexpensive way? Triggering LEDS with drum trigger

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Mbira

Well-known member
Joined
Jun 4, 2004
Messages
2,422
Location
Austin, TX
Hi guys,

I'm using these:
256393.jpg


Going through this to convert the pickup into midi and trigger samples through Ableton:
461469.jpg


So I have these discs that I will hit as triggers, but I really would like to be sure when one is activated.  Can you suggest a way to do that? 

One way is maybe there is a midi to LED box and I could send the midi message from Ableton to light the same event that is triggered by the hit?

Any thoughts?
 
Cheapest way I can think of is tapping off a signal from the trigger before it gets to the MIDI "brain".

Trigger -> Buffer -> Monostable de-bouncer/pulse stretch -> LED

Probably about 5 bucks worth of parts per trigger, although it won't help if you're trying to confirm the MIDI side of things.
 
Mbira said:
One way is maybe there is a midi to LED box

I think the solution can be pretty simple if you don't need individ. LEDs for each pad.

Just split your generated MIDI-data into the original destination and a feed for the monitoring circuit.
Note a MIDI-input circuit is usually an opto-coupler, 'so the incoming part is a diode'. We now just use a diode that's visible outside:

To spot MIDI-activity I've a few simple DIN-5 plugs in the tool/gizmo-box that are just
the connector, a LED, a series resistor (220 Ohms ?) (and an anti-||-diode IIRC).

This can spot single MIDI-notes well, even short hits.

Only complication can be if the MIDI-box already generates a stream for some reason
that spoils the contrast to the drum-hits. If there's such 'background data' check if you can disable it.


Bye,

  Peter
 
Hey-I do need individual LEDS for each.  To clarify, there will be 9 triggers that are each programmed to trigger a specific place on the timeline in Ableton.  IE: Only one will be used at once.  Not sure if that changes things.  The reason I need the light is to be sure:
1. the sample got triggered
2. So I can remember which one I just hit so I hit the next correct one in sequence.
 
For indiv. indication I'd now turn to some software application that you can run alongside Ableton, reflecting the status of incoming MIDI. Would that be OK or do you need some indication away from the screen ?
 
Yes please. ;D Because we are jumping all around the stage and the idea is these things will all be set on the computer so we don't have to (can't) watch the screen.  I'm wondering though about sending a midi signal and triggering an LED?  Seems like that would be semi-common?
 
Mbira said:
Yes please. ;D Because we are jumping all around the stage and the idea is these things will all be set on the computer so we don't have to (can't) watch the screen.  I'm wondering though about sending a midi signal and triggering an LED?  Seems like that would be semi-common?

You'll then need some decoding of the MIDI-notes into separate LEDs.

And you'll need some method to extend the LED-on after the trigger has ceased.
You'll probably want to have the LED on as long as the sample/loop/thing is playing.

Maybe a box like this is cheap there days ? And modify & connect it to more LEDs ?

rol_mks900_sigInd.jpg



A weird alternative that will perfectly work if you happen to have some surplus simple hardware around is to trigger both your wanted sound
and another 'dummy sound'  (say an ongoing  1kHz sinewave/pad/loop) in some other module in parallel, so responding both to the same MIDI-note-number.

Say you need eight indiv. LED-indications and your surplus synth/sampler/soundcard has eight outputs.
Route each of the eight each to a separate audio-output and connect a simple signal-present or signal-clip LED to is, circuit to be found here @ GDIY (don't listen to those sounds)

Bye,

  Peter
 
clintrubber said:
You'll then need some decoding of the MIDI-notes into separate LEDs.
That's the part I'm not sure how to do.

And you'll need some method to extend the LED-on after the trigger has ceased.
You'll probably want to have the LED on as long as the sample/loop/thing is playing.

Not sure if you've used Ableton at all, but what each of these triggers will play is a sort of scene with several clips in the scene.  I could have one of the clips just be sending midi notes out continuously (or as a rhythmic flash ;D).  Each scene could send a different note corresponding to a different led.

So the question is how to make midi notes trigger an LED?

I've seen this:
http://www.jpleisure.co.uk/item117.htm

which is eventually where I want to get (controlling lighting through the Ableton as well...but not sure if there is a quick and fast way to get this aspect working.
 
the peizo elements in those triggers produce a voltage spike when vibrated.  Use some zener diodes, some fets and a handful of resistors and you can trigger the LEDs directly without all of that fancy stuff..

 
But then how would you keep the LED lit after the strike, and how would the LED turn off after the next hit on a different pad?

 
I do NOT go over to that side of the shop, but I believe that lighting systems are (or were) MIDI controlled.

Then you get an old 8-in(*) MIDI light-board, program it so that the code for "pad 1 hit" triggers a "lamp 1 ON" event, etc. Plug in some little night-lights (or LEDs with serious dropping resistors).

(*) You say nine, but 8 may be a stock size, and if you get the first 8 right then 9 may happen OK.
 
Mbira said:
But then how would you keep the LED lit after the strike, and how would the LED turn off after the next hit on a different pad?

Sound like a nice application for a 555 timer-IC per pad: configure it as a one-shot to extend the lighting and reset it when one of the other pads is hit. So the reset for each is looking at the other (say) seven pads, performing in fact an OR-function (might be possible simply by diodes, didn't check) 

While a simple principle, it still looks like quite some soldering though. How about having a look yourselves at the Ableton-output ? Or is what's to come next not obvious from looking at the current clip ? (looks like it's indeed so, otherwise you wouldn't have started this thread  ;))

Bye
 
Since you really want to know if the midi brain recognized the trigger, no variation of detecting the analog pulse raw trigger output is appropriate. I would use a cheap little pic microcontroller to read the midi data and drive LEDs.  Software will interpret which trigger was hit, control how long the LEDs stay lit, etc.. I bet there is some free c source code floating around on the web that would get you started with dealing with midi.  Course, if you aren't already up to speed with this kind of thing it will take a while.

Mike p
 
Ableton can easily send the midi to whatever device will switch on and off LEDs.  Then only rub is finding the controller will turn on an LED with midi info. 

There must be a way to do this without having to run through a midi/dmx converter->dimmer pack->lights.
 
Mbira said:
There must be a way to do this without having to run through a midi/dmx converter->dimmer pack->lights.

Continuing on the 'trigger a dummy sound in parallel'-route I described above, how about making those dummy-samples distinct sinewave frequencies ?

Send them all out from an unused single audio-output from the computer (let's assume that's possible, otherwise add the cheapest USB-audio-I/O). Sinewaves can be one-shot triggered samples having the duration as required.

Make those different frequencies cover the full freq-range of your audio-I/O.
For NINE indicators for instance 60 Hz, 120, 240, 480, 960, 1920, 3840, 7680, 15360 Hz
(taking into account bandwidth-limitations of your audio-I/O, and/or shift if you expect massive mains-interference etc)

Using bandpass-filtering in front of a simple LED-clip-stage (say little more than one transistor & LED & some R & C) you can discern between the frequencies. Adjust the sample-levels to finetune. Simple passive filtering might work, otherwise you might need to add one transistor for making it an active bandpassfilter in front of the usual clip-LED-stage.

This should be pretty simple and have low parts count per LED-indicator:
- one or two BJTs
- one diode
- some resistors & caps

If it works well enough you might want too squeeze in more frequencies. And using an audio-interface running at 96kHz sample frequency instantly doubles your amount of possible indicators while maintaining filter-spacing.

If you go the route of making the bandpass filters pretty sharp (say a twin-T in the feedback of an amplif-stage) then note you can finetune the circuit-response by adjusting the sinewave-frequencies to the actually realized filter-frequency (the top). So keep the circuit with its tolerances like it is and adjust at the sinewave-generation side.


Bye,

  Peter   
 
I like this idea!  This would be doable with my equipment.  If I'm understanding, this would be like building a "feedback detector"?
 
Mbira said:
I like this idea!  This would be doable with my equipment.  If I'm understanding, this would be like building a "feedback detector"?
Nice, good to hear this might suit your needs. Note that if you have one audio-output to spare you probably have a second channel available as well, which will make the filtering less critical (using panning, it allows to spread out the freqeuncies more) and/or allows more frequencies(=LEDs).

yep, it basically comes down to spotting frequencies. If you put them at the ISO-freqs you could do without soldering & watch an RTA-box during gigging'n'jumping. if you can find one cheap you could actually use such a beast for the filtering & mount the LEDs where they suit you best.

Bye,

  Peter
 
Back
Top