[sdiy] How long should a Trigger pulse be?

rburnett at richieburnett.co.uk rburnett at richieburnett.co.uk
Thu Dec 19 10:30:46 CET 2013

I'd shoot for minimum high time of 2ms and minimum low time of 2ms.  I 
think that would take care of the finicky early micro stuff that only 
polls every millisecond or every two milliseconds.  And it still allows 
your gate signal to do a complete cycle in 4ms for a gate frequency of 
250Hz.  Do you need to go any faster than that!?!?


On 2013-12-18 23:04, Tom Wiltshire wrote:
> Many of you have raised the same point, which is that gear that polls
> the input is likely to be much slower to respond than either analog
> gear or gear that uses an interrupt to detect changes, In this regard,
> the early microprocessor-based stuff is probably the worst case
> scenario, where the inputs are polled in round robin style at a fairly
> <yawn> slow rate.
> I'm not particularly interested in designing something that works for
> the worst-case scenario, especially if that's some rare/special piece
> of gear from the 80's that had a desperately overloaded processor. If
> it works on most gear most of the time, that's fine with me.
> It sounds like I can't go much below 1ms without starting to run into
> problems, if I'm summarising correctly. And even that might be a
> problem for the 1980's polled input synths - but I think that's
> probably that's their problem, not mine.
> Thanks to all,
> Tom
> On 18 Dec 2013, at 19:11, "Dave Brown" <davebr at earthlink.net> wrote:
>> I've always used 5 mS for the stuff I've designed.  I don't remember 
>> exactly
>> why I picked this number but it works well with microprocessor-based 
>> modules
>> that poll for a trigger.  My ARP 3604 Keyboard outputs an 80 uS 
>> trigger.  I
>> would suspect any vintage gear would work with a short trigger as it 
>> is most
>> likely edge sensitive or used as a set (or reset) input.
>> My ARP keyboard had another issue that caused problems with some gear. 
>>  The
>> gate is used to generate the trigger pulse so the gate goes true 
>> before the
>> trigger pulse (I don't remember by how much).  I ended up adding a D 
>> flip
>> flop with the trigger as the clock and the gate as the reset-0 to
>> synchronize them.
>> Dave
>> Modularsynthesis.com
>> -----Original Message-----
>> From: synth-diy-bounces at dropmix.xs4all.nl
>> [mailto:synth-diy-bounces at dropmix.xs4all.nl] On Behalf Of Tom 
>> Wiltshire
>> Sent: Wednesday, December 18, 2013 6:38 AM
>> To: synthdiy diy
>> Subject: [sdiy] How long should a Trigger pulse be?
>> Hi All,
>> I'd like the benefit of your combined experience, if I may.
>> I'm decoding MIDI Note On/Offs and outputting Gate and Trigger pulses 
>> to
>> drive envelopes. My own envelope designs don't need a very long 
>> trigger
>> pulse in order to fire, but I thought perhaps some others do.
>> Currently I'm using 500us Triggers, but I don't have a big pile of old
>> synths here to try that on, so I don't know how well that will work on
>> vintage gear.
>> And then of course, there's the problem of envelopes or synths that 
>> don't
>> have a Trigger input...
>> I remember the Kenton Pro-2 MIDI-to-CV convertor had a "retrigger" 
>> mode
>> where it dropped the Gate briefly to allow Gate-only envelopes or 
>> synths to
>> retrigger. It didn't say how long it dropped the gate for though. I'd
>> thought to do something similar with the following circuit:
>> http://www.electricdruid.net/GateAndTriggerCircuit.png
>> This converts Gate and Trigger into a Gate-with-gaps that will 
>> retrigger a
>> Gate-only envelope. And this is the main reason for my asking the 
>> question;
>> In this case, the output gate is essentially delayed by the length of 
>> the
>> trigger pulse, hence my desire to keep it as short as possible. If it 
>> wasn't
>> for that, I'd just use 5ms and reckon that'd trigger anything.
>> So, oh wise ones of Synth DIY, how short can my trigger pulse be?
>> Thanks,
>> Tom
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at dropmix.xs4all.nl
> http://dropmix.xs4all.nl/mailman/listinfo/synth-diy

More information about the Synth-diy mailing list