Author |
Message |
ericcoleridge
Joined: Jan 16, 2007 Posts: 889 Location: NYC
|
Posted: Tue May 24, 2011 10:24 pm Post subject:
Serge Synapse Touch Controlled Voltage Source problems Subject description: I've got one breadboarded and partially working, need some help |
|
|
Nearly three years ago, I started this DIY TKB topic hoping to dig up a way to build a touch controlled voltage source. Over 15 or so pages of dialogue on this topic, Several circuits were proposed (Serge, Buchla, Paia) and a few layouts were even put up. In the end, the most promising was probably the layout for the Serge Synapse Touch Cotroller circuit, the schematic for which you can see in this link. The layout went through a couple revisions, but, I etched each one and I could never get it working.
So, anyway, I still want to build a TKB and I recently went back to the original Synapse schematic, and I breadboarded it. Lo and behold, it actually works quite well! Except for a couple issues that I don't understand:
First and foremost, I can only get a voltage response when I'm holding the shield of the cable going to the voltage input of my oscillator!! When I was first testing the circuit, I happened to be holding the cable by the metal part of the jack, and so I didn't notice this until I set the cable down on the table, and then it would stop working. I thought maybe it was a grounding problem, because I didn't have the touch voltage output jack grounded, but when I tried to ground the output jack, the circuit wouldn't work at all.
What's going on here, and how would I go about fixing this?
Besides that, the voltage output is sort of noisy, like it's frequency modulating my test oscillator at a constant frequency, in addition to the voltage response that the touch surface creates. This is obviously less of an issue at the moment.
I've got to figure this thing out, I've been trying for so long to get a working TKB, any and all suggestions are appreciated. |
|
Back to top
|
|
|
andrewF
Joined: Dec 29, 2006 Posts: 1176 Location: australia
Audio files: 4
|
Posted: Tue May 24, 2011 11:36 pm Post subject:
|
|
|
could you load up some photos/diagrams of your circuit, especially of the touch-plate? |
|
Back to top
|
|
|
ericcoleridge
Joined: Jan 16, 2007 Posts: 889 Location: NYC
|
Posted: Wed May 25, 2011 12:48 am Post subject:
|
|
|
Oh, you just wait, when I get this thing worked out, it's gonna be so cool.
But, how about this-- is it possible that the circuit was only working when my finger, touching the touch-plate (surrounded by ground plane) was connecting the ground from the circuit to the ground on my synth (via the jack)? I'm thinking maybe it will work fine as long as the two circuits share a common ground. Is this probable? |
|
Back to top
|
|
|
andrewF
Joined: Dec 29, 2006 Posts: 1176 Location: australia
Audio files: 4
|
Posted: Wed May 25, 2011 1:30 am Post subject:
|
|
|
sorry if this is obvious;
the touch pad ground plane need to be connected to the ground of your main circuit,
wiring needs to be as short as possible from the pad to the PCB, mine is about 2cm,
the touch pad needs to isolated, thats why Arpad( ) says to spray paint or poly-something on it. |
|
Back to top
|
|
|
LFLab
Joined: Dec 17, 2009 Posts: 497 Location: Rosmalen, Netherlands
|
Posted: Wed May 25, 2011 2:54 am Post subject:
|
|
|
Looking at these touch keyboards for a while now, seems really cool. Maybe I should etch a board and check it out. I have done a Zvex Wah probe clone which is a proximity sensor (piece of copperclad) controlling a wah circuit (resonant bandpass) with a vactrol. Worked right away, used a TL072 as oscillator. |
|
Back to top
|
|
|
ericcoleridge
Joined: Jan 16, 2007 Posts: 889 Location: NYC
|
Posted: Wed May 25, 2011 10:23 am Post subject:
|
|
|
andrewF wrote: | sorry if this is obvious;
the touch pad ground plane need to be connected to the ground of your main circuit,
wiring needs to be as short as possible from the pad to the PCB, mine is about 2cm,
the touch pad needs to isolated, thats why Arpad( ) says to spray paint or poly-something on it. |
Yes, I cut off a very small, maybe 2 Cm rectangle of copper clad. I 'etched' a line in the middle and soldered a wire from one half to the Touch Switch input, and soldered another wire from the other half of the rectangle to ground.
I wired jacks out from the Touch Voltage and Pulse. I attached a cable from these jacks to CV inputs on my synth (MiniMoog).
Only when I'm touching the metal 'barrel' of the cable does the circuit work. I feel like I'm inadvertently connecting the ground of the Synapse circuit (via one hand on the the Touch Switch) to the ground of my MiniMoog (via my other hand holding the barrel of the cable).
Is it possible that this 'Serge' circuit, probably designed for a single enclosed modular system, using banana's no less, was conceived to function in a system where the PS + ground was shared among all of the parts (Keyboard + VCOs, etc). Does this make sense? |
|
Back to top
|
|
|
Luka
Joined: Jun 29, 2007 Posts: 1003 Location: Melb.
|
Posted: Wed May 25, 2011 2:16 pm Post subject:
|
|
|
similar to any other system banana systems need to share the ground to any external systems - since banana cables don't include a shield containing 0V you need a specific ground cable (join as close to the powers supplies as possible) _________________ problemchild
melbourne australia
http://cycleofproblems.blogspot.com/
http://www.last.fm/user/prblmchild |
|
Back to top
|
|
|
ericcoleridge
Joined: Jan 16, 2007 Posts: 889 Location: NYC
|
Posted: Wed May 25, 2011 2:51 pm Post subject:
|
|
|
Thanks Luka, I guess this solves the problem. |
|
Back to top
|
|
|
akrearke
Joined: Sep 08, 2008 Posts: 53 Location: US
|
Posted: Thu Jun 27, 2013 12:25 pm Post subject:
|
|
|
I have sucessfully built the synapse article verion in ten notes. Used Cgs modified driver. Its performance is erratic. Used shielded cable to connect touch plates and dramatically improved performance. Gonna try the paia touch plates with ken's cv buffer on the back end. What I will say is this, generating a gate output and using a programmer or other cv generator from the gates is, I thinks the best was to go. The idea of using cmos to provide touch plate gates seems the best route, as logic is either on or off. Will keep the em forum up to date if I succeed, and provide foils and alterations. |
|
Back to top
|
|
|
LFLab
Joined: Dec 17, 2009 Posts: 497 Location: Rosmalen, Netherlands
|
Posted: Fri Jun 28, 2013 1:06 am Post subject:
|
|
|
Yeah, I recall reading that the distance between touchpads and driving/ receiving circuit should be short.
Thinking of doing this on a single PCB, and like yourself, use the Paia circuit, not sure if I'd go for the one with or without "velocity".
Bit overwhelmed by the programmer circuits, seems to be a lot of components and wiring. (I'd like two Cv's, gate, and some kind of keyboard scanning). |
|
Back to top
|
|
|
|