[sdiy] Two parameter mechanical controller thoughts....

[thresholdpeople]

If I were going to build it, I'd lose the softpot and replace it with a mechanical drive mechanism - either something built onto linear bearings, something on a guide rail, or a telescoping slide - turning either a 10-turn pot, depending on the gearing of the mechanism, or an encoder.

Check out McMaster for different possibilities - https://www.mcmaster.com/linear-bearings. These types of things can be DIYed as well.

With a rigid carriage on some sort of guide rail system it'll be easy to mount an additional encoder, and an LED or anything else you need there. For wire management, if you want it real clean and fancy something like a cable and hose carrier will do it.

Boris

‐‐‐‐‐‐‐ Original Message ‐‐‐‐‐‐‐
On Friday, June 7, 2019 2:13 AM, Pete Hartman <pete.hartman at gmail.com> wrote:

> At the risk of depriving myself of a market advantage....
>
> If you haven't seen videos of the Collidoscope, go to youtube and watch now.
>
> The control mechanism -- encoder sliding in a slot on some kind of carriage that presses down on a softpot underneath -- strikes me as a really appealing and expressive means of controlling two parameters simultaneously. However, they don't provide a lot of mechanical detail at the micro level (that I've been able to find... they call out the encoders and the wiper for the softpot etc, but not the carriage mechanism).
>
> I've been thinking about how this could be accomplished as just a CV generator -- one CV that reflects the position in the slot, and one that reflects the position of the encoder. Would be good to use an LED shaft encoder as well, so you could light the LED with a brightness that reflects the level of the encoder, so you can look at a glance and see what both CV values are, roughly.
>
> The tough problem to solve here, of course, is the mechanical carriage. I need 4 signals and ground to deal with the encoder and an LED. Maybe only 3 signals, but I'm going to go with 4 cos it's symmetrical. Don't want to use a set of wires rolling back and forth for the signals because they will wear out and break off.
>
> My first thought is something to somewhat mimic a slide pot. PCB that holds the encoder, has a hole for the wiper screw. Glue a nut to the PCB on the bottom side, screw the wiper through it and adjust for the height to the soft pot on a PCB underneath.
>
> PCB has 4 spring contacts (two on each side of the encoder) that ride on bare traces on the PCB that comprises the panel These are on the "top" of the carriage PCB, pressing up against the panel. These are the 4 signals.
>
> PCB rides on a pair of rails that are made to be soldered into the panel PCB, tied to ground, and the PCB has a pair of bare traces that connect into its ground. So the rails carry the ground to the carriage PCB.
>
> SO here is where my ignorance of what's out there takes over... is there anything like this rail "out there" that I could buy off the shelf? What about the spring contacts? I'm concerned about the rigidity of the rails -- pressing down on the encoder while it's in the middle of travel shouldn't be able to deform the rails. What about twisting of the carrier PCB? I don't want it to rotate off the rails. Presumably if the encoder has a reasonably loose rotational resistance it won't be a problem, but I don't think that's a good thing to assume. I have a bunch of copper wire of various thicknesses, and my initial thoughts about this were I could prototype using that as springs and rails, but I definitely don't think copper has the necessary rigidity for a 100mm long throw in the slot, which is what I'd want to have.
>
> Is there some other approach that would work better/more reliably? Are there pre-made slide pots that have encoders built into their shafts? Am I missing some other prior art on how this could be accomplished?
>
> Obviously I'm inspired by the Collidoscope; it's not a particularly novel mechanism. But it's not something that "exists" in a form generally usable for synths either.....  At least not that I've seen.
>
> Thoughts?
>
> Thanks
>
> Pete
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://synth-diy.org/pipermail/synth-diy/attachments/20190607/5648db60/attachment.htm>