clickable rotary switch



Moderators: PumpkinPieces, Gunner Recall, metalmariachi

Forum rules
The DIY forum is for personal projects (things that are not for sale, not in production), info sharing, peer to peer assistance. No backdoor spamming (DIY posts that are actually advertisements for your business). No clones of in-production pedals. If you have concerns or questions, feel free to PM admin. Thanks so much!

clickable rotary switch

Postby mysteryroach » Sat Jul 22, 2017 5:02 am

I'm getting a local pedalbuilder to make a patch-bay/interface for me that includes a TRS-splitter/switcher to switch between what each individual pedal receives: expression pedal (accessable front input) or cv generator (permanently plugged in back input). 8+ outputs going to 8+ pedals.

It was originally going to be switching between just 2 TRS inputs, and didn't need an off - so a 2-way switch worked, or even a button. But now I'm deciding what to do with the excess space in the box, and thinking maybe adding more - another front input for another expression pedal, maybe a back input for an envelope device, and maybe even more for other CV generators I have if I go over-the-top with it. As well as maybe an off switch.

So I figure the best way to do that is wiht a rotary switch. (I think that's what they're called?) Thing is, if I want to, for example switch from expression input 1 to expression input 4, I want to avoid (if possible) going through 2 and 3. It could introduce problems, e.g. where 2 and 3 might be noisy settings, or I'm controlling pitch or something. Are there any rotary switches/dials/knobs that double as a button, or that you can pull-out/push-in like some guitars that have switches on the knobs. I'm not really sure what to look for, and I want to get back to him with a more complete proposal of what the device is to include, rather than hound him with too many questions and get him to hunt down specific parts for me. If any of you know off-hand the type of thing I'm looking for, and where I would find it, that would be awesome. Even what forum would most likely be able to help me out with this dilemma (I assume a DIY forum? any in particular?)

Is there any other solutions to the problem? (i.e. the one of having to cycle past inputs to get to the one you want. Apart from, having like 4 buttons per output selecting what input it receives. (too many buttons, unless they're fucking tiny) Or, having it all midi preset controllable, which would be my preference, but that's be asking way too much - when I'm already asking a lot of him and getting him to step outside his comfort zone a bit. Dude doesn't know a lot about midi, and I'm already trying to get him to chuck a midi patch-bay/merger/splitter in there. He's working on a programmable looper, but midi stuff sounded pretty alien to him unfortunately.
mysteryroach

User avatar
committed
committed
 
Posts: 192
Joined: Wed Jul 13, 2016 11:45 pm

Re: clickable rotary switch

Postby crochambeau » Sat Jul 22, 2017 10:39 am

I don't know of a hardware rotary switch that allows disengaging of the contacts so the user can jump positions. Such an assembly would certainly break the connections before making them, which I feel is contrary to your wishes.

Radio buttons will do it in hardware, otherwise you can throw a little logic at it and turn all your switches into relays.

You say you've got loads empty panel space, why not throw more switches at the puzzle? This might pave the way to "wrong" switch configurations, but sometimes those are the most fun.

Isn't feature creep fun? Be warned though, if you're already pushing the envelope on what the builder is comfortable with, adding more shit is a sure way to delay completion forever. This is why people like modular stuff, you can remove and replace function blocks without tossing out others.

Also, MIDI patch bay is just opto-isolators feeding CMOS switches. It's only the generation or manipulation of the data stream that requires heavy lifting.
crochambeau

User avatar
FAMOUS
FAMOUS
 
Posts: 1097
Joined: Mon Jul 20, 2015 12:49 pm
Location: Cascadia

Re: clickable rotary switch

Postby mysteryroach » Sat Jul 22, 2017 3:49 pm

Yeah I figured it was a long shot.

What about the switches that click back into the central position when you switch them. Is there one like that that does it in 4-directions? Up, down, left right. That way if it's going back to center, I have access to any of the 4.

Otherwise. 2 of the bi-directional ones could work?

The alternative: Go back to 2-inputs. Or just deal with not being able to jump positions and roll with a regular switch or rotary switch. I guess it's really not the end of the world if I have to scroll past a couple inputs lol.

Btw. In your opinion, (if you have a rough idea off-hand), how much does adding inputs to a TRS splitter like that complicate it? (when there's lots of outputs) As you said, I don't want to push the envelope too much. I told the dude bout the midi merging/splitting - that if it's too much hassle, then I can leave it out. Sounds like that isn't particularly complicated though.
mysteryroach

User avatar
committed
committed
 
Posts: 192
Joined: Wed Jul 13, 2016 11:45 pm

Re: clickable rotary switch

Postby crochambeau » Sat Jul 22, 2017 6:28 pm

Like a joystick? Those are generally limited in contacts, so to employ that you'd need to be integrating logic and relays of some sort (be they mechanical or solid state).

A break out patch panel could also resolve input.

mysteryroach wrote:Btw. In your opinion, (if you have a rough idea off-hand), how much does adding inputs to a TRS splitter like that complicate it? (when there's lots of outputs) As you said, I don't want to push the envelope too much. I told the dude bout the midi merging/splitting - that if it's too much hassle, then I can leave it out. Sounds like that isn't particularly complicated though.


I'd have to have first hand knowledge of the build to honestly answer that.

On one hand, a TRS input is a hole, a jack, and three wires. On the other hand, that input needs integrated with the rest, and I do not know how that is happening.

Passive?
Active? (how?)
Switching?
etc.

I offer that as an illustration as to why I cannot answer, not my requesting all the build notes. ;)

You're also mentioning stuff like CV and MIDI, so, depending on the application there might be huge issues that don't start out by staring you in the face. Solving those can increase complexity for sure.
crochambeau

User avatar
FAMOUS
FAMOUS
 
Posts: 1097
Joined: Mon Jul 20, 2015 12:49 pm
Location: Cascadia

Re: clickable rotary switch

Postby digi2t » Sat Jul 22, 2017 10:51 pm

Go MIDI.

https://www.robot-electronics.co.uk/htm/midi-rly8.htm

I used something similar from Highly Liquid years back for an A/B/Y stereo switcher. Worked like a charm. If your builder is having trouble getting his head around MIDI, and MIDI is what will make your life easier, than switch builders.
"I like where this is going!
Not sure exactly where it's going but that's beside the point."

Goroth
digi2t

User avatar
experienced
experienced
 
Posts: 501
Joined: Sat Jun 09, 2012 7:07 am

Re: clickable rotary switch

Postby mysteryroach » Sun Jul 23, 2017 9:18 am

crochambeau wrote:Like a joystick?


Nah. I was referring to, for example, the type of switch that's on the Count To Five. Except with both vertical and horizontal operation (rather than just up and down). So kinda like a joystick but still a switch.


mysteryroach wrote:On the other hand, that input needs integrated with the rest, and I do not know how that is happening.


Yeah unfortunately I don't know how it needs to happen. I just know what I needs to do. For each output, have the ability to select the input.

I'm beginning to think that adding more inputs might be a bit too much of a hassle. I might float the idea, and see how difficult it will be. With just 2-inputs, I still have a custom multi-input copilot broadband that can act as sort of a patch bay, so it's not so bad, I can patch between them at the broadband (i.e. change what the back input receives) and still have a decent amount of freedom to switch between stuff.

If I end up expanding to 4-inputs, which is unlikely at this point - do you know of any regular up/down/left/right/center switches? (i.e. like the joystick style, but doesn't snap back).


digi2t wrote:Go MIDI.

https://www.robot-electronics.co.uk/htm/midi-rly8.htm

I used something similar from Highly Liquid years back for an A/B/Y stereo switcher. Worked like a charm. If your builder is having trouble getting his head around MIDI, and MIDI is what will make your life easier, than switch builders.


I don't really wanna jump ship on the dude. He has helped me out with other stuff before, and I've already discussed it with him enough that I don't wanna waste his time. I imagine it would probably also add a lot of internal components to the device that either can't fit, or would stop me from fitting other stuff.

Anyway, while midi is what will make my life easier, I'm already doing a lot of switching via midi - effects presets + bypass loops + CV generator + looper etc. Most of the switching will be on Chase Bliss pedals, which I can already switch between the internal LFO and whatever the expression jack is receiving. If I have to manually change whatever the expression jack is receiving, it's really not the worst thing in the world. As much as I'd love to have stomp control over EVERYTHING, I should probably not fly too close to the sun and chase the absolute ideal.

Maybe I can make it a future goal if I ever get into DIY.
mysteryroach

User avatar
committed
committed
 
Posts: 192
Joined: Wed Jul 13, 2016 11:45 pm

Re: clickable rotary switch

Postby mysteryroach » Sun Jul 23, 2017 10:58 am

Question: I have 2 Voodoo Lab GCXs, a HEX, and going to get a 2nd HEX when a good enough deal pops up. So a fuck tonne of loops basically. I've heard people getting a stereo mod done to their GCX, where some jacks are stereo jacks. I assume this is for stereo effects and not expression pedals.

That said, I'd be remiss if I didn't ask: If I get that mod done, could I potentially send an expression pedal signal through the GCX? Or is it not quite that simple?

Say I have a basic 8-way TRS splitter box (or however many outputs I roll with). If I wired up a few loops in the GCX as if it's doing instrument switching (i.e. for a 2-way expression-input selector, plug the 1st expression unit into the loop input, don't plug anything into loop send, then plug the 2nd expression unit into the loop return, and have the loop output go to the splitter box) - would that work?

Then I would have ground control over what expression unit is operational at each time. I'd only be able to have one going at once though. I wouldn't be able to have 2 going on different boxes, unless I hand-patch it. Which isn't so bad.

I could also do a combination of the 2 ideas. Have a 2-way switch at the splitter box. And have midi control what in1 and in2 is. I.e. in1 could be midi switchable to be either expression pedal or envelope, in2 could be midi switchable to be either the 2nd expression pedal or cv-generator (converted to expression via copilot broadband). And then have the device on the splitter switch between in1 and in2. Essentially giving me the ability to have midi ground control switching between either expression&envelope, or expression&cv, depending on how I've got each output switch set.

Anyway, I'm guessing it isn't possible to do expression switching through a Voodoo Lab unit meant for audio signals. Maybe that's a question for them? I figured I'd float it here first. Cause if it's an overtly/obviously dumb idea y'all will know lol.
mysteryroach

User avatar
committed
committed
 
Posts: 192
Joined: Wed Jul 13, 2016 11:45 pm

Re: clickable rotary switch

Postby crochambeau » Sun Jul 23, 2017 12:15 pm

Okay, as someone who has taken on a project from a friend that started out as a loose and cool but poorly defined concept piece that proceeded to go full nightmare, let me attempt to help you and your builder out.

The root device here is a patchbay for 1/4" TRS connections, which will carry either resistive expression reference or CV expression reference needs (depending on pedal) from the eight "output" jacks, and allow selection via switch per channel to reference either a CV generator or a resistive treadle expression pedal plugged into the "input".

Am I comprehending the project criteria so far?

mysteryroach wrote:What about the switches that click back into the central position when you switch them. Is there one like that that does it in 4-directions? Up, down, left right. That way if it's going back to center, I have access to any of the 4.


No, the switch you describe does not exist. The functional form factor of something with an X/Y, return to center would have to be a joystick supported by logic circuitry - which can latch so the state would not change until you selected another direction, even if the knob has physically returned to center. With eight, I would be concerned about budget....

Otherwise, in order to increase the amount of "input" channels, you need to add THROWS to the switch. If above mentioned logic circuitry is in place, you could add another channel and retain toggle switching, you'd just need to assign the center state to activate another channel and you'd lose the option of "off" - which may be important depending on what effects you are feeding and how well they tolerate having their inputs connected together. :animal:
crochambeau

User avatar
FAMOUS
FAMOUS
 
Posts: 1097
Joined: Mon Jul 20, 2015 12:49 pm
Location: Cascadia

Re: clickable rotary switch

Postby mysteryroach » Sun Jul 23, 2017 1:14 pm

crochambeau wrote:Okay, as someone who has taken on a project from a friend that started out as a loose and cool but poorly defined concept piece that proceeded to go full nightmare, let me attempt to help you and your builder out.

The root device here is a patchbay for 1/4" TRS connections, which will carry either resistive expression reference or CV expression reference needs (depending on pedal) from the eight "output" jacks, and allow selection via switch per channel to reference either a CV generator or a resistive treadle expression pedal plugged into the "input".

Am I comprehending the project criteria so far?


The very basic requirement is for it to just receive only resistive expression signals on its 2 x TRS inputs, and have 8 basic up/down switches for 8 x TRS outputs selecting which input signal goes to the switches corresponding output. That was the original proposal, which he was comfortable with. (IIRC he initially said it was fine, but still needs to look into it, which in the interests of not wasting his time, I told him to hold off on pending giving him a more detailed proposal of requirements).

Yes, one of the inputs will be a CV generator, but I have a CV->Expression converter that the CV generator signal can go through. I bought it because a significant number of my pedals have CV ranges that are not 0-5V or do not accept CV, but do accept expression input. (the drawback is I'm shit outta luck with devices that have only CV input, and aren't 0-5V). Therefore it's not strictly necessary that the device needs to be able to handle CV.

Having said that, it may be nice if can handle both CV and expression and be able to route either type (CV or EXP) to the devices outputs based on what signal it is being fed. Can you enlighten me if this presents any complications so I don't present this as a potentially unreasonable request to my builder (a request I floated in the original proposal, but told him wasn't required - i.e. it only needed to handle expression if both is too much of an ask).

Would it automatically be able to detect whether the input device is CV or expression? Could do it based on whether a TS cable is plugged in or whether a TRS cable is?

It's fine if the device can't handle both CV/EXP. There's a small drawback to such a feature anyway - if not careful I could accidentally send a 5V CV signal to a 3.3V pedal depending on how the switches are set.

Anyway, having either CV or expression being able to pass through the device depending on what is inputted - that would be cool but not entirely necessary to the project. If it's not a difficult feature to implement, I will request it.

Potentially, if it's necessary, if you feel the 8 x switch version is too complicated a task and already "full nightmare", I can scale back the project for the guy by making it a 8-way splitter with only one switch and all the outputs receive what it is set to (so the input 1 and input 2 go through the switch and THEN get split off into the outputs - much easier I'd imagine). This is less ideal, but also likely A LOT cheaper and less complicated.

Or even, if that stereo-modded GCX stuff above works, it could really just be a simple splitter. I imagine that's a long-shot though?
mysteryroach

User avatar
committed
committed
 
Posts: 192
Joined: Wed Jul 13, 2016 11:45 pm

Re: clickable rotary switch

Postby crochambeau » Sun Jul 23, 2017 2:55 pm

For brevity, I'm just going to start calling the "input" a bus, and the "output" a channel.

mysteryroach wrote:Having said that, it may be nice if can handle both CV and expression and be able to route either type (CV or EXP) to the devices outputs based on what signal it is being fed. Can you enlighten me if this presents any complications so I don't present this as a potentially unreasonable request to my builder. Would it automatically be able to detect whether the input device is CV or expression? Could do it based on whether a TS cable is plugged in or whether a TRS cable is?


Okay, that the CV unit is not *generating* a DC voltage that is placed upon a bus is helpful..

..only that it is not exacerbating a potential issue: in my understanding some expression circuits supply a reference DC voltage (CV) that is divided at the expression potentiometer and returned to the pedal as a CV voltage that the pedal comprehends as useful and does its thing. That set up will require the support of DC through your switching matrix, but (in the simplest incarnation of hard wired ins-switches-outs), if we're to treat each "input" as a bus, in that such a line is available to all channels, it will become possible for the pedal generated CV reference to connect to all other pedals that are set to the same bus.

That may cause problems (failures) with the pedals themselves, especially units built around more modern lower voltage (+3.3v instead of +5) digital circuits. SO, separation of the channels becomes a (potential) concern or you'll at the very least want something to clamp each channel voltage to the scale of its effect for protection.

With the primitive physical switch set up, this becomes a substantial issue, so let's pretend that didn't happen, erase the chalk board and now all inputs of the channels (typically ring) will be buffered so bleed from a 5 volt channel will not waste a 3.3 volt channel, and the return (usually tip) will have a voltage limitation to reflect what is provided from the pedal itself (further necessity of the channel input buffer) OR will get exclusive access to whichever bus you have assigned. Ah, shit, wait, some stuff is so fucking primitive it does not want a voltage, just a resistance that an AC signal is working into that will affect an RC time constant. While feeding DC might not fuck things up... the functionality of such a primitive incarnation of an "expression" jack will probably find issue with an input buffer on the (usually) ring of the jack and an output buffer "clamping voltage" on the return path.

At this point, you can walk two paths:

A) maintain a primitive switching set up that will only allow one channel directly connected to one bus (the bus is switched to channel), this will be a two pole (tip and ring) switch of however many channels = throws (with 8 I believe you're stuck with a rotary) with the *potential failure mode* of requiring one such switch for each bus, meaning it will be possible to physically assign multiple buses to a channel. Upside is, that in this version adding another buss is just another switch, another input method, and channelx2 of wiring connections.

B) go whole hog and allow all channels active at once with "sensing" protections in place. I think this would require a microcontroller with an A/D sense on each channel. You'd only need one very low bit analog to digital converter slowly swept across all the pins to collect states, from there limits would be assigned and a higher bit D/A could provide voltage return from a CV source on your bus, or if all pins dead it could just activate a relay that would hard connect to whatever resistive expression pedal you have plugged in. CV generation is available to whatever degree of code you're throwing at it, and actual channel input selection/merging could be employed in pretty much any manner, even MIDI mapping (which has become fairly primitive in the scenario I just described).

Honestly, I'd probably go with option A unless you like to write code. There's certainly option A.5, as I glossed over a bunch of shit. This has just been an illustration of feature creep, and how what seems simple on the surface can get crazy.

For example, you can manually set/configure channels to reflect a given device, but even a small circuit to scale a 5 volt CV to a 3.3 volt one is an added layer of complexity that may or may not be needed, and if it's not needed - don't waste the time & money on it.

Tons of ways to go about it.

This will boil down to you having a very specific road map of what you want the device to do. Compile a final list a wishes and meet with the builder to discuss if they are feasible and how, once you have that meeting DO NOT CHANGE YOUR MIND unless you're expecting a +$500 change order. This comes down to interface, amount of buttons/switches, what they do, what they don't do, etc, etc, etc.

Remember, sometimes limitations foster creativity.
crochambeau

User avatar
FAMOUS
FAMOUS
 
Posts: 1097
Joined: Mon Jul 20, 2015 12:49 pm
Location: Cascadia

Re: clickable rotary switch

Postby mysteryroach » Mon Jul 24, 2017 5:49 am

Ok hold up, you may have lost me in some of that so let's just back up a second:

Are you describing issues (the ones resulting in potential failures in my pedals) related to expanding the scope of the project to include CV? (because if so - I won't. It's as simple as that. And the below is irrelevant)

Or are you describing issues related to having an EXP-only switcher, but voltage still entering the circuit and damaging the pedals due to the wrong type of expression pedal being plugged in? (if my source audio reflex is a problem, i'll switch to a 100% passive expression pedal. And the below will again be irrelevant. Not much I can do if the copilot broadband is a problem though)

Or are you describing issues that will be inherent in any expression pedals being used? (even completely passive ones)

in my understanding some expression circuits supply a reference DC voltage (CV) that is divided at the expression potentiometer and returned to the pedal as a CV voltage that the pedal comprehends as useful and does its thing


I have a Source Audio Reflex expression pedal which I was going to use as 1 of the inputs into the splitter. Unfortunately, the expression-mode (it has a 0-5 V CV-mode as well as either 10k/50k expression) of this device is "semi-active". That's not a concept I'm all that familiar with. Is this what you're describing in the above quote (are you describing a semi-active pedal, such as a Source Audio Reflex), or are you instead simply describing an expression pedal that outputs CV instead of using resistance? If switching to a completely passive expression pedal will solve the problem, that's what I'll do. However, will I similarly have problems with the expression outputs on my copilot broadband? (a CV->EXP converter) I'm not sure if that is similarly "semi-active". Are these questions for Source Audio and Copilot? Unfortunately the copilot is irreplaceable, so if there's a problem with a CV->EXP converted signal being used, then the problems will just have to be overcome as per your suggestions.

Anyway, can the problem be overcome by eliminating "problem pedals" from the switcher/splitter outputs? Or putting permanent failure protection only on individual outputs that need it (e.g. EXP + 3.3V CV pedals).

Are the problem pedals the ones that either don't accept voltage, or don't accept 0-5V CV? If I have an effects pedal that accepts either 0-10V CV, or expression pedal - if I plugged that into my TRS splitter as an output alongside other pedals that are also expression, but only 0-5V CV - would that put my 0-5V ones in danger of receiving a >5 V DC and potentially frying it?

If I can only use pedals that have both 0-5V CV and EXP - if that's a strict limitation (barring solving the potential failure issues), then perhaps I should make it a CV splitter instead so it only has to be TS jacks. May simplify it a little, if the only pedals I can use the expression-switcher with also have the requirement that it must accept a CV of that range, then what's the point in using expression? Reflex can be set to output CV and the splitter/switcher can have mono jacks. Limits the pedals I can use it with, but it's enough.

If I can still use it with pedals that have 0 to anything above 5V, (i.e. if including those higher range pedals won't damage the lower range ones), then it may be worth it to have it remain an expression splitter, if only for the extra pedals it will be compatible with.

There's a problem I misunderstood you somewhere but hopefully if you've read this far I haven't sounded inanely stupid. Sorry I haven't exactly discussed the solutions you've proposed. I had a more detailed reply written up that addressed some of it, which I unfortunately lost. Either way, it's more important for me to be certain of what problem it is that you're describing. E.g. if it's with the reflex (i.e. semi-active expression units), I don't use the reflex. Or if it's with CV - even better - I just don't use CV. If it's with certain pedals, maybe I limit it to just the ones that will work and won't cause failures. These solutions may be preferable to getting the pedalmaker to solve them in the actual device.
mysteryroach

User avatar
committed
committed
 
Posts: 192
Joined: Wed Jul 13, 2016 11:45 pm

Re: clickable rotary switch

Postby crochambeau » Mon Jul 24, 2017 11:57 am

I apologize if I'm sowing the seeds of paranoia, in most all instances there will probably not be a problem with the situation I described. It's just a potential concern that should probably be considered.

Some designs, say a Moogerfooger for example will have +5 volts on the ring so a passive expression pedal can return a meaningful signal (now CV), so it provides a CV source and will accept CV from the outside world. Some designs won't have the on board CV, others (say using the FV1 Spin chip) will operate on 3.3 volts.

Will bleeding some 5 volt reference onto an FV1 kill it? I do not know. I wouldn't be the person to test it - and as a builder/designer I would consider it important.

Basing your design around the criteria that channels do not overlap would resolve it, but will only allow the two devices to be underway at a time. (this might have been your intent from the get go?)

I'm not 100% clear on your vision, in my mind having eight channels and two input/buses sort of hints at wanting to be able to run your interaction (be it expression pedal or CV generator) into more than one pedal at a time.... and that's where complexity comes in.

Complexity can be reduced if you're like me and buy a piece of equipment and use it forever, or at least a really long time. In that scenario you can simply tailor the channel to the effect: putting some overvolt protection on the 3.3 volters and just plugging the real primitive shit through and hoping for the best (I'm having trouble visualizing a scenario in which 5 volts is destroying something primitive - it might not operate as intended... but you should think of me as a really stupid guy right now).

Conversely, if you're pulling a jwar and flipping through every last pedal you can, you'd probably want a really fool-proof set up that will support everything automatically. But that's an expensive path.

I do apologize if I'm over-complicating things, but you did ask about complexity in particular. :lol:
crochambeau

User avatar
FAMOUS
FAMOUS
 
Posts: 1097
Joined: Mon Jul 20, 2015 12:49 pm
Location: Cascadia

Re: clickable rotary switch

Postby mysteryroach » Mon Jul 24, 2017 1:06 pm

Wait, so are you suggesting that the voltage from one of the effects pedals (i.e. one of the outputs), might bleed back through into the splitter box, and into the other outputs? So was I mistaken when I thought you were saying there might be issue an issue with the expression pedal? (i.e. one of the inputs) Were you instead saying there's an issue with the internal voltage of the effects pedal that the expression pedal is controlling (i.e. making too high for the other pedals that the voltage might bleed into).

One of my pedals is 0-10V CV/EXP. If I have that as an output on the splitter, when I use an expression signal on it, if it's making the voltage on the effects pedal go to 10, will that then go into all my 0-5V CV/EXP gear and fry it. I.e. will I have to put overvolt protection on anything less than 10V if this pedal is one of the outputs (no matter if the expression pedal is completely passive).

Yes, I was hoping to run it into more than 1 pedal at a time and was my intent from the getgo. Either setting it individually for each output (preferred), or a switch before it that chooses which one goes to ALL outputs (less ideal).

The absolute most complicated incarnation of this (also the most ideal) is a combination of the 2:
4 inputs total. in1&in2 immediately go through a switch before it goes anywhere (i.e. before it gets split) - lets call the output "workinginput1". in3&in4 also go through a switch - output = "workinginput2". Then for each output in the splitter, have a switch to switch between workinginput1 (in1 or in2 depending on what was switched earlier) & workinginput2 (in3 or in4).

I have emailed the guy at Voodoo Lab about if there's any issue doing switching (pre-splitter) through the GCX. At the very least, if it's a dumb idea he'll get a laugh out of it.

Re: flipping. Most of what I want to use with it will be there forever. Most of that is already 0-5V. There are a couple others that will also be there forever. One 0-3.3V CV + EXP. One 0-10V CV + EXP. Others I'm less certain about. I have my eye on some other 0-10V CV + EXP ones. Might be a while till I can source them, but they'll likely be kept.

If voltage is gonna get into the circuit anyway even if I use 100% passive expression (if not and I've misunderstood, disregard), and I have to have circuit protection anyway, is it any additional complexity to have the device accept CV as an input?

Are the primitive pedals you're referring to ones that only have an expression input. Or are some of these fine?
mysteryroach

User avatar
committed
committed
 
Posts: 192
Joined: Wed Jul 13, 2016 11:45 pm

Re: clickable rotary switch

Postby crochambeau » Mon Jul 24, 2017 2:15 pm

mysteryroach wrote:Wait, so are you suggesting that the voltage from one of the effects pedals (i.e. one of the outputs), might bleed back through into the splitter box, and into the other outputs?


Yes, if there is a conductive path between effects, any self generated voltage will be shared.

mysteryroach wrote:So was I mistaken when I thought you were saying there might be issue an issue with the expression pedal? (i.e. one of the inputs) Were you instead saying there's an issue with the internal voltage of the effects pedal that the expression pedal is controlling (i.e. making too high for the other pedals that the voltage might bleed into).


Yes, in my understanding (which HAS LIMITS) standard expression pedals are just potentiometers attached to a foot treadle. The expression pedal will not itself cause issue. It might cause poor performance IF you have multiple effects connected to the same expression pedal, because at that point you have all these circuit points between different pedals that are connected together. The connection issue here is SWITCH related, and IF there is destructive force that would come from another pedal.

Inclusion of a CV generator may also provide a voltage source that may also be a destructive force in certain situations unless you are careful and prevent that.

mysteryroach wrote:One of my pedals is 0-10V CV/EXP. If I have that as an output on the splitter, when I use an expression signal on it, if it's making the voltage on the effects pedal go to 10, will that then go into all my 0-5V CV/EXP gear and fry it. I.e. will I have to put overvolt protection on anything less than 10V if this pedal is one of the outputs (no matter if the expression pedal is completely passive).


Correct. This is why I opted to call the input "bus" and the outputs "channels". When you say output the mind thinks in a direction. If we are talking about a bunch of hardware connectors hanging off a switch and connected with wire we are talking about a two way street. Every single connector will serve equally as an input or an output, depending on how they are connected.

This is the answer to your previously mentioned Voodoo Labs (IIRC) thing, whether or not it handles signal is based simply on how the connectors are connected. Generally speaking, if it will pass DC it shoud pass AC, but it is not my intent to write a book here.

mysteryroach wrote:Yes, I was hoping to run it into more than 1 pedal at a time and was my intent from the getgo. Either setting it individually for each output (preferred), or a switch before it that chooses which one goes to ALL outputs (less ideal).


Okay. This is a huge aspect of your intended design criteria and you want to be 100% sure that the designer/builder is aware of this.

mysteryroach wrote:I have emailed the guy at Voodoo Lab about if there's any issue doing switching (pre-splitter) through the GCX. At the very least, if it's a dumb idea he'll get a laugh out of it.


It's not a dumb idea or question, the wording used might raise other questions (see: this thread) :lol:

mysteryroach wrote:Re: flipping. Most of what I want to use with it will be there forever. Most of that is already 0-5V. There are a couple others that will also be there forever. One 0-3.3V CV + EXP. One 0-10V CV + EXP. Others I'm less certain about. I have my eye on some other 0-10V CV + EXP ones. Might be a while till I can source them, but they'll likely be kept.

If voltage is gonna get into the circuit anyway even if I use 100% passive expression (if not and I've misunderstood, disregard), and I have to have circuit protection anyway, is it any additional complexity to have the device accept CV as an input?


Okay, I'm going to break one of my rules and suggest how I would go about this. This is just what *I* would do, and make no mistake, there are probably easier and cheaper ways to go about it.

First off, any primitive pedals that don't do CV are treated like children and stuck at the kids table on their own fucking hardwired switch system that invites its own chaos. Sorry, that's sloppy, but it is what I would do if faced with your criteria. Let them cross breed and party, new species will be formed.

I'm going to simplify my thinking and conclude ANYTHING taking CV/expression will also be providing a CV reference of its own to the expression jack. This means that NO MATTER WHAT you use as a manipulation thing that you as the artist can twaddle with it will be turned into a voltage at the effect.

The bus that connects everyone together would operate at 0-10 volts (or whatever is highest in your rig). The bus, since it is doing this, does NOT NEED a reference voltage from ANYTHING. So all rings (that's an assumption on my part, you'd want to double check with all your gear) would simply terminate to nothing. The RETURN path for your CV will then be faced with the task of scaling a 0-10 volt feed to whatever voltages are acceptable at the pedal. This can be figured out via this technical document:

www.ti.com/lit/an/sloa097/sloa097.pdf

With that, you should be able to switch in a "10 to 3.3" and a "10 to 5" or a "bypass scaling" to reflect whatever effect you have plugged at that "output". For the bus, you simply either recreate a ring provided reference voltage (now 10) on an expression jack or refer to that document to scale UP anything that is providing a voltage below 0-10.

With this, your switching becomes simplified in that it becomes a matter of simply switching ONE conductor per channel (output) because ground is ground and your rings are terminated; yet becomes more complex in that you are looking at one switch per channel instead of one switch per bus (that was my interpretation from earlier, and may not be germane).

Adding more buses is simply adding more throws on the switch, another voltage reference/scaler and another source of fun. You'll still be faced with not being able to skip middle inputs, unless you add an interrupt momentary next to the selector switch, or, again, go with logic switching or expensive radio buttons.

Hope this helps, and I am truly sorry if I've made the project worse by planting my greasy foot into the conversation. Cheers!
crochambeau

User avatar
FAMOUS
FAMOUS
 
Posts: 1097
Joined: Mon Jul 20, 2015 12:49 pm
Location: Cascadia

Re: clickable rotary switch

Postby mysteryroach » Tue Jul 25, 2017 1:48 am

Problem is, I already bought a CV->EXP conversion unit a while ago to deal with the problem of damaging 3.3V or no-voltage units and have full access to the sweep of >5 V ones. My only other solutions were to:

a) use CV instead and have set unchangeable CV scaling at each output, which I specifically didn't want because it goes against, as you put it "buy a piece of equipment and use it forever". This is also one of the solutions you're presenting now which I could have originally gone with.

or b) use CV instead and have switches, which I thought also posed a potential danger of overloading the unit (if switch is set wrong) and was maybe a more complicated and expensive thing to do than a conversion (i.e. as one requires different settings for different voltages. More required avenues to include to reach the same outcome. A converter needs one. A conversion). Again, this is another solution you're presenting me with that I could have originally gone with :/ And kinda wish I had now.

Hindsight is a bitch.

My original reasoning was sound, but I never really anticipated that the internal voltage of a pedal operated by resistive expression unit could bleed back into every other pedal. If I'd known I had to implement voltage scaling at every goddamn output in this second unit I'm having built now, I would have just stuck solely with CV, only had to deal with TS splitting, not worry about expression at all, and done it on one unit. If I had to do the scaling no matter what... Oh well, it's my own dumb fault. It made sense at the time. Don't feel bad for being the one to break this illusion and fill me with regret :P If anything you saved my pedals from an early grave. That would have been a far deeper and more crushing regret.

Ok. So either I can make the voltage scaling device I originally should have done that will support EVERYTHING. Tempting.

OR I can make do and simplify. I understand that the former will be much more of a "forever" piece of equipment. However, inevitably, this pedalboard build has been a long time coming, and this is one of the last peices in the puzzle. So making it simple enough for it to get done in a reasonable timeframe has value to me (even though versitility is also valuable too).

I think what I'll do is only use 5V pedals with it. That's 8 pedals already (incidentally, these are the ones I'm most likely to never flip) - the number of outputs in the original 2-in 8-out spec I'm discussing. So, if I'm going to use the device with ALL the pedals that can take it, adding expanded voltage protection will require a device with more outputs anyway (which I'm not sure you're aware of).

OR I could have it support anything UNDER 5V and I could add specific outputs for 3.3V pedals. This means circuit protection will be required on much much fewer inputs (2 or 3 max) comparison to allowing up to 10V pedals (8+).

10V pedals can be handled seperately by the CV->EXP converter, either on its singular input or its split input (the 3.3V pedals may also be on this split input. But they may also be handled at the switcher via circuit protection). As will any non-CV pedals. It won't allow switching though, but these things can be hand patched. The switcher is more useful for the 5V pedals that would require unplugging + lifting off the velcro every time a change to its input needed to be made. (i.e. can't be as easily hand-patched)

Anyway, here's a question: my CV->EXP converter also has splitter. It's 4-in 6-out, with 1 of the input splitting into 3 of the outputs and the rest of the inputs are non-split (i.e. 1in/1out). I'm guessing these same problems will be present in this device? I.e. The 3 split outputs will have to be the same voltage? (otherwise risk overloading) I imagine the other 3 non-split outputs from the other 3 inputs will not experience this voltage bleed as they will be isolated.

I also imagine if one of the split outputs of the converter is plugged into one of the splitter/switcher inputs, that voltage will bleed from the outputs in the splitter/switcher into the input, back into the converter's switched outputs. Again, I assume the non-split outputs of the converter are isolated from such voltage injection/bleed issues.

Thanks for all your help dude. Sorry to waste your time explaining this all to me. You may very well have saved me from frying many of my pedals, so it's much appreciated.
Last edited by mysteryroach on Tue Jul 25, 2017 10:36 am, edited 14 times in total.
mysteryroach

User avatar
committed
committed
 
Posts: 192
Joined: Wed Jul 13, 2016 11:45 pm

Next

Return to DIY Effects



Who is online

Users browsing this forum: No registered users and 4 guests


Sponsored Ad. (Please no inflated/repetitive clicking. Thanks!)

Advertisements help support ILF


ilovefuzz.com is not responsible for user-submitted content. Users participate at their own discretion and risk.