clickable rotary switch

[mysteryroach]

Also, if voltage is passing through this thing anyway (even when using a 10k/50k expression pedal), and thus needs circuit protection (if it is going to support a range of voltages, rather than just 1). Does it matter if I plug a 0-5V CV input into it? Will the device accept the input and split it all the same, without the need for any design changes? Or will it need to be designed differently to cater for both CV + EXP on the input.

[mysteryroach]

ALSO, Regarding: Voodoo Lab GCX. I got a message back stating:

"The (our) stereo mod converts four GCX loops to stereo/TRS.
And, the loop returns are ground (sleeve) lifted.
So, it's unlikely it will work as you desire (if at all)... but, you could try it."

So I'm guessing it won't work?

[mysteryroach]

Even more questions sorry:

I have 2 pedals that operate on EXP and also 0-10 CV (One with separate jacks for CV and expression that you can operate with 2 separate controllers, the effect of which is summed - and one with a single plug that can handle both). I plan to use these both in the 3-way splitter section of my custom Copilot Broadband (the CV->EXP converter). Leaving 1 slot open that can only be filled by a 10V pedal (if there's voltage bleed). Could I fill that with a pedal that doesn't take expression, but does take 0-10V CV? Will the reference voltage from the other pedals bleed into the splitter and then into the non-exp pedal, operating on it as if a control voltage would? That could be a way to turn lemons into lemonade.

BTW: The Broadband also has range knobs effecting the range of each individual output. So I'm not sure if that effects the voltage bleed issue. I should probably ask Copilot about that. I imagine at the very least it could offer circuit protection by way of scaling back the range on the lower voltage pedals.

Also, if I understand this correctly, it isn't the sum of the reference voltages of pedals in the same splitter that will be getting into every other pedal in the splitter. It only the reference voltage of the one with the highest range (e.g. 0-10V)? If I have 8 5V pedals in a splitter, it's not gonna put a 40V load on my pedals is it? Probably an extremely dumb suggestion, but I thought I better ask rather than assume that won't happen, considering assumption is what got me into this mess in the first place. I imagine that's not the case otherwise TRS splitters wouldn't exist at all.

[crochambeau]

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.

Then I would think of the CV bus as 5 volts, scale 5 down to 3.3 with one of those opamp circuits in that document I linked, and scale it up to 10 if you want with another circuit.

The only reason I opted to make the CV line the uppermost voltage is that you could have identical down scale circuits feed ALL the pedals, and a switch to set and forget that stage to meet the needs of whatever pedal you have plugged in.

There's an upscaling circuit as well, so you could hypothetically scale your already owned piece of gear UP to the highest usable voltage only to be scaled back down at each effect. But each of these stages will cost money, so it sounds like your simplification of the application is reasonable as well.

Also, if voltage is passing through this thing anyway (even when using a 10k/50k expression pedal), and thus needs circuit protection (if it is going to support a range of voltages, rather than just 1). Does it matter if I plug a 0-5V CV input into it? Will the device accept the input and split it all the same, without the need for any design changes? Or will it need to be designed differently to cater for both CV + EXP on the input.

I believe, and make no mistake - I am not infallible, that with the set up we're talking about everything can be called CV. EXP is just a localized implementation of CV.

ALSO, Regarding: Voodoo Lab GCX. I got a message back stating:

"The (our) stereo mod converts four GCX loops to stereo/TRS.
And, the loop returns are ground (sleeve) lifted.
So, it's unlikely it will work as you desire (if at all)... but, you could try it."

So I'm guessing it won't work?

Sounds like it's safe to try.

Even more questions sorry:

I have 2 pedals that operate on EXP and also 0-10 CV (One with separate jacks for CV and expression that you can operate with 2 separate controllers, the effect of which is summed - and one with a single plug that can handle both). I plan to use these both in the 3-way splitter section of my custom Copilot Broadband (the CV->EXP converter). Leaving 1 slot open that can only be filled by a 10V pedal (if there's voltage bleed). Could I fill that with a pedal that doesn't take expression, but does take 0-10V CV? Will the reference voltage from the other pedals bleed into the splitter and then into the non-exp pedal, operating on it as if a control voltage would? That could be a way to turn lemons into lemonade.

Yeah, I'm thinking CV in and EXP are the same thing, only with EXP the pedal provides the CV source voltage on the ring of a TRS jack because it's not needing anyone else to provide voltage for it.

That bleed I was talking about goes away if you terminate the rings at the pedals and simply feed them all CV from your device. One voltage source, that you select.

If I have 8 5V pedals in a splitter, it's not gonna put a 40V load on my pedals is it?

No, you would not wind up with 40 volts: unless you went out of your way to intentionally make that happen (it would be difficult to pull off and still have a functional switcher). Again, simply terminating the ring connection coming from the pedal into atmosphere will prevent this from becoming an issue. Sounds like your CV converter will provide all the voltage you need.

[mysteryroach]

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

I kinda glossed over this. Does this mean I need to scale the higher voltages down to a lower voltage (i.e. acceptable by the other pedals) on its return path to the splitter? So instead of, for example, having mainly 5V outputs and a couple of 3.3V that need circuit protectors - I would have mainly 5V and a couple of 10V with circuit protectors? Or am I just getting confused now?

Could I do both? Scale the return path down at 2 of the 10V outputs to 5V. Scale the incoming 5V path down to 3.3 at the 3.3V outputs. Or does it not work that way?

[crochambeau]

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

I kinda glossed over this. Does this mean I need to scale the higher voltages down to a lower voltage (i.e. acceptable by the other pedals) on its return path to the splitter? So instead of, for example, having mainly 5V outputs and a couple of 3.3V that need circuit protectors - I would have mainly 5V and a couple of 10V with circuit protectors. Or am I just getting confused now?

Could I do both? Scale the return path down at 2 of the 10V outputs to 5V. Scale the incoming 5V path down to 3.3 at the 3.3V outputs. Or does it not work that way.

You can do both, we're basically talking about the electronic equivalent of LEGO here. I'm also probably swapping terminology which will increase confusion.

Return path is the tip of your cable (in everything I know, THERE MIGHT BE WEIRDO EXCEPTIONS, you'd be surprised at some of the people wielding a soldering iron (myself included)). So, the voltage on that should be what the pedal needs. The only other connector you're concerned with is ground, but you still want to use a TRS so that little on board supply that was potentially causing all the problems to begin with goes nowhere, or into the air.

If everything on your control side (your CV generator, expression pedal inputs, etc.) is operating on 0-5 volts, all you need is as many 5:3.3 (or 0.666 gain) as you have 3.3 volt pedals. Anything tailored for a 0-10 volt CV range will just work half as deep, unless you scale the CV up (2 gain). No harm will befall any of these things unless they are a 3.3 volt pedal not scaled down.

[mysteryroach]

So, If I did both:


Say I have 8 5V pedals, 2 10V pedals, and 2 3.3V pedals. I would only need circuit protection on the 4 non-5V outputs?

I wouldn't need it on 10 outputs? (e.g. the non-3.3V ones if you're scaling down, the non-10V ones if you're scaling up)

[crochambeau]

So, If I did both:


Say I have 8 5V pedals, 2 10V pedals, and 2 3.3V pedals. I would only need circuit protection on the 4 non-5V outputs?

I wouldn't need it on 10 outputs? (e.g. the non-3.3V ones if you're scaling down, the non-10V ones if you're scaling up)

Correct.

Though to be accurate, the 10 volt stuff does not need protection, it just needs a little gain. If there's a sensitivity control on the pedal, it may operate to your liking on 0-5 volts and need nothing at all.

Total case by case basis.

Protection from over-voltage is the only potentially destructive concern, everything else is just a feature.

[mysteryroach]

Righto. I'm pretty sure I get the jist of everything now.

I'll post back if I have any more questions. Thanks for dealing with my confusion.

[autopilot]

the broadband outs are independent so there wont be any bleed from one output to another (you can ran one out going into a 9v cv device, another out to a 5v cv device and the other out to a 3.3cv device), even when they all share the same source (to sync).

there are various options for your situation, since you want to have multiple inputs that you want to route an specific out and use a footswitch, get some multiplexers ic and have a predetermined idea if you have 8 inputs and have 8 outs you can route it, and it will solve your problem, for more efficient (timely and cost), all the inputs should be converted to cv internally and it will be easier (less parts and design) and feed them to the broadband (so you will have 6 outputs of your 8 devices, and can solve the ones that needs exp, 9v, 3.3v) and the others (ideally the ones that receive 0-5v) will be controlled for from the 8in/8outs device.

the only drawback is that you will have to scroll from each pseudo preset (if input a is a sequencer, and you want to go to envelope, but envelope is input f, the. you will have to go from a to f clicking the footzwitch). is an example because you could configure how they will be routed but once is routed, it will stay that way. but i dont think in mid song you will use (while playing) different inputs with the same out (like 4 bars of the solo has a sequencer of the cutoff in a filter then 2 bars into a lfo next 2 bars envelope last 8 bars noise source)

[autopilot]

another option 2 rotary switches and have a master footswitch so you can select via the footswitch the 2 inputs, but this is per channel so if there are 8 then there will be 16 rotary switches and 8 footswitches

[mysteryroach]

the broadband outs are independent so there wont be any bleed from one output to another (you can ran one out going into a 9v cv device, another out to a 5v cv device and the other out to a 3.3cv device), even when they all share the same source (to sync).

there are various options for your situation, since you want to have multiple inputs that you want to route an specific out and use a footswitch, get some multiplexers ic and have a predetermined idea if you have 8 inputs and have 8 outs you can route it, and it will solve your problem, for more efficient (timely and cost), all the inputs should be converted to cv internally and it will be easier (less parts and design) and feed them to the broadband (so you will have 6 outputs of your 8 devices, and can solve the ones that needs exp, 9v, 3.3v) and the others (ideally the ones that receive 0-5v) will be controlled for from the 8in/8outs device.

the only drawback is that you will have to scroll from each pseudo preset (if input a is a sequencer, and you want to go to envelope, but envelope is input f, the. you will have to go from a to f clicking the footzwitch). is an example because you could configure how they will be routed but once is routed, it will stay that way. but i dont think in mid song you will use (while playing) different inputs with the same out (like 4 bars of the solo has a sequencer of the cutoff in a filter then 2 bars into a lfo next 2 bars envelope last 8 bars noise source)

So, convert all TRS inputs to 0-5V CV (if not already cv) and then make it just be a TS splitter? May as well, as it'll be cheaper. I could make it even just have TS inputs and only accept CV (my reflex does CV). But I may as well allow a EXP-in - for versitility.

What I'll do - is have the switching unit run only my 5vs. And run everything else off the Broadband as a universal conversion unit or sorts. Like you're also suggesting I think.

If I really want to any of my 3.3's off the switcher (in case the Broadband gets full) I can maybe get one of those cables Disaster Area are making (converts 5V->3.3V). If I need to run shit off it in the future, just put the conversion in the cable. Although I'm not sure if you can make a similar cable that will amplify the voltage up to 10V. I assume something like that would need to be powered? (whereas a reduction in voltage is just a resister) It would be nice to know what I would need to do in the future to support >5V CV-only devices - with full voltage range. It's not something I'll solve today (i.e. in the switcher) - but it'll be nice to know what would be involved.

Anyway, when you say "input f" - having that many inputs (6) was an extreme luxury even in the first post, that I had since rolled back. However, your idea of a footswitch never even crossed by mind. I had wondered if it was possible to input the expression units through the GCX to do the switching. But I can input into the patch-bay instead (being that it may not work through the GCX), and use the GCX as a footswitch, then I can have midi control.

I could either just cycle between 2 inputs. Or have 4 inputs, and 2 4-way rotary switches (to determine what inputs get cycled between using the footswitch). Or on/off switches for each input, selecting which inputs are cycled between - i.e. pressing the footswitch goes to the next input that's on (I assume that may be a little more complicated though?) - maybe have a reverse switch too.

Could I have multiple midi operated footswitches to expand the number of inputs I can cycle between? E.g. if I had 2 footswitches, could it be configured to choose input 1 when footswitch 1 is pressed, input 2 when footswitch 2 is pressed, and input 3 when both are pressed at the same time? Tbh, this sounds like a significant step up in complication though.

In this case I wouldn't have a (regular) switch for each output (i.e. the inputs at the splitter will all receive the same CV). While it would be nice to include a per-outlet switch, I'm unlikely to bother including it i.e. the footswitch decides whether [in1-or-in2] OR [in3-or-in4] is selected. However, it's not exactly necessary. It's a pretty staggering amount of control as is, so I think i'll keep it simple:

2x EXP/CV-in
8xCV-out
5V in/out. (so only use 5V input/output devices unless additional conversion steps are added - e.g. cable or broadband)
EXP->CV conversion at the input
Inputs switched BEFORE the splitter
Footswitch control over switch
GCX/GCP interfaces with footswitch for midi control

[mysteryroach]

So this is probably what I'm going to pitch to the guy, if y'all are interested. If it's too big I'll remove a row of CV outs from the splitter. But I'm pretty sure there's enough room in the box for all of this:





EXPRESSION SPLITTER/CV CONVERTER MODULE:

3 TRS EXP/CV inputs. 1 front, 2 back. Accepts either 0-5V CV or EXP.
Thru out before conversion occurs (for front input).
All Converts 5V CV or EXP into 10V CV. (maybe after switch if that's even possible. i.e. so you only do it once. May not be possible though)
12 outputs with a attenuation switch to switch it down to 5V
2 footswitch inputs on the back. One selects between in 1 and [in2 or in3]. The other selects between in2 or in3.
Resulting input that makes it through the signals get split into all the 12 outputs.

2 manual switches. Same function as footswitches. Not sure how this is going to work so will talk it out with him. Will flicking the switch simply switch from one to the next? Could get confusing if it's done that way. E.g. if I flick the manual switch (the in2/in3 one) to up, switching to in2, then press the footswitch, switching to in3, then I will be flicking the manual switch DOWN to get to in2 again. (i.e. the reverse of what it was). Maybe simple buttons are an option. If you guys have a solution that's not too complex, lemme know.



PATCH MODULE:

2 TRS jacks either side. Both isolated from each other. Each individual front jack connects to the corresponding back jack.
For basic patching but also for guest loops. Am I right in am I right assuming that I can patch a guest loop from a single stereo jack? (rather than use 2 TS ones on a single guest loop). Or will it cause problems?

The way I was planning to do it is: MONO LOOP SEND (GCX) ->(merging into Y cable) STEREO TRS in/out (back of unit) -> STEREO TRS in/out (front of unit) -> Y-cable (TRS into 2x TS cable) into a pedal -> out of pedal back into Y-cable -> back into STEREO TRS in/out (back of unit) -> STEREO TRS in/out (front of unit) -> back into original Y-cable the original loop send merged into -> MONO LOOP RETURN (GCX).

Not sure if that made sense. Anyway, can Y-cables not be used this way? (provided that it's wired up correctly).



MIDI MERGER/SPLITTER MODULE:

2 MIDI Inputs (front)
2 MIDI IN/OUTS (front/back. Switchable in/out)
2 MIDI Outputs (back)
MIDI OUT 1 does not go through a splitter. Is patched in such a way that it will carry phantom power from the GCX through the output into IN1 to power the GCP.
Each MIDI IN/OUT has a switch that tells it whether it's functioning as an input (i.e. merged + split into every other output), or an output.
All MIDI INs and all MIDI IN/OUTS selected to input mode go through a splitter and gets split off to MIDI OUT 2 and all MIDI IN/OUTS that are selected to output mode.
3 switches (front) to filter out unwanted midi clock from midi in2 and midi in/out 1&2 (midi in1 doesn't have clock. Also not sure if it can fit another switch)
4 analog beat triggers (back). (or maybe 2 and 2 switches to change the trigger type of each - you think that's necessary? cause 4 is kinda excess anyway)

[crochambeau]

Wow, what a build. I'd love to see pics once complete.

I hate to rattle the cage any more on this project due to the inherent depth/complexity, but how are you going about switch snubbing the MIDI clock on those ports?

Anyway, can Y-cables not be used this way? (provided that it's wired up correctly).

Yeah, a TRS to two TS cable can facilitate a send/return for a loop. Read up on "insert cable" with respect to mixing boards.

[mysteryroach]

Switch snubbing?

Are you referring to cutting the midi clock from certain inputs (ones switched off) out of the output signal? I'm not even sure if it's possible. I'm by no means a midi expert. If it's not possible, those switches can be something else maybe.