Momentary / Non-Momentary footswitch button all-in-one?

[boroman]

Okay. I ran out of ideas... I have a Eventide H90 that I need to control via external non-Midi "regular" footswitch.

NOW

Like with all those effects, they need MOMENTARY switches to operate, but in this case I won't be having LED when it's on or off. LED will stay on until i depress the switch which does not make any sense. So, is there any way to have momentary function switch to operate my external effects pedal AND on the same time, regular, non-momentary ON/OFF signal for led?

[TomCarlos]

You mean something like this? Hotone Dual Footswitch

[Mick Bailey]

I'd make a little flipflop circuit on veroboard to toggle the LED. Use a DPDT momentary switch, one side wired to the FX and the other to the flipflop. Take a look at how Boss switches the LED on pretty much all their effects. Any BJTs will work.

Edit: you may need to use a separate battery for switching.

[boroman]

You mean something like this? Hotone Dual Footswitch

It does not have LEDs

[boroman]

I'd make a little flipflop circuit on veroboard to toggle the LED. Use a DPDT momentary switch, one side wired to the FX and the other to the flipflop. Take a look at how Boss switches the LED on pretty much all their effects. Any BJTs will work.

Edit: you may need to use a separate battery for switching.

Thanks. Don't have that type of knowledge... what are BJTs?

[R.G.]

BJT stands for Bipolar Junction Transistor. This is the NPN or PNP type of transistor; Junction Field Effect Transistors (JFET) and MOSFET (Metal Oxide Semiconductor Field Effect Transistor) are yet other different types of transistors as well.

Mick is referring to the circuit used in Boss and Ibanez pedals to run the LED and bypassing; The circuitry that makes the Boss flip flop work is explained starting on page 2 here:
http://geofex.com/Article_Folders/bosstech.pdf

That circuit, and several others, will reliably convert a momentary switch to an on/off for running an LED. There is a problem though, with the idea of using an external LED flipflop and a momentary switch to be an on-off indicator. The effect and the flipflop do not "communicate" on whether the effect is on or off, so it is possible for the effect and LED to get out of sync with each other, so the LED is on when the effect is bypassed. This can be very confusing. Boss and Ibanez style switching can't get out of sync because the effect bypassing and LED are both running from the same flipflop. Out-of-sync issues are often a result of the mechanical switch bouncing, making multiple switch closures on each "press".

This can be a problem. To me, having an LED that is mostly right but sometimes wrong is even worse than no LED. Powering cycling both the effect and the footswitch resets things usually, but that's a PITA too.

[g1]

I have never seen such a thing as a switch that has both momentary and latching sides with a common actuator.

[Chuck H]

I have never seen such a thing as a switch that has both momentary and latching sides with a common actuator.

I suppose that's what the "flip flop" circuit does. I admit I haven't researched it, but I know BOSS did it. It's just a matter of making the transistors behave in a way that latches with a trigger while the actual momentary trigger changes the midi circumstances. I won't say I'm clear on it. The point is that it has been done and there's a circuit to achieve it done by BOSS (and surely others).

[g1]

I suppose that's what the "flip flop" circuit does. I admit I haven't researched it, but I know BOSS did it. It's just a matter of making the transistors behave in a way that latches with a trigger while the actual momentary trigger changes the midi circumstances.

Yes, that is correct. I wasn't disagreeing with any of the other posts, but I got the impression the OP was asking if it was available as a single physical switch.

[Chuck H]

Yes, that is correct. I wasn't disagreeing with any of the other posts, but I got the impression the OP was asking if it was available as a single physical switch.

Ah. Yep. Got it. That and with an LED indicator.

There may be a market.?.

[R.G.]

That is of course what the OP wanted. No, there isn't a single physical... er, wait. Yeah, there is. If you take a common alternate-action DPST switch and use one half as your alternate-action/latching side, then tie both throws of the other half to ground and pull up the common, you get a switch that is open - that is, the common goes high - for the very brief moment that the common is moving between the two throws. This is probably only for a millisecond or so, but break-before-make switches will all do this in some fashion. So - voila! a single switch that does both latching and momentary.
But unfortunately it has much the same problem due to switch bounce. The "momentary" half still has switch bounce, so it's still possible for the thing receiving the momentary signal to get out of sync with the alternate action. It is an article of good faith for all good digital designers to have a really, really good debouncing routine to stop this nonsense when sensing an external switch, but then you as the switch user has to rely on whatever circuits and/or software the other guy put into his possibly unknown box, and it may work differently on each box.
As you might guess, I've been in this position before. My solution was to use a $1 microcontroller and some really good software to reliably turn a single switch action into two different signals - one a momentary of clean, no-bounce action to send out, and the other to make a clean, no-funny stuff alternate action to express the switch state unambiguously. But I was lucky enough to be designing both the switch action and the circuit receiving it.

[Mick Bailey]

I've had problems with switch bounce in my microprocessor guitar synth projects. Firmware debounce didn't work for my needs as it screwed up the timing but a simple R-C debounce works perfectly. The processor I'm using is the RP2040 which is incredibly sensitive to switch bounce. Mainly though the firmware libraries for debounce work fine for regular applications.

[R.G.]

I have found that it depends a lot on how firmware debounce is done. The simpler variants of firmware debounce will not always work, or generate funny delays. I think this is mostly the case for debounce routines in libraries. I found the "vertical adder" debouncer in a post by Scot D'atallo on his web page, unfortunately no longer up. I have yet for find a switch or situation where it does not work well. It was a pain to dig through at first, but once I did, it opened up a whole new vista for me.

The vertical adder uses (at least) two registers/RAM locations, and loads the first (call it "A") from one to a full register's width of inputs every so often. "Every so often" in my implementations is a 10mS interrupt, although the timing can be messed with for different switches. The two registers count vertically, in that register A is the lsb and register B is the msb. Bit A0 and bit B0 form one counter of 00 to 11. So do bits A1-B1, A2-B2, etc. All of the bit positions vertically are independent, so you can debounce as many inputs as you have bits of width in A and B. I wrote the routine to do 8 bits simultaneously in my implementations. Each A-B position counts 00, 01, 10, 11, and when it hits 11, the switch is considered to be debounced. logic on the inputs reset the counter to 00 if the switch bounces along the way. Every switch transition is debounced, and every real transition takes three counts of the interrupt time. I used XOR and masking logic to allow selectably setting each switch input to be either momentary or alternate action for a programmable footswitch loop controller.

The vertical adder technique been absolutely reliable in practice. For a lot more on the shortcomings of many software debounces, read Ganssle's article on debouncing at https://www.ganssle.com/debouncing.htm. The code on the second page is very close to the vertical adder, but I like D'attlo's approach better.