Building a 3-button TRS Footswitch? UA OX Box

[Bunkey]

Back once again with an enquiry I really should know more about...


I've been tasked with putting together a 3 button footswitch to use with the UA OX Box - to switch 3 software effects on and off during live performances.


It's owner is currently using a simple Marshall 2-button footswitch for the purpose...



...but he wants to make full use of the 3-button functionality along with having LED indication so he doesn't have to play a little between songs to keep track of what's active.

Sure! I said..

Well as it turns out, switching 3 channels independently through a single TRS jack is not as simple as switching 2 

I was hoping to use latching DPST switches so I could keep things simple and separate the LED indication to it's own onboard power supply but I'm at a loss as to how the switching circuit itself would be configured to use latching switches?

The only 3-button footswitch I can find using a single TRS connection is this thing for DigiTech pedals:



I'm guessing this relys on momentary switches? No.1 pulls from both tip & ring, No.2 pulls from tip only & No.3 pulls from ring only; so the pedal processor knows which button has been pressed and activates or deactivates the correct circuit accordingly.

From reading the manual, The OX Box software supports both momentary & latching switches and I think it has some sort of learn function - This makes me think I can use the layout above to control the 3 channels in the same way.

That doesn't solve my issue of LED indication though.

I don't really want to have to build some sort of on-board processing circuit to control the LEDs as this approach just seems way out of my depth right now. I appreciate though that might be the only solution.


Does anyone have anything they could weigh in with? Anything I'm overlooking here?

Of course, if anyone is familiar with the OX Box and knows a 3-way footswitch schematic which works with the unit that would also be very helpful.


I'm not trying to palm you off with my work load 
It's just a learning process and I need some tutoring.

Thank you

[Bunkey]

Doing a bit more head-scratching - it would be possible to use relays to control a simple passive LED circuit for each button.

LED comes on when the button is pressed, turns off when the button is pressed again.

The only problem with this is that it poses the possibility of the LED status indication becoming reversed if say there was a false positive or for some reason the OX registered the button press but the relay circuit did not.

I imagine this situation would be a complete headache if it occurred in the middle of a set.


It looks like latching switches are the only viable candidate for the LED problem but is it even possible to control 3 circuits through a single TRS using latching switches?

[Phoenix]

Simple.
Tip, ring, sleeve.
+V, control, gnd.
Footswitches short voltage divider ladder to control, which is an A/D input to uC. Each switch has a different voltage output so the A/D can identify them within a window.
Have a second uC in the footswitch, which controls the onboard LEDs.

[Bunkey]

So to put that into laymans terms - are you saying that each channel has it's own unique value and the sum of each combination is not equal to the sum of any other combination?

I had this idea last night based on a TS jack:



But you're saying I'd send these from tip to ring on the TRS instead of from tip to ground on the TS?

Then the use of DPST switches would mean I could set up a completely seperate LED circuit that is mechanically connected to the switching circuit.

You'll have to excuse me, I'm electronically illiterate 

[Phoenix]

R-2R ladder
What I'm suggesting is use momentary switches, and micro in the amp and a micro in the footswitch. The micros talk with each other, the one in the amp controls the channel switching, the one in the footswitch controls the indicator LEDs.
You have three terminals on a TRS jack, that allows two for power (you need two for power regardless if you want LEDs), and one for signal. If you want three functions/buttons, but you only have one signal line, you'll need to use multiple voltage levels to indicate different things, hence the R2R ladder.
If you're not familiar with microcontrollers, this can be easily done with an Arduino, which is a nice simple way to start out.
There are discrete digital circuits that can achieve this, but they end up being more complicated than a simple micro, this is the way most amp footswitches are going these days because of the much greater flexibility available and how inexpensive it is to implement this way.

[Bunkey]

Ah I see, thanks for clearing that up.
The only caveat is that I don't have access to the OX myself and neither I nor the owner would be keen on modifying it.

This has opened up a line of enquiry about ready-compatible digital footswitches though so I'll look into that.

The idea of using latching DPST switches meant I could keep the switching passive and run the LEDs seperately with their own power supply instead of relying on drawing power from the OX Box.

It's all just a bit more complicated than I was hoping this was going to be.


Cheers.

[Phoenix]

Ah, sorry, somehow missed that it was for an Ox, thought it was a more general question.
I looked up the manual, but they are incredibly non-specific, and don't appear to offer an "official" footswitch. But the way it's implemented in the top box will be some fancy variation of what I've described, just offering more features.
They imply that it can work with just about anything, but make no guarantees. If I were you I wouldn't bother making my own, I'd just buy whatever generic TRS three button footswitch I could find, with the correct settings on the Ox it should hopefully work, and if not then at least you'e got a box and hardware to modify (and this would probably be a cheaper way to get the parts than buying them individually anyway).

Honestly that's pretty crappy documentation from UA, no pinout or voltage information for the TRS, no suggested schematic(s), no official footswitch and no list of vetted third-party products. All it would've taken is an extra paragraph and diagram in the 73 page manual. That sort of thing really puts me off wanting to buy things.

[Bunkey]

My thoughts exactly.

Unfortunately there don't seem to be many 3-way footswitches with LED indication available.

UA are unresponsive to enquiries and most of the UK dealers I've asked didn't know you could use a footswitch with them, let alone suggest one that works 

It's meant to be great at what it does too but never mind.

Thanks for your help man.

[Bunkey]

As if by magic I just got an email from Julius @ Universal Audio.

He suggested that DigiTech setup (2nd schematic) is the go-to and that's the one they use.

Maybe it is just a passive setup after all and that's why it's limited to 3 channel switching.

Looks like I'll be rigging it up with the relay-controlled LED status indicators then.


It's gonna be a little while but I'll try and post back here with the results if I get it working well.

[Phoenix]

Ok, keep in mind that for proper functionality with that scheme you MUST use momentary switches, which complicates the indicator LEDs. While it's quite easy to toggle LED indicators with momentary switches and a flip flop circuit, you may have issues synchronizing the footswitch and Ox. And even if you get them synced, if there are any switch bounce issues they can lose sync again.

You could use latching footswitches, but then when number 1 is engaged, 2 and 3 will not function, because the way it works is tip and ring have pull-up resistors, 2 pulls down tip, 3 pulls down ring, and 1 pulls down both.

[PRR]

> you MUST use momentary switches

Not clear to me. A deep-hidden page of the manual shows a "latch" option. I suspect if set not-latch, it follows the switch. If the switch is latching (dumb stomp-switch or something clevererer) that should work without sync trouble.

[Phoenix]

> you MUST use momentary switches

Not clear to me. A deep-hidden page of the manual shows a "latch" option. I suspect if set not-latch, it follows the switch. If the switch is latching (dumb stomp-switch or something clevererer) that should work without sync trouble.

I guess if it can differentiate between pulled-down to 0.7V and 0V, then it should work with latching switches, but that's not at all clear to me.

[Bunkey]

I imagine it just reads where the current is being pulled from, tip or ring or both, as opposed to any specific value.

I think the latching options are just for 1 or 2 button setups - It's a single TRS jack after all and there are no microcontrollers in the DigiTech pedal.

Maybe I'm just over simplifying my understanding of how the OX works but given it only supports up to 3 switches despite offering more effects channels in-software I'm guessing it's not configured to accept digital signals, else it'd support any number of switches for any number of effects.

Either way that 2nd schematic is the recommended method so that's what I'll be using and yeah it relys on momentary switches - I'd rather the option of latching so that the LED indication is mechanically syncronised as mentioned but it looks like that's just not an option here.

Instead I'll be using momentary DPST switches, one pole to do the switching as per the second schematic, the second pole to control a separate LED circuit using something like this:

https://www.brimal.co.uk/latching-indicator-circuit-for-momentary-switch.html


As long as the LEDs in the footswitch aren't lit when the OX is turned on it'll be syncronised fine - There's a reverse-sync option in the software if not.

The only problem is if it reverses sync in the middle of a set for some reason and Keith get's confused about what's active but its a possibility we're both aware of so we'll just have to keep it in mind and see how reliable the system is when it's put together.


Should be a fun project and I've learned a bit about switch circuits from this if nothing else.

[Phoenix]

The Digitech uses momentary switches.
The Ox would monitor the voltage (probably 5V or 3.3V) at the tip and ring, when tip is pulled down, it's button 2, ring it's button 3, both at the same time it's button 1. That's why latching would be a problem - if button 1 is latched, there's nothing for buttons 2 and 3 to pull down, unless it can differentiate between being pulled to ~0.7V (the diode drop on button 1) vs all the way to 0V. It's not impossible, but I would think quite unlikely.
Sync could definitely become a problem mid performance due to switch bounce, so make sure to properly debounce the LED circuit and test thoroughly.

[Bunkey]

Ah I see what you were driving at with the 0.7v now, sorry.

I'll pop them an email just now - might as well ask!

[Phoenix]

Yeah I'd just be surprised if they used two analog inputs or four digital inputs plus window comparators for this task rather than two plain digital inputs, but I've been wrong before.

[Ripthorn]

You could do the LED indicating with a micro controller and let the voltage signaling be according to the schematic up above.