Externalizing switches and much more

[manson]

Hi,

I was wondering the following: I'd like to put the guts of five existing stompboxes into one 19" rack. Next, I'd want to externalize the 5 switches by connecting the wires to a midi connector plug (which has 5 pens) in some way. Next, make a little stompbox with 5 DPDT or SPDT switches or whatever needed, and also a midi plug. Then connect the 5 wires in the "floorboard" to those switches. Effectively creating an analog all-in-one rackmount thingie with midi cable and floorboard. Sort of.. You know that an amp's channel switching box uses a regular guitar cable, so I thought that could be possible in this way.

Is this a completely silly idea or could it be done? Comments are more than welcome!

Thanks, Peter.

[axeman010]

Hi

You may experience some noise pick up if you just extend the leads of the switches to a floor switch.
It might be a better idea to use relay or logic chip to do the switching so that the cable to floor unit is only
carrying a control voltage rather than potentially an audio signal.

Axeman.

[GibsonGM]

Sounds totally possible, but like Axeman said there's a potential for things to get messy if you just ran the switches out to a floor box.  You'd have to figure out a pattern of relay switching (a job in itself) to send your inputs to the right circuit...simple SPDT or DPDT switches in the floor box would work great, and let you have LED indicators (Mill bypass or DPDT).  Or you can go digital and use "soft switching" - I prefer solid stomp switches myself, and simpler...read up on it, it's been done before!   

[R.G.]

Oddly enough, there is a build-it article on GEO about this very topic.

See http://geofex.com/Article_Folders/pffootsw.pdf dated 2/27/03.

[Mark Hammer]

There are a great many ways to do remote switching.  What they share in common is that the remote switch floor unit essentially provides "control signals".  That is, voltages and switch-closures that control something else far away.  As noted, because the audio path remains entirely in "the box" where the effects reside, there is little risk of contaminating the audio path with noise or hum by running long leads.

A couple of things to note:

• A great many guitar amps actually carry what you might call "audio signal" to remote footswitches.  While things like channel-switching are generally done via FETs or relays, you'd be shocked to see how many amps turn reverb on and off by having a wire directly tied to the input of the reverb recovery stage running out to the footswitch.  The footswitch then grounds that wire to cancel reverb.  Ungrounded, however, it is essentially an "antenna" for noise coming in through the reverb channel.


• Remote switching is often pretty easy to accomplish andneed not involve terribly sophisticated technology.  Remote switching with local status LED indication, however, can be another matter entirely.  That is, switching something at the front of the stage, with an LED going on or off to show you what just took place at the side or rear of the stage can seriously up the complexity of design or number of conductors needed for the connecting cable.  The CMOS switch detailed in the first Craig Anderton EPFM book is a personal favourite because it places an LED in series with the footswitch, which means you don't have to add any extra conductors to have status LED indication.  It tends to be alone in that regard, though, and has some limitations on what you get to use it for.


• ANY Boss, DOD, or similar pedal that uses FET-switching can often be easily retrofitted for remote switching.  The internal switch is a simple momentary SPST unit that temporarily grounds a connection.  There is little reason why a person couldn't install a mini-phone jack, and plug in a cable to a parallel momentary switch.  Certainly all the individual grounds can feed a single wire, and the remaining pedals each contribute only a single additional wire each.  So, imagine the arrangement whereby a person has a half-dozen FET-switched pedals seated atop some sort of pedestal/pedalboard at waist or chest height.  A "snake" takes those retrofitted control lines down to a switchboard on the ground, where the user has strategically situated momentary switches (big push-buttons) that they can step on, say, two or three at a time or individually if they feel like it.  Many commercial pedals make themselves more amenable to remote switching than you'd think.


• A remote box containing the effects themselves, regardless of how switching is done, can often lend itself to easier "reassignment" of effect order without the burden of patch cords.  The user can create multiple "stations", using rotary switches, and (for example) assign effect A to station 2, effect B to station 1, and effect C to station 4.  The remote footswitches, meanwhile, determine if stations 1, 2 3, or 4 are "active" or bypassed.  Alternatively, the user can simply install a little patch-panel on the front or back of the master rack unit, and assign effects to a given "station" in that way.  Though not as efficient, one of the advantages of assigning effects to stations via patch cords is that one can lump several effects together in series within one station, or assign a given station to something off-board (e.g., another rackmount unit).  Which brings us to...


• Nesting of effects.  A good remote switching scheme will allow one to "nest" combinations, such that clusters can be switched en masse.  For instance, effects B+C, and D+E, may be treated as clusters, such that one can engage cluster B+C with a single button press, or even bypass B-thru-E with a sinle button press.


• Not only is remote switching feasible, but so is remote panning.  The Anderton "Volume Pedal Retrofit" is one example of a VCA-based gain-control module where the foot-pedal doing the controlling can be situated at a distance from the thing whose volume it is controlling.  So, a person could remotely fade delay in and out as they play, again without incurring noise.  Wire up two such modules with a dual-ganged 250k or 100k pot and you can choose to either raise the volume of two things simultaneously, or flip a switch and fade one in as another is faded down - "morphing" if you will.


• Finally, mounting effects circuits in some sort of larger rackmount chassis (or simply a large chassis of some sort) can often bring with it improved opportunities for power regulation/distribution, and more options for situating controls and switches once liberated from the oppressiveness of small individual chassis and finding a spot for a knob amidst the jacks, stompswitches, and batteries.

[manson]

Thanks for all your answers. I did not realy think about the fact that your signal will have to travel all the way from rack to stombox and back, possibly picking up noise etc.
But I would very much prefer to keep things as simple as possible. I'm quite the electronics noob. I can manage building a fuzzface, but remote switching as described in R.G.'s document is way over my head. I could never make that or even interpret the schematics correctly.

What is the most basic, but working solution for this?

[Mark Hammer]

I'll see if I can find an example of a 4016 or 4066-based CMOS switch on-line somewhere.  Each of these chips has 4 SPST switches on board.  Give 'em +5v, they turn on and let stuff pass.  Take the +5v away and they close, blocking stuff.  You can set them up to be just like a DPDT bypass switch, the important difference being that the applying/removing +5v can be done at a distance.

You can easily build up a couple of those switch modules, and stick 'em in a box.  That's exactly what I did almost 20 years ago.  I got a rackmount chassis, and made a bunch of Anderton modules, plus two of the CMOS switch modules.  The switch modules were "nondedicated", meaning that the in/out and effect in/out connections could be used by anything I wanted to plug into those 4 jacks on the back panel.  If I wanted one of those switches to have 4 effects in series patched into the send/return loop, then that's what I'd do.  VERY very flexible system, and not a smidgen of MIDI or microprocessor control anywhere.   One of the things I like about those switches as well is that they require a latching type switch (i.e., step on it, and it stays on until you step on it again).  What I would do is wire up a latching and nonlatching (momentary) switch in parallel such that I could either leave an effect on, or else switch it on for as long as I stepped on the switch, and revert back to normal the moment I lifted my foot off.  Tremendous power in that arrangement.

The switch circuits themselves are no more complicated than a Fuzz Face, so if you can successfully make one of those, you can certainly whip up a trio or quartet of the these switches.   I'm headed out of town on the weekend and so trying to tidy up a lot of local loose ends before then, but I'll see if I can find something to direct you to that is painless and cheap. 

If memory serves, you would need one wire per switch, plus a ground wire, in order to run a cable from the central switchbox to the thing on the floor at your feet.  Most electronics distributors will likely be able to sell you, say, 25 feet of 6-conductor cable for a modest price, and that would likely be sufficient for at control of least 4 switches. 

[manson]

Yes, please enlighten me on this. I've read some articles about these chips, but I don't see how to wire/make this.

|------------------------|
|Switch 1 -\             |
|Switch 2 -\             |
|Switch 3  ---- Chip |-------------- Rack
|Switch 4 -/             |
|Switch 5 -/             |
|------------------------|

Is this one chip all I need? I definately need someone to draw this out for me.. 

[Mark Hammer]

No, you need one chip per effect.  The chip itself simply does what the stompswitch would normally do mechanically.  Its advantage is that a much simpler mechanical switch (SPST, just like a lightswitch) can make the chip accomplish multi-pole switching from a distance.  Think of it like having (as per your drawing) 5 "experts" to accomplish some unique task in another location.  In between you and those 5 experts are 5 morons who do your bidding.  When it's time, you tell the relevant moron "Go tell the expert to do it now", and the moron rushes off and tells the expert "Manson says to do it now!".

[manson]

Yes I see Mark, thanks for clearing that up This looks nice too:



But the effect's input and outputs are wired to this board.. I think. So this could perhaps be usefull for making a multi effect stompbox..

[R.G.]

See also: bypass switching with the CD4053 at GEO. That's going to work much more quietly if you bias the inputs and outputs to 1/2 the power supply to the CD4053.

You can also use DPDT relays with only the coil switching remote on alternate action footswitches, but you'll need to put a big capacitor across the coils to keep the coil voltage changes from causing clicks in the audio path.

[Mark Hammer]

No reason on earth why that couldn't be a standalone circuit in a rackmount box with re-assignable ins and outs.  Pretty much exactly what I had in mine.

[manson]

Hmm re-assinable ins and outs, won't that result in a lot of cables running from the footswitch to the rack?

I have a pretty consistent setup. The only thing I'd like is:
a) to put 5 existing (boss, mxr, etc.) effects into a neet 19" rack with the control in the faceplate
b) make a footswitch with 5 switches, to switch each of the rack efx on and off (basically extend the stombox switches)
c) 1 (or as few as possible) cable running from the switch to the rack
d) keep things as simple and low-level as humanly possible, without too much noise and things

I have no experience at all with controller chips and such. I really appreciate Mark's advice and R.G's articles, but I don't think I'm capable to translate those into a piece of hardware I could build. I realy need someone to tell me exactly from A to Z what to do, what to get and how to make. And I know that's asking a lot from your time...

[Mark Hammer]

You're making it more difficult than it is.

Imagine you have two modules in your rack box, each on their own PCB.  One houses, oh I dunno, let's say a Tube Screamer (hmm, fancy that?  ), and the other houses Laurier gendron's 4053-based sswitch.  The TS is wired to two jacks: in and out.  The switch, however, has 4 jacks associated with it: in, out, send (to effect), and receive (from effect).  One patch cord goes from the send jack to the TS input, and another goes from the receive jack to the TS output.  The guitar goes to the "in" jack on the switch, and the "out" jack goes to the amp.

So far, we have one effect with a bypass switch.  Where normally an Ibanez TS would have electronic switching on the PCB and the effect itself would be directly connected (via copper traces) to the electronic switch, what we've done is build the effect and switch on two separate boards, and are connecting them with wires.

I said YOU were making it too complicated.  So, how come what I just said is NOT making it more complicated?   Here's why.  Let's build another effect, say a Dynacomp, stick it on a board, and stick it in the same rack chassis, again with an in and out jack.  We can now run a patch cord from the send jack of the switch, to the Dynacomp, from the Dynacomp to the TS, and from the TS back to the receive jack.  Voila, we now have the capacity to turn two effects on simultaneously with one single switch.  What lets us do this is the fact that we have separated/isolated the switch circuit from the effect circuit.  THAT'S where the magic is.  All of those cables are actually on the rackmount unit, or even inside it.  All we have going out to the front of the stage are two wires to the remote switch.  One wire will come from the 200 ohm resistor and the other will go back to pin 11.

In fact, when I had my own rackmount multi-FX, I only had two such switch modules on mine, even though I had 7 or so effects.  Because I had the flexibility to patch them in different orders/sequences, and could switch them in clusters, I didn't really need a lot of switches.  I think if you made yourself 3 or 4 switches using Laurier's layout (consider making two side-by-side modules on a single board and stacking them; I'll explain why at a later date), that could handle an enormous number of remote switching possibilities; the sort of thing that would make your friends envious.  If you had 4 such switches, it would involve only 5 wires (within a single cable) going out to a remote footswitch floor unit: one from +9v (distributed to 4 separate 200 ohm resistors and then to one side of the four switches) and individual wires running back to pin 11 on each of the switch modules.

A tip:  one of the things that allowed me to get so much done with just 2 switch modules was the use of toggle switches for bypass on the rackmount chassis.  So, I could use one switch module to control a compressor, EQ, and distortion module all at once, but I could toggle the EQ and distortion off leaving the compressor on, or vice versa, without having to unplug anything.  The one remote switch engaged that loop, but within the loop I could still pick what I wanted with toggles.

[manson]

Okay, you've cleared up a lot here. I kept thinking that the cmos switch was to be placed inside the footswitch. With 2 cables per effect running back and forth between the footswitch and the rack. Making five of these switch cicuits and placing them in the rack together with the five effects makes a lot more sense indeed. Did I have my coffee today..? I did make things much more complicated then they are.

What you're saying about little toggles on the rack.. Hmm it sounds like a good idea, but I also do like the idea of flexibility and not having to walk to the rack unit all the time flicking switches when I'm experimenting with different sounds.

Thanks a lot Mark, I'm starting to get the whole idea. By the way, I'd love to see a pic of that old rack of yours Now all I need is to find the best suitable switching circuit for this application. Recommendations?

[Mark Hammer]

I wish I had a picture...sigh.  I sold the thing to a guy for $200 about 15 years ago.

The toggles are actually helpful, even if you have complete control from the remote unit, and the toggles are redundant.  For instance, I could stick a parametric EQ section in front of an overdrive, tune it up, listen with the resonant boost, toggle, without the boost, toggle, with, toggle, without, toggle, readjust the boost frequency, toggle, test, etc.  Kinda like the way animators could do cell-by-cell flip comparisons.  Going back and forth between the floor unit and control panel is annoying, so the few extra dollars for the toggles is worth it.

As I noted before, and many times previously, I personally love the Anderton thing because of how it let me use momentary switches.  In gigging situations, having to step once to engage, and again to disengage can be disruptive to one's playing.  The Anderton (and Gendron) circuits let you stick a latching and momentary SPST switch in parallel, giving you the choice of stepping on the switch and the effect staying on until you step again, OR keeping the effect on only for as long as you hold the switch down, and reverting back the instant you lift your foot.  One of the things I used to do was use the momentary switch to briefly send a feed to a delay line.  The delay line remained patched into a mixer, such that I could momentarily press the switch and get a single riff delayed, "studio punch-in style", and continue playing without delay.  Meanwhile if I had the regeneration up a bit, the riff or note would continue to echo a bit until it died out without any new content added to the delay.  The FET-based switching designs that HAVE to use momentary switches don't provide that sort of choice.

And, in case I didn't mention it enough, routing the in/out and send/receive jacks to the rear skirt of the chassis for at least one of the switching modules allows you to incorporate other external effects into the same stagefront floor control unit.  So, for instance, if you have 5 stompswitches in the floor unit, 4 might be for running things inside the rack chassis, and one might be reserved for controlling some other rackmount unit like a fancy schmancy delay line, or even a digital multi-FX you like for some stuff but not for everything..

[manson]

Happy new year! Hmm I cannot find anything about the 'Anderton' thing you're talking about. What exactly is that ?
Having an extra send/receive is a very useful idea, I'll keep that in mind!

[sfr]

I believe he's referencing one of the projects in Craig Anderton's "Electronic Projects for Musicians", a very interesting book for the DIY'r. 

[manson]

That's going to work much more quietly if you bias the inputs and outputs to 1/2 the power supply to the CD4053.

R.G., about the picture of the switch I posted: What should I do to bias this circuit? Does this thing work noiseless as is, or does it realy need this half of the powersupply biassing? Is it a working schematic at all? I read here somewhere that some of Laurier's other switching circuit contained errors..

[Barcode80]

i have 3 old stompswitches which are just open close, only two contacts, could they be used for something like this? basically like the 5 volts hits the switch circuit when they are closed, and viseversa. would it work?