ampage Tube Amps / Music Electronics	For current discussions, please visit Music Electronics Forum.

ampage archive

Vintage threads from the first ten years

Programmable effects switchers and tapdancing.

12/31/1999 4:33 PM
R.G.	Programmable effects switchers and tapdancing. Just thought I'd make the topic easier to get into - the original got so long. Over the holidays I had some time for introspection and I came up with an any-effect, any-order scheme that is buildable without microcontrollers if you're careful about how you set it up. It goes like this: 1. The input jack and each effects return is connected to one entire side of its own dip switch. Each contact on the other side of the dip switch selects which effect the input jack/FX loop output goes to. One position of each dip switch is connected to the output jack. With this setup, you can get any effect order. Some of them are things you shouldn't do, like connecting the output of the FX loop to its own input, so those connections can be left off. One dip switch per input jack/FX output makes one "program". and can connect the effects in any order. If you like rotary switches, the dip switches can be replaced by a 1P/multi throw rotary switch. 2. Between each output and its dip switch there is a remote switchable contact, like a relay or a CMOS analog switch. All the contacts are turned on and off in parallel. For the one-program case, all the FX outputs and input jack are either all connected to their dip switch or not. 3. To add multiple programs, you duplicate the relay contact-dip switch setup once per program. Turning on all the relays in one program activates that program through it's respective dip switches. Only one program should be active at a time. This can also be done with one CMOS analog multiplexer per dip switch instead of relays. 4. The remote footswitch is unchanged - it just selects one and only one "program" when the corresponding footswitch is pressed. 5. For remote amp settings, the relay at one or more of the effects loops needs to have an uncommitted contact per amp setting to do the switching. I think this thing will do anything a CPU-based setup will with the exception of - it won't check for and prevent settings you shouldn't make, like circular loops of effects or multiple selections in one dip switch - it won't make programming the selector matrix easy; a human still has to know what he's doing to do that.

12/31/1999 4:33 PM

R.G.

Programmable effects switchers and tapdancing.

Just thought I'd make the topic easier to get into - the original got so long.

Over the holidays I had some time for introspection and I came up with an any-effect, any-order scheme that is buildable without microcontrollers if you're careful about how you set it up.

It goes like this:
1. The input jack and each effects return is connected to one entire side of its own dip switch. Each contact on the other side of the dip switch selects which effect the input jack/FX loop output goes to. One position of each dip switch is connected to the output jack. With this setup, you can get any effect order. Some of them are things you shouldn't do, like connecting the output of the FX loop to its own input, so those connections can be left off. One dip switch per input jack/FX output makes one "program". and can connect the effects in any order. If you like rotary switches, the dip switches can be replaced by a 1P/multi throw rotary switch.
2. Between each output and its dip switch there is a remote switchable contact, like a relay or a CMOS analog switch. All the contacts are turned on and off in parallel. For the one-program case, all the FX outputs and input jack are either all connected to their dip switch or not.
3. To add multiple programs, you duplicate the relay contact-dip switch setup once per program. Turning on all the relays in one program activates that program through it's respective dip switches. Only one program should be active at a time. This can also be done with one CMOS analog multiplexer per dip switch instead of relays.
4. The remote footswitch is unchanged - it just selects one and only one "program" when the corresponding footswitch is pressed.
5. For remote amp settings, the relay at one or more of the effects loops needs to have an uncommitted contact per amp setting to do the switching.

I think this thing will do anything a CPU-based setup will with the exception of
- it won't check for and prevent settings you shouldn't make, like circular loops of effects or multiple selections in one dip switch
- it won't make programming the selector matrix easy; a human still has to know what he's doing to do that.

1/4/2000 7:51 PM
Jsh	Uh... I looked at it a few times now... Would you be so kind to make use of your drawing/artistic side? ...I mean, could you make a fast scetch and scan it? I tried to draw it from what you explained here, but I think I missed the point...or uhhh.. yeah! coffee, I'll get some coffee, anyone else... Kind regards Jsh

1/4/2000 8:52 PM
R.G.	I've been drawing it up. I'll post it as soon as I can get it clear. I'm drawing up a 4x4 with three independent amp switches - I think a reasonable design to implement.

1/5/2000 12:11 AM
R.G.	Re: Programmable effects switcher (long) OK, take a look at the effects" target="_blank">http://www.eden.com/~keen/fxselex3.gif">effects order selection matrix and the footswitch" target="_blank">http://www.eden.com/~keen/fxselex4.gif">footswitch and program selection relays. Looking at the effects order selection, it works like this: the rotary switches at the input jack and at the output of each FX loop select the NEXT thing in the chain they're connected to. Assume that Program 1 is active, and all the other Programs are off. That means that P1IJ, P1FX1, P1FX2, ... P1FX4 make contact. So the input comes in the input jack and the first rotary switch selects whether it goes to the output, FX1, Fx2, ... FX4. Let's say the switch is set to FX2. The signal goes through the input jack rotary switch to FX2's input jack. FX2's output jack then has a rotary switch that can connect it to the output jack or any other effect. Let's say that's FX4. So we then go to FX4 and set ITS rotary switch. This way we kind of select a "path" through the four FX loops to the output - yeah, one and only one should go to the output jack. So we have one program. If we imagine that we turn off all of the relay contacts that connect the input jack and the FX outputs to Program 1's rotary switches, and turn ON the relay contacts (P2IJ, P2FX1, P2FX2, P2FX3, P2FX4) that connect up the Program 2 set of rotary switches, we get to set up a new "path" that can only work when all the Program 2 relay contacts are connected. The programming matrix allows for four independent paths through the FX loops. To select a program, we only have to turn on all the relay contacts for that "program". Now look at the footswitch" target="_blank">http://www.eden.com/~keen/fxselex4.gif">footswitch and program selection relays. There are five signals PGM1 through PGM5 that each turn on one BS170 MOSFET. The MOSFET pulls down on the relay coils to close the contacts associated with that "program". I've shown five separate relay coils, but you could use a single 6P relay, three DP relays, etc. The point is that all the relays for a single "program" turn on together. The fifth "program" just turns on the bypass relay, giving a true bypass around the whole mess. All we need to drive the program selectors is five signals PGm1... PGm5. The optional amp switching is also shown. The external/amp switches only need to be turned on/off with each program, not to have a destination selector, so you can use a dip switch or four discrete switches to connect the single relay that connects the external contacts together. Diodes keep the programs from interacting. I've shown three isolated contacts, one for channel switch, reverb, and tremolo. If you keep the external switches separate from the signal ground, they can be used to switch almost anything you like. I've shown SPST relays, you could use any kind of relay within reason. The (presumably remote)footswitch is also shown on the same sheet. This is the same as my earlier design for in-line switching. The momentary footswitches are debounced in the Schmit trigger CMOS gates and drive their data into the 74C373 latch. Only one of the latches will be set high, the others will be set low. The outputs drive an LED per output for a program indicator, and then drive the BS170 MOSFETs to run the relays. I've shown only a four channel plus bypass setup. The stuff shown in gray is available for expansion. I decided that for my setup, I'd use rotary switches, not dip switches for the path programming. That was based on 1. I hate dip switches, and being sure that one and only one was set would be a pain IMHO. 2. A six position dip switch is about $1.25. A rotary switch is $1.51 plus a knob if you don't have knobs in the junk box. I figured the difference in money was small. Makes for a bigger control panel, though. So - that's how it works; all true bypass, no silicon switches, no CPU or clocks, any effects order at all, and with external switch contacts for controlling amps, lights, you name it. It's expandable to seven programs and as many FX loops as you want to pay for. This thing is properly considered to be part of an advanced pedalboard setup. You'd want to build it as part of a whole FX setup.

1/5/2000 12:11 AM

R.G.

Re: Programmable effects switcher (long)

OK, take a look at the
effects" target="_blank">http://www.eden.com/~keen/fxselex3.gif">effects order selection matrix and the
footswitch" target="_blank">http://www.eden.com/~keen/fxselex4.gif">footswitch and program selection relays.

Looking at the effects order selection, it works like this: the rotary switches at the input jack and at the output of each FX loop select the NEXT thing in the chain they're connected to. Assume that Program 1 is active, and all the other Programs are off. That means that P1IJ, P1FX1, P1FX2, ... P1FX4 make contact.

So the input comes in the input jack and the first rotary switch selects whether it goes to the output, FX1, Fx2, ... FX4. Let's say the switch is set to FX2. The signal goes through the input jack rotary switch to FX2's input jack. FX2's output jack then has a rotary switch that can connect it to the output jack or any other effect. Let's say that's FX4. So we then go to FX4 and set ITS rotary switch. This way we kind of select a "path" through the four FX loops to the output - yeah, one and only one should go to the output jack.

So we have one program. If we imagine that we turn off all of the relay contacts that connect the input jack and the FX outputs to Program 1's rotary switches, and turn ON the relay contacts (P2IJ, P2FX1, P2FX2, P2FX3, P2FX4) that connect up the Program 2 set of rotary switches, we get to set up a new "path" that can only work when all the Program 2 relay contacts are connected.

The programming matrix allows for four independent paths through the FX loops. To select a program, we only have to turn on all the relay contacts for that "program".

Now look at the footswitch" target="_blank">http://www.eden.com/~keen/fxselex4.gif">footswitch and program selection relays. There are five signals PGM1 through PGM5 that each turn on one BS170 MOSFET. The MOSFET pulls down on the relay coils to close the contacts associated with that "program". I've shown five separate relay coils, but you could use a single 6P relay, three DP relays, etc. The point is that all the relays for a single "program" turn on together. The fifth "program" just turns on the bypass relay, giving a true bypass around the whole mess. All we need to drive the program selectors is five signals PGm1... PGm5.

The optional amp switching is also shown. The external/amp switches only need to be turned on/off with each program, not to have a destination selector, so you can use a dip switch or four discrete switches to connect the single relay that connects the external contacts together. Diodes keep the programs from interacting. I've shown three isolated contacts, one for channel switch, reverb, and tremolo. If you keep the external switches separate from the signal ground, they can be used to switch almost anything you like. I've shown SPST relays, you could use any kind of relay within reason.

The (presumably remote)footswitch is also shown on the same sheet. This is the same as my earlier design for in-line switching. The momentary footswitches are debounced in the Schmit trigger CMOS gates and drive their data into the 74C373 latch. Only one of the latches will be set high, the others will be set low. The outputs drive an LED per output for a program indicator, and then drive the BS170 MOSFETs to run the relays. I've shown only a four channel plus bypass setup. The stuff shown in gray is available for expansion.

I decided that for my setup, I'd use rotary switches, not dip switches for the path programming. That was based on
1. I hate dip switches, and being sure that one and only one was set would be a pain IMHO.
2. A six position dip switch is about $1.25. A rotary switch is $1.51 plus a knob if you don't have knobs in the junk box. I figured the difference in money was small. Makes for a bigger control panel, though.

So - that's how it works; all true bypass, no silicon switches, no CPU or clocks, any effects order at all, and with external switch contacts for controlling amps, lights, you name it. It's expandable to seven programs and as many FX loops as you want to pay for.

This thing is properly considered to be part of an advanced pedalboard setup. You'd want to build it as part of a whole FX setup.

1/5/2000 12:34 AM
jason	It's been said before but I'll say it again...you 'da man RG! Thanks, jason

1/5/2000 1:44 AM
R.G.	...aw, shucks... The real trick will be whether anyone but me builds one. I'm probably going to fold this into an even bigger article on pedalboard design. There are some folks here in Austin that will happily tell me where my ideas about pedalboards are trash and where they're useful. Meantime, I've ordered a slew of switches and I'm gonna hack one of these together. The two pictures are just the first wave.

1/5/2000 1:44 AM

R.G.

...aw, shucks...

The real trick will be whether anyone but me builds one.

I'm probably going to fold this into an even bigger article on pedalboard design. There are some folks here in Austin that will happily tell me where my ideas about pedalboards are trash and where they're useful. Meantime, I've ordered a slew of switches and I'm gonna hack one of these together.
The two pictures are just the first wave.

1/6/2000 12:06 AM
Steve Newton	WOW! I've just managed to check in for the first time this week and look what I've found. This is a great expansion of the basic idea. I haven't got a lot of use for the re-ordering thing but it's great to see that it could be easily, if not dirt cheaply, implemented. And you've added the amp control relays, nice touch. Again I've got no use for them but I'm in a minority in guitar land in that I use FX processors and a stereo power amp as opposed to a "good" guitar amp suplemented with a few pedals. As was discussed elsewhere on this subject once the logic part is sorted anything that can be controlled via logic 1's is fair game for inclusion in these switchers. I'm writing out the order for the parts for the initial prototyping of my design this evening, if the god of the British postal system is smilling on me I'll have them by the weekend and can start hacking. In the first thread on this subject somebody said there wouldn't be much monetary benefit in DIYing one of these, viewing the ever expanding total column in my order list I'm begining to agree, the duplication of parts soon adds up. One question before I go, what are you planning on using for the momentary footswitches? Years ago I used a heavy duty microswitch for punch ins on my old 4 track and I was planning to use the same for this, but they're like �4.50 ($10-ish) each! Steve.

1/6/2000 12:06 AM

Steve Newton

WOW! I've just managed to check in for the first time this week and look what I've found. This is a great expansion of the basic idea.

I haven't got a lot of use for the re-ordering thing but it's great to see that it could be easily, if not dirt cheaply, implemented. And you've added the amp control relays, nice touch. Again I've got no use for them but I'm in a minority in guitar land in that I use FX processors and a stereo power amp as opposed to a "good" guitar amp suplemented with a few pedals. As was discussed elsewhere on this subject once the logic part is sorted anything that can be controlled via logic 1's is fair game for inclusion in these switchers.

I'm writing out the order for the parts for the initial prototyping of my design this evening, if the god of the British postal system is smilling on me I'll have them by the weekend and can start hacking. In the first thread on this subject somebody said there wouldn't be much monetary benefit in DIYing one of these, viewing the ever expanding total column in my order list I'm begining to agree, the duplication of parts soon adds up.

One question before I go, what are you planning on using for the momentary footswitches? Years ago I used a heavy duty microswitch for punch ins on my old 4 track and I was planning to use the same for this, but they're like �4.50 ($10-ish) each!

Steve.

Page 1 of 4

Next>

Last Page>>