Digital vs analog switching

[propadog]

Hi, all.

One more and then I'll go away for a little while! 

You will recall I recently asked about remote switching of racked effects.

At RG's hint, I investigated servos for use with pedals. Given that the servos are a closed loop device, and position is dependant on the number of pulses per unit of timeframe, it appears that all the control I want can be communicated in (for want of a better word) binary.

The schematic for the servo driver shows is controlled by a pot, so installation into a traditional pedal is a no brainer. It also means that any effect can have an expression pedal added to it should I so choose.

So, having stripped everything the control signals to a whole bunch of ons and offs, and having decided that optocoupling will be A Good Thing in this because I think it will clean up the signal, protect against voltage drops and keep the noise out, I have one lot of questions left.

I am staying with racking up the traditional pedals, there could be up to 10 effects, plus amp switching, volume and wah controls, and control of my wireless link from my guitars (switch channels on the receiver when I switch guitars) and that number could easily change. Because of my experience in my day job I always run a three-case scenario- best case, worst case and just in case. With the pedals, I would set the default to mid point so in case of some failure in the switching circuit, it will default to 50%.

I've had a look at a few resources, including RG's ASMOP page, and on this forum (http://www.diystompboxes.com/smfforum/index.php?topic=42518.0).

I'm asking:

Once it gets to the rack, is it better to stick with traditional analog (or mechanical) switching (e.g. relays) or move across to digital (e.g. PICs).

If I stayed with an analog switch, is there an alternative to a relay to consider? Would you stay with latching relays to give at least some sound if there is a failure? Is there a failsafe using a PIC?

And overall- If you were considering my project, and not looking at it from the point of view of a DSP person, what would you do, and what would you recommend?

Thanks,

(Sorry to be a noobie pest,  and I'll go away after someone replies!  )

Cheers,

The Propadog.

[RaceDriver205]

I probably read you incorrectly, but if your thinking of using remote-controlled servo-actuated potentiometers in ten rack-mount effects then you've got a screw loose!
But apart from that, the kind of switching you want to do is best done with a PIC or an AVR (and I'd say an AVR). Presuming you write a completely working program for it (i.e. well tested for bugs), it won't suddenly disconnect effects or anything like that. It will be inherently failsafe.
My recommendation would be don't underestimate the complexity and time that will be involved.
How do you want to be able to switch your effects? is the key question.

[propadog]

I probably read you incorrectly,

Defintely incorrectly.

but if your thinking of using remote-controlled servo-actuated potentiometers in ten rack-mount effects then you've got a screw loose!

In my twenty five odd years of playing, I've never seen anyone bend down and change a setting on a pedal mid song... So why is there a need to alter an effect  remotely apart from turning it on and off?

This is solely about getting the volume and wah pedals talking to the rack. That's all. I was also seeing if anyone had any more ideas. However, the response was nothing. Absolutely nothing. Not even a "we've tried this, check it out over at www.something.com." Plus I've got a wireless connection between guitar and gear. The receiver is near the rack. Why would I then run a cable out to the stage and back again???

I've worked out how to do what I want it to. I'm recycling a non-related Anderton concept, then inverting a signal twice to clean it up, and a few little things. Nothing new, just fresh uses of simple, old ideas.

The proof of concept has already worked. So, I've now got six individual pedal remote bypasses, volume and wah pedals, a channel selector and a "pick a bunch of effects with one stomp" selector happening for when I want a Brian May or Easybeats sound. It's all now a matter of neatening up the design, making it a little bit more compact, sturdy and beer-proof.

But apart from that, the kind of switching you want to do is best done with a PIC or an AVR (and I'd say an AVR). Presuming you write a completely working program for it (i.e. well tested for bugs), it won't suddenly disconnect effects or anything like that.
It will be inherently failsafe.

On catstrophic signal loss, it defaults to volume pedal at 50%, reverb, chorus and overdrive on, rest off, which will get me through 80% of songs. It's all a matter of alignment. Simple solutions are the best.

I've moved over to a IC-based design.  I even spent a few days playing with a volume pedal using a commercial digital volume control chip, evaluating whether I could play with a pedal that didn't give "positional" feedback. No was the result. So the pedal could not return, meaning therefore it had to be chattering away contantly to the other end.

I do a lot of redundancy and testing. I've spent 12 of the the last 15 years living in a three case scenario. Best case. Worst Case. Just In Case. I had to. I was involved in the design the second largest cell network in Aus.

[RaceDriver205]

Um, good.

[puretube]

err, I recently saw a video, with this RHCP guitarist bow down to his pedal(-s),
and twisting some knob, live, in a song...

[David]

I probably read you incorrectly,

Defintely incorrectly.

but if your thinking of using remote-controlled servo-actuated potentiometers in ten rack-mount effects then you've got a screw loose!

In my twenty five odd years of playing, I've never seen anyone bend down and change a setting on a pedal mid song... So why is there a need to alter an effect  remotely apart from turning it on and off?

This is solely about getting the volume and wah pedals talking to the rack. That's all. I was also seeing if anyone had any more ideas. However, the response was nothing. Absolutely nothing. Not even a "we've tried this, check it out over at www.something.com." Plus I've got a wireless connection between guitar and gear. The receiver is near the rack. Why would I then run a cable out to the stage and back again???

I've worked out how to do what I want it to. I'm recycling a non-related Anderton concept, then inverting a signal twice to clean it up, and a few little things. Nothing new, just fresh uses of simple, old ideas.

The proof of concept has already worked. So, I've now got six individual pedal remote bypasses, volume and wah pedals, a channel selector and a "pick a bunch of effects with one stomp" selector happening for when I want a Brian May or Easybeats sound. It's all now a matter of neatening up the design, making it a little bit more compact, sturdy and beer-proof.

But apart from that, the kind of switching you want to do is best done with a PIC or an AVR (and I'd say an AVR). Presuming you write a completely working program for it (i.e. well tested for bugs), it won't suddenly disconnect effects or anything like that.
It will be inherently failsafe.

On catstrophic signal loss, it defaults to volume pedal at 50%, reverb, chorus and overdrive on, rest off, which will get me through 80% of songs. It's all a matter of alignment. Simple solutions are the best.

I've moved over to a IC-based design.  I even spent a few days playing with a volume pedal using a commercial digital volume control chip, evaluating whether I could play with a pedal that didn't give "positional" feedback. No was the result. So the pedal could not return, meaning therefore it had to be chattering away contantly to the other end.

I do a lot of redundancy and testing. I've spent 12 of the the last 15 years living in a three case scenario. Best case. Worst Case. Just In Case. I had to. I was involved in the design the second largest cell network in Aus.

Sorry about the perceived lack of response.  Short of ASMOP and the Crossbar projects, this is a very new frontier for most of us.  I have some PIC expertise, but I haven't taken on any control systems yet.  I doubt I can help much because it sounds like you know more than I do, but I'll lend what expertise I have.

You can do your switching electronically if you use RG's 4053 switching circuit driven by an ASMOP affair.  A forum member named Chico also implemented an ASMOP-type system that incorporated wireless transmission some years back.  Those could help, as well as reading everything RG has on GEO about the ASMOP.  Searching for it here is also useful.  Look on www.piclist.com for code that reads the position of a pot.  There is a site called www.ucapps.de where they are all about building complex, incredible MIDI controllers.  They measure pot position every day of the week.  There might also be something on the winpicprog.co.uk site.  Read Myke Predko's books and hunt around on his web site.  Also look for stuff from Ross Bencina, Tom Scarff and Marc Bareille.  I refer to PICs only because that's where my experience is.  I'm not about to invest in the development infrastructure for the AVR at this time.

[David]

Hey, Prop!  You still out there?  Any news?

[idlechatterbox]

Don't give up. I think it's a cool idea, but then I love Rube Goldberg type getups anyway.

I don't remember where I read it, but I recall an interview with a guy who did the remote thing, and he really did it remotely, as in, he actually used remote-controllers from model airplanes to adjust his knobs (pedal knobs that is). It all struck me as pretty elaborate in a cool way, but I guess it goes to show that it is possible. Another thing I remember was that he wasn't a guy known for being a tone-merchant, or even a guitar-god. I.e., this wasn't an SRV or an "Edge" trying to make sure he had his signature sound. Just a normal guy who wanted to be able to adjust everything remotely during the show I guess.  I can try to find the link. 

[idlechatterbox]

In case anyone's curious:

http://members.aol.com/guitarron/jmcp/jmcp.htm

happy new year