Bypass idea... am i out of my mind?

Started by lontano, July 17, 2008, 06:33:20 AM

Previous topic - Next topic

lontano

hi!

This is my very first post in this forum, i've been lurking in the shadows for a while getting some info here and there (i'm fairly new to stompbox doityourselfing).

I've been building a fretzer valve starting from the schematic, and came to the eternal problem of true-bypassing the circuit. I finally came to the conclussion of the schematic, but i'm no sure it's a ridiculous idea or not. The idea was to use a DPDT switch to true-bypass the circuit but i also want a on/off  LED, so i'm using one of the switches to select the output signal between the input signal and the FX output, and the other switch to connect/disconnect the FX from its mass connection. This way i think there will be no "tone sucking", but i'm wondering if there will be other side effects... Am i a genius or a fool?



PD. i'm sorry if there's a topic going on about bypassing, couldn't find it

petemoore

  Jfet inputs can be configured to be very high impedance.
Convention creates following, following creates convention.

MarcoMike

but no-ground on the circuit means no power, and it may take some time to wake up after you turn the switch on....
this time may be low enough... or not. just try!
Only those who attempt the absurd will achieve the impossible.

petemoore

   Whoops...yupp: it may take some time to wake up after you turn the switch on....
  Probably easiest to keep the bypass from popping by keeping the power supply connected.
  [lol?] See how long a 'Supercap' [I read about these a while back, have never seen one implimented...] holds voltage up on the circuit, my guess is that it may not hold a voltage long enough to avoid popping when bypass recharges it though.
Convention creates following, following creates convention.

lontano

Maybe it would be better then to use the second switch to insert a very big resistence between the FX and its ground? That way the impedance of the circuit will be big enough for it not to suck tone, and the circuit will be powered, even if at a lower degree... i'll give it a try. Thanks for the answers!  :icon_biggrin:

R.G.

Welcome to actually participating in the forum, lontano.

You need to start by reading all of "The Technology of Bypasses" at Geofex, http://www.geofex.com.

It's OK to do the signal portion of your circuit if and only if the effect circuit has a very high input impedance, 1M or preferably over. The LED switching portion is unnecessarily complex. Just leave the circuit connected to ground all the time, and connect only the LED to ground when it should be on. This will sidestep the big problem with popping when you press the switch.

Finally, there is already a couple of ways to do either buffered bypass (which you would have if you had a high impedance FET as the first thing inside the circuit, the Clinton bypass, or real, true bypass with the Millenium bypass and only one DPDT. They're in the article at Geofex.
R.G.

In response to the questions in the forum - PCB Layout for Musical Effects is available from The Book Patch. Search "PCB Layout" and it ought to appear.

lontano

Quote from: R.G. on July 17, 2008, 09:12:09 AM
Welcome to actually participating in the forum, lontano.

You need to start by reading all of "The Technology of Bypasses" at Geofex, http://www.geofex.com.

It's OK to do the signal portion of your circuit if and only if the effect circuit has a very high input impedance, 1M or preferably over. The LED switching portion is unnecessarily complex. Just leave the circuit connected to ground all the time, and connect only the LED to ground when it should be on. This will sidestep the big problem with popping when you press the switch.

Finally, there is already a couple of ways to do either buffered bypass (which you would have if you had a high impedance FET as the first thing inside the circuit, the Clinton bypass, or real, true bypass with the Millenium bypass and only one DPDT. They're in the article at Geofex.

Thank you very much  :icon_biggrin:
I had already seen that excellent article (that's where i got the inspiration for my awesome image  :P). I was trying to keep it simple because i'm trying to fit the circuit in a TINY box, but i guess that sometimes you can't have it both ways: good and simple.

Roobin

You can use a Millennium Bypass - it's really small, especially if you do point-to point(ie no board involved). This will get you - no tone suckage from the effect and a LED. I believe RG has a PCB layout somewhere on his site - if I remember it's smaller than a coin!

selectortone

You can also build the Millennium Bypass on a small piece of vero - 4 x 4 holes.

Thanks R.G.!

lontano

Quote from: Roobin on July 17, 2008, 09:52:48 AM
You can use a Millennium Bypass - it's really small, especially if you do point-to point(ie no board involved). This will get you - no tone suckage from the effect and a LED. I believe RG has a PCB layout somewhere on his site - if I remember it's smaller than a coin!

Something that works based on the imperfections of a device... that seems to be the norm in this music world!  :P

Roobin

Quote from: lontano on July 17, 2008, 11:07:19 AM
Something that works based on the imperfections of a device... that seems to be the norm in this music world!  :P

Imperfection or odd characteristic? How do you create a really cheap, gnarly fuzz? Put a pencil through a speaker cone. Imperfect, but it works! Coiled cables have a higher capacitance than standard straight cables for a given length (please, no argument over cables!), but some may use them to tame that rogue high-end of a strat or tele. Imperfect? Yes. Useful? Yes. So long as the imperfection isn't inconsistent, it may be highly beneficial to sue that 'imperfection'.


lontano

Just in case someone was wondering, the "wake up" time of the circuit was too long for my liking, the signal fainted for a moment, it wasn't much but enough to be noticeable. i guess 3pdt or millenium switch is the way to go. Thanks for the feedback  :icon_biggrin:

I'll have to keep on thinking to invent something myself instead of keep stealing ideas around here  :P

Radamus

Something I've recently tried (and it hasn't completely worked yet, but everyone said it should) is using a DPDT to control a 4066 switching IC and the other half of it to control an LED. If you wire it right, you only need an SPDT, but I'm unaware of any foot switches like that. Check out http://www.geocities.com/thetonegod/switches/switches.html for ideas. I found that going through the hassle of adding two or three more IC's and then building an actuator to be too much, but a foot switch can easily control the 4066, too.