Explanation for DOD footswitch section with 4007 IC

[Nikenzzy]

Hi, can someone please explain to me the purpose of switching section containing 4007 ic in DOD pedals? Why is this used instead of 3PDT switch?

[Rob Strand]

Hi, can someone please explain to me the purpose of switching section containing 4007 ic in DOD pedals? Why is this used instead of 3PDT switch?

The alternative to mechanical switch is solid-state switching with a push-button switch - like Boss, Ibanez etc.

The DOD circuit is just another version of solid-state switching.

[Nikenzzy]

Thank you.

[Mark Hammer]

The 3PDT switch is a comparatively recent device.  To the best of anyone's knowledge, Mike Fuller - the Fulltone guy - arranged with a manufacturer to have them made, some time in the early '90s.  Prior to then, mechanical switching in pedals was done with either SPDT switches (found on many early MXR, EHX and other pedals) or if you were lucky, DPDT footswitches.  DPDT could, at least, provide full bypass of the effect circuit, but could not provide for a status indicator (not enough contacts).  SPDT switches would select between the input jack and circuit output, leaving the circuit input hard-wired to the input jack all the time.  If one had several pedals patched together, their combination would load down the guitar signal like crazy, launching the concept of "tone suck".  YOunger players who get hold of vintage pedals and hope to "improve" them by replacing a SPDT switch with a 3PDT switch, will often find an unexpected and annoying volume difference between effect and bypass mode, that stems from what happens when a circuit designed with the expectation of constant loading, is UNloaded.

In the absence of a simple mechanical solution Boss/Roland came up with the use of a transistor flip-flop circuit, combined with FETs that could be turned on and off, to cancel/bypass their effect circuits.  DOD eventually came up with their own version, using the 4007, to turn targeted FETs on and off.  Not only could this functionally duplicate what a DPDT switch could do, but it was also able to turn an LED on or off at the same time.  Back in the days when a "pedalboard" might consist of a wah, fuzz, and maybe one more pedal, status LEDs were largely unnecessary, since there was little mistaking what was "on" or "off".  But as players began to employ more and more pedals, it became more important to have some way of knowing which were on and which were off.  The solid-state flip-flop/FET combination solved that dilemma. 

But it did not do so without introducing its own host of concerns.  All solid-state-switched pedals require an input buffer and an output buffer, since one cannot simply stick an FET as the "first" and "last" components around an effect circuit.  The search was launched for better buffer stages, and players began to have concerns about what a half-dozen buffer stages, in series, might be doing to their tone (that's what three electronically-switched pedals would provide, two for each pedal).  Enter loop selectors that would completely bypass several pedals at once, removing all the various buffer stages from the signal path.  Finally, many current pedals and loop selectors will employ miniature relays, and circuitry to drive the relays, to provide full bypass while permitting use of "soft touch" momentary footswitches.

[amptramp]

I have a thread on noiseless switching here:

https://www.diystompboxes.com/smfforum/index.php?topic=120006.msg1122294#msg1122294

This shows several switching methods using both unbuffered controls (like the 4007) and buffered control switches (like the 4066) to achieve near silent switching.  You may have to go hunting (and risk fake devices) to get the 2SK30 devices used in Boss pedals whereas CMOS is sold everywhere and is usually genuine.

[ElectricDruid]

I think the "noiseless" element is a big part of why Boss developed their FET switching.

Many of the SPDT or DPDT-switched pedals of the 70's and 80's produced a decent "Clunk!" when switched, not only mechanical noise from the switch, but an electrical "Thump!" out of your amp too. "Noiseless switching" was the holy grail at this point in the same way that "true bypass" has become now. Boss's FET switching was not only a way to avoid the problems with switches, but also designed to provide the noiseless switching everyone wanted. The first time I heard it (or rather, *didn't* hear it!) it was something amazing and I wanted to know how it had been done.

[R.G.]

The 3PDT switch is a comparatively recent device.  To the best of anyone's knowledge, Mike Fuller - the Fulltone guy - arranged with a manufacturer to have them made, some time in the early '90s. 

Minor correction. 3PDT stomp switches existed before Fuller did a custom manufacture. They were just expensive. If I remember correctly, the only pedal-sized ones were also Carling, but a fancier appearance than the DPDT.  I sold the Carling 3PDTs one and two at a time back then, but they were too pricey for mainstream pedals. Fuller did go get knockoffs of the Carling 3PDT made cheaply in quantity. Later other companies joined in and the prices came down further.

@ Nikenzzy: Mark has done a great job of boiling down about three decades of pedal history for you. There is a fair amount of deeper techie background on bypass switching at http://www.geofex.com/article_folders/bypass/bypass.htm, over several articles boring into the various methods of actual switching, whether hard contact or solid state switches. There is one article on just the signal routing - literally where the signal goes through various switches, and the effect on the signals. I've used nearly all of the switching methods over time. The one that has stuck is the use of a CD4053 CMOS multiplexer for the actual switching of signals and a controller switch, which can be either a foot switch, or a $1.00 microcontroller to sense a momentary footswitch and flip the CMOS.

The DOD CMOS switching with a CD4007 is one way to get two CMOS transmission-gate switches and a flip-flop out of a single CMOS IC. It's an elegant way to do it. It combines the need for a flip-flop, whether electronic or mechanical (that is, a non-momentary foot switch) with the actual switching elements all in one. 

CMOS switches have an issue with control feedthrough. The multi-volt control signal bleeds through to the signal switches a little. This has to do with the internal capacitances of the MOSFETs that make the switches. The trick is to bias the in- and out-pins of the multiplexer switches to half of the CMOS IC's power supply voltage. This beings the amount fed through to the audio path down to well under 20mV; 20mV is important because less than that tends to be inaudible in even a high-gain pedal setup.

Slow switching is a way to make even noisy switching work OK. The audio path for guitars has a distinct high pass characteristic; signals below 40-80Hz are heavily cut. Your ears also don't hear things slower than about 20Hz. So slowing down the steepness of a switching edge both gets it lower than the amplifier passes and lower than your ears are set up to hear. This is what the Boss/Ibanez FET switches do. See http://www.geofex.com/Article_Folders/bosstech.pdf

@amptramp: I never had the nerve to try diode bridge switching in a pedal because of the need for good matching of the drive signals. I'm a fan of diode modulators as used in the Thomas Vox amplifiers for tremolo. Great trick!

[Mark Hammer]

I think the "noiseless" element is a big part of why Boss developed their FET switching.

Many of the SPDT or DPDT-switched pedals of the 70's and 80's produced a decent "Clunk!" when switched, not only mechanical noise from the switch, but an electrical "Thump!" out of your amp too. "Noiseless switching" was the holy grail at this point in the same way that "true bypass" has become now. Boss's FET switching was not only a way to avoid the problems with switches, but also designed to provide the noiseless switching everyone wanted. The first time I heard it (or rather, *didn't* hear it!) it was something amazing and I wanted to know how it had been done.

Noiseless until combined with true-bypass pedals.  In an effort to minimize loading, a great many pedals that use electronic switching will omit a "bleed" resistor to ground on the input, and have the input jack go directly to the input cap.  One may hear a "thunk" upon plugging in, but none thereafter...IF one is using ONLY e-switched pedals on the board.  Such pedals always include a terminating resistance of some sort on the output buffer which, if connected to another e-switched pedal, leaves nothing "hanging".  When such pedals are combined with a TB pedal just ahead of them, however, the mechanical switch will duplicate the "thunk" of suddenly disconnecting and reconnecting the input cap of the e-switched pedal.  The illusion is that the TB pedal is providing the popping noise, when it is actually coming from the e-switched input buffer.  That's why I referred to it as "pedal ventriloquism".

And yes, I should have known, and stated, that footswitches with more than 3 poles were "available" prior to Fulltone's  use.  But as RG rightly points out, they weren't the sort of thing whose dimensions and price were compatible with inexpensive and compact pedals.  I know that in the early '90s, I was paying $20@ for a DPDT stompswitch (though I imagine they were cheaper if one was ordering thousands).

And since my site on ampage.org seems to be inaccessible, here's a scan of the DOD circuit, as presented in POLYPHONY by former DOD engineer David DiFrancesco.