mod VCF to VCA? (convert a DOD440 to a dynamic gate)

[mr_doyle]

Hello guys,
i have an old DOD 440 and i want to do some experiments: do you think it would be possible to modify it so that, instead of cutting frequencies, it cuts volume of signal?
TIA,
D.

[Mark Hammer]

Aren't you just describing a compressor?

[mr_doyle]

quite the opposite, i wish that signal is removed when i'm not playing...

i know, you ask "what signal you have if you don't play?"
so, i wish to remove noise/hiss out of a distortion pedal with a sort of vca, as done in the a/da flager, boss ce1... i'm on the wrong path?

TIA,
D.

[toneman]

A VoltageControlledFilter (VCF) and a VoltageControlledAmplifier(VCA) are two different animals.

So, I guess, the answer is:  no

T

[mr_doyle]

found the anderton ultra fuzz, it has a gate too, right?  
i wish to understand what is the gate part of the circuit and if it can be "attached" to a different circuit.



TIA,
D.

[Mark Hammer]

quite the opposite, i wish that signal is removed when i'm not playing...

i know, you ask "what signal you have if you don't play?"
so, i wish to remove noise/hiss out of a distortion pedal with a sort of vca, as done in the a/da flager, boss ce1... i'm on the wrong path?

TIA,
D.

Well, you're not entirely on the wrong path, but certainly trying to modify a DOD440 to do that job seems to me to be an unnecessarily hard way to do it.

I'm glad you brought up the CE-1 and A/DA Flanger, because they illustrate a very useful principle.  Both of these tap the input signal, extract an envelope voltage from that, and apply the envelope voltage to gate the processed (delayed) signal.  When people use a separate noise gate pedal either placed before or after an effect pedal, it can create certain problems.  Maybe problems is not the right word, but certainly it doesn't perform as "invisibly" and responsively as it could if the gate circuitry is "wrapped around" the effect the way it is in the CE-1 and A/DA.

The ideal is the have some kind of input buffer stage in the distortion (and most will have this), and feed the buffer output to the clipping circuitry and sidechain (the part that detects the amplitude of the input signal).  The sidechain then controls a FET or LDR near the output of the distortion.

Why is this "better"?

• If the noise gate is placed entirely after the distortion (i.e., both sensing and gating take place after the distortion output), it will need to be set to be able to distinguish between the background noise produced by the distortion (and they ALL produce at least some noise - it comes with the high gain) and the distorted guitar signal.  That means the threshold for opening the gate will need to be higher to do the job right, and that risks missing the initial attack of notes or the ending of sustained notes.  Sticking the sensing circuitry in front of the distortion means that input noise will be low so the threshold can also be set fairly low.
• If the gating action takes places before the distortion, you risk not being able to eliminate the noise produced by the distortion pedal itself.  So ideally, the gating action should occur after the effect or point in the signal chain where the noise is most likely to occur.

You will note that in both the CE-1 and A/DA, the gate is only applied to the delay portion of the signal and not the dry/clean portion.  By targeting only the noisy portion of the circuit, the gating action feels more "invisible" to the user.

Although the voltages are all wrong for a 9vstompbox, the circuits in the A/DA and CE-1 are ideal examples of what to do.

You know, this is an excellent opportunity to ask people to think about designing some kind of noise-gate retrofit.  That is, a board that contains an input buffer, an envelope follower, and a gating FET, which can lend itself to fitting in the same case (presumably a 1590BB or equivalent) as a distortion.  The gate sensitivity would presumably be set via a trimpot, and would apply its action to the distortion only.

[Pushtone]

You know, this is an excellent opportunity to ask people to think about designing some kind of noise-gate retrofit.  That is, a board that contains an input buffer, an envelope follower, and a gating FET, which can lend itself to fitting in the same case (presumably a 1590BB or equivalent) as a distortion.  The gate sensitivity would presumably be set via a trimpot, and would apply its action to the distortion only.

I'm no where near designing circuits,  but if I may plant a seed for a concept...

Is it practical to build this into a loop box so any pedal, dist or other effect, could use this gate concept?

[Mark Hammer]

Absolutely.

If I'm not mistaken I do believe someone drew up a DOD noise gate pedal (FX-30?) in precisely this manner.  Take a peek at RG's FXbus project as well: http://www.geofex.com/Article_Folders/FXbus/fxbmod06.htm

Following up on the Behringer SNR2000 thread for a moment, a pedal-specific noise-reduction add-on would ideally target hiss, and have an action like a dynamic treble cut.  An optimised unit might have a fairly quick treble reduction and a somewhat slower volume fade.  That way, the brunt of the signal isn't chopped out immediately but the hiss is moved out of the way efficiently.

[Ben N]

Mark:

What you describe COULD be done in a separate noise gate pedal with loop, like, say a Boss NS-2?!?  Still not a perfect solution for single coil players and other sources of noise, but pretty effective at curbing pedal noise.

As far as a retrofit project, how about just the EA Fuzz minus the fuzz? I.e., a compander with precompression sidechain to control a post-effect expander/gate, buffered sidechain tap to control envelope effects, trim to vary compression at the input (more for modulation effects & fuzz, less for OD or anything where "touch sensitivity" is expected), second trim to vary expansion/gating at output. And while we are at it, it seems like that sidechain would be a great place for a sharp 60Hz filter for us Strat/Tele/P90 types, so even nasty stage hum doesn't trick the thing into opening up.

Whadya think?

Ben

[Mark Hammer]

ATTENTION TONE GOD!!   ATTENTION TONE GOD!!   ATTENTION TONE GOD!!

Here is the next monthly design contest:  Design something to remove undesirable components from a processed or unprocessed guitar signal.  That could be a gate, a compander, a noise filter, a hum filter, or any sort of thing that makes what goes to the amp 99.99% wanted guitar signal, and what stays behind 100% crap.

Ben (my younger son's name is Ben.  You have no idea how many times a day I start off saying something with "Ben,"),
The asymmetrical compander is used more often than one might realize.  It is what makes the PAiA Rocktave behave just a bit better.  The compression and expansion are set up slightl differently such that the tails of notes are faded out quickly.

Strikes me that, in an abstract sense, the ideal gate would be something that compares the summed amplitude of the undesirable elements (hiss + hum + static or clock), and the summed amplitude of the desirable elements (everything else), and turns the gate on when the ratio of desirable to undesirable exceeds some amount.  Here, the goal would be to adjust the gating depending on how easily heard or masked the undesirable components are rather than some absolute level.  Normally, gates operate on the principle that there is some absolute overall signal amplitude below which it is unlikely there would be any signal and above which the likelihood is that there is desirable signal accompanying the noise.  That works up to a point, but....

A simple dynamic hum filter might be reasonably easy to construct.  In contrast to a more broadly-targetted noise-rejection circuit, here the offending content is easily identified by being below some frequency (and note that 120hz stuff leaks through too).  When you play, there is more than enough signal content above, say 240hz, to compare against the amplitude of what's below 240hz. When there isn't enough "real signal", the filter kicks in and sharply attenuates anything below the cutoff.  Voila.  Single coil pickups with plenty of bass, but much less annoying hum when you stop playing.

So what do I think?  I think you got me thinking!

[Ben N]

I got you thinking? Now that is something! 

As for the content of your post--afraid I need some time to chw it over and se if I understand it, and I wont make my boss pay for the time that will take. It looks really interesting at first glance, but I am still suspicious of the idea of a filter that "kicks in" at some threshold point (evn if it is dynamic, if I  understand you correctly) being audible in its operation. I guess my working predicate (which may or may not be valid) is that a notch filter that is audible in the signal chain may not be if all it does is control the sidechain.

Your model of an ideal gate sounds really interesting, too, and quite doable, at least in the dgital realm. But it also raises a question of psychoacoustics. Imagining an ordinary, noisy signal chain, it may be that there are places where the signal falls below the noise threshold, and should therefore be gateable, but the brain fills in the information, whereas if you gate, and there is just a silent gap, it could be more disturbing than the noise. (Does this make sense?) If so, even a more sophisticated gate control which sets the threshold more intelligently might still have this problem. OK, I guess we don't have to solve all the world's problems at once, right?

[toneman]

In addition to the Roctave Dividor kit, PAiA also *still* sells the Gator kit.

www.paia.com


T

[Mark Hammer]

I got you thinking? Now that is something! 

As for the content of your post--afraid I need some time to chw it over and se if I understand it, and I wont make my boss pay for the time that will take. It looks really interesting at first glance, but I am still suspicious of the idea of a filter that "kicks in" at some threshold point (evn if it is dynamic, if I  understand you correctly) being audible in its operation. I guess my working predicate (which may or may not be valid) is that a notch filter that is audible in the signal chain may not be if all it does is control the sidechain.

Your model of an ideal gate sounds really interesting, too, and quite doable, at least in the dgital realm. But it also raises a question of psychoacoustics. Imagining an ordinary, noisy signal chain, it may be that there are places where the signal falls below the noise threshold, and should therefore be gateable, but the brain fills in the information, whereas if you gate, and there is just a silent gap, it could be more disturbing than the noise. (Does this make sense?) If so, even a more sophisticated gate control which sets the threshold more intelligently might still have this problem. OK, I guess we don't have to solve all the world's problems at once, right?

Right.

1) The Gator is a nifty circuit.  Does a couple of tricks, including "reverse tape" emulation à la Slow Gear.

2) Gates tend to be predicated on making stupid decisions.  A bit like telling a child to rigidly adhere to the rule "Don't talk to strangers", with the result being they stay away from police, doctors, new kids in their class, substitute teachers, kids they meet in the playground, and friends of the family or even relatives who come over to the house for the first time.  The trick to making a gate behave smarter is to give it some more skill and reasoning in making decisions; essentially to incorporate multiple factors or sources of information in making decisions.  Of course, at the same time, what it does in response to the decision/judgment made has to be intelligent too.  This is why expansion+filtering works so much better than classic gating; because it is a "titrated" response (i.e., the adjustment is calibrated to the signal properties, rather than just on/off).  It may well be that, even if this can be done in the analog domain, it is more cost effective and compact to do it in the digital domain as you suggest.

3) The notion of a signal sensitive hum-notch filter is no different than any other dynamic filtering or gating, except that instead of restoring treble when the signal exceeds some critical level (which is what noise filters normally do), it would restore content in the hum-zone.  Actually, the filter wouldn't have to specifically be a notch filter.  For instance a simple 2-pole highpass filter whose corner frequency slid up to 250hz during quiet passages, and went down to 50hz once you started playing again would probably make a noticeable impact on hum at gigging volumes.

Now that I think of it, since the hum comes from the instrument and everything else up to the first pedal (and we're assuming the wallwart is well-regulated and injects no hum into the pedal chain), removing hum from the input signal to the first pedal would make the job of differentiating hiss-type noise from signal that much easier for anything placed after the pedal chain.  Essentially, our ideal system tracks and defeats hum on the input and hiss on the output.

Make sense?

[Ben N]

Yup. As long as that slide trombone HP filter doesn't sound like a HP filter. 

Now, Andy TTG, how 'bout that contest? 

[dirk]

Maybe a stupid question: Isn't this simular to Buchla's lowpass gate?

[mr_doyle]

You know, this is an excellent opportunity to ask people to think about designing some kind of noise-gate retrofit.

Good, let's ask 
Who wants to take honour to put this important brick to build the world of quiet pedals???

TIA,
D.

[Ben N]

I think we are stuck in the OT thread title/content disconnect, here. Perhaps there is some way to transfer/copy the portion of t6his thread tht is not really responsive to the OP to a new thread?