Simple noise gate recommendation?

[jfrabat]

Hi, guys.  I started making my own pedals some months ago, and got hooked into it!  Here's my pedal board so far:



Only the Ditto and the power supply were not made at home (I am proud to say!).  The green one without the label is an EA Tremolo (I have to do the sticker for that one!), and I will remake the Vulcan Rat on a 1590B enclosure (I sent a PCB to OSH, and it should be here soon) to make some space in the pedal board (I do have a phaser, a clean boost, and a chorus coming up!  Guess I will have to decide what stays and what goes!).

Anyway, the issue is that, especially when using the compressor and the Vulcan Rat or the Screechy Cat Overdrives, I am getting some noise out.  So I wanted to make a Noise Gate for the end of the chain.  But I am not sure which one to make, so I figured I'd ask you guys.  So, what is your recommendation?  Criteria is:

• Has to work well
• As low a part count as possible
• Must fit in a 1590B
• Should try to stay away from specialty chips (if possible)

[EBK]

I'll point you here, but I have no firsthand experience with this circuit:
http://tagboardeffects.blogspot.com/2013/01/mxr-noise-gate.html?m=1

[blackieNYC]

The DOD fx-30B is simple, and it has a loop insert, which means you can stick the noisy pedals in the loop and thus turn on your dirt and gate with one stomp. And have no gate when you are in clean mode. Schematic can be googled. If you don't want the loop, there is the fx-30, which is slightly different.
And there's the Q&D compressor at the AMZ site.  Gate and comp chip.

[Mark Hammer]

"Simple" and "end of the chain" do not go together.  The further along the signal path a gate is inserted, the more sophisticated the gate needs to be, simply because noise becomes more complex as it is accumulated, amplified, etc., over pedals.

Units like the DOD one described (and a great many others) include a send/receive loop, as Alan/Blackie described. The underlying strategy is to detect the difference between music signal and mere background noise at the point where they are most easily differentiated, but apply the gating after/later in the chain.

If one is aiming for "simple", however, such units are generally best placed right at the start of the chain, where the threshold for turning on can be set low.  Once additional gain stages are inserted into the signal path, they amplify noise, such that being able to tell the difference between the gentle decay of a note, and amplified hum/hiss, is hard to do, and setting a pleasing threshold point is harder.

[jfrabat]

"Simple" and "end of the chain" do not go together.  The further along the signal path a gate is inserted, the more sophisticated the gate needs to be, simply because noise becomes more complex as it is accumulated, amplified, etc., over pedals.

Units like the DOD one described (and a great many others) include a send/receive loop, as Alan/Blackie described. The underlying strategy is to detect the difference between music signal and mere background noise at the point where they are most easily differentiated, but apply the gating after/later in the chain.

If one is aiming for "simple", however, such units are generally best placed right at the start of the chain, where the threshold for turning on can be set low.  Once additional gain stages are inserted into the signal path, they amplify noise, such that being able to tell the difference between the gentle decay of a note, and amplified hum/hiss, is hard to do, and setting a pleasing threshold point is harder.

You bring up a very good point...  So, would you guys recommend the gate at the beginning or end of the chain?  I ordered a PCB for a MXR Noise Gate clone, by the way; so, where should I place it?

[MetalGuy]

The NG should be placed after the noisiest source which could be for example a high gain pedal or similar. There's no point in placing the NG right after the guitar - this is completely pointless unless your main goal is to cut the hum from a noisy pickup.
DOD30 should work fine. I've built many times the EPFM one (modified to work with a J175) and it works also very well.

[Mark Hammer]

....and that's sort of the problem. WHERE one ought to insert a gate will depend on what's coming in from the guitar, and what's getting added along the way.  If you have a SC-equiped guitar and are using it in a place that contributes to hum then you'll want to cut that hum out before it gets amplified by anything.

Part of the dissatisfaction  many players have with gates is that they stick them after their noisiest pedals.  At a superficial level, that would seem to make sense, but the threshold has to be set so high that it chops off the decay.

[jfrabat]

....and that's sort of the problem. WHERE one ought to insert a gate will depend on what's coming in from the guitar, and what's getting added along the way.  If you have a SC-equiped guitar and are using it in a place that contributes to hum then you'll want to cut that hum out before it gets amplified by anything.

Part of the dissatisfaction  many players have with gates is that they stick them after their noisiest pedals.  At a superficial level, that would seem to make sense, but the threshold has to be set so high that it chops off the decay.

I do have single coil pickups in my gutiar (PRS SE)...  So I guess this will go at the beginning of the chain then, before the compressor.

[Ben Lyman]

I used to use a noise gate as my first pedal after the guitar. I had a Marshall stack and a TS9, everything cranked, that's all, that was my punk rock tone for every gig with that particular band. I got the idea to add the gate to the TS9 because I only experienced excess microphonic squealing when I turned my guitar (tele single coils) volume knob from off to on. It sounded fine as long as I was playing but when I stopped I had to turn the guitar volume knob off all the time or stomp the TS9 off. So by adding the noise gate first in line, it only cut my guitar signal if I wasn't playing but when I was playing I could have all the distortion I wanted and long sustaining notes that didn't get cut off. So, as it turns out, the gate as first pedal worked for me for about 1000 gigs.

[amptramp]

Before you commit yourself to a noise gate with its abrupt cutoff of a signal, consider the "Magic Monitor" circuitry in the upper right of the picture that RCA used as a precursor for Dolby.  It amplifies the signal, rectifies the audio to get a level and uses this level to bias a pentode reactance amplifier toward the "off" state:





If the signal level is high, the negative bias is large and the pentode gain is low.  If the signal level drops, the bias is reduced and the pentode gain gets higher.  The gain of the pentode stage is negative or in the opposite direction of the applied signal.  It acts as a reactance tube where the gain multiplies the 27 pF capacitance from plate to output plug.  This acts as a lowpass filter that reduces high frequency gain and the capacitance is multiplied by pentode stage gain.  Change the gain, change the rolloff.

This circuit gives a smooth reduction of noise (which is mainly high frequency) as the signal level drops.  It does not cut off suddenly like a noise gate.

[Mark Hammer]

....and that's sort of the problem. WHERE one ought to insert a gate will depend on what's coming in from the guitar, and what's getting added along the way.  If you have a SC-equiped guitar and are using it in a place that contributes to hum then you'll want to cut that hum out before it gets amplified by anything.

Part of the dissatisfaction  many players have with gates is that they stick them after their noisiest pedals.  At a superficial level, that would seem to make sense, but the threshold has to be set so high that it chops off the decay.

I do have single coil pickups in my gutiar (PRS SE)...  So I guess this will go at the beginning of the chain then, before the compressor.

It doesn't HAVE to.  I'm just saying that any noise inherent in what the guitar feeds the pedalboard stands a very good chance of being amplified by your pedals, and exaggerated enough that it can be next to impossible to eliminate by inserting a gate between pedalboard and amp.

Keep in mind that a compressor blindly interprets low-level hum between notes as a fading signal that desperately needs a shot of gain to bring it up to audible level.    There have been many threads here over the years from people who thought their DIY compressor wasn't working properly or that one they had bought was of poor quality, simply because it was amplifying a noisy feed from the guitar during the quiet parts.

Blackie/Alan mentioned the Q&D Compressor, which is made using an SSM2166 chip.  I made one and it is a delight to use.  The 2166 was intended as the heart of a mic control strip, so it incorporates downward expansion (similar to gating, but smarter).  The result is a dead quiet compressor when you stop picking.  For me, this illustrates the advantages of using gentle gating (whether via this means or some other) ahead of everything else.  Keep the little crap out and you won't have to contend with bigger crap downstream.

I've noted here on several occasions that for me the "ideal" noise-control system would involve a send/receive loop, but instead of simply gating the outgoing signal to the amp, such a system would remove the hum going into the pedal chain (since hum generally enters at the start), and remove the hiss coming out (since hiss is added as one goes along).  In theory, that ought to provide the least obtrusive noise-control by splitting the task into its two components, and applying them where they can do the most good.

[jfrabat]

Mark, so in that sense, then the MXR clone (the PCB I ordered) is not ideal because it is a simple gate and has no loop?  Or could I use it at the beginning of the chain just to cut the single coil noise?  How would it work with the compressor I already got?  That one is an Engineer's Thumb, by the way...

[EBK]

Mark, so in that sense, then the MXR clone (the PCB I ordered) is not ideal because it is a simple gate and has no loop? 

Perhaps, take a look at this:
http://www.diystompboxes.com/smfforum/index.php?topic=76903.msg629483#msg629483

[Mark Hammer]

Mark, so in that sense, then the MXR clone (the PCB I ordered) is not ideal because it is a simple gate and has no loop?  Or could I use it at the beginning of the chain just to cut the single coil noise?  How would it work with the compressor I already got?  That one is an Engineer's Thumb, by the way...

As someone who purchased one of those grey boxes in the late 1970's, and was pleased with it back then, I won't speak ill of it. 

The adaptation shown in the linked-to thread seems pretty sensible to me, and would likely be feasible with the board you purchased.  If the adaptation is too tricky, then I'd suggest sticking it ahead of the compressor and see if you find it works better for you there than further along in your pedal chain.

I don't want to convey that there is some sort of orthodoxy that MUST be adhered to.  It's simply a matter of identifying where in your signal/pedal-chain it is going to be easiest to differentiate the wanted from the unwanted, and where it will be easiest to "put the kybosh" on the unwanted.

[blackieNYC]

I'm just repeating Mark but I "got it" when I did a show with lots of active pedals: If one happens to only have one gate, and no loop insert, try it at the front. Then, pickup related noise can be muted when not playing and all you will hear is the hiss of gainy pedals. Like turning down the guitar's volume. IOW, you can put "silence" thru a high gain pedal or you can put a little buzz, RF, or beer-light noise thru a high gain pedal.  See if that's enough - don't pursue absolute quiet.
If that doesn't cut it and you still just have the one noise gate around, try the end of the chain. If you can find a spot there where a) the gained-up pickup noise does not open the gate, and b) you can get your notes to come out when you need to, you might like that just fine.
Not everyone will want two gates. You don't have a vast pedalboard, and I admire your restraint.  That's boardspace that could be filled by something much more fun.  Maybe you'll get in the habit of "waking up" the gate so the first note of your solo does not get upcut.
Matter of fact, if you add a second gate, make it something fun like a Slow Gear, or a triggered input option where a drum mic or keyboard output triggers your guitar.

[Mark Hammer]

Question to all:

Have any of you ever experimented with filtering the signal going to the rectifier circuit?  That is, the sensing part is looking for, say, ambient content below 200hz (hum and some multiples thereof) or above 2khz (accumulated and amplified hiss).

[blackieNYC]

Yes.  Sorta. I've been unscientifically looking at how subharmonic intermodulation can affect thresholds. Filter them out. And not all designs do

[robthequiet]

Question to all:

Have any of you ever experimented with filtering the signal going to the rectifier circuit?  That is, the sensing part is looking for, say, ambient content below 200hz (hum and some multiples thereof) or above 2khz (accumulated and amplified hiss).

Sort of what a de-esser does, in multiple bands?

[Mark Hammer]

A second cousin to de-essing.  De-essers pull the gain back for the passband that holds much of the sibilance of speech so that those Ss, Ts, Zs, etc don't blow your tweeters out.  So content higher than that (which would include hiss) is unaffected.  I'm thinking more in terms of there being any low-level content in the hum range that would need to be blanked out.

[PRR]

> a noise gate with its abrupt cutoff of a signal

There are many variations of "gate" and thanks for another one.

However an "abrupt cutoff" gate is, for most musical purposes, "wrong". The fade-out should be long. You don't want hiss when you are sitting silent, but hiss right after strong sound is not so offensive.

Yes, low-pass is another good technique, especially since the decay of many musical instruments is not too bright. You can let the fundamental ring out while shaving the highs and the hiss.