Easy three transistor noise gate (schematics)

[Carlos]

Hi everybody!

As soon as I've got some time to kill I'm going to put this simple noise gate after my distortion, but NOT in the bypass signal path.
I would replace the trimpot with a pot, the BC109s with 5088s and adjust C3 which seems to influence the attack/decay time.
BTW Eingang= input; Ausgang: output.
What do you think about it?
http://www.geocities.com/kobraelectronic/index/12/noisgate1.gif


Regards
Carlos

[Mark Hammer]

That's a pleasantly simply and compact circuit.  Hell, you could even build that into your otherwise irritating single-coil equipped guitar! 

As a standalone, it needs a couple of things, some of which you have correctly pointed out.  R3 would appear to set sensitivity or turn-on threshold.  C3 does set the decay time.  Like all half-wave rectified side-chains, it WILL be susceptible to audible envelope ripple unless you increase the decay time and value of the smoothing cap (C3).

T3 and R6 form a voltage divider, very similar to the role played by the 82k input resistor in the Orange Squeezer and the 2N5457 tied to it.  To be a gate, normally the resistance of T3 would be low such that R6 and T3 act like a volume pot turned way down.  When the input signal exceeds some minimal amplitude, T3 is then turned off, bringing the resistance high, and mimicking a very high-resistance pot turned almost all the way up.

R5 sets the "minimum volume", such that if the drain-source resistance of T3 were near zero, the "pot" would consist of R6 and R5.  So a 2.42k pot turned down to 10%.

What this means is that, as with the Orange Squeezer, the degree of attenuation can be adjusted by varying the input resistance (R6) and or the resistance to ground (R5).  Once you play around with it, R6 can be a fixed value, but R5 will benefit by being replaced by a 100R fixed resistor in series with a 5k pot.  This will allow for a broad range of attenuation, from decidedly silent to light reduction in level.  As with any noise gate having less attenuation tends to minimize the annoyingness of the envelope ripple.

Finally, as a standalone unit, both input and output culd benefit from some terminating resistors, and this unit could probably benefit from using that nice FET input buffer stage that Jack Orman tacked onto the Dr. Quack.

[Carlos]

Thanks Mark!

I think I'll try this simple circuit as I have got some benchtime.
As for the ripple I don't mind on my OS, if there is any audible ripple at all. At least for my 40-something-year-old ears 
If I remember correctly, there was another very basic noise gate in the BBD signal chain in the CE-1 Chorus Ensemble I used to have.

Carlos

[Mark Hammer]

You are correct about the CE-1 gate.  Keep in mind that the gate on the CE-1 simply cuts out the effect not the primary signal, so any effect of ripple becomes minimal (especially given that the effect is modulated).  Second, envelope ripple on gates is almost always more noticeable than or compressors, so don't go by your experiences with the OS.  Worst case scenario on the OS is that it sounds a little "Scottish" (slight burr on the decay).  On a gate it starts to sound like an intermittent stompswitch.   More importantly, the decay is part of what determines the "naturalness" of a gate with respect to how it impacts on your playing/phrasing.  Fast decay = no sustain or finger vibrato allowed.

Actually, it would be kind of nice to have some means of adjusing the decay in real time so as to facilitate an optimum balance between noise rejection (which a fast decay accomplishes) and picking-style flexibility (which a longer decay facilitates).

[Carlos]

Mark,

I think I'll modify this noise gate in the way you explained so that it reduces the level significantly, but not completely. There is a Behringer Denoiser (great, downward Expander + VCF low pass) in the loop of the amp anyway.
The Denoiser is carefully set for the different channels of the amp, but when I use this particular boost/distortion in front of the overdriven amp, the noise floor is above the threshold of the Behringer, so it doesn't cut the noise.
This three transistor noise gate may reduce the noise level during pauses so that the Behringer denoiser can work.
That's what our keyboarder does, too: using a noise gate for a single vintage Roland Vocoder and its submixer channel AND a Denoiser for his stereo mix that goes the FOH mixing desk.

BTW There are power lines 12 metres up in the air above our rehearsal room, so I put Bill Lawrence PUs in my Tele and a Denoiser in the loop

Thanks

Carlos

[free electron]

There is one cool trick to improve the working of the noise gate. It can be done e.g. with popular Boss NS2, almost every noise gate can be modded to work in that mode. The idea behind this trick is to feed the envelope follower and threshold detector, known also as side chain with the bear signal, just from the guitar. Then place  the variable gain part as the last one, before the delay/reverb so it won't cut out the fading repeats:

The changes of the noise floor in the input signal are almost no significant when compared to output of the overdrive channel or pedal. So the threshold can be set once and work for all sounds, from clean to heavy overdrived. I've convinded some of my friends musicians to use that methode with their NS2's. It worked great. I've also used that idea in one of my pedals equipped with noise gate.
Measure the signal at the input and regulate the gain at the output.

[MetalGuy]

Here's another one taken from a german company pedals:



You can use the other section of the opamp in place of the transistor.

[Mark Hammer]

Apparently the DOD FX30 includes a loop insert so that the user can do what free electron describes.  Still useful to keep an eye on the decay, though, just the same.

[Carlos]

Thanks Mark, free electron and metal guy.
I think free electron's idea is great.
I wonder what the input impedance of the side chain is, if the guitar is connected to the side chain without any buffer in front of it.

Carlos

[col]

I built this exactly as drawn over the weekend before any of the other replies had been posted. It works OK but makes a strange ticking noise just before it cuts off. Also, if the battery isn't full you have to hit the strings hard to get it to come on. I will make some of the changes suggested and see if it improves but it won't be for a few days,

Col

[Carlos]

Great Col!

Keep us posted. Have you seen the other noise gate I posted:

I'd love to smell some solder fume now instead of meetings all day, but at least I can check this forum once or twice a day.

Carlos