Noise gate

[Mick Bailey]

I've seen the schematic kicking around for this noise gate for quite a few years and decided to see how it functions. Not very impressed initially, but with a few tweaks it's showing some promise. It needed a 4.7uf cap form Q3 gate to ground to give it a smooth release, otherwise it's very abrupt and kills the note decay. The cap coincides with my Tele's natural note decay and it's just right. It works much better with BC109C for Q1 and Q2. The threshold control doesn't do much - turned right down it introduces distortion, but slightly turned up and the circuit operates without distortion even on high signal levels. It needs a series-resistor to set the minimum undistorted threshold. Also, 1M is way too large and 100k is better.

It's super-quiet and transparent but there are problems;

1. To open the gate it needs a high signal level, but following this the guitar can be played normally - even the volume control can be rolled off and it will still play until you finish the last note and then it closes nicely. Then it needs the high initial signal again to operate.
2. The attack is very abrupt; so fast that it's clicky. It needs just a little easing-off to soften the initial response.

Overall this could be a nice circuit that could be built as an add-on to an existing pedal, or stand-alone. I have it breadboarded right now as a side-project

[Mark Hammer]

Yeah, the absence of a terminating resistor on the output, and the lack of any gain feeding the side chain suggested to me it was intended to be an add-on to an existing circuit.

But simple, though.

I've been waving the flag for dual-gate approaches to noise control for a number of years: one at the head of the signal path to clean up any hum coming from the guitar, and a second one at the tail, to clean up any accumulated hiss. High-gain stages/devices generate hiss, but can also amplify hum coming into them. Killing the hum at the outset gives you less to clean up at the end, such that thresholds for both noise-control devices can be set sensibly low and not interfere with hearing the instrument.

[cbarrow7625]

Do you think adding a transistor gain stage between the input & C2 would alleviate the issue of requiring a high signal level to open the gate? I imagine that may make the threshold control work better as well.

[J M Fahey]

Thatīs a very crude noise gate; Iīd keep searching for something better.

[Mick Bailey]

Do you think adding a transistor gain stage between the input & C2 would alleviate the issue of requiring a high signal level to open the gate? I imagine that may make the threshold control work better as well.

I was thinking along those lines for it to be useful as a standalone unit. To remove as much input loading as possible I went with an opamp, though a FET may do just as well. I've just removed R1, R2, Q1 and C3 and used a non-inverting opamp stage to feed the rectifier. C1 is now 1uf, and a 1M pot in the feedback loop sets the gain and therefore the threshold (or perhaps more accurately the sensitivity). The circuit now works much better than I thought it would. I've tried it with some really noisy, high-output, high-gain pedals and it's fairly transparent. It holds for just long enough so that it isn't chopping between phrases. I've experimented with different caps on the gate of Q3 and its fine as stock, though some variable control may be desirable. I'm going to see how it holds up with a Marshall stack tomorrow at gig volume with a few pedals in front.

The only thing is the parts count has gone up and for just a few more parts the MXR circuit may be a better bet, though I have no experience of that. I'll draw up the schematic of this circuit when I get chance.