AD SSM2166 as a simple noise gate?

[Bill Mountain]

I am looking at including a noise gate as part of a new project.  I want it to work well on both clean and dirty signals so I fear it'll need to be one of the more complex designs.  I'm considering it as a way to control unwanted noise in CMOS gain stages when I'm not playing.

Looking at the SSM2166:
http://www.analog.com/en/products/audio-video/audio-amplifiers/micpre-amp-audio-products/ssm2166.html

Pin 9 allows you to adjust the noise gate in the chip.  Can I simply use one of these chips and set it up as only a noise gate?  Or that a misuse of a more powerful IC?  Is it even good noise gate for our purposes?

Thanks!

[R.G.]

The 2166 is a handy little chip. It's a bit expensive, but it does a lot in one package.

I would not even blink at using a bigger chip for one of its functions. Mis-use is what we do here.   

Downward expanding noise gates (this is one of those) perform very well, as there's no sudden cutoff.

You'll need to read the app literature for the 2166 to get it set to a "compression" ratio of 1:1 between noise gating and limiting. I think this is possible, although I haven't read the 2166 spec for a couple of years.

[Bill Mountain]

The 2166 is a handy little chip. It's a bit expensive, but it does a lot in one package.

I would not even blink at using a bigger chip for one of its functions. Mis-use is what we do here.   

Downward expanding noise gates (this is one of those) perform very well, as there's no sudden cutoff.

You'll need to read the app literature for the 2166 to get it set to a "compression" ratio of 1:1 between noise gating and limiting. I think this is possible, although I haven't read the 2166 spec for a couple of years.

You didn't immediately tell me I'm heading down the wrong path so that's progress!

I read through the datasheet once just to get the gist of it.  Now I'll have to do it again to work up a schematic.

Thanks!

[blackieNYC]

See the quick and dirty compressor at AMZ

[R.G.]

See the quick and dirty compressor at AMZ

IIRC, that's the SSM2165, a brother-cousin to the 2166. Could be similar enough to help with the 2166, though.

[Bill Mountain]

Looking at it now it seems there a Quick and Dirty 2 that uses the 2166.  And he sells boards which is great.

Thanks everyone!

[amz-fx]

Bill,

To use the 2166 as a gate, you want to set the rotation point as high as possible (1k resistor) so it doesn't go into limiting on small signals. You also want to set the compression at 1:1 (100 ohms) as previously mentioned, and you want the input to have gain = 1.  A gain adjust resistor of about 1k will give unity gain on the internal vca. The noise gate set resistor can be a 500k pot so you can adjust the threshold of the gate.

Small Bear Electronics sells the SSM2166 smt chip on a thru-hole adapter if you want to get one to breadboard.

Best regards, Jack

[anotherjim]

CMOS stages you say? Inverters?
There is a trick.
Parallel the input resistor with back to back diodes in series with another resistor.
Both resistors are double the value of the original Rin. This will reduce gain when signal is below the diode thresholds. It's a simple noise reduction.
You can adjust via those resistor values. No signal & the gain is set by the direct resistor path. Signal big enough to switch the diodes and the gain is set by the diode resistor in parallel with the direct resistor.

[Bill Mountain]

Interesting.  So when the voltage gets high enough the diodes become an easier path than Rin?

This will lead to some crossover distortion which may not be an issue in the "dirt stages".

Thanks for the tip!

The 2166 is also appealing because of the on board opamps.  I'm looking to see if I can easily replace other stages in my existing circuit such as a mixer stage on the output.

[anotherjim]

Xover isn't that audible, it depends how severe the resistor difference are.
Inverter noise is a funny thing - at the bias point it has the highest gain, near clipping it compresses down the harmonics as it effectively looses a lot of gain, but then amplifies them more around the zero crossing. It sounds like noise, but I don't think it's just noise. Gating the gain about the zero crossing seems to smooth that out.