SSM2166 compressor; AMZ awesome

[kvb]

Smooth, bright, long, steady, consistent, sustain, etc.

I am very much enjoying the AMZ SSM2166 compressor. I bought the DIP chip in 2010, or something - before Jack Orman sold the PCBs. I didn't feel like building on paper and cardboard, then I misplaced the chip, and finally I stopped building pedals for a good while.

When I found the PCB (I forgot I even had it), I also located the chip and, during a break from other things, was finally able to put the circuit together.

Knowing that there are other compressors on a chip available, I wonder if any others implement a noise-gate. The noise-gate (downward expander) on this one makes me think that this is the reason that Jack worked with this chip over others.

The first time I heard the noise level shoot up on a compressor I didn't know what to think. I inquired here as to whether or not my Orange Squeezer was working. I heard a sigh, audible for a few thousand miles, that came from somewhere in Canada. Then, Mark Hammer graciously explained what was going on in the circuit, and I then knew that the noise was a good sign.

The noise-gate on the SSM2166 is a great feature. One can even set the response to kill the sound only when the strings are muted. Or the response can be set so that the circuit brings the volume down right where there isn't much of a note left anyway.

*Before going much further, I would tell anyone interested to build the circuit just as Mr. Orman presents on his site.

However, I do have some thoughts about this circuit and will mention some changes I made.

1. Staying on the subject of the noise gate function, I noticed that there was a "psshht" as the gain shot up at the end of a note and the gate shut. No matter what, the sound and speed of the gate shutting remains the same. I wondered how to speed it up.

It turns out that the AVG cap (for the RMS level detection / compressor speed) also determines how fast the noise-gate shuts. A smaller cap here speeds up the reaction of the gate. This might not matter depending on where the other settings are placed by a user, but I like the gate to close quickly. I have a 3.3μ cap here. The data sheet say 2.2μ is low.  Schematic has 4μ7; Jack says 10μ for a bass/guitar compromise. I also put three caps on a switch to experiment, and I just left these inside the enclosure instead of drilling and mounting the switch.

2. Right now my gain pot is 5K. This is larger than necessary. What I did was to turn the compression and rotation to a minimum, and the gain to max. This turns the compressor into a distortion pedal (not the point). I observed that the distortion goes away as the compression and rotation knobs are turned and the signal gets squeezed. I also see in the data sheet that the VCA input noise is designed to be lowest when the gain is highest. So, maybe a 5K pot is not a terrible thing.

3. I noticed that the circuit at idle, with the gate shut, still produces noise. I suppose this is only a problem with a gain producing pedal afterward. I also wondered about the diagram that shows a compressors ability to bring a signal above a noise floor. Then I tried cranking the gain, as mentioned above, and then added an output volume pot (30k). That way I can turn down the noise of idle. I can then use the gain and volume pots to get the circuit to a good level. With the volume pot turned down a bit, the hiss at idle is much reduced with an overdrive engaged. There is a bit of treble lost, but because this circuit has an early treble boost (c13) the volume pot doesn't kill too much. It is also why I used a fairly low value pot.

I am finding that I can crank the gain, bring the rotation and compression down, and get some seriously long notes. Bright and clear. With regular playing it just sounds very, very good.

** Thanks Smallbear for the chip. Thank you much Mr. Orman, et.al. for all of the info and the PCB.

[MaxPower]

I had planned to build a compressor based on those sm chips (2166/2167?) but never got around to it. Still on the back burner.

[amz-fx]

Thanks for your feedback and build report, Kevin!

I am glad that you are enjoying the pedal.

Anyone who wants to see the schematic and other notes can find them at:

http://www.muzique.com/tech/2166.htm

Best regards, Jack

[rankot]

I've build this too, will report soon!

[Mark Hammer]

The averaging cap on pin 8 can be changed to different values to achieve different response times.  Unfortunately, we are generally confined to the time constants defined by the standard cap values available: 2u2, 3u3, 4u7 6u8, etc.  As near as I can tell (and if I'm wrong please do correct me), the rectification inside the SSM2166 is not all that different from what we normally see in most autowahs and compressors.  In which case, we have the option to parallel the averaging cap with a variable resistance to ground to bleed off the charge stored in the cap at a rate we prefer.

[PRR]

> As near as I can tell (and if I'm wrong please do correct me), the rectification inside the SSM2166 is not all that different from what we normally see in most autowahs and compressors.  In which case, we have the option to parallel the averaging cap with a variable resistance to ground to bleed off the charge stored in the cap at a rate we prefer.

I'm not sure this is correct. It calls this an "RMS detector". The similar thing from dBx/THAT is NOT like our everyday peak catchers.

However 12 cents will tell you if it does anything useful.

[blackieNYC]

I gotta say, that "psshht" sound drove me up a wall and I put it aside. Just as described above.  I'll be watching to see if the cap change helps.

[Mark Hammer]

> As near as I can tell (and if I'm wrong please do correct me), the rectification inside the SSM2166 is not all that different from what we normally see in most autowahs and compressors.  In which case, we have the option to parallel the averaging cap with a variable resistance to ground to bleed off the charge stored in the cap at a rate we prefer.

I'm not sure this is correct. It calls this an "RMS detector". The similar thing from dBx/THAT is NOT like our everyday peak catchers.

However 12 cents will tell you if it does anything useful.

Hence the tentativeness of my post.
I had a full-featured Q&D, but sold it to a guy a while back.  Still have a pair of 2166 DIP chips left, so I may build another and put the hypothesis to the test.  Just not today.

[kvb]

"psshht" sound drove me up a wall

The sound was bothering me also. However, it was while the circuit was not yet in an enclosure. Also, it turned out that my guitar was having a grounding issue - so as the gain came up, nasty came with it.

After completing the pedal I tried a different guitar and noticed less noise. I did some testing with an external attenuator and decided the volume pot wouldn't be a bad idea.

*** I forgot to mention above, that I also put a high quality 10pf cap to ground at the output - in place of the pull-down resistor, on the PCB.
The volume pot can be used to take out a bit of treble, and the cap to ground should take care of any HF noise coming from the chip.

I tested the noise level against my Orange Squeezer. Very similar noise levels; both with maxed OD250 - OS blasting hiss, 2166 at least stops. And I just noticed that the 2166 stays more consistent through to the end of the note, and at these settings was blooming into some nice feedback tones.

So as far as the pssht goes, It is noticeable partly because the compressor sounds so good right up until the noise becomes more present.

Try settings with the gate set loosely, where it might only set with the strings muted. I suppose the 2166 sounds just like another compressor, except that one can get the noise to stop by muting the strings.

And don't forget about C13 on the present schematic. This compressor is basically emphasizing treble - which is partly why it sounds so nice.
If C13 was socketed (I did not socket mine), then one could remove C13 to see what it contributes to the final output of noise. Or, bypass the external buffer to see how the final output sounds with the internal buffer alone.

Every pedal has its quirks, and as far as DIY and all that goes, this compressor is just what I needed.