Anderton Compressor

[vanessa]

Anyone have the schematic? All the old links seem to be dead...

[LP Hovercraft]

I have an old EPFM book.  I'll scan you a copy tomorrow.  It is a soft knee optical compressor that Joe Meek may have even liked to use. 

[Mark Hammer]

Here is what Stellan used to have up at his site: http://img.photobucket.com/albums/v474/mhammer/ANDERTONCOMPR.gif
As far as I know, this is his drawing, so in deference to him, I'll take this down very shortly, and ask that people not serve it from any other site.

EDIT JUNE 1: The file has now been taken down.

Stellan's drawing shows a Vactrol where Craig originally used a Clairex CLM6000.  It also shows a TL072 where Craig used an LM/XR4739 and an LM301 where he used a 748.  Stellan's substitutions are fine.

One change I will suggest is to place a .022uf cap and 4k7 resistor in series with each other and in parallel with the 2.2uf/8k2 network.  This will provide some modest boost above 1.5khz to offset the generally dulling effect that compressors can have.

The Whisper compressor that Jack Orman and Thomas Henry co-designed (or was it one of those Lennon-McCartney things where one did it but both names are attached?) was based on the EPFM compressor with some additional well thought-out bells and whistles.  Given that Midwest Analog Products is more or less defunct, I'm wondering if Jack would mind posting the schematic.

[vanessa]

Once again, Thank you Mark!

Any vero layouts (or PCB)?

[MartyMart]

I built mine from the EPFM book with a few "tweaks" from Mark too.
Very nice, almost more of a "studio" compressor quality
I use an original CLM6000 from smallbear, RC4559P's and a 301

MM.

[Bernardduur]

Great pedal!

[vanessa]

Hey Marty! I read you built the Flatline too. How do you conpare the two? Is the Flatline still on your pedal board?

[MartyMart]

Hey Marty! I read you built the Flatline too. How do you conpare the two? Is the Flatline still on your pedal board?

Flatline is on the board, CA comp is one of "many" .......... as yet unboxed !!
I would also need to solve the PSU for it, at the 'mo it's 2x batteries 
( +/- 9v supply )
If that were solved it would be "comp No1" for sure, great quality and loads of range
but a little "lumpy" when squeezed too much ( ripple on decay )

MM

[Gyoon]

Marty, when you are talking about ripple are you referring to a distortion type sound?  I am using a B curve 100k pot for the compression knob and I notice that there is distortion, or something like it, when it is cranked about half way.  I think the schem called for an audio taper pot.  Just checking to see if my build is working correctly.  A great unit nonetheless.

Glenn

[Cliff Schecht]

Pot taper shouldn't matter, that just affects how the pot "feels" when you are turning it.

[MartyMart]

Marty, when you are talking about ripple are you referring to a distortion type sound?  I am using a B curve 100k pot for the compression knob and I notice that there is distortion, or something like it, when it is cranked about half way.  I think the schem called for an audio taper pot.  Just checking to see if my build is working correctly.  A great unit nonetheless.

Glenn

No, not distortion at all, it's what many compressors can suffer from, it's a little volume "pumping" in
the decay, sort of odd sinewave up/down of the signal , hence "ripple"

Hey, Vanessa did you ever build the CA Comp ??

Marty.

[Mark Hammer]

One of the things that the use of the optoisolator accomplishes in this unit is that it smooths out any remaining ripple in the envelope signal.  It uses the same simple half-wave rectification that the Orange Squeezer does, but where the FET in the Squeezer is able to change resistance value almost instantaneously in response to every little smidgen of ripple in the envelope signal fed to the gate, the LDR in the CLM6000 (or whatever you use) is "a little slow to get out of bed", and essentially ignores small fluctuations in envelope signal.  This leads to a much smoother trill-free decay.

Why decay?  Remember that while the initial transient peak is chock full of harmonic content, the lingering note afterwards is not.  If you "chop" the first 100msec of the picked note such that you only retain positive half cycles, many of the "dips" between the positive peaks of the fundamental will be all full of lower order harmonics.  That ripple is at a high enough frequency that it will not be noticeable, if audible at all.  As those harmonics die, there is nothing to fill the gap, and the contrast between the peaks of the positive half cycle and the dips will be big enough that you WILL hear those rapid fluctuations in volume IF the control element can respond to them quickly enough to represent them, and/or IF the averaging cap (the one to ground that normally helps to set attack and decay parameters) is small enough.  It also doesn't help that players generally introduce some finger vibrato later in the lifespan of the note, and that finger vibrato will often introduce "beats" that show up as lower frequency ripple.  The point is that in circuits that use simpler rectification (for cost savings and smaller footprint/board size), ripple is an issue if you aim for anything with the potential for slower release/recovery times.

If you set the averaging cap value quite high, it will "smudge" all the ripple and make the envelope signal seem smoother than it really is...but at a cost of much slower response time.  The saving grace of the Anderton/EPFM circuit is that it uses an LDR as the control element and the lag introduced by ths LDR allows the user to get away with a smaller averaging cap value for faster response time (if they want) since the LDR essentially finishes the job that the half-wave rectifier didn't quite complete.

Make sense?

And, as probably noted in other threads, one of the things I like about this circuit is that adding a bit of treble emphasis or adapting the unit to be a ducker is a piece of cake.

[Tubebass]

Is there a PCB layout for this, using "modern" opamps?

[oldschoolanalog]

There is a "modernized" PCB layout for this at GGG.