570 Compander tip

[Mark Hammer]

I've been organizing all my printed documents this past week, and was working on one of the last two binders this morning before heading off to work.  In one of them I had a schematic for a Noise Gate from a very old issue of Professional Crafts, a Japanese audio/hobby mag.  The gate used a 570 compander.

As at least some of you know, there are individual rectifiers for each channel on the 570/571.  The time constant cap, which smoothes the rectified signal and also sets how fast the amplitude envelope voltage rises and falls, is tied between pins 1 (or 16) and ground.  It is typically the case that if you go to use one of these chips for noise reduction (one channel for compression, and the other for complementary expansion), the same cap value is used on both channels.  Again, note that it is selected as a compromise, since the 570 does not permit independent selection of attack/rise and decay/fall time; you pick a time constant for the goose and the gander gets the same.

What I was pleased to find in this Japanese article (and thank goodness the schematics are largely in English, because the accompanying article would not have helped at all! :lol: ) was the use of a simple parallel pot to adjust cap discharge time.

In other words, just like you've seen on so many devices (compressors, auto-wahs, to name but a few) using a standard FWR or HWR circuit with a cap to ground, if you stick a resistance to ground in parallel with that cap, you can make it bleed off faster, shortening the decay time.  In this particular case, it was being done on only the expandor half of a companding circuit so as to make the audio signal "shut down" sooner and produce a gating effect.  Craig Anderton said he used this trick on the Rocktave Divider, but I had never really understood it before now.  In the case of the Rocktave, pins 1 and 16 "share" a bleed cap to ground, but the expandor FWR circuit goes to that cap via a diode from pin 16.  If you have the schematic, you will also see a 47k resistor that goes to ground from pin 16.  A-HA!

What that also means is that if you find you don't like the gating of the Rocktave, whether because it is too abrupt or doesn't happen soon enough, you can change it by altering that 47k to ground.  It also means that if you have anything using a 570 (the old PAIA limiter comes to mind) where you might find it useful to vary the decay time by this means.  One possible application might be on some BBD-based effects where a compandor is used for the delay path only.  Here, a slightly faster decay on the expandor half would shut down the delay path so that any clock noise cioming on the delay path would be cut out when the S/N ratio starts to get too poor to tolerate.

Yup, that's Hammer for ya.  Always ahead of the curve by reading stuff from 25 years ago!  :roll:

[Bill Bergman]

That's some good stuff :wink:

[Gripp]

Thanks Mark!!

I'm just about to perform final tweaks to my Rocktave before it being boxed up in a hammond 1590 DD (lotsa room for switches). One of the things I've been wondering about is the dynamic section and this answers my questions beautifully.
Btw, being born in -73 and having picked up this hobby 3 years ago makes it perfect to start digesting knowledge from say around -80 for me now :wink:
Best!
/ Pelle

[george]

Ah Mark Hammer

So intelligent and so accessible - you are a rare gem.

you awesomely so rock dude!

[Peter Snowberg]

Very nice Mark! Thank you.

[davebungo]

Hi Mark,
I don't know if anyone picked up on this idea when I posted it but it may be of interest to you if you're working in the field of envelope followers:
http://www.diystompboxes.com/smfforum/index.php?topic=32395.0

Dave.

[Mark Hammer]

Hmmm, 66 views and not one of them was mine!  Sorry I missed that one.  Dead clever.

[gez]

I don't know if anyone picked up on this idea when I posted it

No I didn't, but thank you for drawing our attention to it now Dave!