Hi Mayo!

I had this confusion too, but it works out quite straightforward.

The part of your compressor that actually does the compressing (the "Gain Block" which can be a VCA or Opto element etc.) needs to know when to increase or decrease the audio level. This is controlled by the detector stage which allows you to set the threshold, attack, release etc. This in turn has to "hear" your incoming audio signal so that it can respond to the changes in level.

The audio input to your compressor splits into two- one goes to the gain block to be level-controlled and then out again to the outside world, and the other is the sidechain which the detector uses to "listen" to the input signal.

Normally in a compressor you want it to "listen" to the audio signal coming in exactly as it is so that you can compress over the entire audio range equally (or equally-ish, all compressors impart their own characteristic!). The sidechain jacks on the back of your unit allow you to break into this signal and change it using an EQ unit.

If you connect the "sidechain out/send" to the input of your EQ and the "sidechain in/return" to the output of your EQ, you can start having fun with "frequency dependent compression"!

This means that you can change the frequency band of your signal that the compressor is dealing with. So for example, you now have a de-esser that you can use for vocals by making the compressor respond only to the top end of the audio spectrum. You usually have access to the sidechain audio so that you can "tune" it to the frequency you want to compress. There's usually a "key listen" switch on the front- punch this and you'll hear what your compressors hearing!

Hope this helps you out, I found it really helped patching in a graphic EQ and cutting out different bands whilst listening to eg a drum track. (the sidechain signal should remain at the 0dB level to keep sense of your threshold setting- so do EQ cut rather than boost- but hey, anything goes!) I'm sure others out there will have more to say on the subject, but here's a start!

Have fun with audio!!

Mark

Thanks for the response Mark Burnley. I will experiment on my next mixdown. I do need a de-esser and with this new knowlege i will have alot of fun. I
normally have to use plugins to take care of that situation but any chance to utilize my analog gear is always welcome. I might as well try it with a 1/3 octive eq. Any suggestions? :w:

Wait!!! That's not all. You can also use the side chain to duck with. This is done by running one element through the compressor and taking a "mult" from a second element and running it into the side chain. The second element will now drive the compression detector. Example .... A guitar is run through the compressor. A vocal is "multed" and run into the side chain. Every time the vocal comes up the guitar will be turned down by the amount of gain reduction you set with the threshold. Like an automatic gain ride. Good answer Mark. ....... Fats
------------------------------------------------------------------------
Tannoy, Dynaudio, Blue Sky, JBL, Earthworks, Westlake, NS 10's :D , Genelec, Hafler, KRK, and PMC
Those are good. …………………….. Pick one.
------------------------------------------------------------------------

Hi again Mayo!

Glad the sidechain info was useful, have you tried Fats' idea of ducking? Works a treat.

Here's some more ideas...

1. Compressors are great for controlling the dynamics of a signal, but an analog comp can only work in real time, it can't "guess" whats heading towards it. However, if you've got a good sounding digi delay you can try this.

Take the signal you want to process from the board and split it 2 ways. Send one feed to a digi delay. Patch this delayed sig to the main compressor input. Patch the other feed from the board(non-delayed) to the "sidechain in/return" jack.

Now, what happens is the comp detector "hears" the signal before it arrives at the VCA. So you have a kind of "negative attack". Say you've got a sound with a sharp attack, and you want to smooth it out. What happens is the compressor responds to the signal in the sidechain first, before the actual signal it's going to compress arrives! So you can see the gain red. meter reducing the gain just before the main (delayed) signal arrives. So the initial attack is shaved off without clicks etc. Obviously, the delay can't be too long otherwise you end up with sync problems, but just a few mS or so is enough for the comp. to kick in and start doing its thing. You can get a kind of "swell" sound with guitar, bass or synth. Good for pads too. Play with the thresh, release and delay time to get different effects. Quite a high ratio works well, and time the release to suit.

2. (mad one this...) Connect a radio tuner to the "sidechain in/return" jack. Set it to a talk show or something with a steadily changing audio level. Whatever signal is passing through the comp will respond to the changing levels of the radio show. Sounds pointless, but try putting a flat drum machine hihat pattern thru, and if you set the comp subtly, it can impart a bit of human feel! No really, it does work!!

3. Most gate/comp/limiter units use the same sidechain/detector to trigger the gate part of the channel. If you haven't got a midi gate and need to "chop" up a sound, program a snare pattern or any sound with the kind of dynamic you want, send this audio sig. to the sidechain return, and pass the signal you want to "slice" thru the comp channel. Switch off the comp (max. thresh) and adjust the gate threshold instead. You may have to adjust the gate ratio and/or attack rel. speed to get it right. Cool with distorted guitar..."heavy trem"!!

:D While you're round the back of your rack, check to see if you've got a sidechain on any old analog or digi delays. Some units have one in the feedback loop (regen). If you patch an EQ in, you can get crazy resonator-type effects- good for percussion etc....

Just thought I'd add a bit more, you may just be left with time to record music after battling with patch leads...

More fun with audio!!

Mark

:) I love these threads, I thought I would throw this in, a different kind of sidechain. Enjoy!

http://homepages.paradise.net.nz/~rhulse/Side%20Chain/sidechain.htm

--Rick

I really apreciate the replies Mark, Cedar Flat Fats & Rick. I apologize for the delay in getting back. These techniques are amazing I have a break from this show I am doing and just now getting to my email. So i stole a few more minutes to read post's from my favorite forum. Surprise a few more responses. Mark I have done some shows reciently and figured what the heck lets try them in a live situation. I am a producer that fell in to engineering so I never intended to get as deep as it gets. I had to use my clients equiptment and your points for all sorts of frequency
issues in thier sound web set up. Never used sound web before & figured might as well have fun. The guy had 2 clusters of 7 Meyer cabnits that we had to hang from the high steal beams of the hall we are in and 12 apogee cabinets for satilight speakers. I have pleanty of 1/3 oct eq's & comp/gates available so I just pached one 1/3 & eq to each sound web in put. I could have used the eq's on the sound web and did and saved my setting just in case but I wanted to have a littel fun. It worked well. I want to thankyou personally and will experiment with the methods that Cedar Flat Fats and Rick Hammang have posted. Thanks again guy's
mayo

I've never heard of using the side chain for frequency dependent compression before. Sounds like a cool idea. I have used it for ducking. I do some occasional radio voice over work and it is really nice for that. Whenever the announcer speaks, the music lowers. Once you get the attack, threshold, and decay right, it's super. If your DAW compressor plugins have a sidechain capability, you can used them for this also.

themidiroom