RMS Detector VS Full Wave rectifier for compressor sidechain?

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

elskardio

Well-known member
Joined
Sep 29, 2010
Messages
564
Location
Montreal - Canada
Hi Guys!

I'm working on a compressor based on various THAT design notes using their RMS Detector. What I have so far sounds pretty good... but even with a fast Time Constant it doesn't react aggressively enough!

Could I use a full wave rectifier instead of the RMS Detector in the sidechain?

Here's the circuit I had in mind...

Thanks for your help!
Gabriel

7azn.jpg
 
I do not know how fast that posted circuit will be, with the .47uF integrator cap across the opamp.  Generally fast attack side chains combine fast attack with slow release.

======

Did you experiment with making the capacitor smaller in the THAT RMS circuit?

You should be able to make it arbitrarily fast, while release will be faster too. 

I posted a possible approach to make that RMS cap variable** with an opamp wrapped around it, in a discussion on another design forum.

JR

** Small cap with an opamp configured as a capacitance multiplier, with a pot to vary how much it gets multiplied. Note: I do not know if anybody actually took my suggestion and tried this.
 
The THAT RMS detector IS full-wave. (A half-wave RMS detector would be very odd.)

The diagram you show is clearly an Averaging detector. Slow.

And finally: the THAT VCA and THAT detector have complimentary LOG conversion factors. They use this fact to implement FeedForward gain control. Simple peak detectors are Linear. The output is much larger than the THAT VCA can stand. You can pad this, but in Feedforward mode it will do The Wrong Thing. Linear peak detectors "usually" are used in FeedBack topology.
 
elskardio said:
What I have so far sounds pretty good... but even with a fast Time Constant it doesn't react aggressively enough!
Waddya mean 'doesn't react aggressively enough'?  Do you mean it doesn't compress enough?  Or that it doesn't give that raunchy distorted noise like Radio Gaga with every cycle peaking at 0dBfs?
This is nothing to do with Time Constants.  Twiddle the compression ratios instead.
 
While I don't want to go too far down the rabbit hole of discussing subjective aspects of dynamics controls, a fast release rate may deliver more of a squashed sound than speeding up the attack.  A slow release will sound more dynamic than a fast release.

Of course with effects YMMV.

JR
 

Latest posts

Back
Top