Doepfer

Building a Simple Mono Compressor with Doepfer A-100 Modules
==========================================================

Hi folks,

I was thinking about building a compressor with the A-100 modules. Because I don't have all necessary modules to realize it yet, maybe someone of you who have them, could try it and tell my how it sounds and of works anyway.
I'm desperate to know, how it sounds. Maybe some could send me an audio sample, not processed and processed.

Modules needed:
A-119 Ext.In/Envelope Follower
A-140 ADSR or Quad-ADSR
A-133 Dual VC Polarizer
A-132-3 Dual VCA or any other exp. VCA

You can find my patch suggestion here:
http://homepages.physik.uni-muenchen.de/~Oliver.Gretz/a-100/Doepfer-Simple-Mono-Comp.jpg

The Envelope Follower generates the gate signal to trigger the ADSR and an envelope which is used to control the compression ratio. Here, the threshold knob on Envelope Follower function just as well as the threshold knob on a compressor. You just have to start from value 10 instead from 0.
The ADSR triggered by the Envelope Follower generates the envelope, how you want your signal compressed. Usually attack-decay volume attenuation envelopes are sufficient, but of course you can make it more complex with sustain and release or with any other modules. 
Although the ADSR signal is positive, it will be inverted by a voltage controled polarizer. The ratio how strong the ADSR signal should be inverted is controlled by the inverted envelope, originates from the Envelope Follower, which represence the volume of the input signal. Therefore the higher the volume of the input signal is, the higher the volume of the original signal will be reduced by the VCA. And right so how you have set the threshold and attack and decay values.
Depending on how heavy you compress the audio signal, you could need a second VCA to get the overall volume louder. I don't know, try it out.

You could easily expand this patch suggestion to a stereo compressor with a second A-119 Envelope Follower and a second VCA. 
You could also make the compression smoother by low-pass filter the A-119 envelope signal or using a slew limiter.
If the A-119 module would have had a seperate envelope follower input jack, you could high pass or low pass filter the audio signal to easily realize a sidechain compressor (to get the pumping effect as well no pumping at all).

This really just a very simple implementation of a typical compression process. So don't be to strict to me. ;)
As I have already said I don't have the modules at home to try it for my self. There are good chances that there is something wrong. So don't hesitate to tell me the mistake.

Have a lot of fun,
Ollie

Hi Ollie

This may be an interesting effect, but it is not a compressor.


The attack and delaytimes are a direct "portamento" on the envelope 
voltage of the original signal. They are NOT attacktimes of a separate 
envelope circuit. You may find a suggestion here: 
http://fa.utfs.org/diy/a100/A100_compressor.jpg


The signals envelope voltage is inverted in the A175 (then fed through 
the VCS A159 which I will neglect at the moment(*)), then the A170 slew 
limiter adds the slew rate on the fast changing original envelope, which 
controls the VCA.


Florian

(*)The VCS decides, with the threshold of the gate out when the signals 
envelope voltage starts to become valid - also this is not what a real 
compressor is working like).

Hi Florian,

It is very interesting how much you learn about sound processing with a modular system. 
You are right, in real compressor you don't have an envelope generator. Your slew limiter solution comes a real compressor more closer. I used an ADSR because I would get very well defined volume attenuation characteristics.

Have you thought about how to realize the ratio/limiting behavior? How do you like the sound of your compressor?

Ollie


--- In Doepfer_a100@yahoogroups.com, Florian Anwander <fanwander@...> wrote:

>
> Hi Ollie
> 
> This may be an interesting effect, but it is not a compressor.
> 
> 
> The attack and delaytimes are a direct "portamento" on the envelope 
> voltage of the original signal. They are NOT attacktimes of a separate 
> envelope circuit. You may find a suggestion here: 
> http://fa.utfs.org/diy/a100/A100_compressor.jpg
> 
> 
> The signals envelope voltage is inverted in the A175 (then fed through 
> the VCS A159 which I will neglect at the moment(*)), then the A170 slew 
> limiter adds the slew rate on the fast changing original envelope, which 
> controls the VCA.
> 
> 
> Florian
> 
> (*)The VCS decides, with the threshold of the gate out when the signals 
> envelope voltage starts to become valid - also this is not what a real 
> compressor is working like).
>

Couldn't you also call this a compressor? I mean even though you use an ADSR the volume is also reduced above a specfic threshold; not in that manner your know from a usual compressor, but in an ADSR-wise manner.

Ollie 

--- In Doepfer_a100@yahoogroups.com, Florian Anwander <fanwander@...> wrote:

>
> Hi Ollie
> 
> This may be an interesting effect, but it is not a compressor.
> 
> 
> The attack and delaytimes are a direct "portamento" on the envelope 
> voltage of the original signal. They are NOT attacktimes of a separate 
> envelope circuit. You may find a suggestion here: 
> http://fa.utfs.org/diy/a100/A100_compressor.jpg
> 
> 
> The signals envelope voltage is inverted in the A175 (then fed through 
> the VCS A159 which I will neglect at the moment(*)), then the A170 slew 
> limiter adds the slew rate on the fast changing original envelope, which 
> controls the VCA.
> 
> 
> Florian
> 
> (*)The VCS decides, with the threshold of the gate out when the signals 
> envelope voltage starts to become valid - also this is not what a real 
> compressor is working like).
>

Hi Ollie

I will answer both postings in one mail.

> Have you thought about how to realize the ratio/limiting behavior? 
The ratio is simply the modulation amount at the VCA.

> How do you like the sound of your compressor?
I found, that it acts quite crude. It definitely requires a linear VCA 
instead of a exponential. Also: If you put a passive highpass in the 
analysis signal path, then the action becomes more musical.

I use it not often, but it is interesting if you split a signal with the 
A121 in different frequency bands, and the different bands then are 
handled by different compression ratios (VCA modulation amounts).


> Couldn't you also call this a compressor? 
> I mean even though you use an ADSR the volume is 
> also reduced above a specfic threshold; not in that 
> manner your know from a usual compressor, but in an ADSR-wise manner.
This is called a Ducker. Some commercial noise gates and also some 
compressors can act as duckers (LA Audio did some).

Florian