Optical compressor project

[CSlee]

As part of my research I am looking into optical compressors and have drawn up a schematic to start playing around with in the lab.

At the moment I guess it is just a limiter, but Id love some input on whether this will work before I progress the circuit.

So I guess the questions are, will this work? Am I using the right Vactrol for a compressor? Also, is there any need for a grid leak on the first stage? Any problems with leaving the stages open loop?

As always, all feeback is really appreciated.

Cheers

Charlie

[Frank Berry]

The diode and LED will produce distortion on the output due to loading.
You may want to use a third stage to drive the LED and isolate it from the CF output stage. You may also want to drive the LED through a bridge rectifier so that it will respond to both halves of the audio waveform.
I've made such a circuit using ICs. It worked very well.

[bigwill]

Seconding Frank's suggestion - even an opamp would do the trick, in fact it would probably be preferable since you wouldn't have to devote an entire triode section to something that isn't in the audio path

[CSlee]

Thanks for the comments.

Just a single transistor or opamp stage followed by another coupling cap?

Cheers

[Frank Berry]

If you use an Opamp and power it with a bipolar supply, you can eliminate the cap on the output of the Opamp.

[richwalters]

About 25 yrs ago I worked in the audio industry as a designer and photoswitches as analogous devices were quite rife within mixing desks in "non click" cross channel switching applications.
Not sure links work but I know the Vactrol still exists.

Vactrol Analog Opto-isolators

P873 datasheet and application note, data sheet, circuit, pdf, cross reference, pinout | Datasheet Archive

Remember, one will most likely need 2; i.e a series-shunt arrangement as the off capacitance in the series may create bleed through with high impedance circuits.
A Jfet solution: Certainly works but I never got the thd on transition down to an acceptable level which needed a synchronised gate timing circuit to eliminate the worst on the "pinch-off" which can vary by 20%.
Even earlier for organ work, an ORP12 in a tin can excited by a cycle bulb was used.

richy

[ruffrecords]

Originally Posted by richwalters

About 25 yrs ago I worked in the audio industry as a designer and photoswitches as analogous devices were quite rife within mixing desks in "non click" cross channel switching applications.
Not sure links work but I know the Vactrol still exists.
richy

They are available from Farnell.

Cheers

Ian

[ruffrecords]

Originally Posted by CSlee

As always, all feeback is really appreciated.

Cheers

Charlie

I would decouple the cathode of the first stage to maximise the open loop gain.

Cheers

Ian

[CSlee]

Thanks for the comments.

So a few changes...

Decoupled first stage with a 47uF cap. And a Jfet stage before the diode, just using similar resistor values on the E, C and B? Dont want any gain there do I?

Thanks Richy, if this setup works it will allow me to experiment with different devices.

In my eyes the only way this would be able to be a compressor is if the LED was variable, which it isnt. I did see an old compressor with a bulb in like Richy said, looked pretty cool but maybe a little overkill for these times....

Cheers

[ruffrecords]

Originally Posted by CSlee

In my eyes the only way this would be able to be a compressor is if the LED was variable, which it isn't.

Are you sure about that? Check the data sheets, then look at the VTL5C3 and think about what happens if you change the first 220K resistor to 30K.

Cheers

Ian