would someone kindly explain how the hollis compressor actually compresses?

Started by ulysses, June 01, 2007, 09:56:05 AM

Previous topic - Next topic

ulysses

hey guys

ive been looking over the john hollis flatline compressor over the last couple of days. im of the opinion that it doesnt actually compress heavy signals down, only raises low signals up. so in essence it is a sustainer only, and not a compressor.

the schem is here-
http://www.hollis.co.uk/john/flatline.jpg

the input signal goes into the first opamp, the signal is then boosted in a feedback loop then directly outputted. the feedback gain on the first opamp is based on the resistance of that feedback loop. the resistance of the feedback loop is based on the LDR resistance. the LDR resistance is based on the overall volume of the output of the first opamp via the lit LED lowering the LDR resistance.

so.. as you play your guitar louder the feedback of the first opamp is lowered to compensate for the fact that you are playing louder. which would give you percieved compression but really its just the low volume playing that is raised up - up to the "flatline" output volume.

so my question is - does it really compress the loud guitar signal? or simply not amplify the input signal very much when you are playing loud?

cheers
ulysses

edit:typo

markm

This would be a great topic for Mr. Hammer as he modded this circuit into the PunchLine which seems to do what your does.
I have to say mine seems to act like a compressor should......or at least a stomper compressor.
It sounds very clean and has a response IMO between that of a Ross and OS but, recovers relatively quick.

Paul Perry (Frostwave)

Decreasing gain when the signal is louder, gives the same result as increasing the gain when the signal is softer.
Try it with a volume control!
True, you need to set the "midway" gain to match the two cases, but that's it.

Mark Hammer

Quote from: ulysses on June 01, 2007, 09:56:05 AM
so.. as you play your guitar louder the feedback of the first opamp is lowered to compensate for the fact that you are playing louder. which would give you percieved compression but really its just the low volume playing that is raised up - up to the "flatline" output volume.

so my question is - does it really compress the loud guitar signal? or simply not amplify the input signal very much when you are playing loud?
Well, my son, you have answered your question.  The missing piece, which I guess had never entered your consciousness (or perhaps never at the right moment so as to make things click into place) is that this is how ALL compressors work, every single one of them.  They drop the gain in response to louder signals and allow it to drift back up when all the kerfuffle has died down.  The differences between them are largely in the method used to reduce some default gain setting.

Well, I lied, actually.  Not EVERY single one of them.  For instance, the Orange Squeezer leaves the gain fixed but adjusts the level of the input signal fed to that gain stage.  Same effective outcome, though: turned down for the loud stuff, turned up for the soft stuff.

A particularly clever circuit that appeared in Electronic Musician back in the early days was Bill Berardi's "Fuzztain".

Here, the two LED halves of a pair of optoisolators were used as clipping diodes, and the LDR halves were situated upstream to provide gain adjustment in an earlier stage.  Of course, the harder you "pushed" to get distortion, the more gain reduction there would be at the input, and the less likely it was that the LEDs would be pushed into clipping, thus yielding a smooth consistent and less dynamic distortion.  Not what everybody wants, I'll grant you, but a clever idea with musical validity nonetheless.  At first blush it might seem that the gain levels are not all that high, especially where the LEDs are situated.  Keep in mind, though that gain is multiplicative.  With a gain of 4.7 in stage 1, and a potential gain of 61 in stage 2, you have x286 right there, which finds its way to the LEDs.

markm

I love Comps!
Thanks Mark for the explaination and example, as always, I learned something from you!  :icon_cool:

Meanderthal

 Hey, I really like that Fuzztain!  8) Thanks, and for the explanation too, that's a really cool way to get a synth like sustain out of a distortion... it would be great driving an overdrive- the way I imagine the note bloom as the signal dies down getting nastier... interesting! I'm thinking that something like this, with the proper balance of splatter and snarl would be an enormous amount of fun to play! Led clipping (to the point where they light up) does have an agressive bite to it...
I am not responsible for your imagination.

ulysses

Quote from: Paul Perry (Frostwave) on June 01, 2007, 10:38:15 AM
Decreasing gain when the signal is louder, gives the same result as increasing the gain when the signal is softer.
Try it with a volume control!
True, you need to set the "midway" gain to match the two cases, but that's it.

true true.. didnt think of that :)

cheers
ulysses

ulysses

Quote from: Mark Hammer on June 01, 2007, 10:56:53 AM
Well, my son, you have answered your question.  The missing piece, which I guess had never entered your consciousness (or perhaps never at the right moment so as to make things click into place) is that this is how ALL compressors work, every single one of them.  They drop the gain in response to louder signals and allow it to drift back up when all the kerfuffle has died down.

well there you go.

i'll be honest and say i thought they worked like this:
1. input signal gets passed to a boost stage with large headroom to cleanly boost the input signal.
2. some "magic mojo" is done to compress the high points down to the flatline after the boost
3. the "softer notes played" are left at the same level in the boost stage

this way any "sustaining" notes of low volume will not drop off when you hammer on the guitar

you learn something new every day.

thanks everyone :)

cheers
ulysses