Forssell Opto Compressor

[TomWaterman]

Hi!

Boards from Gustav would be awesome but I think your right, it would be sensible if we wait until the prototypes are ironed out berfore doing a batch run.

Do you still require the larger resolution meter? I have some time now - but I was just following the LM3916 datasheets.

What is the maximum gain reduction of this box - around 40dB??

A larger meter may only give more readout of massive GR approaching 40dB but not extra accuracy in the top of the range if we follow the LM3916 datasheet.

Anyway great stuff guys - looking forward to your prototype results with this.

Is anyone adding a sidechain HP filter like the Drawmer "Big" switch....also is there any other mod that could be switched, like adjusting the opto or something to produce tonal variations? I know its supposedly transparent but a feature to change its sound subtly may be cool.

Sorry I don't really know enough to make any hardcore design suggestions.

Cheers Tom

 

[Gustav]

Well..Just let me know.

Gustav

 

[tommytones]

So does this mean I should stop what I started? I've been sitting back waiting for you guys to iron out some of the details that popped up. I've already got the layout(s) going.

 

[Gustav]

Im not making the layouts, so...

Gustav

 

[TomWaterman]

Hi Tommy!

Were you going to get the boards made as well??
I hope you don't stop what you've started - I really appreciate what you're doing, as do many others I'm sure!

Do you still want to look into the larger resolution meter? If so, I guess people would prefer the VU spec rectifier circuit as opposed to the PPM one? PRR had one somewhere with the values tweaked to provide a slihtly faster PPM response.

Cheers Tom

 

[TomWaterman]

Hi again,

These are the ideas from the LM3916 datasheet.....I have no idea if it will work so maybe somebody with a good brain could cast an eye over it.

It says the meter is 10V full scale - that may be completely wrong here.....

The VU rectification circuit apparently complies with the DIN spec in terms of rise time and decay....these can be tweaked I guess. There is a PPM version in the notes and PRR had one with tweaked values (see below).

Here is the VU rectifier, it should use LF353 dual opamps but I think you can use TL072s here too.

For a gain of 1:

R5 = 100k
R6 = 43k
C2 = 2.0F
C3 = 0.56uF

http://myweb.tiscali.co.uk/ivyterrace/VU_Rectifier_Schematic.bmp

This 15 segment VU meter uses the LM3916 and LM3915 cascaded and is in dot mode......I figured this may be nice for GR ala CraneSong Flamingo. Im not sure if you can get the 15 segment to work in bargraph mode...if you can I think its just a case of tying the signal in to the mode pin.

However there is a bargraph version with 17 segments in the notes which would be better suited to input and output levels.

http://myweb.tiscali.co.uk/ivyterrace/15-Segment_VU%20_Meter.bmp

Here is PRRs PPM version from another post....

We need a rectifier and filter to smooth out the audio so the slow eye can make sense of it. There are a zillion ways to do this, but I am liking this idea stolen directly from National Semiconductor's LM3916 data sheet:

http://music-club.rutgers.edu/headfonz/3-LED-c.gif

Originally this was unity gain and DIN PPM balistics. The red scribbles are my suggestion for a gain of 5 ........and a somewhat faster (more VU-like) decay for more lively action. You can fiddle this to your taste, or use another rectifier and filter.

Cheers Tom

 

[tommytones]

That was kinda the plan (laying out and making the boards). I've been reading some of the problems and just waiting until butta got everything together to make sure it's all gonna work out before going too far with the pcb.

 

[Gustav]

Why dont you just finish the layout, get an idea of a price, and ill see if I can do better - Or are you not gonna make the layouts unless you can have the boards fabbed as well?

Gustav

 

[tommytones]

Sure, we can do that...no prob.

 

[TomWaterman]

What do you guys think to that meter circuit???

Is it totally wrong for this application?

Cheers Tom

 

[hitchhiker]

hi guys,

I got the correct opto's today and just tried them in tommytone's latest posting of the schematic using buttachunk's make up stage.

one change,
R7 I have 100 ohms instead of 1K. I thought I saw Fred say R7 would be
between 100 and 200 ohms. That's what I had in so I left it .

The side chain circuit seems to work as Fred says it would in all regards.
I think this compressor will be especially good on finer things,bringing the good stuff out without changing the sound too much.

Maybe the make up stage has considerably more gain available than needed.
I barely turn it to get level back after compression is set to full on.
I'd say a half of the gain available would more than enough.

Buttachunk, How are you making out with your proto?

Lance

 

[Svart]

sweet! i can't wait!

 

[Svart]

i guess the question still stands, what Op-Amps are we using for each board. i remember tommytones mentioning pads for both discrete and IC op-amps, right? wouldn't that make most of the difference between colored and transparent or are we talking something else here?

edit: I'd still like the boards to be as small and compact as possible in case i just happen to "need" to stick them in my console's channel strips... :green:

 

[hitchhiker]

hey Buttachunk,

Looking forward to your results.
Do we get to see your discrete circuit so we know what your testing?

I'm using a tl072 and for a chip I'm happy with it. I didn't like the 604,
too much in low mids and anemic in high mids.YMMV

I'd love to see a discrete with color as an alternative to Freds 992/3.
But I'll like it best if it's got mid range personality .

cheers,
Lance

 

[hitchhiker]

two beers are better than one :guinness: :sam: :grin:

cheers,
Lance

 

[Gustav]

[quote author="buttachunk"]i guess my question is; do people want this version to be colored or transparent ? :green:[/quote]

I know nothing about design, but since you ask like that id say colored. I could use a good colored unit.

Gustav

 

[Svart]

we can color it any way we want.. just pick the right cap on the can! :green:

 

[tommytones]

I'd vote for use of the good ol' 5534 in butta's circuit. This footprint can allow the user to drop in a few other brand/model opamps without changes.

As for color, transformer in and out should do something I would think. Fred and butta's circuits will surely sound different anyhow, both offering thier own color even if only slight. I don't think there's a need to go out of the way to create color.

 

[hitchhiker]

hi Tom ,

5534 was my choice at first and then I remembered that Fred said the
high impedence of a fet input opamp was needed to not interact with the
sidechain. tl072 is fet input and isn't too much of a change in sound.

cheers,
Lance

 

[Svart]

one of the BB op-amps then? i think a few of them are fet input too...?