Forssell Opto Compressor

[hitchhiker]

I spent the day bread boarding Fred's design using Buttachunks gain make up (w no blocking caps). I think some folks are gonna like this
compressor.
I used OPA2604's and OPA604 in the make up amp.I also used the LM3916
LED circuit that Fred suggested but with seperate LED's. I found that I had to use a 2k7 current set r. I also had to leave out the whole meter trim section. Maybe I'm way off with the LED circuit because my circuit
turns the LED's out when it is hit with the peaks.

Anyway, I fed it a voice and guitar track,and a voice - guitar w drums track and I quite liked the effect. Smoother, bigger, slightly fatter and no loss of highs and natural tones.( I was using the unbal send /return on my Maxcon2 console )

thank you in advance Fred because I know I'll be building a number of these.

cheers,
Lance

 

[soundguy]

It's alive!!!

If there has ever been a candidate for using fred's opamps I think it would be this one folks!

dave

 

[Svart]

oh baby, somebody make some PCBs!!! quick!

 

[daArry]

great work fellaz :thumb:

 

[FredForssell]

I also used the LM3916
LED circuit that Fred suggested but with seperate LED's. I found that I had to use a 2k7 current set r. I also had to leave out the whole meter trim section. Maybe I'm way off with the LED circuit because my circuit
turns the LED's out when it is hit with the peaks.

When I read this I went back and reviewed the schematic. There is an error on the schematic and therefore on the pcb. Good catch Lance.

The LDR that controls the meter feed needs to be tied to +24 VDC not to ground. You will probably want to install the meter trim pot/resistor when you hook this up correctly. With the VTL5C1 opto you need about 1k ohm to ground, but this varies from cell to cell, hence the trim pot. You'll have trouble getting the meter to accurately indicate gain reduction levels without this resistor. The value will probably need to be adjusted for other type/brands of optos. I'll post corrected schematics and gerbers later today.

If you have already made pcb of the side chain, you'll need to cut the trace making the ground connection and install a jumper connecting one side of the LDR to +24VDC.

Sorry for any inconvience. Thanks Lance.

 

[FredForssell]

Ok... The schematic and gerbers have been updated (rev B).

http://forsselltech.com/gerbers/optopcb_B.ZIP
http://forsselltech.com/SideChain_B.pdf

 

[hitchhiker]

hi Fred,

I tied the LDR r to +24v and went back to your designated LED circuit .
Everything works great now. also I am using VTL5C1 optos.

thanks again Fred, this design is very good .

How would we go about a stereo link?

cheers,

Lance

 

[PRR]

> How would we go about a stereo link?

Opto-matching will drive you crazy. But....

It may work to just short-together the attack caps in two mono boxes. That ass-umes that the optos are matched, which in general is far from true. It also assumes you set all the controls the same on both boxes.

I would prefer to have both rectifiers dumping onto one attack-cap for both channels, one side-chain, and separate trim-pots for each opto. Counting the meter we have 3 optos, so the original 2-way GR/Meter balance plan may need to be converted to a 3-pot plan:

 

[FredForssell]

Another way to do this is at this link...

http://forsselltech.com/OptoLink1.pdf

This approach sums the side-chain input feeds together and also ties together the voltage followers that follow the attack time RC. There is some inter-action with the attack time controls (they sum). The ratio controls remain independent of each other allowing you to use the ratio controls to trim either the left or right side for the same amount of gain reduction. When used with this configuation, on channel is placed in "bypass" or "out" mode. This means that only one channel's threshold, attack, and release time pots control the settings.

 

[PRR]

> This approach sums the side-chain input feeds together

For most work, that is fine. Also commonly done (it is about how the LA2a was hacked into stereo).

Nit-picking: a strongly out-of-phase stereo signal may be +2V left, -2V right. With threshold at 1V, this should be cut 6dB. But (+1V)+(-1V)= zero, and no gain-reduction will happen.

Strongly out-of-phase signals are generally rare in true stereo. And would rarely be so equal-but-opposite that they would sum lower than threshold. So I don't say it is a real problem. And it is a minimal hack to a pair of mono limiters to "work as stereo".

 

[hitchhiker]

that's very interesting PRR. As a rookie, it makes me think that flawless stereo optos could get rather complicated.

I plan on a number of these units mono but I'd like to have one as stereo.

Hi Fred, I looked at your optolink schematic and I wonder can I leave out the side chain circuit from the in/out switch up to just before the 2meg r .
I'm aiming at a dedicated stereo version.

cheers,
Lance

 

[Svart]

I'm itching to get some of these made! anybody working on pcbs? I would love to get some proto pcbs and play around with this!

 

[hitchhiker]

I'd like to see a group order for PCB's . If someone could take the lead
and check a sample for errors. I've no experience in PCB manufacture.

The sidechain works just as Fred described early on in the thread. It's very transparent unless you are squishing stuff lots. If heavy compression
is what you need then the Gssl or the G1176 are gonna be better.

I'd love to try this sidechain with Fred's 992/993 JFET for makeup because as much as the make up circuit of Buttachunks works great it has the sound of a chip. I'm using an OPA604. To me 604's scoop out a bit of the high mids and the lowish highs, which is not good on snare drum and vocal among other things. So far as amps I only like 604's as mic pre's.

The other thing I'm wondering about ,if Fred's around to respond , is the reverse pot. I bread boarded mine with a reverse and this results in the ratio increasing as you turn it CCW. I put in a regular pot and it seems to work fine but now CW.

Lance

 

[tommytones]

I would gladly take on the pcb making/group order thing. I'd like to put one of these together myself. We'd have to decide on certain components and type(s) of make amps, etc... but it's no sweat for me to handle.

 

[Svart]

i'd be in for a bunch... maybe we can try different op-amps for flavor?...

 

[owel]

I'm in for the PCB too.

 

[soundguy]

It might be nice as a show of support for all the time Fred has spent with us on this one to use his opamps in the circuit, I know Ive been wanting to try them out for a while now and this seems like the perfect opportunity to do so. A discrete versio of this is definitely the way to go though if we get boards made up. Id defintely be in for a pair though!

dave

 

[Consul]

I know a stereo opto compressor interests me. I'd be in for PCBs. Maybe we could do a group buy for some 992s? Fred?

 

[hitchhiker]

I'm with you on the Fred's opamp Dave. I too have been wanting to try them out.

Lance

 

[Svart]

i would like to see something like the SSL setup where you have the footprint/pads/holes for a discrete op-amp but also include pads/holes for an IC op-amp such as the x604 or 5532(s). Not much more work for the amount of flexibility that could be had.