freestompboxes.org

My take on using the TDA7052A chip as gain cell in a compressor pedal:

Project page link

Description:
the compressor is a feedback type one, uses the TDA7052a's two opposite phase low impedance outputs to drive a full wave rectifier and an output buffer. This is a simplified and optimized version with, IMHO, the most useful features available as knobs.

Controls:
Input Gain - adjust it according to the input signal amplitude to avoid the clipping. Lower settings for humbuckers, active bass electronics, higher for single coils passive bases etc.
Attack - response time
Compression - ratio control
Level - output volume.

Project files: Pics and video:

Enjoy!

Thanks for this great project, pz! The knob controls make it look very versatile!
Can I ask you something, though? How did you manage to etch a 2-sided board?

Hector

Grounded top layer, uh?
That's a good idea.

Ettore_M wrote:Thanks for this great project, pz! The knob controls make it look very versatile!
Can I ask you something, though? How did you manage to etch a 2-sided board?
Hector

Duckman wrote:Grounded top layer, uh?
That's a good idea.

I have a lot of 2 sided rest offs from other projects, i used the top layer for a ground plane and jumpers. But it is not necessary, the pcb design is one layer only.
How do i etch 2 sided boards? This way:

Is that a laminator I see?

phibes wrote:Is that a laminator I see?

Yep, removed the original temp fuse and added external temp controller:

Great way, pz! You line up the "X"s in the two layouts! Thanks!
But how do you apply them on the board and don't lose the accuracy you already have?
I'm really excited about this method of yours!

Ettore_M wrote:Great way, pz! You line up the "X"s in the two layouts! Thanks!
But how do you apply them on the board and don't lose the accuracy you already have?
I'm really excited about this method of yours!

As you can see on the photo, once the layers are aligned, i'm using a masking tape (made of paper, does not melt in higher temps) to make a kind of an envelope. It keeps the both sides in correct position. Then the pcb is put into that envelope, secured with another two tape stripes and gets into the laminator.

Ok! Now I get it! Thanks again!

pz wrote:

phibes wrote:Is that a laminator I see?

Yep, removed the original temp fuse and added external temp controller:

Nice! It's good to see others using laminators. I'm still surprised that people aren't catching onto how easy and accurate they are to use. So it works out with gloss paper and I'm guessing laser toner alright? I've always just used the TTP from Pulsar.

phibes wrote: Nice! It's good to see others using laminators. I'm still surprised that people aren't catching onto how easy and accurate they are to use. So it works out with gloss paper and I'm guessing laser toner alright? I've always just used the TTP from Pulsar.

It works out with a plain magazine paper and laser toner. Successfully tested few hundred times over the last years.

Enough tech talk - what about the sound, eh? How do you guys like that one?

meffcio wrote:Enough tech talk - what about the sound, eh? How do you guys like that one?

To me it sounds great! So far I liked Chris realy cheap compressor the most but this took its place since it has nearly the same nice sounds but with more controll over it. As it is another not that far complex circuit I realy would recommend it!

I breadboarded this, and it's excellent. There are a few mods you can easily do (like "jangle" mode, and a decay time control) but it works very well just as it is.

I built a 7052A compressor recently, using the IC and a handful of transistors - the rectifier part was more or less the well-known Dynacomp circuit (if it's not broken, why fix it?), and I used one amplifier output for the effected signal and the other to drive the rectifier. It has the advantage of separating the rectifier completely from the audio path, and initial tests suggest that it's going to be a pretty good compressor, with low noise and distortion:
The "Tadek" idea has a better sidechain than mine, and the compression is much smoother, but at the cost of more components and a bigger board. I'm going to work out a Vero layout for the Tadek thing and everyone can try it for themselves!

mictester wrote: I built a 7052A compressor recently, using the IC and a handful of transistors - the rectifier part was more or less the well-known Dynacomp circuit (if it's not broken, why fix it?), and I used one amplifier output for the effected signal and the other to drive the rectifier. It has the advantage of separating the rectifier completely from the audio path, and initial tests suggest that it's going to be a pretty good compressor, with low noise and distortion:

It is the original circuit from the "No-CA3080 compressor" Elektor article (by Ian Field), i believe. Fits your description exactly. Tried it, but i didn't like it, especially the gain control input hanging on the 100k from the V supply when there is no signal at the input. It was faaar from being low noise and distortion IIRC. One way of keeping the noise lower was to reduce the maximum gain available on the TDA chip and use it only as an attenuator.

pz wrote:

mictester wrote: I built a 7052A compressor recently, using the IC and a handful of transistors - the rectifier part was more or less the well-known Dynacomp circuit (if it's not broken, why fix it?), and I used one amplifier output for the effected signal and the other to drive the rectifier. It has the advantage of separating the rectifier completely from the audio path, and initial tests suggest that it's going to be a pretty good compressor, with low noise and distortion:

It is the original circuit from the "No-CA3080 compressor" Elektor article (by Ian Field), i believe. Fits your description exactly. Tried it, but i didn't like it, especially the gain control input hanging on the 100k from the V supply when there is no signal at the input. It was faaar from being low noise and distortion IIRC. One way of keeping the noise lower was to reduce the maximum gain available on the TDA chip and use it only as an attenuator.

I saw the Elektor thing (someone's put a pdf of it on here somewhere) after I tried something similar. That bit of Vero has a couple of "tacked-on" resistors and a capacitor under the board, in an effort to get rid of the hiss and reduce the maximum gain. The sidechain works OK (with the usual caveat that it can be saturated), but it's just the same as a Dynacomp (which I also don't like) and has all the problems of the MXR thing.

I'm going back to transconductance amps in the negative feedback loop around good quality op-amps! Two cascaded gain cells like that can really work well, with low noise and distortion.

Hello, i'm gathering the parts for this nice project. i have a question about the shottky diode. I can only get the 1N5819 on short notice (it's 40v 1A vs the original 1N5817 witch is 20V 1A)
My consirn is the folowing: is the 1N5819 a worthy sobstitution?

Yes, you can use the 1n5819.

It's really hard to record a good sample of an compressor, clearly showing the sound of it, mostly because you rather feel it while playing the guitar, how the pedal changes the reaction to picking and dynamics etc. Nevertheless, i tried to record another sample, using a strat like guitar (Schecter VS-1). Its percussive sound makes a good testing signal for a compressor. The signal chain was:
Guitar->Compressor->TubeAmp set to clean->112" box (Grennback)->SM57->Mixer->PC
The 2nd part is a mix of four guitar tracks and a bass, all processed with the pedal.
Here it is:

everything is put together, but now i got some weird distorsion going on. basicly it is affected by the level and gain pot and the distorsion is constant no mather what level. I use it on my passive bass. Can anyone sugest me what to try to change?
here's what i have now: