better compressor pedal??

[merlinb]

Here are the numbers I am getting (25K threshold pot and 1K threshold resistor, 2N5458 Fet):
with threshold maxed and gain at minimum
-30db in / -37db out
-20db in / -30db out
-10db in / -24db out
0db in / -14db out

OK, those numbers don't look too bad, since you're getting 14dB compression max. In which case, I'm surprised you want more squish...

[Ripdivot]

Part of the issue is that I am playing a strat which doesn't have all that much output so at times it seems like there isn't much compression. Thanks for all your help Merlin! I will still report back when I try the J112 Fet.

[merlinb]

Part of the issue is that I am playing a strat which doesn't have all that much output so at times it seems like there isn't much compression. Thanks for all your help Merlin! I will still report back when I try the J112 Fet.

I suggest you just fiddle with the FET bias, it's very sensitive. May get you what you;re looking for.
As for the noise, I had a brain wave yesterday. With one extra resistor you can get a major noise improvement. Down side is you sacrifice headroom, but with a strat you probably don't need 6V headroom anyway! Been playing with it today, I think with some tweaking I might be able to eliminate the level knob, potentially making a one-knob compressor! The extra resistor is the 10k feedback on U1a. As you can see, as you lower the threshold it simultaneously increases the pregain.

[Ripdivot]

Very cool idea Merlin, I'll give this a try in the next few days. Too late to make noise at the moment...

[Ripdivot]

Merlin, could I not increase the sensitivity of the compressor by increasing the 100K resistor in the feedback loop of U1b? This should cause it to start compressing sooner with week pickups or lower level inputs would it not?

[merlinb]

Merlin, could I not increase the sensitivity of the compressor by increasing the 100K resistor in the feedback loop of U1b? This should cause it to start compressing sooner with week pickups or lower level inputs would it not?

Yes, you could try that. But I still think even your 'weak' pickups shouldn't be so weak that it wouldn't work as shown. Let me know how you get on.
Incidentally, I'm currently using a 4.7k feedback resistor on U1a, 1k+47k input to U1b and a 47k feedback resistor on U1b.

[tubelectron]

Hi,

I have a 1980 DynaComp "mint" block logo unit, an Ibanez CP-9 of the early 80's, and an "somewhat unknown" EXAR Compressor (stompboxes from Poland, probably from the early 90's). The CP-9 is not worth - forget it ! The DynaComp is nice but sounds nothing more than correct, and the best is the Polish EXAR : better transparency, better sustain, less noise... It has DBX IC inside, and the stompbox is crammed...

A+!

[earthtonesaudio]

Brilliant method of adjusting the threshold/pregain, Merlin.  I like that a lot.

From the look of the circuit, it seems you've biased the JFET so the difference between ON and OFF resistance is not very large (some J112s won't turn fully off).  What's the reasoning behind using the J112 versus say a J113 or J201?

[merlinb]

 What's the reasoning behind using the J112 versus say a J113 or J201?

Simply because that's what I had, and most of them worked equally well. J113 and J201 should also work, probably better.

[Astronaurt]

Hey folks, I'm gonna raise this page from the dead for a bit, hope that's chill. I've been working on the Valve Wizard Compressor, and trying to come up with a way to meter how much Gain Reduction is going on and settled on using an LM3915 LED Driver chip. The question on my mind though is how and where do I have the Chip monitor from? I could take it off from the input to the JFET, but that wouldn't include the rather important effect of the Ratio control. Otherwise I could take it off of the source of the JFET, but I'm concerned that in order to not interfere with the Ratio control I would need to have a very heavy resistance going to the LM3915; and I remember reading that it becomes less accurate/reliable with a particularly high input impedance. So what would people suggest? Thanks for any input folks!

[earthtonesaudio]

If you want accuracy the "right" way would be 10log(output/input) which means an instrumentation amplifier, rectifier, and log converter all to drive the input of the LM3915.

But that's easily just as complex as the pedal itself.  How about settle for a little less accuracy and simply use the CV?  Ah but the ratio control...
Okay then, make the Ratio control a dual ganged pot and use the other half to mix in an offset to the CV you're feeding into the LM3915.

[liquids]

Merlin - your solution certainly looks easier than building the Carl Martin.

Do you see any real benefits (or trouble) making it work off a bipolar supply?  9v supply is easy, sure, but bipolar isn't much more trouble for me, so if there are benefits, I'll do it.

I guess if I just breadboarded it myself, I'd know....but always nice to ask the designer in these rare cases when that is possible.   =)

[merlinb]

Merlin -

Do you see any real benefits (or trouble) making it work off a bipolar supply? 

Yes you could use a bipolar supply, which would certainly give you more headroom.
I should say, though, that I have gone off this design. I don't like the way it sacrifices headroom for lower threshold. I am still playing with topologies to get low noise and low ratio, with simple parts...

[Jorge_S]

Merlin -

Do you see any real benefits (or trouble) making it work off a bipolar supply? 

I don't like the way it sacrifices headroom for lower threshold. I am still playing with topologies to get low noise and low ratio, with simple parts...

Do you plan on sharing your new design when it's done?

I just discovered this thread (I'm kinda new here) and I like all the control on compression/sustain your last design permitted. I hope your new design will have that same flexibility.

thanks!

[merlinb]

Do you plan on sharing your new design when it's done?

Of course!

I will be a while though, as I'm currently working on a signal generator, delay pedal and phono amp.

[pedalgrinder]

hi everyone,

Has anyone built this compressor : http://www.sugardas.lt/~igoramps/article60.html on this russian site and tried it on bass. One thing i believe it would be nice and responsive being fet. But i have tried different compressor and i have found in general i should say there always variations to the rule. But VCA compressor seem to be hissy. Fet compressor tend to be pumpy if not controlled properly as the fet triggers so quickly. I was quite pleasantly suprised with transistors used as attenuation. Cause there a bit slower than fets they tend to be warmer and not so trigger happy. The one i haven't played with which i intend to do some prototypes with is the TDA7052a this chip acts similar to a vca with the dc volume control option on it. basically you have to control the variation between 0.3 and about 1.4v but the it can happily ride above 1.4 and just dip down as far as needed to compress by using some type of pot to contol the threshold. Well this is as far i assume for now. Only time will tell once i do the prototype. Of what i can work out or hoping it should be the most hifi of all or the other word used for hifi is transparent. as being it is a dc volume control the distortion levels due to the chips charecteristics should keep distortion levels essentialy not audible. If anyone has any suggestions on what else could be done i'd love to hear from you. I would also love to know what that russian fet compressor is like especially on bass. Cheers 

[liquids]

I just ordered a 4301 chip last week and I am surprised at how much I like it.  Of course it is an expensive chip to blow up, so it should be treated carefully on the breadboard, and is best when run off bipolar supply, but it's a chip more or less designed for a compressor, and it's the best I've tried outside or alongside the RNC.  No hiss even when maxed out, from what I am hearing.

Familiarity with THAT's design notes helps a lot though.  The best feature in a compressor (that is lacking from the stomboxers) is a threshold control.  I use my guitar's volume control a lot.  With the threshold set properly, the compressor will affect my signal with the guitar's volume maxed, but when I dial it back, the compressor's threshold is no longer met and I can retain the ability to 'cleanup' overdrive pedals that follow it.

[pedalgrinder]

the only thing the that 4301 has going for it versus other vca's is the rms or averaging of the signal other than that it really doesn't do much for me other than hurt my wallet theres lots of chips that easily sound as good.

[liquids]

I would also love to know what that russian fet compressor is like especially on bass. Cheers 

You should breadboard it, then you'll know more for yourself.

[pedalgrinder]

i have and i think there average