DiSCO - Dirty compressor and overdrive

[ggedamed]

Samples. please?

[samhay]

It certainly is compressing quite nicely.  I think I know where I was wrong last night, but is there any way you can zoom in on that last shot so we can see a few individual cycles when it's squished?

Have updated the simulation output with the early and late waveform below - click on the picture to make it bigger, but I don't think you will see much difference.

Samples. please?

Working on it, but struggling to find enough quiet minutes to get something set up - might have to wait until Monday, or if that doesn't work, next weekend.

[ashcat_lt]

Thanks.  Not exactly what I expected to see, but I'm not really surprised to be wrong. 

I find it difficult to visually extrapolate the response curve when it's applied to a sine wave because wines are already curvy.  Triangle waves just kind of show you the in/out function as long as there isn't a lot of filtering going on around the non-linear portion.

[samhay]

I tried to record something last night but was scuppered by a little problem that only manifest it when I tried to record direct to my audio interface. I think that IC1A was oscilating as it was making an unpleasant craking sound when I played certain notes. There is nothing unusal about this stage, so I guess I just need to lay everything out again on the breadboard and perhaps take a look at the way I generate Vb. This isn't going to happen until the weekend.

In the mean time, I have updated the schematic to show a few minor tweaks I have made.

[deadastronaut]

you little teaser...

[samhay]

I am probably being a bit picky, but this is supposed to sound good without the gain turned all the way up (which does sound rather nice). At the moment, it looks like it either needs a bit more attention to breadboard layout and/or a design tweak. I don't think there is anything majorly wrong with it, but I'm away for the week, so you will have to spend the next few days squirming in anticipation.
However, if anybody wants to throw it on the breadboard and give it a try in the mean time - it's neither particularly big, nor uses any mojo parts - then that would be great.

[deadastronaut]

i'm squirming...

[sajy_ho]

i'm squirming...

Me too

[samhay]

Seems the distortion I didn't like was a grounding problem with my audio interface and/or pedals rather than the circuit. It has started raining again and I guess the humidity sorted it out, as it has gone.

Here is a very quick and dirty audio sample recorded with a touch of reverb direct to the audio interface. Having listened to it, I can surmise that it probably needs some EQ* and I can probably dial back the compression a bit. I guess I should have also have recorded something with the gain turned further down too. Anyway, some of the settings sound pretty good I think, but there will be a few more tweaks before it is ready to box.

*I have flatwound strings on the guitar at the moment so everything sounds dark, but I should probably drop the value of C1 and/or make this switchable.

https://dl.dropboxusercontent.com/u/11996927/DiSCO_audio1.mp3

[sajy_ho]

Sounds good man! I think the clipping is a little strange but compression is awesome. Also you're right about the dark sound, lowering the caps would be nice too.
IMO adding a three band EQ or maybe a little mid scoop will make it a killer pedal

[samhay]

I gave it a few tweaks last night and added some more aggressive HP filtering to the first clipping stage, which makes for a significant improvement in both tone and clipping behaviour.
I also realised that the compression doesn't work very well when the symmetry pot is at minimum as the half wave rectifier doesn't get much left to work with - I can/will fix this with a redesign of IC1B.
I have also been trying passive tone controls tacked onto the output of 1C2B - mainly SWTC and variants. Not sure I have found anything I like yet, but it is proabably going to make for a useful addition. Any suggestions?

I should have a little quiet time this evening to update the schematic and audio clip.

[samhay]

Schematic updated - works significantly better now with only a somewhat modest increase in parts-count.
Still haven't settled on a tone control, but the Drive / gain stage has some variable tone control built in now - essentially some pre-emphasis when the Drive pot is centred.

Recording / audio interface problem is back, so no new clips for the moment.

[sajy_ho]

Nice job man, How about adding this one as tone stack:http://www.runoffgroove.com/tonemender.html
I used it in some of my builds and I love the marshall mod.

[lion]

Sam - a bit OT and probablya bit dumm too - but can you explain how the comp indicator works and how you've adjusted it.

Erik

[samhay]

Nice job man, How about adding this one as tone stack:http://www.runoffgroove.com/tonemender.html
I used it in some of my builds and I love the marshall mod.

Thanks - I was thinking 1-knob and passive, so that it will still fit in a 1590B., but it's a good suggestion and would make for a more complete amp-in-a-box effect.

[samhay]

Sam - a bit OT and probablya bit dumm too - but can you explain how the comp indicator works and how you've adjusted it.

Erik

No problem. The indicator monitors the voltage of the buffered clipping threshold, so as the compression / sag increases, the voltage drops and the LED turns off.

You may be confused by the use of 2 resistors? If so, this approach is taken from the 'Tremulus Lune', but I'm sure it has been around longer than that. R14 is the current limiting resistor for the LED and R15 acts as the other half of a voltage divider to bias the LED's anode to about half way between the monitored voltage and ground. This makes the LED much more sensitive to the changes in the clipping threashold.

If you are wondering why I chose to use 2k2 resistors, then there are 2 reasons - the bright red LED I am using lights up nicely with them and I ordered more of them than I was intending, so tend to use them when I can. If you use another LED that needs more/less current then these resistors can be tweaked to taste.

[jtn191]

I really like your idea for this. Would be interesting to do a fet/fet/fetzer based compressor for line level stuff for recording...

[lion]

.......so as the compression / sag increases, the voltage drops and the LED turns off.

Ahh - LED turns OFF (obviously) 

Thanks Sam.

Erik

[samhay]

I really like your idea for this. Would be interesting to do a fet/fet/fetzer based compressor for line level stuff for recording...

Thanks.
The idea of this design is to use diodes to add non-linearity, but some of the ideas could be used elsewhere. How were you thinking of doing envelope-control of a FETzer?

[samhay]

.......so as the compression / sag increases, the voltage drops and the LED turns off.

Ahh - LED turns OFF (obviously) 

Thanks Sam.

Erik

I guess this is only obvious if you look at the schematic - one typically sees an LED light up as a compressor compresses. However, I haven't found a good way to do this here without significantly adding to the parts count.