Tiger Driver; something between Tim and OCD, help required

[Mr. Lime]

I would like to build myself a new overdrive pedal that is baesd around the Tim and the Fulltone OCD.

It features:

Charge pump +/-9V via IC7660S
Clean Blend
Soft- and hardclipping options
Overall feedback switch
Dual gang saturation/gain control to blend between Tm and OCD style
Buffered Bypass

My question is, how should I connect Vref?
Does it work connecting leads to Vref to ground instead?



Here is the Tim with the original Vref:


I'm glad reading any suggestions, I'm thankful for anything..

[midwayfair]

Your treble pot also affects the clipping. This may or may not be desirable. If you want it to be independent, out it after the MOSFETs, but before your 10nF cap. Increase R1 a little.

With a small change, you can make the tone pot a SWTC (Stupidly Wonderful Tone Control), which will make the volume more consistent. Attach the 10nF to the wiper of the pot, and connect lug 3 to R1 and lug 1 to IC2A.

Your MOSFETs as hooked up are working as silicon diodes (MOSFETs conduct in both directions). They might clip very slightly different from 1N4148 or 1N914 (they're "slower"), but possibly not in a way you can hear. If you're going to vary the saturation with a pot anyway, you might as well make them MOSFETs alone. Either that or you need a diode in series with each one, OR you can simply remove one of the MOSFETs. Read here:
http://www.diystompboxes.com/smfforum/index.php?topic=90474.0
http://www.muzique.com/news/mosfet-body-diodes/
http://www.muzique.com/lab/zenmos.htm

This is a typical connection of mosfets used as clippers for a guitar distortion pedal. You might think that with the gate tied to the source, the gate threashold is the clipping level of the setup but this is not the case. The reverse body diode that is inherent in this type of mosfet has a threshold of around 0.7v. With one of these in each of the mosfets, the body diodes will prevent the gates from ever becoming forward biased. You will only hear clipping from the body diodes and the signal limits will be about +/-0.7v.

The body diode junctions are slower switching and have a high input capacitance. This results in a softer sound than some people like. Any small signal mosfet like the BS170 or 2N7000 can be used for this method of signal limiters.

Back to the treble pot affecting the saturation: First, for some reason, if you remove the tone cap, you won't hear any hard clipping. So don't remove the tone control entirely, and leave the cap connected to ground instead of Vb like in the OCD. (I'd love to read an explanation for why that is. Also, if you remove the MOSFETs from the OCD, it doesn't sound any different. Seriously. I have recorded it flipping a switch back and forth and can't hear when I flip the switch.) Anyway, your saturation pot is a voltage divider with whatever resistance precedes it. If you have the tone pot/cut up all the way and the saturation pot at halfway, you'll end up with the diode clipping being mixed about 50/50 with the dry signal. If you turn down the tone pot, you're going to end up with less saturation because the impedance will be lower and divide less with the saturation pot. So when the tone pot is at 0 resistance, the only divider is the 1.5K, and if you have the saturation pot at halfway, you'll get 1.5/50K -- hardly any clipping at all! This is why I suggest moving the tone pot.

[Mr. Lime]

Thanks for your great advice, midwayfair!

The tone control is really a good point, to be honest I haven't thought about the influence on the hardclipping..
About the sound difference of the MOSFETs switches in/out, I hope that there is a difference when the Saturation control and the Feedback control of IC2 interact with each other.
In my experience I don't like hardclipping too much but I would like to try it in this circuit again.

I also followed your recomendation concerning switch just one MOSFET out of the clipping circuit.
I would switch one MOSFET and one softclipping diode via DIP switch.

Still wondering if the virtual ground thing left out will work?


Thanks!

[midwayfair]

The switch is a good idea. However, you went too far when you moved the MOSFETs -- the output impedance of the op amp is tiny and your diodes won't conduct much, so you want the 10K resistor (R1) between the output of the op amp and the MOSFET connection. Just leave the tone pot as-is, you're only moving the 10K.

[slacker]

Still wondering if the virtual ground thing left out will work?

Yes it will work, Vref is just there to create a "virtual ground" in the middle of the supply, so 4.5 volts for a 9 volt supply. With a +/-9Volt supply ground is in the middle so you don't need a virtual one. 
Your schematic is missing a resistor from the + input of your first opamp to ground, same as the 510k on the OCD schematic you linked.

[Mr. Lime]

Thank you very much for your help, guys.


I really missed that 510k resistor at the input, the schematic is now corrected.
Glad to be sure that the power supply will work that way.

I got one last point so far.
I would like to use the pedal parallel with a modded Ross compressor, so the signal should better be in phase for this application but the last stage is inventing the signal.
My first tube screamer derivate was similar to this circuit conserning the clean blend and the last stage.
The solution was an AMZ MOSFET Booster at the output, which inverts the signal again.

I want to leave the booster away because there is already a feedback booster in the first stage.

Any suggestions how to get the signal in phase?
What if I turn the last stage to a non-inverting stage?
Are there any disadvantages?

Thanks again!

[Mr. Lime]

I came across Mark Hammer's Roseyray circuit, which is based on the OCD/Voodoo Lab OD too.

It has an interesting blend control between stage 1 and stage 2.
Don't know if it would make a sin throwing this blend pot into my circuit as there is already a clean blend.

There is much more behind Mark's idea, cause he is using a low-pass filter at the first stage and a mid scoop filter on the second stage, that's why I'm unsure if this is going to work at my arrangement.
Any ideas how I could implement the blend successfully?
It is important to me to keep the Timmy tones accessible..



Thanks!

[Mr. Lime]

Here's my updated version of the Tiger Driver.
I simplified the boost section by leaving it's tone control away.
The input buffer is changed to a inverting buffer to aviod phase issues with other effects, I want to use the pedal parallel with a compressor.

Because of the inverting input buffer it's better wiring a true bypass instead of S3.
If someone wants a buffered signal, he's just got to turn the clean blend to maximum and level to minimum.

With the old non inverting buffer I had an input gain of 2, now I've got unity gain.
Should I better raise R9 to 2M to get the gain of 2 again?

Please feel free to post any suggestions.
Thanks