New Distortion Design

[Joe Hart]

www.joestump.com/eBay/JoeHart'sSchematic.JPG

Of course, very rarely is anything "new" so I'm sure this idea has been done already, but before I attempt to build this, does anyone see any glaring errors? Also, any comments? Am I rolling off enough highs early on? Too much? Too much gain early on? Not enough? Would a third stage be too much? What would happen if I lowered the gain early on and added a third stage? Or a fourth? Where would be a good spot for a tone stage? Right before the output? What would you suggest for a tone section? Big Muff?

As you can see, I have too many questions. And I know that if I try to build it and start messing around, I'll end up with a billion options and never get the pedal done. I certainly don't mind messing with caps and resistors, but I would rather have something "in the ballpark" before I start messing around.

Thanks!!
-Joe Hart

[cd]

That 330k/1M in the first opamp stage, I would change it to 33k/100k (or even 3.3k/10k), that will lower the noise level (lower DC path) without altering the gain level (gain of approx 3) or frequency response.

[petemoore]

I don't see why it would'nt work, adding some features would make it easier to find out what it's capable of.
 Two clipping stages, using some tailoring, Swiwtch/ifts, pots or sockets might be a good idea to at least leave room for so you can 'tune' it.

[Joe Hart]

cd, I did my calculations wrong! The first stage should be more than a gain of 3. I will change the schematic.
-Joe Hart

[Joe Hart]

I changed the schematic. I meant 150K, so a gain of 6.667. Although that sounds kind of low???
-Joe Hart
P.S. That lower DC thing is cool. I never knew. And I don't have to lower the cap in the feedback loop to follow along?

[Joe Hart]

I'm a boob. I meant 10K. I was trying to guess in my head and I don't know what happened!! So, that's a gain of 100. That sounds better!
-Joe Hart

[Marcos - Munky]

Looks interesting. I never saw the diodes-to-ground in this place of the circuit.

[RDV]

I for one am interested to hear what your breadboard tests sound like. It's kind a cascaded dual Dist +. I can dig it.

RDV

[Joe Hart]

It is a cascaded Dist+! Well, technically a pedal I built for a friend of mine (similar to a DOD250) into a DODYJM pedal. Same difference. I happened to run the two together and was like, "wow, that's almost like my amp tone" so I decided to persue it further. The problem with the resistor discussed above was that I was trying to guess what I had the first one set on for a drive setting without taking it all apart to measure the pot's resistance and got confused trying to imagine it upside down (from looking at the guts of the pedal). I just was imagining it counterclockwise instead of clockwise or something!

I hope to get it done up in the next week or so. I'll let you all know.
-Joe Hart

[stm]

Looks like you are punishing the OpAmps, you don't have a current limiting resistor between the output of the OpAmp and the clipping diodes.  Sound characteristic will be very depending on the OpAmp type used, since there is a lot of difference in the way OpAmps clip and limit current when stressed.

[Joe Hart]

stm, I kind of like opamp distortion (and I was going to audition a bunch of different ones but probably use a 4558 or a 4559 -- I seem to always go back to those two).

What would you suggest for a resistor before the diodes? I just toss one so the output goes through the resistor and then on to the diodes, right? What value might you suggest and what would this do to the sound?

Thanks!!!!!!!!!
-Joe Hart

[Paul Marossy]

I'd like to look at that circuit with a scope...
You're whacking the signal hard in two places with clipping diodes. You'll probably end up with something that looks a lot like a square wave at the output. I'm guessing that it will produce a lot of odd order harmonics and sound pretty fuzzy/buzzy.

Here's an idea to try: instead of red LEDs, you could use MOSFETs connected as diodes. Maybe even 1N34s after the first set opamp section and MOSFETs on the second opamp section. I think this would give a little different sound with a touch of compression.

[aron]

Right now, the signal is being really clipped. Between the first stage and the 2nd (plus germanium diodes), that sucker will really squash.

How did you want this pedal to sound?

[Joe Hart]

Aron, thanks for the site! (I figure if I always thank you when I "speak" to you, then maybe I can, in fact, thank you enough for this site!)

Anyway, how do I want it to sound? Heavy, fuzzy, loads of harmonics, like a cross between "Rust in Peace" era Megadeth but with the "life" and "openess" of Van Halen I. I guess....

All I know is I ran two DOD250 clones together and it sounded very close to what I was looking for.

The first one has the drive all the way up (with four LED's to ground), the tone (this one has a tone control) all the way down, and the volume set for about unity. The second one was the YJM pedal with the drive all the way up and the volume set at about unity.

And I was shocked!

So, I figure if I build two DOD250's into each other and mess around I could get something I totally dig. But I am a little concerned because if I design and then build it and it sucks right off the bat, I don't think I would know what to do with it. So, I jut want to get an idea if it should work (in theory), then if I get close, I can take it from there.

But so far, this is a lot of food for thought. Any more ideas? I would also like to toss together a page like a "technology of the Dist+" thing for beginers. It may be helpful?

-Joe Hart

[Transmogrifox]

I did happen to notice one glaring error:

look at the '+' terminal of the second op amp.  Notice that it is effectively referenced to ground through the diodes.  This effectively turns the opamp into a half-wave rectifier...the 10 k resistor doesn't do anything either.

I recommend putting that 10k in series with the output of the first opamp, then the 4.7 uF cap, then the diodes, another large valued cap, then a V/2 reference to the second opamp.

Finally, you need a large resistor in series with the head-toe diodes to reference that node to ground, or you may have some strange floating voltage effect that may or may not sound good.

Otherwise, it looks good.  Give it a try on breadboard and start tinkering with caps and resistors until you're satisfied.  It looks like it'll be a real scorcher....probably a bit mushy, too, but mushy is great if the sound is fat (IMHO).

[Paul Marossy]

Well, I am interesting in hearing a soundclip. Sometimes things you think ought to sound good don't and things you don't think will sound good do...

[aron]

Good catch on the 2nd op amp. I think it might be really, really squashed at the output especially with the germaniums at the output - but who knows?

[Brian Marshall]

I was just going to say that the 2nd opamp input was at ground.... damnit i missed out another opportunity to point out someone elses mistake.

the first opamp stage has a gain of 1000 no 100.... 1000000/1000=1000.

anyways, depending on the opamp, even with the 100pf cap in the feed back loop, i think you might have oscilation problems with that much gain.

Since the opamp works independantly of the clipping stage that follows it (unlike feedback loop diodes)  the opamp will clip too, uless you are using a lot more than a 9 volt supply.

This could be good or bad.  

The large electro caps after each stage may not really be necessary.. you could probably get away with much smaller, depending on the load.  If you use another 470k resistor to bias the 2nd stage you could get away with a .022, or .01 cap.  the .01 will cut some lows out.... the .022 shouldnt cut any, or at least not much.  If you use a 100k pot for the volume pot... try an output cap between .022, and .1uf.

Also your caps to ground after each clipping stage arent doing much, because there is not a resistor in series with them.... opamps have some output impedance, but not enough for a .001 cap to do much.  you probably want to put a 47k or so resistor in series with the first .001 cap.

for the 2nd one id recomend removing it and stealing a tone conrol from another circuit.  you will definitely still need some series resistance as well though.


The biggest thing i can see in this circuit that i would change, is the gain control, and the germanium diodes.

I would definitely put thegain control in the first stage.... it will give you better over all contol.  maybe put the germaniums in the first stage, and some the leds in the 2nd stage.  you have to be kind of carful on these kind of circuits, because you usually want the first stage to clip arround the same time as teh second stage..... almost at the same time. so you really have to sit down with the calculator, and figure out your clipping voltages, your gain, and maybe even signal losses.

[Joe Hart]

It looks like I really opened up a "can o' worms" here! And it seemed so simple!  :cry:

Mr. Marshall, where could I research this info:

"you really have to sit down with the calculator, and figure out your clipping voltages, your gain, and maybe even signal losses"

I had no idea it would be this involved!

Are there some (hopefully easy) formulas to use?

Thanks!!
-Joe Hart

[Joe Hart]

Mr. Marshall, I think the gain is still 100. 1M in the feedback loop and 10K to ground. So, isn't that 1,000,000/10,000=100? I think you're missing a zero on the 10K.

Does this change your thoughts on the gain structure?

Also, with the calculator stuff (I'm really liking the idea of figuring stuff out -- like I'm a real effects designer!!), what results would I be looking for? Like should I match up the resulting voltages with the forward voltages of the diode pairs?

This seems like it could be interesting.

Thanks.
-Joe Hart