Less "presence' on 250 Clone

[leroylanes]

Hello all.

I have a DD250 clone.  It works perfectly. I like the 'big' sound I get with it, it is just one notch less "presence or high treble" away from working perfectly for me. 

I imagine there is a resistor adjustment that can take away just a bit of the high end and still retain that huge sound.  Can you please point me in the right direction?

For reference, I have decent soldering skills and have built pedal kits.  So while I do not design circuits, I can follow directions.  Here are some pics.

[blackieNYC]

The best way to do this is a small capacitor to ground.  But a good question is: should you roll off some highs before the distortion happens or after?  Two very different sounds.   If you roll of a little treble at the guitar, does that give you the sound you want?

[antonis]

Hi & Welcome..

Without a schematic of your particular klone it's hard to tell but you could try to make the NFB cap (the one between op-amp output and inverting input) bigger or make the resistor before clipping diodes bigger or make the capacitor in parallel with clipping diodes bigger..

[Steben]

Welcome.

First of all: Great news it works.
Antonis summed up the classic simple mods to tailor the treble sound.
It is easy to understand why a treble or tone control came to live in drive pedals.

[leroylanes]

The best way to do this is a small capacitor to ground.  But a good question is: should you roll off some highs before the distortion happens or after?  Two very different sounds.   If you roll of a little treble at the guitar, does that give you the sound you want?

Rolling down the tone on the guitar does not work for this particular high frequency.  SO after the distortion is definitely the place. 

[leroylanes]

Hi & Welcome..

Without a schematic of your particular klone it's hard to tell but you could try to make the NFB cap (the one between op-amp output and inverting input) bigger or make the resistor before clipping diodes bigger or make the capacitor in parallel with clipping diodes bigger..

And there is my problem.  I know how to follow directions but do not know where the caps and resistors you mention are located.  I assume that brings us back to the first part of your post regarding the schematic. 

Kinda like a 5 year old with good penmanship tracing letters on this. 

[leroylanes]

I guess I should add - For lead boost, love my '00 Fulldrive 2 (pull comp cut) and a Klone I built from a kit. 

The reason I would really like to solve this issue is because I am finding it difficult to like a pedal for rhythm distortion/ drive.  Have tried and sold several.  This one does it exactly how I would like except for the high presence.  So I'd really like to try something before punting. 

If it matters, strat, tele, or McCarty through a Carr Rambler.

[idy]

Weclome
Just tell us where you got it. If it has a "build doc", post the schematic and board diagram. Other wise it's hard to spot what is what...but...

The treble cut cap is probably a 1nf (.001uf) that is near the volume control... is that grey box cap marked 102? That would be it. You could take it out and add a tone control of your choice, or just try adding a bigger cap. A tone control could just be a pot (100k linear would work) in series with a cap (bigger than what you have, you need to experiment) in place of the old cap. The order of cap and pot does not matter here.

You will connect pins 2 and 3 of the pot, then connect the cap to either pin 1 or 3. If you connect it to 3, then the loose end of the cap and pin 1 go to the pcb.

[Mark Hammer]

Make the 25pf cap in the feedback loop a little higher, like 47-68pf.  Make the 1nf cap in parallel with the diodes 3n3, and you should be very near the territory you want to be.

[amptramp]

The schematic for the DOD250 that I have found shows a 1 megohm feedback resistor shunted by a 22 pF capacitor for a -3db rolloff at 7238 Hz.  There is really nothing you need to hear from a guitar at that frequency.  The highest key on a piano is 4186 Hz and you don't really need anything much higher from that in most music.  One of the remarkable (and successful) albums of the 1980's was No Jacket Required by Phil Collins and on that album, the recording was done so that there is nothing above 5000 Hz.  As you go up in capacitor value, you will go down by an equal proportion in rolloff frequency.  The standard capacitor values would be 27 pF and 33 pF which would give rolloff frequencies of 5898 Hz and 4825 Hz.  You could still afford to go higher than 33 pF.

At the output, before clipping, you have the 10K resistor in parallel with the 100K volume control for 9.0909 K parallel resistance and a 1 nF capacitor across it.  This gives you a rolloff at 1572 Hz before the diodes conduct and a higher frequency once they do conduct since the resistance of the diodes is in parallel with the 90909 ohms, reducing the parallel resistance and increasing the rolloff frequency.

You should note one thing about the guitar and the harmonics that are generated by any distortion / fuzz circuit and that is, the guitar has an output with partials which are very close to exact harmonics of the fundamental frequency but not exact due to detuning of the string at various frequencies due to power delivered to the air and to the pickups, so the second partial of the string may be 1.002 times the fundamental whereas the harmonics generated by  a fuzz / clipper will be an exact multiple.  Due to the effect of the diodes, the exact harmonic and the partial will mix to provide a difference frequency of a few Hz that will modulate the output like a tremolo.

Welcome to the rabbit hole!  I fell down it many years ago and never really want to get out.  There are just too many interesting things going on here.

It is an addiction that leaves no needle marks.

[bluebunny]

It is an addiction that leaves no needle marks.

Just the occasional solder burn...

[amptramp]

Caught a mistake in what I said yesterday.  The second partial may be 2.002, not 1.002 times the fundamental.  Let us know how the capacitor changes work out.

[PRR]

Caught a mistake in what I said yesterday. 

Well, and also the sharp harmonics are not-much "due to power delivered to the air and to the pickups". Instead the classic explanation for sharp overtones is string self-stiffness. A limp string with all stiffness due to tension would have harmonic overtones. A steel string has so much self-stiffness that the end of the string vibrates less. This is a small effect on a whole-string wave but gets significant on higher overtones which are only part of a string long. A similar effect on pipes (flute, sax, brass).
https://www.phys.uconn.edu/~gibson/Notes/Section4_2/Sec4_2.htm

IAC: Yes, this is musically significant, fuzz-tones not lying ON string/pipe-tone frequencies.

[Mark Hammer]

Harmonic distortion is essentially multiples of the components of the original signal at a higher amplitude than in that original signal.  The relative amplitude of the various multiples may be different than in the original signal.  It pays to remember that the guitar signal is a) complex, and b) changes its relationship to the clipping threshold with time.  Some components of the guitar signal  may remain over the clipping threshold for a bit, while others fall below.

So, while increasing the cap in parallel with the diodes will remove unwanted fizz, the best strategy is to use suitable lowpass filtering BEFORE the diodes, to remove harmonics in the original signal that you don't want to amplify and "enhance", such that the cap in parallel with the diodes essentially "cleans up", and isn't totally relied on.

It is also not out of the realm of reasonable to split that 10k resistor ahead of the clipping diodes, and use that to impose a SECOND layer of pre-filtering.  A 47pf cap in the feedback loop of the op-amp, with the 1M resistor, will roll off the top end at 6db/oct, beginning just below 3.4khz.  If you replace that 10k resistor with 4k7 and 5k1 in series (4k7 comes "first", close enough to a 10k total), and run a 10nf cap to ground from their junction, tha will provide a second treble rolloff at the same frequency, for a 12db/oct total.  Your diodes will now provide multiples of *some* content above 3.4khz, but not nearly as much.

Why the use of 4k7 + 5k1?  The behaviour of clipping diodes apparently depends on the current of the signal feeding them.  Using two series resistors adding up to about the same resistance as the original preserves thatl evel of current.

[Steben]

Harmonic distortion is essentially multiples of the components of the original signal at a higher amplitude than in that original signal.  The relative amplitude of the various multiples may be different than in the original signal.  It pays to remember that the guitar signal is a) complex, and b) changes its relationship to the clipping threshold with time.  Some components of the guitar signal  may remain over the clipping threshold for a bit, while others fall below.

So, while increasing the cap in parallel with the diodes will remove unwanted fizz, the best strategy is to use suitable lowpass filtering BEFORE the diodes, to remove harmonics in the original signal that you don't want to amplify and "enhance", such that the cap in parallel with the diodes essentially "cleans up", and isn't totally relied on.

It is also not out of the realm of reasonable to split that 10k resistor ahead of the clipping diodes, and use that to impose a SECOND layer of pre-filtering.  A 47pf cap in the feedback loop of the op-amp, with the 1M resistor, will roll off the top end at 6db/oct, beginning just below 3.4khz.  If you replace that 10k resistor with 4k7 and 5k1 in series (4k7 comes "first", close enough to a 10k total), and run a 10nf cap to ground from their junction, tha will provide a second treble rolloff at the same frequency, for a 12db/oct total.  Your diodes will now provide multiples of *some* content above 3.4khz, but not nearly as much.

Why the use of 4k7 + 5k1?  The behaviour of clipping diodes apparently depends on the current of the signal feeding them.  Using two series resistors adding up to about the same resistance as the original preserves thatl evel of current.

Given the behaviour I always found the current vs voltage driven discussion intriguing when talking diodes. Ladders of the stuff and ideal diode calculations always felt so comfortable...
I still haven't built the elektor soft clipper of 30 years ago. Though I do wonder whether it sounds that much different than classic soft clippers.

[leroylanes]

Thank you all.  I have plenty to go on.