Asher Guitars WD Music Products Amplified Parts Mod Kits DIY Nordstarnd Pickups Warmoth.com
Asher Guitars WD Music Products Amplified Parts Mod Kits DIY Nordstarnd Pickups Warmoth.com
Asher Guitars WD Music Products Amplified Parts Mod Kits DIY Nordstarnd Pickups Warmoth.com
Join TDPRI Today

Fuzz experts...

Discussion in 'Burnt Fingers DIY Effects' started by JD0x0, Aug 12, 2017 at 7:33 PM.

  1. JD0x0

    JD0x0 Friend of Leo's

    Age:
    27
    Feb 22, 2009
    New York
    I just built a cheap DIY Fuzz kit from a supplier in china. First, I have to say, for a $20 kit, shipped, it's pretty darn good. IC sockets, decent quality parts, good PCB. I initially thought it was a Muff Fuzz, but I believe it to be something different, due to various differences I've found.

    Anyway, it sounds really good. I should mention it's not totally stock. I changed/added some diodes, and this pedal is running a mixture of silicon, germanium and Light Emitting diodes, and higher forward voltages.

    My one peeve is that at higher 'sustain' settings it starts to get some strange gating and artifacts. It's most easily repeated by doing pinch harmonics, which it will gate the initial attack a bit.

    Could this be due to the slew rate of the IC chips? Or perhaps I should address the diodes more. I believe the final pair of diodes could potentially use a higher forward voltage, as the LED's in front of them are probably hitting them with a lot of voltage compared to the original design. Unfortunately have no schematics, and this layout is the best I can offer.

    IMG153.jpg

    As far as the current diode situation. The bottom pair (as viewed in the picture above, as oriented)
    is ~.6V silicon - .5V silicon in series w/ ~.28V germanium
    the pair of diodes second from the bottom are both LED's one approx. 1.8V the other approx. 1.65V
    The top pair, are stock, both .6V silicon, I'm thinking this pair perhaps is being 'overloaded' at high gain and could perhaps use a bit more FV.
     

  2. luckett

    luckett Tele-Afflicted

    Jun 14, 2011
    .
    Looks like an IC big muff.

    It's probably not the diodes assuming you have them oriented correctly.

    Check your IC voltages, particularly the bias voltages. Make sure you put a 47 ohm resistor at the 9V source..

    [​IMG]
     

  3. ranjam

    ranjam Tele-Holic

    535
    Dec 29, 2015
    Canada
    That 47Ω is not what you think it is. It's part of the DC filter for 'less than well-filtered' AC adapters. That's all. Because the resistor is in series with the supply, and then a 220uF filter in shunt, it's never going to 'bias' anything. Maybe the poor bias set up is part of the Muff sound?
     

  4. Forum Sponsor Sponsored posting

  5. luckett

    luckett Tele-Afflicted

    Jun 14, 2011
    .
    I know why that resistor is there. I never said that resistor was biasing anything. It's obvious that resistor is in series with the supply and there is nothing else going on there. That's why it needs to be verified, because if a much larger one like a 47k was put there inadvertently it would be starving the circuit for voltage.

    "Poor bias" is not part of the muff sound. It's not supposed to have a "poor bias". That should be obvious from the schematic.
     

  6. JD0x0

    JD0x0 Friend of Leo's

    Age:
    27
    Feb 22, 2009
    New York
    The values are correct. Since the kit was Chinese, I decided to measure every resistor to ensure I wasn't supplied with any of the wrong parts by accident. It doesn't sound voltage starved as when I hit big chords with a lot of lows, the pedal doesn't get the 'fart' out effect, but does sound compressed/limited, not in a bad way, most fuzzes do this, IMO. It appears there's a lot of lows being fed into the clipping circuit. I might try adjusting the cutoff and see if it helps

    That schematic is hugely helpful, I believe you're right that it's an IC Muff, now that I see the schem.

    Other than the slight gating effect at high sustain levels, the pedal sound beautifuls. Better than most off the shelf fuzz boxes IMO.
     

  7. ranjam

    ranjam Tele-Holic

    535
    Dec 29, 2015
    Canada
    I read you completely wrong. My mistake, and I retract that statement 101%.

    Well, maybe not obvious, but there is bias via the voltage divider of two 220K resistors on the first IC, and the 820K and 1Meg resistor on the 2nd IC. You can absolutely bias any op-amp 'wrong'. I just wasn't thinking when I said what I said, and chose my words poorly.
    What I could have said, and appeared slightly intelligent, is the Sustain control is part of the feedback loop to IC1b. I suppose there's no harm-no foul in toying with the values or P1, R7, and R9 so at maximum value it isn't acting 'weird'. But if can say that most EH pedals were notorious for having 'weird' sounds at maximum setting of certain controls, then I can also say there is either a desire for 'stock' circuits, or circuits that are tweaked to suit your tastes. The choice is yours.
    I apologize for any confusion caused by my misspeaking. Age will do that to you. Wait and see. ;)
     

  8. telemnemonics

    telemnemonics Poster Extraordinaire

    Age:
    57
    Mar 2, 2010
    Maine
    Lots of fuzz lovers seem to like the gated effect, but I can't stand it.
    In a Muff, it's possible the ge is the source of the gating, as typically si does not have that feature/ problem.

    It's almost wrong to call a Muff a Fuzz, but I haven't heard this one.
    I have a bunch of Muff variants and for example the typical Gilmour smooth singing tones are not really fuzz tones.

    Re: lots of bass, there are ways to put a bass cut pot in there so you can regulate the bass content, and adding a mid control also helps get useful sounds.
    But I'm coming from a buyers perspective, only know how to choose someone elses build.

    Any switches?
    Or just three knobs?
     
    JD0x0 likes this.

  9. JD0x0

    JD0x0 Friend of Leo's

    Age:
    27
    Feb 22, 2009
    New York
    Just three knobs. The way they have the enclosure/PCB setup means you have to run wires from all 3 pots, the LED, both jacks, and 9VDC Battery, as well as the 9VDC walwart input all to the PCB. In a standard size enclosure (Dynacomp size) it's a little messy and any more added switches would just make matters worse.

    The pedal definitely has some fuzzlike qualities especially at higher sustain settings, but definitely has more amp like OD qualities, probably due to the high FV in the clipping circuit. I like how responsive it is. I can get it pretty clean, just by backing off my picking dynamics, even at relatively high 'sustain' settings.

    I ordered a modern TL071 to replace the 741 chip, I'd like to see what the difference in the chip performance gives audibly. I know lots of people seem to prefer the worse performing chips for distortion devices, but like the gating effect, or 'sag' from tube rectifiers, it isn't necessarily better for everyone's tastes.
     
    telemnemonics likes this.

  10. NotAnotherHobby

    NotAnotherHobby Tele-Afflicted

    Oct 27, 2015
    Da' Magic Mittin'
    R12 and R11 are awfully high values for bias on that last op amp. I'd try putting some 470K resistors in parallel with those to see if it improved things.

    I think current is being starved on that last amp and that might result in the gating effect.

    Most op amp clipping stages use much lower resistors to provide a bias voltage. Opamps tend to be more voltage-dependant with signals, but current is a factor. But those values are in the upper end for what most opamps can handle.

    Not saying this is the issue, but it may be a contributing factor.

    A couple of cold solder joints may also be in the mix. Hard to say without reconstructing the circuit.
     
    Last edited: Aug 13, 2017 at 10:47 PM

  11. NotAnotherHobby

    NotAnotherHobby Tele-Afflicted

    Oct 27, 2015
    Da' Magic Mittin'
    Also, that last clipping stage is set up for OD. So that's why you are getting an OD vibe from the pedal.
     

  12. NotAnotherHobby

    NotAnotherHobby Tele-Afflicted

    Oct 27, 2015
    Da' Magic Mittin'
    Another thing, the clipping diodes - 3 back-to-back means that the feedback voltage needs to exceed 1,8v to conduct (for silicon). Ideally speaking, your quiescent voltage on an average bias is something like 4.5v (and you are around that with the r11 and r12). Now I'm not 100% sure what such a high voltage threshold voltage would do in that circumstance, but it does seem like this may also be part of the issue.

    I've seen that back-to-back configuration in other designs, but if I remember correctly they were using voltages in the range of 12V - 18V.
     
    JD0x0 likes this.

  13. BB

    BB Friend of Leo's

    May 17, 2003
    Great Pacific NW
    I understand zero percent of whats being said in this thread, but I love it!
     

  14. JD0x0

    JD0x0 Friend of Leo's

    Age:
    27
    Feb 22, 2009
    New York
    Noted. Though, the above schematic suggests increasing the 820K(R11) to 1M if you use a TL071, which I intend to, so maybe I'll leave those alone for the time being.

    I believe I'm closer to 3V since I'm using LED's mixed with silicons (and a germ) the fV from the LED alone would be ~1.8V
    I've got a 1N34(~.27V)>Schotky(.5V)>Red Led(~1.75V)>silicon(.6V) on one side
    Sili(.6V) > Yellow LED (~1.85V) > Sili (.6V)
     

  15. NotAnotherHobby

    NotAnotherHobby Tele-Afflicted

    Oct 27, 2015
    Da' Magic Mittin'
    Yeah, I don't get that. R11 and R12 set up a voltage splitter to give you a "bias" voltage on the one input of the opamp. So, that means one side should, ideally, be clipped more than the other side of the waveform. But, the amount of current that input has is pretty low given the values provided.

    There should really be no reason why one flavor of opamp would requite a different bias than the other. For what's being done, and the chips being used, they are functionally the same. If there is a concern about offset voltages (meaning that when your output SHOULD be at 0V, you're actually off by a bit one direction or the other), then the design should have used an opamp that allows you to tweak this to zero volts - usually that is a leg on the chip that is normally unused on most packages. You don't do that at one of the inputs. You'll usually see this in s schematic, where an extra line comes off the triangle roughly neat where one of the power-supply leads is located, and that goes to a trimmer that is set up like a voltage divider.

    Many opamp circuits I've seen usually create a single "bias" voltage with a couple of 100K resistors, or something like 220K - and that will go a long way feeding many different opamps in a circuit.. But nothing as high as 1M. Hence the reason why I think you may be on the fringe current-wise on that bias input. The input current to one of the opamp inputs tends to vary by flavors. Usually the FET or BiFET chips are pretty low-current, but that has it's limits.

    Now, I could be wrong about starving current at the input, you may have enough there, and that may not be the problem with the gating at high gain. But the design notes you mentioned sounds more like someone's experimentation notes than actual *design* information to me.



    Still, given your available swings (theoretically +-4.5V from the bias), that's a lot. It very well may be that the design considerations did not include using LEDs as they have a higher forward voltage, but instead assumed silicon signal diodes (and this is almost always the case).

    If I am reading what you wrote correctly, mixing LEDs, regular silicon diodes, and germanium really doesn't give you any sonic distinctiveness in my opinion. Mainly because part of the sound you get from a fuzz is determined by where you clip on the signal. Germanium clips WAY down, where the change in voltage over time is much higher (look at a sine wave where it crosses zero, and you'll see what I mean), making the wave look somewhat rectangular. As you move up into the higher forward voltages, the signal moved more to a square-ish appearance. Hence the different sounds.

    Different diodes tend to react differently when they start to conduct, and that also somehow figures into to what you hear. But, when you mix types, you run the risk that some of the distinctiveness of the clipping type.

    If you like the sound you've got, then it is what it is. But it could be that the arrangement of diodes you have is contributing to the gating effect you don't like.

    I should note that the loop from the output of the clipping stage, to the input of the opamp is a negative-feedback loop. In short, this is what tames the amplification factor of the opamp. Otherwise, the circuit acts as a comparator, and basically rails the different in voltage one way or the other (think of an amplifier having a gain of 50,000, and that's what a comparator is). If you play a guitar though a comparator, you get that 8-bit-sounding bleeps and bloops with that same gating effect (yes, I've done that). So, it is reasonable to assume that the gating you have going on is because there is something amiss on your feedback loop.
     

  16. JD0x0

    JD0x0 Friend of Leo's

    Age:
    27
    Feb 22, 2009
    New York
    Thanks again for the input. The op amp came in today, no real help. I decided to trim the bass before the clipping stage, and what do you know, that apparently fixed the issue entirely. Changed the high pass from 4hz cutoff to something closer to 300hz. Lows sound tighter, and absolutely zero of that gating effect on the pinch harmonics. It was kind of interesting, to a point, the filter was passing so many lows that the sound attack of some notes almost had this effect as if you're making a 'PR' sound with your mouth. It reminded me of Trombetta pedals, which almost get this trombone effect on the attack, but less pronounced.. Cool effect, but not really my thing..

     

IMPORTANT: Treat everyone here with respect, no matter how difficult!
No sex, drug, political, religion or hate discussion permitted here.