IC Muff Fuzz Issue - Will an Enclosure Fix Excessive Noise?

[nerdbot]

First time builder here in need of some help (go figure).

I've built an EHX IC Muff Fuzz on vero with a couple modifications (pictures/explanation at the end). Upon completing the vero and testing it, I was concerned with the excessive amount of noise (mostly hum and buzz, 90% of it goes away when I turn the guitar’s volume knob down) and a sort of crackling that comes in waves as notes decay and the noise comes back in (no gating of notes and it seems to be separate from the actual clipping)

However, I’m not testing within an enclosure and am using alligator clips, a battery, and a couple jacks. Once I moved my testing to a part of the house that has grounded outlets and is away from computers, I would say that the issues were reduced by half. Below is a video of the noise, with the odd decay being more prominent on the 2nd chord:

I’m hoping that putting this in an enclosure will take care of these issues, but I am also getting op-amp (TL072) readings that seem low:

1) 3.19
2) 3.18
3) 3.03
4) 0.00
5) 3.04
6) 3.17
7) 3.24
8 ) 6.34

Other than the noise, it sounds good to me. Not a ton of volume when dimed, but it's louder than bypass and seems on par for this circuit. I built it using a combination of a vero layout and a schematic done by Jack Orman and J_Flanders. Essentially, I added a capacitor and resistor in parallel with the diodes to the vero layout (see below: vero layout with schematic that I believe it’s based on, the alternate schematic with added capacitor and resistor highlighted).



Here's some pictures of my build:





You’ll probably notice that I’ve made some other modifications (too many sockets, 1M bias resistors, asymmetrical BAT41, TL072 op-amp), but I started at the values in the vero and the part changes have had little to no effect on the issues with the noise. One exception: I tried out D9J germanium diodes, which had a forward voltage of 0.28-0.3 (slightly lower that the BAT41s currently in the circuit). This eliminated a lot of the noise and crackling, but there was very little clipping going on. Not sure if it’s relevant, but another user measured these diodes as having an extremely soft knee.

Thanks in advance for your help! Let me know if any other info would be helpful.

[antonis]

Hi & Welcome..

What is your power supply..??
What is its voltage reading without feeding the circuit..??

[nerdbot]

Hi & Welcome..

What is your power supply..??
What is its voltage reading without feeding the circuit..??

I'm using an unused 9v battery, which of course, measures at 6.83v. Tried other unused 9v batteries and they all had readings below 9v. Maybe I'll run to the store in a bit to grab a new 9v today and set a reminder to never by Amazon Basics batteries again. I do have a one-spot, but I heard it's safer to test circuits with a 9v.

Thank you! I'm curious how it will sound when it's not voltage starved.

[iainpunk]

an enclosure helps a lot in combating noise. try a temporary shielding with some aluminium foil, being careful not to short anything out
the voltages on the IC are quite low tho, they should be around 9 or 4.5 v not 6 or 3

if you are careful, testing with a power supply in just fine, most modern ones have some current limiter in it.

cheers, Iain

[nerdbot]

Just tried it out with a new battery that is in fact 9v. Volume and signal to noise significantly improved and it strengthens my belief that the odd decay is related to the noise starting to overtake the signal, as it significantly decreased and happens much later on. The op amp voltages also increased as expected.

A byproduct of the proper voltage is that the gain seems to have increased and/or feels more saturated. It lost a little sensitivity/clean up as a result, but the exchange is well worth it and I may mess around with the 470k resistor to bring the gain down a touch or add a gain control.

I now feel comfortable moving forward to placing it into an enclosure and finishing my first pedal. Thank you all!

[antonis]

A 47 - 100 pF capacitor across 470k resistor (between pins 6 & 7) could also help.. 

[iainpunk]

or add a gain control.

you realize you have a gain control on your guitar right? its mislabeled as volume.
turn down for less gain, i prefer using the guitar volume control over having a gain knob, because lits right at your finger tips

cheers

[nerdbot]

or add a gain control.

you realize you have a gain control on your guitar right? its mislabeled as volume.
turn down for less gain, i prefer using the guitar volume control over having a gain knob, because lits right at your finger tips

cheers

Haha for sure! I have a strong preference towards drives/fuzzes that respond well to the volume knob or pick attack, which the muff fuzz does pretty well. Getting this circuit into the sort of "goldilocks" range of gain where I can go from a jangly clean to palm mute riffing with the just the volume knob is the goal. It's pretty close. Just going to try some different op amps, diodes, and maybe bring the gain down just a touch. In another thread here, I read that "the 470k feedback resistor between the output of IC-A and the negative input controls the amount of signal that gets sent to the clipping circuit." So, I'll start there in terms of altering the gain.

So far, I changed out the 680k bias resistors for 1M as the clean up tone was better (more highs preserved) and also changed the 100p cap to 47p for a similar reason (more highs/presence). I also tried upping the value of the 470k resistor (in parallel with the diodes), which made it more saturated and smooth, but the dynamics and clean up suffered, so I kept it at 470pf.

Regardless, I think it'd be a fun learning experience to see what I can do to provide more control over gain, maybe a rotary with different caps, a footswitch to go between a couple settings, or just a normal pot.

[Marcos - Munky]

I'm using an unused 9v battery, which of course, measures at 6.83v.

A 6.8V battery is actually a dead battery. But on the other hand, your voltages are ok. I mean, you have half-ish the "full" voltage on the pins (but gnd). So they're ok. I'd use just wall warts dc adaptors.

On the noise, there are two things I noticed that can make them excessive:
1- not using an enclosure. Yeah, it makes a huge difference. Your circuit is exposed to lots of electromagnectical fields and, since it's a high-ish gain circuit, it'll amplify those em fields. As you noted, putting it closer to a pc gets it to be noisier. It may even getting affected by your amp transformers' em field. When you add a shielded enclosure, it'll protect the circuit from external em fields.
2- wiring. You're using long wires onn random positions. Once you put the circuit inside a shielded enclosure, it's usually the best idea to keep them as short as possible. Sometimes, it's also a nice idea to use shielding wire for the input (and gain pot, but you don't have one), because they can be exposed to em fields inside the box (from other wires and the circuit itself). And there are situations where you don't want to run some wires in parallel. But, for now, just trim them as short as possible after you put the circuit inside a shielded box.

I think those two things will solve your noise problem. Or at least will reduce it a lot.

And last but not least, are you using single coils or humbuckers?

[PRR]

> putting this in an enclosure

Here, I would not need a metal box if I went a half mile south. Into the woods. Away from all electricity.

(Until summer lightning storms.)

(Half mile north puts me *under* the cross-county 100kV power line and even a metal box may be problematic.)

Yes, don't judge a pedal wide-open in a room full of electric power buzz, or with a nearly-dead battery (6.8V should get you a refund).

[nerdbot]

And last but not least, are you using single coils or humbuckers?

A Zexcoil! It's supposed to sound like a single coil, but it's noiseless and has 6 individual coils.

Thank you all for the advice! I'm still a bit worried about the noise but am encouraged to move forward. For all the help provided, not just in this thread but the many others I looked through, I felt I should share my uneducated findings.

The TL072 has been the only op amp I've used in the circuit that doesn't gate. It is the noisiest, but probably for that reason. NC1458, NE5532, LM833, and even the JRC4558 recommended in schematics all exhibit gating, some worse than others. For these opamps I adjusted the 470k resistor in B to 1M, as I read this helps, but it didn't have a significant effect.

The 470k resistor in both A and B seem to adjust gain, but feel a bit different. I don't know why it happens, but to me, increasing the 470k in B resulted in more compression, which I actually liked when cleaning up the pedal with the volume knob, as there was a larger sweet-spot for a nice low gain sound.

After a couple days of testing, here's the choices I'll be going with:

• 1M biasing resistors - made it a little brighter, but did it because the tones from turning down your volume sounded much better.
• 22nf input cap - a little more bass
• 340k in A and 680k in B - to my ears, it is a similar amount of drive with a little more compression than when they were both 470k.
• BAT41 diodes (series on one side for asymmetrical clipping) - saw from someone's measurements here that they generally had a similar forward voltage and soft knee of 1N34As and I couldn't find those. Mine measured around 0.32. Also, I just like asymmetrical clipping
• Went back to the 100p cap after testing at volume as going lower was a bit too icepicky.

Lastly, do you think it's worth adding a capacitor from output A to the - input of B? I see this in nearly all dual op amp circuits and in other muff fuzz schematics. It seems a bit redundant to the input/output caps, but I figure there must be a good reason.

[iainpunk]

i bet that if you put that cap there, your gating will be way less, just give it a shot. if you want a more vintage buzz-tone sound (flatulent fuzz), add diodes to the first opamp, change R2 470k to 4M7 as well and use a tiny cap in between stages, like a 4n7. this gives buzzy flatulence, but really cuts through the mix.

im going to build this with some extremely slow opamps, and if my calculations are correct, it will be mid-range galore

cheers, Iain

[ElectricDruid]

What happens if you put C2/10uF on the bias supply back? Tweak the value? 2u2? 47uF?

+1 agree with Iainpunk about the 100n between the two stages. It'll reduce the offset on the second stage, so it might help with gating. Without it, you might well see more differences between different op-amps, since many audio op-amps have pretty poor DC performance and bad offsets. That's not really a fault, since they're designed for AC signals and expect to be AC coupled.

[nerdbot]

i bet that if you put that cap there, your gating will be way less, just give it a shot.

Sounds good. I'm planning to do another vero build without the sockets and I'll add that cap. Seems that it could be placed in either of these two circled positions (would move the 10k to the jumper if I were to replace it):

im going to build this with some extremely slow opamps, and if my calculations are correct, it will be mid-range galore

Sounds neat! I'm curious to see how that goes.

What happens if you put C2/10uF on the bias supply back? Tweak the value? 2u2? 47uF?

I still have the 10uf cap in there. I assumed this was a power filtering cap. I have a 100uf that I was going to use on my next build, but if that isn't power filtering, then I guess I should figure out what that does.

Thanks!

[iainpunk]

im going to build this with some extremely slow opamps, and if my calculations are correct, it will be mid-range galore

Sounds neat! I'm curious to see how that goes.

i am curious too! i'm gonna order them as soon as my salary comes! (always the 28th) MCP6042 dual opamp. its slew rate is 6600 times less than that of a TL072...

cheers, Iain

[Phend]

Noise...often bad cable or loose jacks. Of course those have been checked. Just a thought.

[nerdbot]

Going to be ordering an enclosure soon, but in the meantime I've been messing around more trying to see if I could reduce any of the noise and/or eliminate the gating that occurs with any of the other op amps I have. The TL072 works fine, but if other op amps don't work out, then I feel like something must be off with the circuit. Also, it'd be nice use different op amps.



The only parts I haven't adjusted at this point are R1, R4, C2, and C3, though I'm not too concerned about C3. I've added a 100nf cap behind R4; I've added a cap and resistor in parallel to the diodes; I've adjusted the resistor values in both feedback loops; and I've used 10k, 680k, and 1M for bias in R3 and R5. My op amp voltages look fine, but none of the changes have reduced the gating in a noticeable way.

My biggest question at this point is C2, mostly because I have no idea what it's for. I thought it was power filtering, but after closer inspection, I realized it's not. In other schematics of this circuit, and other similar circuits, this cap goes to ground. See the 0.1u here:



So what is C2 and is it worth messing with?


[In other news, I've ended up flip flopping and going with 1M in A and 100k in B for the feedback resistors. I'm pretty happy with the general sound of the pedal now.]

[PRR]

I think the way C2 is drawn is a mistake. (Rare for Beavis.)

[ElectricDruid]

I agree with Paul. I think the way C2 is drawn is a mistake and it should go down to ground, not up to 9V.

Its function is to make the "virtual ground" 4.5V bias point that you've created with R3/R5 act more like a genuine ground from the point of view of AC signals.

[nerdbot]

I agree with Paul. I think the way C2 is drawn is a mistake and it should go down to ground, not up to 9V.

Its function is to make the "virtual ground" 4.5V bias point that you've created with R3/R5 act more like a genuine ground from the point of view of AC signals.

Got it. Is it as simple as grounding that capacitor or will it's position have to change? I'm using the vero layout below and I can handle just grounding it (+ to pin 3, - to ground), but if it needs to be in a specific spot in relation to R3 and R5, I'm having a hard time figuring out where it should be relocated.