Static sound (Fizz?) in the decay of chords

[drchek]

I just finished a build of the Sweet Honey Overdrive and I'm getting a static buzz as the sound decays. If you listen to the sample I recorded you'll hear that the static sound cuts out when the decay gets below a certain volume. This leads me to believe it's one of the sets of clipping diodes. Is this what causes that? Any guess as to which ones are the likely culprits (the 1N4007s I'd guess...)?

Sound Sample: here

[Quackzed]

sounds like it could be clipping the rails of one of the opamps... opamps can really square off the peaks and this adds alot of high frequency/ high order harmonics that can sound fizzy, or like static. you might try another led of opposite polarity with d5 ,same back to back setup as d3 and d4, if that solves it, you know what your fighting with and you can add another led in series with the new one, see if it crops up again if so add a silicone in series with the new led instead, then a germanium etc. the idea being to adjust the threshold for that side of the signal so its juuuust below the opamp fizzy clipping point... or if you like the sound of the back to back leds in the second stage, you're done.
but it'd be easy to see if thats the problem with another led , back to back with d5...

[PRR]

Check the DC voltage at each op-amp output. In this circuit it should be very close to 4.5V.

A way-off opamp bias may "work" on big signals (such as when the distorter is working full) but garble small signals.

[drchek]

I did some audio probing and the buzz is present at the output of the first opamp (pin 1). I checked the voltages and everything looks fine, but I did notice that the voltage at pin 8 (high supply voltage) is ~8V due to the polarity protection diode D2. If I jumper that I'll gain an extra 1V of headroom on the opamps. I'm hoping that will help when I try it out later tonight.

Good idea?

[Cozybuilder]

Change D2 from a 1N4007 (silicon) to a 1N5817 (Schottky), and gain back most of the dropped voltage, while maintaining reverse polarity protection.

[drchek]

Change D2 from a 1N4007 (silicon) to a 1N5817 (Schottky), and gain back most of the dropped voltage, while maintaining reverse polarity protection.

Will try this tonight...

[PRR]

Fizz on decay is not from ~~10% drop of supply voltage.

[Groovenut]

I did some audio probing and the buzz is present at the output of the first opamp (pin 1).

D3 & D4 are supposed to prevent rail clipping at the first opamp from happening (which could be causing the buzz). What color LEDs are you using here and what is the Vf?

[dschwartz]

I hate when that happens..
What intrigues me is that the fizz is present when the signal is pretty small..the clipping fades out and then the ugly fizz appears.. That could mean that there's crossover distortion maybe..

Have you tried it removing the "focus" feedback part? Feedback can make weird things..and sharpening clipping is one of those things

Other thing to check is hf oscillation..,try increasing the value of the 100pf cap..

Also check your soldering ..maybe there's a cold joint...

And at last..try other opamp..maybe you got a lemon

[drchek]

And at last..try other opamp..maybe you got a lemon

BINGO!

After spending all night with it, turns out the batch of TL072s I got cheap from China are crap. I switched the IC out with a known good Texas Instruments chip from another pedal and the sound is perfect!

[dschwartz]

I got the same crappy fizz when breadboarding my last design (which has a similar gain stage) tried everything to minimize it to no avail..

Then i built the prototype with TI smd 072's i got from mouser and the fizz was gone..yup..lemons everywhere

[Kipper4]

It seems the op275 and tl072 might be very differant beasts

Op275 data.
http://www.analog.com/media/en/technical-documentation/data-sheets/OP275.pdf

Tl072 data
http://www.ti.com/lit/ds/slos080m/slos080m.pdf

Then there's this......
https://www.muffwiggler.com/forum/viewtopic.php?p=482993

The sweet honey with the op275 is one of my favourite pedals.
Dare I say I'd choose it over a tube-screamer most days.

[drchek]

To confuse things more..

Actually the bad batch from China are TI chips too, but the number on the chip is different from the working chips I got from a US dealer:

Static Chips from China: 


Clear Chips from US: 


Whats the meaning of the top number on the chip? I can find no reference to it online anywhere. I had 10 chips of each one. ALL 10 of the ones from China have static, ALL 10 of the ones from the US are clear. What gives?

[dschwartz]

Probably they are from a defective batch. For IC's always use serious dealers

[Cozybuilder]

Did you try the OP275 in this circuit? As Kipper noted, its not the same as a TL072.

[Kipper4]

Very diferent bandwith specs and the op275 has a hybrid input...

[drchek]

I don't have a OP275, but my other batch of TL072s don't have the static. Clear now that it was a bad batch of chips, not the spec difference.

[duck_arse]

the "4AA4DOM" is the manufacturers tracking code. some makers even publish the marking meanings in their datasheets. it will include a date code, and a factory or foundry code, or simply a batch number that references a huge list of what country and when the die was made, and what country and when it was packaged. and other stuff.