Aion Flare Kit - Beautiful Fuzz - Awful Squeal

[Mistborne]

Hi all,

Just finished my Aion Electronics Flare (Fuzz Factory) and plugged it in. Pedal works beautifully - wonderful fuzz and sustain.

BUT - there is a high-pitched squeal present, and being someone an amateur, I'm not sure what may be causing it. Any suggestions on things I should check for?

I followed the Aion Documentation/schematic with no mods. It can be found here: https://www.dropbox.com/s/wo40kkfnb9gug1l/aion-flare-fuzz-factory-documentation-v1.1.pdf?dl=1

Thanks to any who can help this pedal rookie.

[antonis]

Start populating transistors (from in to out) with 47 - 100 pF caps between Collectors and Bases..
(56 - 68 pF should be most reasonable values but they are strongly depended on particular transistors..)

If you manage to stop squealing with only one cap and simultaneously lower highs gain, try to "split" it in more stages with the use of lower value caps..

[Mistborne]

Typically this is done on Q2 IIRC?

[Mistborne]

For what it's worth I am using germanium AC128s for Q2 and Q3, and not silicon transistors

[antonis]

Typically this is done on Q2 IIRC?

"Typically" this is done on 2nd stage of a 3 stage audio amplifier (VAS dominant-pole compensation) but fuzzes are (intentionally) far away from Good Designer Practice implementation.. 

For high pitch oscillation, NFB caps are the easiest/quickest/cheapest cure..

[Philo]

I'm having the same issue with the Flare kit.  High pitch oscillation and glitching, even with the controls set to the default settings in the manual.

A 100pf cap across Q2 base-collector and a 47pf across the Q3 base collector helped a lot.

Next step may be a cap across the feedback resistor (R3, 47K).

Advice elsewhere suggests this might range from 200-4700pf, best to start low and work up until the squeal is eliminated.

Even with just the Q2 and Q3 caps, the pedal is much more useable, though still noisy.

I also see the issue that the soft control doesn't seem to make much if any difference.
I have not tried to put a base-collector cap on Q1, the input boost for the main circuit.

This thread was very useful in pointing toward a solution. 

[Philo]

I wound up using 100 pf capacitors across the base-collector pairs of Q2 and Q3, and 150 pf across the 47K feedback resistor (R3).

This seemed to provide the best balance of stability and treble response given my Q2 and Q3 transistors.
A 350 pf on R3 (nominally 470 pf!) gave more stability and less noise, but at the cost of significant less treble response.  Too much of a sacrifice for me.  Didn't try 200 pf.

I soldered the capacitors on the component side of the PCB, directly on the transistor leads and the wires of R3.

FWIW, Q2 was 67 gain (51 leakage); Q3 was 102 gain (47 leakage).  Q3 seems a little lower than optimal.

Obviously the value have to be tuned to the transistors and the desired stability/treble response; I used capacitors I had on hand.

[Philo]

An update for the record, since my previous post had some wrong info.

I clearly can't read capacitor values, having confused a 103 (.01 uf/10,000 pf) value for a 101 (.0001/100pf).

As a result, I had WAY too much capacitance on Q2, Q3, and the feedback network, which suppressed gain dramatically.

I think my original problem was using a power supply designed for a digital pedal, which generated a loud ticking oscillation when powering the flare.

Anyway, I pulled all that out and just put a 102 (.001uf/1,000 pf) across the feedback resistor (R3) and switched to a OneSpot power supply.

Now there's a lot more gain, and I only get the ticking sound if I maximize the drive control (which I don't have to do, because there's enough gain at lower settings).

This seems workable.

I still don't hear much impact at all from the tone and soft controls.

[Mark Hammer]

Never used one, but my understanding was that this was a feature, rather than a bug, and could be produced, at will. via the Stab (stability) control, IIRC.

[eh la bas ma]

I found a real Zvex Fuzz Factory on the used market few years back.

The person who sold it to me thought it was broken too...How lucky for me !!

It squeals, that's the spirit ! "Awful" ? Of course ! This is the real fuzz spirit unleashed, the real crazy thing. Enjoy !

you can incorporate squealings and control the squealing pitch. All controls help to fine tune and get a hard noise in A or D, or E...

It's very fun to incorporate such noises in your guitar playing. Play with the squeals. It's kind of beautiful.

I like to play with modulations and deep reverbs on that squealing. It's a good starting point for some sound experiments... like a drone.

You can either shut it down completely by increasing gate and comp, or with stab fully CW. A noisegate works too.

I guess it's called factory because you can dial almost any fuzz sound with this circuit. Some lists of settings are available on the net :

https://www.thegearpage.net/board/index.php?threads/my-take-on-a-zvex-fuzz-factory-settings-list.176319/

[Philo]

I understand the squeal is a feature, but the loud clock noise like ticking is not!

[eh la bas ma]

I understand the squeal is a feature, but the loud clock noise like ticking is not!

This circuit is sensitive to what is connected before it. Squealing can disappear for some reason when it's at the end of an effect chain, for example.

I guess you still have the issue with the guitar directly plugged in the fuzz, and fuzz to the amp ?

I use onespot too, and I don't get any clock noises. Might be the electrical configuration at your home... or an issue with the build.

I know many circuits can get noisy when there are many effects, several power supplies on power srip and such. I wouldn't be surprise if this circuit is very sensitive to that kind of things.

[Philo]

I was originally testing on a desk near a computer, an Apollo twin, and a Fractal FM3 and with a Source Audio wall wart.  That makes for a noisy environment, electronically speaking. 

Seems to be doing better with a one spot and in a guitar, Flare, amp configuration.

[Philo]

Got some quality time with this build in the basement this morning.

Using a one-spot power supply, and away from the computers, the clock noise sound is much less and only appears with the drive pedal fully dimed.
With the way excessive feedback caps I had to dime it to get enough gain; with those gone I don't.
So the solution is simple--don't dime the drive control!

I may remove the remaining cap in the feedback circuit to see how the pedal does in the basement without it.

Plenty of gain and solid fuzz sounds just working the volume and drive knobs.
The tone control (added to the original design by Aeon) does add nice treble shaping at the end.
The soft control (also added) doesn't seem to do anything audible.

When I dip into the "factory" knobs, they produce ripping noises and squeals aplenty, which is what they're supposed to do.

Also tried it out in comparison to my silicon fuzz face built from a kit.
Both had some nice and similar sounds, but I liked the core tone of the (germanium) Flare better.

Thanks for the help here, and sorry to have steered things off course.

[anotherjim]

I'd try adding a resistor in series with +9v power input. It will make the power filtering capacitors work better and it will be more tolerant of lower-quality power supplies. This is a major omission on many FX PCBs since if it isn't on the board, where do you find a good & secure place to fit it? In this case, it looks like the resistor would replace the +9v link from DC jack to PCB, but not being too flexible, it's not ideal.
The series resistor will drop some voltage to the circuit, but this should require little current and anything from 100R to 1k will work, the higher the better. *BUT* the LED feed would be better if it was from before any series resistor since the LED current will cause more voltage drop over the resistor. Then I always find 4k7 LED resistors are too bright and would fit 6k8 which will reduce the volt drop.
A bonus is it will reduce the fault current if a reverse polarity power is used and causes the protection diode to conduct - and maybe stop anything more valuable than the resistor getting damaged if a high current supply was used.

...and, did you try battery power? That's a go-to test with weird noise/stability issues.

[Philo]

The Aeon kits use headers to convey the wires from board to board, so there isn't a good way to add a resistor in series with the 9 volt in.

In any case, this now seems to be working well.  Good basic sound with tonal variations in the stable region, and lots of ways to use the other knobs to creatively destabilize the sound.

I kept the 1000 pf cap in the feedback circuit.

So long as I don't totally dime the gain, the pedal is stable when I want it to be. 

[sopapo]

This is something that I dont understand, why the schematic or diy layouts often omit the resistor in a dc filter?? knowing that is more effcient this way.... they are too much aprehensive to voltage dropping??
I know that with the  quality of the modern power supplys this is not much of a concern, and I can understand it a little in the case of commercial makers (one resistor save ) but I think a 50 ohm and 220uf is a safe bet for a lot of design..not too much thinking in this...my opinion..

[ElectricDruid]

This is something that I dont understand, why the schematic or diy layouts often omit the resistor in a dc filter?? knowing that is more effcient this way.... they are too much aprehensive to voltage dropping??
I know that with the  quality of the modern power supplys this is not much of a concern, and I can understand it a little in the case of commercial makers (one resistor save ) but I think a 50 ohm and 220uf is a safe bet for a lot of design..not too much thinking in this...my opinion..

I totally agree. If you're going to put a filter capacitor on the power supply, then you need a resistor to make it into a "filter" and not just a "capacitor"! Of course, there's always *some* resistance, but we assume our wires are made of decent stuff and our solder joints are good, so that intrinsic resistance should be negligible if we did it right. So much easier to put 47R in there as well to make sure. If the volt drop is killing you, redesign the circuit to use less juice, or use a bigger cap and smaller R. It's not hard. Like you said Sopapo, not too much thinking in this.

[FSFX]

... I know that with the  quality of the modern power supplys this is not much of a concern ..

It is actually the opposite of that - it is a big concern as the source impedance (ESR) of modern power supplies is so low that the so called 'filter' capacitor is not providing any filtering at all without that additional series resistor in the power feed.

[sopapo]

Electric druid, yes you are very right, you can even put a 25r resistor and a 460 uf, this electros still can be small enough.... Any ways, the voltage drop in most gain effects is negligible...
FSFX, why is so low the esr of the modern supplys vs the old ones? its something related to certain conmuted supplys that are being using more and more??
regards