Death By Audio - Supersonic Fuzz Gun

hardstatic · Post by **hardstatic** » 28 Apr 2012, 02:22

I desoldered my buddies original and measured the trannies (don't say anything):

Fairchild BH40 2n2089 - HFE measured 620ish. Vbe 0.8V, Ib 4.60mA.
Fairchild BJ19 2n5309 - HFE - wait for it - 21,369. Vbe 1.45V, Ib 3.89mA.

I measured my Smallbear 5306's and they are in the HFE 10,000-16,000 range. My Tayda 5306's measure in the HFE 60,000 range.

Plugged everything back into the original and it fires up as normal (phew!)

muzboz · Post by **muzboz** » 28 Apr 2012, 03:07

Now THAT is interesting! So what are you saying?

We should look to buy some 5309's instead of 5306's?

I'd LOVE to get this thing working properly!

Might be getting somewhere now!

- Murray

okgb · Post by **okgb** » 28 Apr 2012, 03:18

Fuzz newbie question , but how does this compare to the fuzz factory ?

hardstatic · Post by **hardstatic** » 28 Apr 2012, 04:43

Of course I meant 5089, not 2089. And it's supposed to say 5306... good grief my typing. What this seems to indicate is the 5306 transistor is an inconsistent asshole and for this circuit we should be looking for very specific 5306's with a gain in the 20k range.

muzboz · Post by **muzboz** » 28 Apr 2012, 06:26

Haha, OK, glad you got some hard stats off the original pedal, anyway, HardStatic!

How do you measure the HFE of your 2N5306's? Do you just use the NPN slots on an HFE-capable multimeter?

If I plug mine into my multimeter in the NPN slots, I have to wiggle the transistor around the get a reading, and it jumps all over the place.
Doesn't seem to fit neatly into the slots. (When I've done HFE readings in the past, it's been relatively easy!)

Sometimes it jumps up to around 1924, and that seems to be the highest number I've seen. At other times, it just says 0, or 1, or -1. Often it flicks between number like 581, 497, whatever.

Would a reading of 1924 mean 19k?

Those are the ones I bought from MusikDing. I don't seem to be able to find a datasheet on their site, unfortunately!

I also have some from Tayda, and some from Futurlec, which went into my original build, but they even have a different leg order!

JustinFun · Post by **JustinFun** » 28 Apr 2012, 07:18

muzboz wrote:
How do you measure the HFE of your 2N5306's? Do you just use the NPN slots on an HFE-capable multimeter?

If I plug mine into my multimeter in the NPN slots, I have to wiggle the transistor around the get a reading, and it jumps all over the place.
Doesn't seem to fit neatly into the slots. (When I've done HFE readings in the past, it's been relatively easy!)

Sometimes it jumps up to around 1924, and that seems to be the highest number I've seen. At other times, it just says 0, or 1, or -1. Often it flicks between number like 581, 497, whatever.

I never have much luck measuring hfe on my multimeter.

If you can spare the cash, you will never ever regret buying one of these: http://www.peakelec.co.uk/acatalog/jz_dca55.html

IvIark · Post by **IvIark** » 28 Apr 2012, 10:39

Yes transistor measurement in the low cost meters is notoriously crap. The Peak is one of the best things I've bought, and a must for accurately measuring transistors, particularly germaniums. You don't even need to know the pinouts, it identifies everything for you and tells you what type it is.

hardstatic · Post by **hardstatic** » 28 Apr 2012, 14:06

IvIark wrote:Yes transistor measurement in the low cost meters is notoriously crap. The Peak is one of the best things I've bought, and a must for accurately measuring transistors, particularly germaniums. You don't even need to know the pinouts, it identifies everything for you and tells you what type it is.

Ditto. I gave up with the breadboard method and the cheapo multimeter - ordered my Peak DCA55 a few weeks back. For this hobby it's the best money I have spent on a tool since my Hakko iron.

muzboz · Post by **muzboz** » 28 Apr 2012, 14:29

Dang. Well another thing to buy, I guess.

In the meantime, can anyone tell me if they have any 2N5306's with a roughly 20k measurement, and where they got them?

I guess each batch canner differ (?), but does anyone have a spec sheet for one within the right range, I might just buy some...

Also, do you think that reading could be the key to a pedal that doesn't work properly, and one that does? Or is the difference not likely to be so great as that?

My pedal I built has always made SOUND, from day, one, but it just seems to be "tuned" all wrong, in terms of how much sound gets through...

- Murray

roseblood11 · Post by **roseblood11** » 28 Apr 2012, 20:13

@glass_hero: I think the filter pot is a 10k log:
download/file.php?id=14370&mode=view
in your schematic it's linear...

roseblood11 · Post by **roseblood11** » 02 May 2012, 14:09

This is a corrected version of my layout, with the changed pot orientation, as pointed out by hardstatic. I built it a few days ago, so it's verified.

muzboz · Post by **muzboz** » 02 May 2012, 15:29

Excellent! Did you need to make any other changes, or just the switch in pot type?

Could you make us a little video, or MP3 of it in action on various settings?

Great work!

- Murray

muzboz · Post by **muzboz** » 02 May 2012, 15:45

Groovy design.

And so much more compact than the original DBA pedal.

Jarno · Post by **Jarno** » 02 May 2012, 17:40

did anyone check the datasheet on that high hfe transistor? probably a darlington, if so, you could piggyback two transistors to get exactly the right hfe.

roseblood11 · Post by **roseblood11** » 02 May 2012, 19:43

muzboz wrote:Groovy design. And so much more compact than the original DBA pedal.

I'm not sure if that was a good idea... If the switch is set to oscillation mode, the circuit bleeds into the bypass signal.

muzboz · Post by **muzboz** » 03 May 2012, 02:49

Oh, that sounds annoying. Does it work effectively apart form that?

The only pedal I've built successfully was a one knob fuzz, that I then added 2 additional knobs.
I built it on breadboard, then created a layout, and built a new one into a case.

They sound quite different from one another! The overall functionality is the same, but the subtleties are quite noticeable.
My breadboard version, with longer wires that zig zag around the place, has a nice sort of "interference" to the sound. A kind of fizzle that's not present in my final built (which was built more efficiently).

It inspired my to make further alterations, and I added a voltage starve knob (to try to simulate the "inefficiency" of my breadboard pedal's excessive wiring!). It adds a nice range of sounds, but doesn't really emulate the breadboard model still.

This is probably not news to you guys who have built many pedals, but it has made me aware of how small differences in layouts, components, wiring, etc, all make noticeable changes to the final sounds the pedal makes.

And this Supersonic Fuzz Gun seems to have plenty of it's own little quirks to defeat those who would harness it's power!

- Murray

roseblood11 · Post by **roseblood11** » 03 May 2012, 07:22

I think, from an engineers standpoint the SSFG simply is a bad design. It's highly dependent from hfe values, everything interacts with everything, the oscillation is hardly controllable... But on the other hand, its crappy sounds can be interesting in some situations - that's why I don't care what the engineers think.

One solution for my problem with signal bleeding into the bypass could be to switch the circuit into gated mode, when it's in bypass.

Apart from that, it sounds as expected. The oscillation is better in some of the demo video, but I think I can improve that when I get better trannies (I only had 2n5088's with hfe=450, maybe that's a bit too low).

roseblood11 · Post by **roseblood11** » 03 May 2012, 13:43

I just realized that the oscillation is different from the demos. Listen to this video: (I know that he uses delay and reverb, I tried that as well...)

He plays the chord, mutes the strings and then the oscillation starts, with the frequency going up. With my pedal, the oscillation is always there, even if I play a chord, and the frequency doesn't rise as well, it stays the same. And the frequency of the oscillation always stays too low, I would have to turn the guitars volume knob to reach that note (sounds like a D flat).

I had a look at the picture in the first post, and there is no 10k resistor at all. It looks like 1k or 3k. Maybe R5 has to be tweaked to work with the pair of transistors?

muzboz · Post by **muzboz** » 03 May 2012, 15:03

Is that a 10k in the bottom left corner near the LED in this photo?
https://www.freestompboxes.org/download/ ... &mode=view

Or is it a 1k? I was umm-ing and err-ing over those a bit when scrutinising the images initially posted by Glass Hero.

- Muz

roseblood11 · Post by **roseblood11** » 03 May 2012, 15:25

That one is definitely 10k. Glass_hero and hardstatic aren't stupid...

Death By Audio - Supersonic Fuzz Gun [traced]