NPN silicon Fuzz Face delight

This project started with my curiosity about how Scott Asheton got his fuzz sound on the first Stooges album. I've been teaching my girlfriend how to play rock guitar, and the songs on that album are great for three-note power chord technique.

An old Usenet posting indicated that it was a Dallas Arbiter Fuzz
Face, so I did some research and saw that it was a pretty simple circuit, one that I could easily build myself. Further research (at the excellent Fuzz Central http://fuzzcentral.tripod.com/ and The
Technology of the Fuzz Face http://www.geofex.com/Article_Folders/fuzzface/fffram.htm) suggested that the Fuzz Face design that was current at the time The Stooges was recorded was the NPN silicon version.

I was also attracted to this design because it sounds like there were thermal problems with the PNP germanium transistors. The characterization of the NPN silicon version as "less mellow" sounded right up my alley. I've since listened to samples of the PNP 'Face, and while it of course sounds beautiful, I still like mine better.

I had originally hoped to find most of the parts at Radio Shack, but apparently they have cut back on their stock of electronic components since the last time I was interested in doing projects. Too bad. My local electronics emporium, Al Lasher's in Berkeley, had everything else that I needed except for the traditionally-impossible-to-buy-in-a-store DPDT footswitch. An NTE dealer, their substitution book indicated that the NTE123A was a functional equivalent for the specified BC109C transistors. I was suspicious, but figured I'd give it a shot, and if I didn't like how it sounded, I'd find some real BC109C's to swap in. The NTE123A is packaged in a nice little metal can with a tab for the emitter. I liked the retro factor in the looks, at least.

I built it on a small piece of plated-hole perfboard I got at Lasher's for a buck, and wired the pots and jacks. After a couple of (inevitable) wiring mistakes, I had just plugged my guitar in and gotten a fuzz sound when my girlfriend phoned from her recording session. I told her to listen, and blasted through a few bars of "I
Wanna Be Your Dog." Suitably impressed, she told the guitarist at the session to pick up the phone and listen, and he said that he loved the sound and wanted to build one of his own. Excellent! I had gotten immediate positive responses from both a cute girl and fellow musicians.

Given more time, I listened some more and was quite inspired by the sound. It made me want to play more, which is the best criterion for a new musical gadget. I compared it to the record, and it sounded pretty close to Asheton's tone (except I don't have a CryBaby....yet).
Further reading on the www indicated that I might be better off using a mylar film cap for the final .01uF output cap, instead of the ceramic I had first installed, so I swapped them and heard no difference.

The final piece of the puzzle was what kind of case I was going to put this in, because I wanted something round, like the original Fuzz
Face. Thinking that a food tin would be good, I took a late night trip to Walgreen's to scope out what goodies they might be selling in a large flat can. The Fuzz Fates (sorry) were smiling upon my quest, because I found the perfect enclosure in a Cavendish and Harvey Candy
Drops container. I bought a can of Sour Cherry and a can of Kiwi and
Guava. They came with replaceable metal lids, which was perfect for my intended use (the candy wasn't bad, either). The lip on the lid was just big enough to put in some sheet metal screws to secure it, and I think it's gig-worthy.

I've since built 2 more of them. One as a birthday present for my girlfriend (lightning struck again; it sounds great), and another as a sort of testbed, because I wanted to try different components to see what effect they'd have on the tone.

First thing, I substituted tantalum caps for the electrolytics, and
2N3904's for the NTE123A's. Yuck. What I got was a harsh, nasal, unpleasant fuzz. It didn't sound that different as far as overall frequency content, but I just couldn't stand to listen to it for more than a minute or two. I found another pair of NTE123A's, this time
NTE123AP's, cheaper and in plastic packaging, swapped them in, and got the same tone that I had enjoyed from the earlier ones. Next I want to try 2N2222's. Just to see what happens. And I still want to get my hands on a pair of BC109C's if I can.

Thanks to Phillip Bryant and R.G. Keen for the excellent information regarding this classic circuit.

Sounds like you are on your way!

I have some BC109C work-alikes (U.S.-made transistors tested for me to the spec of the European devices) as well as stomp switches and other things you might want, at better prices than most retailers.

I now regret not ordering from you; I phoned Antique Electronics (who've been fine in the past) and ordered a few of the $5 knockoff
DPDT's (which I believe you sell as well?) along with some 12AX7EH's for my Mesa Studio Preamp.

Got an order number and everything, then waited for a week and called back to check. Their computer system had eaten the order, so they placed it again. The damn switches are STILL not here and my girlfriend's birthday was 3 weeks ago (I bought her some other presents). Antique Electronics said that they'd send me a Jensen
T-shirt as compensation, but I won't be surprised if the guy forgot it. He was pretty mystified that I had an order number that didn't show up in the system.

Anyway, I saw the "BC109C work-alikes" on your site and they piqued my curiosity. I've downloaded transistor spec sheets and combed them to see what the possible differences were among NTE123's, BC109's and
2N3904's, and I couldn't find anything that stood out. That's why I went ahead with building a 2N3904-based one.

It sounded like crap, and I don't know enough to know why. The NTE's sing; the 2N3904's grate. I figured that it was possible that rather than differences in Hfe (or other specs), maybe the different models of transistor just distort differently.

I imagine that the specs you test to determine your "work-alikes" are a trade secret. I am curious as to whether the NPN silicon Fuzz Face circuit is as sensitive to gain variation as the PNP germanium seems to be. The three of them that I built (one of them using NTE123's in plastic) all sound pretty much the same.

I'm going to see what my test one sounds like with 2N2222's in it.

While I was looking about, I couldn't find any indication in the DIY effects literature on the web as to the effect of different cap construction on audio performance. When I was working with engineers, they all loved tantalums vs. aluminum electrolytics (they said that the tantalums last longer and are more stable), but nobody in the DIY community seems to use them.

The first thing I would do is stick those NTE devices into a decent
DVM with an Hfe scale. That will give you some idea of gain, a figure that you can use when looking for other types that have the sound you want.

Nope. If I ask my vendor for a "BC109x work-alike" they sort and test to the original spec for that device.

I am curious as to whether the NPN silicon Fuzz Face

Can't tell you much, because all my experience is with germanium.
"Success" has left me too little time to experiment.

The best info I have (from R. G. Keen and others) is that tantalum caps will sound dirtier than comparable aluminum electrolytics. YMMV.