The Ministry of Tinkering

Wednesday 6-5-2013

Poking around for more RCA 2N270 transistors, I sent a RFQ to one of the online electronics clearing houses. Ummm…I think $175 a piece might be a tad beyond my pain threshold.

2n270wow

Wednesday 5-29-2013

Have been back tinkering with my Fz-1a recently. While it worked, it didn’t work as well as it could. The fuzz knob had little range and the overall volume was barely unity at max. I knew the volume problem was because the collector of Q3 was near full supply voltage (I think that’s referred to as saturation), so not a lot of current was being passed to the output. If I could get that down nearer half the supply voltage…around 0.75 volts….I would have a pretty big increase in current flowing through the collector and a much louder pedal. I did not, however, want to swap transistors out or change resistors…or do any intrusive modifications at all, really.

I pondered this a while and then remembered I might be able to force it into proper bias if I tied a resistor from the power rail to the base of Q3. It seemed, studying lots of schematics in the standard transistor books of the 50’s and 60’s, that the rule of thumb was to make that resistor about ten times the value of the the one from base to ground. So, I wound a 100K resistor in from the power supply point on the board to the base wire of Q3. Instantly my collector voltage for Q3 dropped to just a hair under 0.7 volts and I had a HUGE increase in volume. Success!

I knew from experiments with the Fz-1 that the 470K resistor from power supply rail to the base of Q2 was just too large to allow effective bias adjustment with the Attack knob. In those experiments I found that you had to have something over 1000µA leakage in Q2 to start to get any effective adjustability out of Q2! Lowering that 470K to something in the neighborhood of 180K made the Attack knob actually useful. I wound a 470K resistor in parallel with the original (effectively making it a 235K resistor) and it gave the Attack knob a pretty decent range. I may try replacing that with a 180K so I can get closer to 130K total resistance. I took notes for the collector voltage readings of Q2 at minimum and maximum attack before and after this mod, but I can’t seem to lay my hands on them at the moment.

The end result is my Fz-1a just kicks ass now! Tons of volume on tap and a Attack knob that goes from fairly mild fuzz to all out sonic mayhem. All from simply twisting two resistors into the original circuit, a totally unobtrusive and reversible modification. I can’t help but think there are quite a few Maestro Fz-1a Fuzztones out there that would benefit from something similar. I’ve tested a few 2N2613 and 2N2614 transistors from this time frame and they generally had quite high gain and moderate leakage, but by in large it’s uncommon to find them with enough leakage to make both Q2 and Q3 operate at their best in the circuit as it’s designed. The same can definitely be said for the Fz-1, though it’s much more likely the 2N270’s they used in those pedals would have leakages closer to optimal for that circuit. I can say that most of the sound clips I’ve heard of original  Fz-1 and Fz-1a pedals you can hear the real beast hidden inside struggling to get out. A properly tuned one is a real blast to play and much more versatile than people give it credit for.

Here are a couple crappy cellphone camera gut shots of mine as it sits now:

IMAG0164 IMAG0163

Friday 5-10-2013

Few quick things. Here’s what I’ve been reading recently…little bits at a time. It’s from 1966 and quite interesting in that it deals with germanium transistors in depth.:

IMAG0104c

Also, I got this kit for a DIY capacitor meter https://www.sparkfun.com/products/9485

It’s quite nice, was easy to assemble and is more than accurate enough for pedal building or even amp building. Highly recommended, especially for $13.95

IMAG0110c

 

Have also been busy with the breadboard learning, testing measuring and…of course…building some pedals. New one listed today and more to come fairly soon. Plus some more updated clips as I keep promising!

Friday 4-26-2013

A very happy confluence of events for me. Scrimping, saving and taking my lunch every day finally paid off when this little monster went at auction on ebay right when I had enough spare cash to be the winning bidder. This is a Gibson Maestro Fz-1a Fuzz Tone.

Right Front Fz-1a. Original knobs?

Right Front Fz-1a. Original knobs?

Is this the second ever fuzz pedal made? I really don’t know the exact time line of all the early fuzzes, but it’s got to be close to number two. You may or may not know the story…the first fuzz was the Gibson Maestro Fz-1, released sometime in the 1961/62 timeframe. They didn’t sell well and by 1963 or so they where dropped from production. Allegedly not much more than 5000 where made…which is a lot and isn’t a lot, if you follow me. They are fairly rare today and nice examples sell for thousands of dollars.

Right Front Fz-1a. What's the jack doing there?

Right Front Fz-1a. What’s the jack doing there?

For all intents and purposes, fuzz was dead…but then something cool happened. Keith Richards heard a catchy riff in a dream and out of it was born a song. He and the Rolling Stones cut a demo where Keith used a Fz-1 Fuzz Tone with his guitar to stand in for what was going to turn out to be a saxophone part. But something about the demo was cool with the end result being the saxophone never got used…and “Satisfaction” became a hit….and suddenly fuzz was NOT dead…which resulted in Gibson getting orders for a pedal they no longer made. Thus the Fz-1a.

Would this mean there where 6498 Fz-1a's built before this one?

Would this mean there where 6498 Fz-1a’s built before this one?

Mine is quite “reliced”…that’s “baby boomer” for beat to shit….serial # 6499, which I think may make it a pretty early Fz-1a. The date code on the Attack/On-Off pot is the 38th week of 1965, so it likely was made not too long after that. Those with a keen eye will also notice it’s not stock. Which is one of the reasons I was able to afford it, I’m sure..the other being it was advertised as not working. The mod is pretty nifty in my book…somewhere along the way somebody took that stupid guitar cable out of it and put a nice Switchcraft  1/4″ mono open frame input jack in the side instead. This allows you to plug what ever cable you want between the guitar and fuzz pedal and you don’t have to deal with a cable hanging out of the side when you want to move it around from place to place. Other than that, it’s all stock with the original 2N2614 RCA transistors, carbon comp resistors, ceramic input and output caps, etc. The two electrolytic interstage coupling caps (those black cylinders with one red and one yellow end) are also still stock…which piqued my curiosity vis-a-vis it’s not working anymore…could the fix be that easy?

Gut shot!

Gut shot!

It arrived yesterday and I had a little time between getting home from work and needing to leave for rehearsal, so I plugged it up after installing a fresh AA battery. Not functioning was correct….you could barely hear a fuzz tone with everything on the pedal cranked up, but it was very weak. The bypassed signal was fine though. Hmm. Got my volt meter out so I could see the voltages on the transistors. It was still pretty cold from having sat outside a good portion of the day, but here’s what I found (Attack pot at maximum):

 

Not bad...and a Switchcraft just like the output jack.

Not bad…and a Switchcraft just like the output jack.

Q1E -.392

Q1B -.435

Q1C -1.527 (same as battery)

Q2E   0

Q2B -64.5 mV

Q2C -1.09

Q3E   0

Q3B -10.4mV

Q3C -1.467

With the Attack pot at minimum, the Q2C was -1.468 volts.

So, the emitter voltage on the first stage is on the low side, the voltage range of the second stage is not bad and the collector voltage on the third stage is a bit high, but it should still be functioning. Obviously the problem lay elsewhere. That’s as far as I got till I could turn my attention back to it this morning before work. I dug out my audio probe and started tracing the signal path. Everything was fine up to Q1 and the signal coming off the emitter of Q1 was as it should be, just slightly below unity gain. The signal was getting through the first coupling stage fine and was strong going into and coming out of Q2, but as soon as it hit that second coupling capacitor it choked off to almost nothing. I don’t keep much in the way of axial caps around, so I grabbed a radial 1µF and physically twisted it’s leads around the wires of that second coupling cap so it was in parallel (no solder). That did the trick and gave me strong signal all the way to the output.

"The Fix!"

“The Fix!”

I wanted to post the new voltage readings now that it’s warmed up and functioning more or less properly, but it sounds so damn good I just kept playing guitar before I had to leave for work. I will update when I can. The Attack knob has some decent functionality and it get’s pretty fuzzy at max setting, though I suspect if I where to drop that Q3 collector voltage down closer to around 1 volt it would get even crazier. Unity seems to be around 6-7 on the volume knob, so there’s plenty of volume on tap. Both pots still function fine, the jacks are still great as is the foot switch. I need to decide how far I need to go in this repair…even though the pedal has been modified, I hesitate to do much more to it…do I remove and replace those electrolytics? Or just the bad one? Do I replace Q3 with a period correct NOS 2N2614 I have that has just the right leakage to get the collector voltage in range? I dunno…it’s very tempting to tune it properly, but it’s also tempting to stop right where I am and enjoy it for what it is.

 

Another angle of "The Fix".

Another angle of “The Fix”.

I may try to get a clip up soon too…

One thought on “The Ministry of Tinkering

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>