Ohhhhh Kay!!

[Mark Hammer]

Having played around with mine for a little while now, I think the use of 10uf electros may retain a little too much bass for the circuit's own good.  This is especially true when using single-coils in any non-hum-rejecting arrangement (e.g., one at a time).  I know I recommended subbing 10uf for the 100nf cap to ground on the emitter of Q1, but I will now suggest it is problematic in some situations.  Increasing that cap to the 1uf range is fine, but larger values (like 10uf) introduce enough gain to the low end, that any 60-hz hum bleeding into the input of the circuit produces some rather annoying sideband products.

Conceivably, not going much higher than 1uf for the Q1 emitter cap, and possibly dropping the values of the caps on each side of the diodes, will let enough mids and lows through without introducing too much of the sloppy stuff.  certainly use of hum-rejecting pickups sidesteps that issue, but for the most universally-usable circuit, keeping the problematic bass out seems like a good strategy.

[pinkjimiphoton]

Daniel Lanois is in town tomorrow, and that got me interested in the Kay fuzz.  The Kay is essentially a Superfuzz, with a simpler front end, no variable drive, no output level control, and no gain recovery stage after the mid scoop.

Looking at it, I thought it might benefit from a few things.  First, the emitter bypass cap on the input stage is kind of small at 100nf.  I installed a 3-way toggle that patched in a 680nf or 10uf in parallel with the 100nf.

Next, I stole a trick from ZVex and inserted a 10k Lin pot between the diodes and ground.

Next, I stuck a 10k pot between the 100nf cap in the mid scoop and ground.

The  50k foot treadle pot was replaced with 100k and the 200k ground resistor replaced with 150k.

The 3way toggle is KILLER.  Feel free to experiment with other values but there is no reason to go beyond 10uf.  And at  10uf, the octave blooms beautifully, like a feedback simulator.

The diode pot helps soften the clip and vary the degree of growl vs sizzle.  Useful.  Just note that reducing the clipping action raises the volume.

The variable scoop works as expected, but interacts with both the foot treadle control and the other installed options...in a good way.  Because these options alter the volume level, I'll swap out the 100k output resistor for a 100k volume pot.

The Kay fuzz is a very nice unit.  But this....this is better.  Much better.  Jimi, I hear your soldering iron calling.

'



bastid. now i gotta revisit this circuit again

thanks mark

[Mark Hammer]

Hey Jimi,

The scoop and clip pots can probably be as low as 2k and still do everything they need to do.  And in retrospect, Kay made a good call by making the foot treadle pot 50k, rather than the 100k I used.

Note that the variable clip can work on *anything* using a diode pair to ground.

[pinkjimiphoton]

thanks for the heads up mark!!!!!!!!!!!!!

look forward to getting back to some build time soon.... a lot of jobs built up over the move period,
and i've gotten guitaraholic.... 35?36? hard to tell... it seems like i sell one and buy 4 more.

i've revistied some of my last designs, you may like the miss kay amp sim/overdrive/fuzz mark. it's a small kay vanguard all ge amp converted to
a stompbox.

it kinda does what the suzy q does, but sounds like an old transistor amp more. you may like it.

but the flying spaghetti monster with the ge fuzz as a switchable diode clipper/fuzz in the feedback path of an opamp and especially the ancient prophet all ge monstrosity
are the one's i need to focus on the most.

sorry... no coffee yet, and pre doob my mind actually works

[kaycee]

My build from Mark's schematic



Probably my favorite Superfuzz variant so far, and I've built a few over the past months. Thanks for the ideas MH , all the suggested mods give usable parameters to this fuzz.



Made a few changes as you do. I changed the cap over collector/base of Q1 to 1nf as it gave it more sizzle. I have a 3 way switch on the Q1 emitter resistor which is off, 100n and 1uf. it gives progressively more gain and on the 1uf gives sizzle/oscillation and gated effects. I have silicon diodes, the diode control to ground is a great mod by the way. I found that I wanted a bit more output volume so have added in the usual SF post gain stage after the volume control.

sounds great! only issue I'm having is that I'm getting switch pop?? Using the opto tron for bypass, grounds the input when not engaged, added a 1M to ground before the input cap, the output booster has a 150K to ground after the out cap, so a bit puzzled there?

[Bret608]

This sounds really good. Could anyone share the layout with me? Or is it already posted in another part of the forum?

Thanks!

[Mark Hammer]

Here's one project layout, that can be easily adapted to the various offboard changes: http://ustomp.com/?p=5

[Bret608]

Thanks Mark! That looks doable for sure. Let me see if I can work it out.

[Bret608]

Actually I wonder if this might be an even better jumping-off point:

http://tagboardeffects.blogspot.com/2012/05/kay-f1-fuzz-tone.html

Kaycee, is that what you used?

[Bret608]

Yep, this vero layout at Tagboardeffects will be a much simpler starting point for Mark's mods.

Could anyone walk me through how to physically make the offboard changes? I more-or-less understand what changes are happening to the vero in terms of Mark's schematic; I'm just not sure in which cases I need to lift the end of a cap and connect that leg to a pot with a wire, for example.

[kaycee]

Bret, I drew up my own one over a cup of tea and a biscuit. Mines a little different in that it has a further gain stage. You could mug the mods onto the tagboard layout, the clipping and the scoop look a little tricky. If you can wait a few days I'll post up a cleaned up vero for this from my scruffy build notes.

[Bret608]

Thanks, sounds good! I can wait. You're right; the clipping and scoop were the ones I couldn't figure out, at least not without making my build pretty ugly! 

[kaycee]

Vero layout for this one. I have a confirmed build from it.




Notes: I changed the cap over collector base of Q1 to 1nf from 2nf. For the switch you need a SPDT switch with 3 positions, so it latches in each. This gives you cap lift, 100nf and 1uf with the lift in the mid position. Lift is lo fi, gets progressively gainy. Feel free to change cap values.

I used silicon diodes, I prefer them to the germanium ones. Volume and tone work as you'd expect. Clip increases clipping as you turn it CW so volume decreases. Scoop, more scooped so sound thins as turned CW.

Match the pair with your meter hfe function.

The capacitors in the tone path don't need to be 10uf, I used 4u7 for the in and output caps and 1uf either side of the clippers, experiment.

Added post tone boost section.

[Bret608]

Thanks!    I PM'd you a couple of noob-like questions already but I think I could make this happen. Just need to order a bit of vero and I'll be off and away...

[kaycee]

One error, there should be a cut to the left of Q2 collector between it and the link between Q1 collector and the base of Q2 isolating it from Q1 collector.

[kaycee]

As I can't edit my earlier post here's a revision of the vero drawing with the missing track cut added and notes for the connections on the right hand side of the board.



And a soundclip of it in action, riff played into a looper and then controls adjusted. Starts on the lo fi setting and around 1.20 the high gain setting is engaged.

https://soundcloud.com/retro-electro-effects/kf-riffmp3

Lead line to show the octave effect.

https://soundcloud.com/retro-electro-effects/kf-lead-linesmp3

[Bret608]

These clips sound great!  The only barrier I have to going ahead and building this is that I can't find a source for vero with this many columns. Am I right that there are about 40 columns here? The most I can get from Smallbear for example looks to have about 35. If anyone knows of a source with more different sizes, let me know!

[Mark Hammer]

Go perfboard.  When it comes the transistor-based circuits like this one, it's really not all that hard.  You have components running to ground on one edge, others running to V+ on the other side, and a bit of space between successive transistors for other components.

Admittedly, vero provides a nice solid base where components don't wiggle.  And it is always nice to know that all connections have been worked out ahead of time by someone else.  But if apprehensiveness is all that's stopping you, don't worry: take a stab at it.

[kaycee]

These clips sound great!  The only barrier I have to going ahead and building this is that I can't find a source for vero with this many columns. Am I right that there are about 40 columns here?

Just counted my second build of this, and your right, it's 40 holes wide (with 10 strips), so good estimate! Mark is right, the way I've laid out the vero closely follows the schematic, so you could use it to build on perf if you feel adventurous. The amount of transistors if you socket them is a bit off putting though as they are fiddly to do.

[Bret608]

Let me dig a little deeper on the vero--they may have a good variety in the electronics lab where I work (a technical college); they certainly have perf as well.

Also, this prompted me to finally download DIYLC and see if I can at least recreate your layout in the program. Wish me luck!