Quick Zonk Machine layout to suit an enclosure that I just picked up from MW Pedal Parts - pretty much one row difference in terms of where the mounting holes are drilled compared to old layout that I have been using, and I decided to move a couple of components around while I was at it.
MW are great by the way - no hesitation in recommending them based on my experience. The ordering process is a little unusual, but they deliver on the goods.
I thought a post was required for the silicon Zonk II, as looking at some schematics online, some appeared to be a little different. Being a pedal from the sixties, variations and tweaks were not uncommon.
At the moment, this is probably best known as being the inspiration for the Reeves Zo.
This is fun to play with on the breadboard, and it's actually a really good sounding fuzz.
JHS ZONK II - SILICON
I know the mounting point in the centre is a little odd, but this is how the orginal is mounted, and the one I'm amking will be mounted the same way.
SILICON JHS ZONK II SCHEMATIC
Image from the D*A*M forum datadump - a gold mine for digging through vintage effects. Photo credits to Graham, who perhaps has the most impressive collection of vintage English effects ever assembled.
Here's a better shot of the 4k7 resistor across the 5k fuzz pot
In the shot below, the resistor across the 5k fuzz pot is 3k9 AND there is no pulldown resistor on the input. Image from Freestompboxes, photo credit to beedotman, who has done an awesome number of layouts for a lot of interesting and/or old circuits.
Essentially though, a 3k9 across a 5k pot compared to a 4k7 is only a matter of a few hundred ohms difference. You could probably hear it, but it's unlikely to make a big difference?
I would not recommend a 10uf input cap without a pulldown resistor or being grounded when the pedal is bypassed - unless you like some switch pop. Sometimes, not all things old-school are good ideas.
I saw this on the Gray Bench Electronics youtube channel, and given the cost and apparent popularity of Reeves pedals, I thought I'd post a layout. The schematic is also on screen for a lengthy period of time.
In depth explanation provided at Gray Bench, but in short, it's a silicon two-transistor Zonk II with mods. There were multiple versions of the Zonk II back in the sixties, with distinctly different circuits - just scroll through the page on fuzzboxes.org to see what I mean.
Be sure to read the comments on this post, as there is some valuable info regarding changes / updates to the circuit - and like almost any simple fuzz, get it on a breadboard and have some fun with it before committing it to a board.
REEVES ELECTRO ZO - VERO LAYOUT
I added a polarity protection diode to the circuit, cause why not when there's room.
From Mr Reeves in the comments: "as of the facelift version with new transistors (2n2926) the output pot is 100k"
ZO NOTES
You can read the notes on the schematic in the video, but here's a short list
the 10k bias pot is sometimes replaced with a fixed resistor
the 470pf cap is not always present (value is potentially much smaller - 47pf)
and read the rest on youtube...
Or if you're really brave, instead of the using vero, follow along the instructions to build the full point to point version in the enclosure. Hope you have plenty of blue tack and coffee at the ready, and perhaps be ready to laugh at yourself after your first attempt.
ZO x ZONK II
Just a quick reference to the schematics - the Reeves schem is redrawn from the Gray Bench video & the silicon Zonk II is from all the usual places - but with the addition of the pulldown resistor that's visible in vintage units, but missing from a lot of schematics.
As you can see, they share the same bones, but with a few key differences. Notice the output cap differences - the Zo rolls off a bit more bass than the Zonk, and with the 470pf cap, a bit of treble goes missing too (apparently not all Zos have this cap, and it may be much smaller).
They both should have a decent amount of output, but the Zo should have more than the Zonk by a fair margin.
Spice did not have the right silicon transistors, so I just used the plain old 2N390X in each one for a fair comparison between an NPN & PNP circuit.
In my experience, the Zonk can be a little easier to make than the MKI in terms of transistor selection, and just getting a sound out of it - but it's still not one to rush into without a good stock of transistors to test. If it sounds thin and nasty, it ain't right yet.
Transistors
Q1 needs leakage to work. Q2 doesn't need leakage to work, but it uses an OC75, which are usually quite leaky compared to other transistors. Q3 should also need leakage to work, but it has an OC44 here, which are generally quite low in terms of leakage - this is partly where the gated sound comes from.
A good way to test Q1 is to play through it on the breadboard. Just take the output directly from the 22u cap (disconnect the next stage) and see how it sounds. It should sound clean and sustain. Listen for noise. Any noise here is just going to get a lot louder.
For Q2 I like to bias it so when the fuzz is turned all the way down, it’s either not working at all, or barely working. Then the pot will have some range.
Dave Main of D*A*M fame was kind enough to share some voltages and transistor hfe on a schematic, which is as follows.
Q1 Texas Instruments AO2650 hfe 30
Q2 Mullard OC75 hfe 156
Q3 Mullard OC44 hfe 69
Voltage on Q2 collector
-5.1v on full fuzz
-9.6v at zero
Most batteries start life at 9.5v so I have no idea what kind of freak battery Dave had plugged in to get -9.6 volts. Anyways, it will be a high voltage in any case, a range of -5 to -7v would be pretty normal.
And here's another set of quite different readings from an original Zonk, care of acidfuzz on freestompboxes. Great to see some leakage measurements as well.
and here's a demo, of what might be the very same pedal, as the video is provided by acidfuzz.
Input capacitor
The 1nf input capacitor is really the standout component value on a Zonk - this is really what sets it apart from the MKI Tone Bender. In the video demo above, the Zonk sounds quite fat - it's not easy getting this tone. I've found that a higher hfe Q1 helps and in some cases a lower gain Q2. This is not a hard and fast rule by any means.
If your Zonk is sounding a bit thin and nasty, it's not the end of the world if you increase the size of the input cap until you get it to a place that you're happy with. Try a 3n3. There's a lot of builders out there that offer this circuit with a MKI switch, which basically swaps or adds another input cap to increase the value to something like a MKI (10nf).
The sizzle
The what now? Zonks sometimes have an annoying sizzle on the decay, some people say it sounds like bacon cooking, and you don't really want that - or maybe some of you do? Some sizzle is not unusual on the decay of a MKI style circuit, but the Zonk can have way too much at times.
First, try different transistors - I've found lower hFE on Q2 to be helpful. If you can't fix sizzle with transistor swaps, try a low value cap directly on the input of the circuit. You want something big enough to reduce the sizzle down to manageable levels, but not so big that it changes the tone too much, although this can sound kind of cool, it's a bit like a cocked wah sound if you go too big. A 1nf cap often does the trick without affecting the sound.
Similar to the previous Zonk Machine, this has a MKI switch for a larger input cap. Have had this on the breadboard recently, and decided to jump to a vintage-style veroboard.
There are of course more differences between a MKI Tone Bender and a Zonk than just an input cap, but for convenience's sake, the input capacitor gets you pretty close.
MODDED JHS ZONK / MK1 HYBRID - 0.15" VERO LAYOUT
HYBRID ZONK ON THE BENCH
Some minor changes to keep some mojo caps on the board. 12n instead of a 10n input cap, as I got a reasonable deal on 100x 12n caps. 10n tropical fish are sometimes expensive.... same for the 100n, went for something larger, which will have almost no impact on the sound, but not a common value, meaning substantially cheaper.
OC76 for Q1 and Q3. Q2 is an old silicon glob top from Fairchild.
This is my version of the Zonk Machine, based on one of the early vero layouts. I added a switch for the input cap, which is possibly the most unoriginal mod known, and increased one of the input caps from 1n to 2n2.
I used an OC76 for Q1 and Q3 and an OC44 for Q2. I found this provided the best balance between noise and the aggression of the fuzz. It was very similar to using OC75s, but with noticeably less noise.
How does it sound? It's a nasty primitive fuzz with a lot of character - I really like it. The standard 1n input cap is not nearly as bright as you think it might be, but it wasn't entirely to my liking which is why I went up to 2n2. The 10n which is stock for a MKI is really meaty - again, perhaps more than you might expect if you have not tried a MKI before.
The switch essentially functions as a fat switch. 2n2 is good - 10n is about as much as you would want.
ORIGINAL ZONK MACHINE VERO LAYOUT
This is one of the original layouts, which is a little cramped for my liking.
The image below is attributed to Graham and his seemingly endless supply of vintage English pedals on D*A*M forum Note the 25k Fuzz pot and the vero construction.
I started copying the original layout for veroboard and gave up toward the end as it's a cramped mess - this is a similar but more practical version.
The germanium John Hornby Skewes Zonk II is similar to a Fuzz Face but without the 100k feedback resistor, and note that the output cap is taken directly from the collector of Q2, and there's a resistor in series with the cap to take a bit of edge off the output level. There's also a 680k resistor to help bias the base of Q1, probably due to the missing 100k feedback resistor.
It's a clear departure from the original Zonk Machine, before the silicon model, which also has two transistors. To make things even more confusing, a hybrid silicon germanium version was thrown in for good measure.
I’ve read that these were made by Sola Sound for JHS in approx 1966-67. Check the thread on the D*A*M forum for photos, as this is slightly different from one of the schematics.
Also important to note that there were multiple versions, all called the Zonk II, but with very different circuits. Check out Fuzzboxes.org as a reference to this.
As far as I know, only a handful of D*A*M FZ-678 were ever made, maybe 5 in total - it's a variation on a Zonk Machine. It's the same structure as a Zonk / MKI, but with a filter attached to the input.
I haven't made a layout for it, as it's all there to see - and I could not imagine that anyone wants to build one using this format. If you want to make one, I'd say just use your favourite Zonk layout, and tune the filter to your liking. Like all things Zonk / MKI - get it on the breadboard first.
D*A*M FZ-678
There are still a couple of things I'm unsure about on this, as there are very few photos available, and the colour coding on blue metal-oxide resistors is misleading. example: the 470k resistor looks like 5m7
Link to Stompboxes site. Transistors: Mullard OC42, OC44 and OC83 (Q1, 2, & 3).
I've already made a point-to-point layout for this, and as I'm currently on a 0.15" D*A*M vero roll, I thought, why not... and it's a Zonk Machine with tweaks.
Remarkably similar to the original D*A*M layout, I did move the output from the board a little and swapped pins on the swell pot to a more standard configuration. If you want to change it, I'm sure you'll work it out.
The D*A*M layout has a standing axial capacitor, which doesn't really work on DIYLC, so I used a standard cap and added polarity markers.
