Knight KG-389 Fuzz

Chuck D. Bones

Circuit Wizard
This pedal was available in kit form only in the late '60s and early '70s. LucifersTrip did a great write-up on DIYstompboxes, check out the link.

I breadboarded LT's modded version and it sounded pretty damned good. Didn't even need a Tone control. :p

Knight KG-389 Fuzz unit.png
Freakin' Hippies!! Adjusted for inflation, $12.95 in 1970 is $88 in today's money. We can sure as shit build one for less than that. Check out the stomp switch. Looks hazardous.

Knight KG-389 advert.jpg

Original version:
Knight KG-389 Fuzz schematic.gif

LT's modded version:
knight KG-389 fuzz finale (LT mod).jpg
I used a NOS low-gain Si tranny for Q1, a 2N1308 for Q2 and 1N4004's for the diodes. I changed the FUZZ pot to A250K. After playing with the 50K bias trimmer, I ended up replacing it and the 8.2K with a 15K resistor. Q2's collector - base resistor is 680K. I bumped the last capacitor up to 47nF. With only 13 parts, this one qualifies for the Express Lane.

One mod of my own I'm playing with is replacing the FUZZ pot with a 10K series resistor and adding a C50K pot in series with Q1's emitter as the new FUZZ control. Still undecided which I like better. The advantage of my mod is I'm not turning the signal down, I'm turning the gain down when FUZZ is rotated CCW. This reduces the noise at lower FUZZ settings. It gets pretty muddy when FUZZ is dimed. R1 might be too small.

Knight & Day Fuzz.png
 
Last edited:
It is. Simple and to-the-point distortion. I just realized I omitted a part in my schematic. It's correct now.

Have not tried the buffer. A buffer up front will drive Q1 harder, making it even more dense.
 
No power filtering or pull down at the input sounds like fun.

And hey here's a thing: sometimes you see the fuzz pot with a large value capacitor (10uf say) in parallel, or coming off pin 2 to gnd. How come this one doesn't need that?

Also is a 22nf beefy enough as an input cap do you think?
 
Power filtering and a protection diode are always a good idea. Anti-pop resistors on the input are optional if the bypass switch grounds the board input in bypass.

This FUZZ pot arrangement will scratch when rotated. The pot provides the DC path to ground for Q1's emitter, similar to the GAIN pot in a SHO. The same function could also be realized with a large cap, similar to the FuzzFace. I was going minimalist here.

LucifersTrip thought 47nF was too big on the input and I came to the same conclusion. I may reduce C1 further, not done testing yet. If you want more bass at high Gain settings, then make C1 bigger.
 
Hey, I'm about to get my hands on some hfe appropriate NOS transistors for this fuzz, what are people's experiences with biasing this thing?
 
Experiment with different transistors for Q1 & Q2.
Plan on fiddling R4 depending on Q2's leakage.
D1 & D2 limit Q2's collector voltage swing, which means as long as Q2-C is above 2V you should be ok.
If you build to the 2nd schematic, then Q1's collector voltage will vary as the FUZZ pot is rotated, this is normal.
 
Experiment with different transistors for Q1 & Q2.
Plan on fiddling R4 depending on Q2's leakage.
D1 & D2 limit Q2's collector voltage swing, which means as long as Q2-C is above 2V you should be ok.
If you build to the 2nd schematic, then Q1's collector voltage will vary as the FUZZ pot is rotated, this is normal.
Thanks for the quick response. I totally missed the notification on this thread which is why I'm only responding now.

I'm planning on drawing up a schematic based on the 2nd schematic, with PNP transistors instead of NPN, as good, high gain NPN Ge transistors are hard to find. Have you figured out what R1 resistor value works best? If not, is there a way to find out what it should be? (I know it's adding input impedance in series with the transistors) Or is it just mess around and find out what works?. I will just play with it till I find a good value if that's the case, no big deal.

Thanks as always, Chuck!
 
Assuming that this is the first pedal in the chain, R1 controls the touch sensitivity of the circuit when FUZZ is dimed. Increasing R1 makes the circuit less responsive to picking and guitar knob setting. If you plan on driving this pedal with another, then you should probably leave R1 as-is. As you turn FUZZ down, the circuit becomes less sensitive to the value of R1. There are only a handful of parts in this circuit and they all interact.

The bottom line is there is no "best" value for any of the parts. You can use the values given in the schematic as a starting point and experiment from there.
 
Assuming that this is the first pedal in the chain, R1 controls the touch sensitivity of the circuit when FUZZ is dimed. Increasing R1 makes the circuit less responsive to picking and guitar knob setting. If you plan on driving this pedal with another, then you should probably leave R1 as-is. As you turn FUZZ down, the circuit becomes less sensitive to the value of R1. There are only a handful of parts in this circuit and they all interact.

The bottom line is there is no "best" value for any of the parts. You can use the values given in the schematic as a starting point and experiment from there.
Thanks for the tip. maybe it would be interesting putting an emitter-follower buffer on the front end the pedal to make it more consistent. I read somewhere that the low output impedance of a guitar and the low input impedance of something like this or a fuzz face means that your guitar is kind of acting like a current source instead of voltage source as with a properly buffered and high input impedance pedal. Maybe I have that backwards, I can't remember. And don't quote me on that - "I read somewhere online," isn't exactly an accurate source :p Maybe I should just do out the maths on that.

I feel like on top of all the tweaks mentioned here, adding a buffer and trying to compensate for that sound would certainly start making this more into a, 'we're designing a pedal,' kind of scenario instead of merely a, 'we're modding a pedal,' one. :ROFLMAO:
 
Thanks for the tip. maybe it would be interesting putting an emitter-follower buffer on the front end the pedal to make it more consistent. :ROFLMAO:
You can easily try that by putting a buffered bypass pedal in front of this circuit.

I read somewhere that the low output impedance of a guitar and the low input impedance of something like this or a fuzz face means that your guitar is kind of acting like a current source...
Guitars with passive pickups have a relatively high, and frequency-dependent, output impedance. Pedals with a low input impedance, FF for example, are current-driven whether the preceding device has a low output impedance or not.

And don't quote me on that - "I read somewhere online," isn't exactly an accurate source
If you can't vouch for the accuracy, then maybe you shouldn't re-quote it here either.

Maybe I should just do out the maths on that.
Have fun with that. Running a simulation would be easier.
 
Guitars with passive pickups have a relatively high, and frequency-dependent, output impedance. Pedals with a low input impedance, FF for example, are current-driven whether the preceding device has a low output impedance or not.
Brain fart. Typing as fast as you think and not double checking your work is a recipe for mild internet embarrassment. Would the FF be current driven if you had an output impedance of a couple of ohms instead of a couple to a dozen of kilo-ohms? Seems like that would be enough change to make something different happen. (I know that's not really a practical output impedance. Just wondering aloud haha)

Have fun with that. Running a simulation would be easier.
You're right again. Dunno what I was thinking, actually doing the maths on circuit that's more complicated than an LED and some resistors? Not reasonable. I'll try and throw something together in a SPICE program when I get the time.
If you can't vouch for the accuracy, then maybe you shouldn't re-quote it here either.
Fair enough (y)
 
Would the FF be current driven if you had an output impedance of a couple of ohms instead of a couple to a dozen of kilo-ohms? Seems like that would be enough change to make something different happen. (I know that's not really a practical output impedance. Just wondering aloud haha)
This is drifting off-topic so I'll give you the long answer and be done with it.

The line between voltage-drive and current-drive is gray. The accurate way to think about it is as energy transfer because that's what's physically going on. The side with the highest impedance throttles the energy transfer. If it's the driving side (guitar or pedal on the upstream side) that has the higher impedance, then the voltage at the interface doesn't change much, but the current does, so we can call it current drive. The converse holds true if the receiving end has the higher impedance. It gets complicated when one or both sides is non-linear (like the FF input) or frequency-dependent (like a guitar's output). Also consider that the guitar's output impedance changes when we turn the volume or tone knobs. That's why a lot of guitar players dime the Fuzz knob on a FF and use the guitar's Vol knob to control the tone. Another thing to consider is stress. Transistors can take just so much base current before they are damaged. If you use a pedal with a low output impedance and high current-drive capability to drive a FF, you could damage or destroy the first transistor. Ge transistors are particularly vulnerable.

Bottom line is that voltage-drive and current-drive are just names and what matters is whether it makes the right tone.
 
This is drifting off-topic so I'll give you the long answer and be done with it.

The line between voltage-drive and current-drive is gray. The accurate way to think about it is as energy transfer because that's what's physically going on. The side with the highest impedance throttles the energy transfer. If it's the driving side (guitar or pedal on the upstream side) that has the higher impedance, then the voltage at the interface doesn't change much, but the current does, so we can call it current drive. The converse holds true if the receiving end has the higher impedance. It gets complicated when one or both sides is non-linear (like the FF input) or frequency-dependent (like a guitar's output). Also consider that the guitar's output impedance changes when we turn the volume or tone knobs. That's why a lot of guitar players dime the Fuzz knob on a FF and use the guitar's Vol knob to control the tone. Another thing to consider is stress. Transistors can take just so much base current before they are damaged. If you use a pedal with a low output impedance and high current-drive capability to drive a FF, you could damage or destroy the first transistor. Ge transistors are particularly vulnerable.

Bottom line is that voltage-drive and current-drive are just names and what matters is whether it makes the right tone.
That makes a lot of sense. Thank you. If I have another question I'll create a new thread, like you said, this is getting quite off topic. I certainly do have a habit of, shall we say, enthusiastically following up different lines of inquiry at the drop of hat. Best stick to best practices though.
 
Back
Top