WEM - Project V

[nightraven]

you make me really happy when you compare your breadboard to a Maestro FZ-1 check the little fuzz guitar melody approx halfway through this song and please give us a comparison!!! i have always assumed this was the same setup as in the infamous title-track but without all the ridiculous studio magic and layers. it does have a similar buzzy character as the Maestro but much much thicker with proper sustain.

[grizzlytone]

Well, I was intrigued enough to breadboard this baby. It's a very raunchy fuzz. It can go from really thin Maestro fuzz, to a more Muffish tone. Pretty good spread.

Just some things that i noticed though; Studying the schematic vis a vis the solder side shot, there maybe an error on the schem. Correct me if I'm wrong, but I believe that the 100K / 470pF / 22pF trio should be going from the base of tranny 7, to ground. That's what I see on the PCB pic, but I don't see it reflected that way on the schem. Can someone confirm this?

As for the trimmers, a 250K in position A seems to do the job, getting the voltage to 4.5v at the 4R7 junction. I'm still playing with trimmer B. This one is touchy, but the best results seem to be when it's right on the edge of oscillation. One thing I'm not happy with right now is that there is a bit of gating going on just as the note fades away. If I can figure out how to fix this, this one is a keeper. Anything that can range from buzz saw to muff, in one box, can go on my board, anyday.

For trannies, I'm playing with a mixture of 108's and 109's, with a 2N2907 in the PNP spot. I also tried 2N2904, 2N2905, and 2N5087, but all give similar results.

Great to hear that someone is making progress on this thingie and Thanks for spotting the error in my rev2 schematic. Here's a corrected rev3 version of my schematic.
What is the voltage drop across the 390 ohm resistor? Measuring the voltage at the collector of T7 or T8 will give the relevant V+ witch divided by 2 will give you the set point for the voltage at the 4R7 junction. If you need the A trimpot to be in the 250k range it tells us that there is little voltage division going on between the 10k and 250k so that the bias voltage at the base of T3 is almost equal to the 47R junction voltage...
Assuming a negligible voltage drop over the 4R7 resistors the voltage at the base of T5 should be in the [4R7 junction voltage minus 0,6V] ballpark, the voltage at the base of T8 should be in the [4R7 junction voltage plus 0,6V] ballpark, and the voltage at the base of T7 should be in the [4R7 junction voltage plus 1,2V] ballpark.
Assuming that the effective V+ is +8,0V and the 4R7 junction voltage is +4,0V the voltage at the base of T7 will be about +5,2V so there is a voltage drop of 8-5,2=2,8V over the two 1k5 resistors constituting a 3k resistor added together.
Ohm's law gives I = U / R witch will give the current I through the two 1k5 reistors = 2,8 / 3000 = 0,93 mA
Since the voltage at the base of T5 needs to be [4R7 junction voltage minus 0,6 as previously discussed] +4 - 0,6 = +3,4 we get that the voltage drop across trimmer be needs to be 5,2 - 3,4 = 1,8V with a current of 0,93mA running thru it (for simplicity sake I've neglected the parts of the current that gets shunted through the 100K resistor to GND, the small base currents on T5 and T7 will cancel eachother out). Now what resistance meets our demands of 1,8V drop with 0,93mA running thru it? Ohm's law again R = U / I = 1,8 / 0,00093 = 1935 ohm > 2k2 trimpot (coinsidently the only markings that I can see on trimmer B is two red dots = 2 and 2 something...)
It seems strange that you need such a high bias at the base of T3 as your 250k trimpot suggests. When I build a protopype on a perfboard it is a rule more than an exception that I need to go back and scratch my head over my circuit topology...
I hope this helps.
Good Luck and keep posting/ M

[grizzlytone]

Just one last thing I forgot to mention:

Trimmer B sets the bias current for output stage transistors T8 and T6. The 390 ohm resistor protects them from passing to much current > if you set trimmer B to high the effective V+ at the collector of T8 will just drop protecting T6 and T8 from frying (hopefully?)

Trimmer A sets the voltage at the 4R7 output junction to V+/2

If you adjust B so that you get a V+ of +8V with a +9 supply there will be 2,7 mA bias current through the output transistors and they will each need to loose P = U x I = 4 x 0,0027 = 11 mW as heat (piece of cake)

I guess I was wrong assuming that the circuit would drain a 9V battery in no time at all !!

Signing out again

[digi2t]

Thank you soooo much for the info. I won't have time tonight, but tomorrow I should be able to make some more headway.

I'll pay close attention to the voltages. I think I was taking the voltages for the B trimmer from the wrong place. I go over it again.

As i mentioned, it has great promise. If I can solve the gating (very little, at the end of the decay, but there) this thing is going to be killer.

Another thing that bugs me is the attack pot. It really acts as another volume knob, which I find totally useless. I might mod this to maybe bypass the fuzz section instead. That way, you can go from a simple boost, to all out mayhem. It would certainly be more useful.

Muchos gracias!!!

[grizzlytone]

Thank you soooo much for the info. I won't have time tonight, but tomorrow I should be able to make some more headway.

I'll pay close attention to the voltages. I think I was taking the voltages for the B trimmer from the wrong place. I go over it again.

As i mentioned, it has great promise. If I can solve the gating (very little, at the end of the decay, but there) this thing is going to be killer.

Another thing that bugs me is the attack pot. It really acts as another volume knob, which I find totally useless. I might mod this to maybe bypass the fuzz section instead. That way, you can go from a simple boost, to all out mayhem. It would certainly be more useful.

Muchos gracias!!!

I'm glad if I can help!! I'm so busy I don't have the time to start my own build but watching someone else getting it together is almost as rewarding....

Yesterday I was calculating of the top of my head while writing the posts. I've been thinking some more and it just came to me that I might as well calculate all the voltages for the post-Attack-stage. So this is what I have done -I'll present it with a schematic ver4. I made a misstake when i omitted to subtract the current draw from the preciding stages when calculating the bias current for T8 and T6 yesterday but it has all been corrected in the schematic ver4....

This is what I think concerning your build:
1) There is something wrong (most likely in lower half of the o/p stage) making it clip the waveform at all input levels -that is why the Attack control don't make no differance, it is also most likely the cause of the gating (some transistors stop conducting when the signal decays -they wouldn't do this if they were properly biased)
2) The easiest way to confirm the above hypothesis would be to measure the current conductance of T8 and T6 -they should be equal. You can easily measure it indirectly by using a DMM set to the DC sillyvolt range and see if the voltage drop over the two 4R7 resistors are equal (+/- depending on the tolerance of your resistors). The schematic gives a theoretical value of 7,66 mV.
3)The theoretical value for Trimmer A is 22k (use 50k trimpot). This is what you should arrive at +/- 20% say when the circuit is working properly.

All calculations is based on a couple of assumptions that will make real world measurements skew from the calculated:
Forward B-E voltage drop may vary between say 0,7 to 0,4V depending on base current -I've used 0,6V for the calculations.
The effective V+ measured at the collector of T8 will vary with the bias current setting. If you have access to a lab PSU you can hook it up in place of the battery and adjust it for V+ = +8V as you go along. When calculating I arrived at a battery voltage of +9,23V just to make the algebra true witch it ain't anyways since I don't know the current draw in the first stage (it should be little though).
So the message is: take the calculated voltages as nice round figures and look at the whole > look at Proportions.
As previously mentioned Trimmer B sets the workpoint of T8 and T6 -you'll be able bias the output "cold" to work in Class B or even C producing lots of crossover distortion all the time (that is probably where you're at right now) or you can bias the output to Class A and overload the input (by adjusting the Attack knob) and hopefully produce some nice clipping with the right content of overtones (that is what was originally intended for the circuit I think)

Go for it and Good Luck !!
/M

[capricorn_1]

great info
i ordered a bigger breadboard for this which came today so can have a go at it soon.
digi2t did you use an inductor or leave it out?

[digi2t]

I left out the inductor. On a lark, I had an old wah inductor, that was giving me 18 ohms on my dmm, so I threw it in there. It only served to produce a lot of noise, so I'm going direct. As was mentioned before, I don't believe it will give any audible advantage. Jumping out directly after this section yields a fairly clean boost.

I'm also going without the 1000uF / 47nF caps for the time being, since those only serve to filter the sheilding. On the breadboard, without no chassis, they too were giving me a lot of noise when taking them to ground. I might try going sans the 6.8nF for the time being as well. My first priority is to get the voltages settled out. As a test, I put aside all the 108's for the time being, and have put in 2N3904's in their place (with a 2N3906 in the PNP spot). Just on a hunch, I'm thinking that the production quality is a bit more linear with these, hence factoring down whether one tranny is better/worse than the other. If I can get inside the numbers (and tone) with these, then I'll start going back to the 108's, one by one, watching how the circuit reacts.

My 4.7 ohm junction is actually 2 pairs of 10 ohm resistors in parallel, giving me 5.1 ohms on each side. That's all I had on hand at the moment.

I'm using a power supply for the breadboard, which is giving me 9.14v, so I'll take that into account (proportionally ) when I go hunting.

I'm also thinking that I may have to adjust some resistor values to get into the zone here, more specifically some of the 1.5K resistors. I have lots of 5K trimmers on hand, so I'll swap them out, and then have an easy way to tweak things along.

[digi2t]

Hi /M,

Just a verification; The voltage posted after the 27K resistor, is that a "4v"? I'm assuming that's a 4, but it's not that clear. Of course, 4v makes perfect sense, considering what's comming down the pipe.

I can't seem to get the 0.6v at the base of tranny 6. I've managed to get all the other voltages close, but this one doesn't seem to want to coorporate, usually hanging around 10mV. On the other hand, watching it while I play, I do see it come up while playing, so something is happening. Maybe the 2N3906 isn't quite up to the task.

What helped enormously was using a trimmer in place of the 390 ohm, the 100K at the base of tranny 7, and a trimmer in place of the 27K just below it, went a long way in helping tweak the voltages.

Back to work....

[digi2t]

Hi /M,

Just a verification; The voltage posted after the 27K resistor, is that a "4v"? I'm assuming that's a 4, but it's not that clear. Of course, 4v makes perfect sense, considering what's comming down the pipe.

I can't seem to get the 0.6v at the base of tranny 6. I've managed to get all the other voltages close, but this one doesn't seem to want to coorporate, usually hanging around 10mV. On the other hand, watching it while I play, I do see it come up while playing, so something is happening. Maybe the 2N3906 isn't quite up to the task.

What helped enormously was using a trimmer in place of the 390 ohm, the 100K at the base of tranny 7, and a trimmer in place of the 27K just below it, went a long way in helping tweak the voltages.

Back to work....

I forgot to mention, the A trimmer is now a 50K, works great. The B trimmer is a 5K, since i didn't have any 2K2 on hand, but I do get my 3.4V on tranny 5. Just that 0.6v that's eluding me. I have 9.14v from the PS, and via the trimmer, I've tuned it to 8.0v, in keeping with the math

Right now, the sustain is greatly improved, but still not to my liking. This is a very "fizzy/buzzy" fuzz. More tuning to be done. As soon I get a chance I'll post a video, probably on the weekend.

Cheers,
Dino

[grizzlytone]

Hi /M,

Just a verification; The voltage posted after the 27K resistor, is that a "4v"? I'm assuming that's a 4, but it's not that clear. Of course, 4v makes perfect sense, considering what's comming down the pipe.

I can't seem to get the 0.6v at the base of tranny 6. I've managed to get all the other voltages close, but this one doesn't seem to want to coorporate, usually hanging around 10mV. On the other hand, watching it while I play, I do see it come up while playing, so something is happening. Maybe the 2N3906 isn't quite up to the task.

What helped enormously was using a trimmer in place of the 390 ohm, the 100K at the base of tranny 7, and a trimmer in place of the 27K just below it, went a long way in helping tweak the voltages.

Back to work....

I forgot to mention, the A trimmer is now a 50K, works great. The B trimmer is a 5K, since i didn't have any 2K2 on hand, but I do get my 3.4V on tranny 5. Just that 0.6v that's eluding me. I have 9.14v from the PS, and via the trimmer, I've tuned it to 8.0v, in keeping with the math

Right now, the sustain is greatly improved, but still not to my liking. This is a very "fizzy/buzzy" fuzz. More tuning to be done. As soon I get a chance I'll post a video, probably on the weekend.

Cheers,
Dino

Great work Dino!
Yeah, the voltage at the 27k resistor is indeed +4V. I think you're close to the solution now, If the emitter of T6 is connected to common ground AND the the 2N3906 in place for the PNP (T5) is conducting you should get about +0,6 V at the base of T6 (with respect to common ground). Could you have reversed the emitter and collector of T5 on your board? Sometimes it's a bit confusing when the emitter on PNP's takes the higher voltage compared to its collector while it's the other way around on NPN's -Collector high, emitter low. EBC order with the pins pointing downwards looking at the flat surface side goes for both 2N3904 and 3906

I can't really draw any conclusions from the 390R and 100k observations at this point (the 100k shouldn't effect the voltages much at all -it is about two orders of magnitude larger than just about any other resistor in the o/p stage so it shouldn't affect currents and the resulting voltages to any large extent. It is just there to complete the filter that it comprises together with 470p + 22p > Low Pass filter attenuating signals above 3,2 kHz with 6 dB/octave. If you need to compensate the DC voltages by changing the 100k resistor I think you're just compensating a primary fault with a secondary fault).

If the PNP has the right polarity -what is the emitter-base voltage drop on it?
Anything lower than say 0,3V means it's not conducting and anything higher than 0,7V means it is broken (provided that the polarity is correct). Red probe on emitter, black probe on base when measuring.

You could also try replacing T6. Remember that without the protection that the 390R resistor affords and a battery adapter in place of the battery you could easily kill T6 and T8 by accidently letting them pass to much current while trying to tweak the circuit...

Gotta go. Good Luck. Keep us posted.
/M

[digi2t]

OK... Success (I think )

I went back to the resistors, keeping only the trimmer subbing for the 390 Ohm. On a whim, I swapped out all the NPN trannies for 2N5088's and kept the 2N3406 in the PNP. After several hours of poking and proding, and head scratching trying to understand the inter relationships between the trimmers, I FINALLY got all the voltages to within +/- 0.1 volt (for the major base voltages) of what you laid out on the schematic. And yes, tranny 5 is conducting, and sitting at 0.585v now. All the other voltages are very, very close.

This is one fizzy fuzzy buzzy crispy sizzly fuzz circuit. Almost a la Fuzz Factory, but crispier. Did I mention sizzly. Oh yeah, and it's real crispy too.

Was probably way to radical for anyone back in the day. Hell, sounds pretty radical to me now. It almost sounds like I'm playing my guitar through a Tesla coil amplifier. Think the Stones "Satisfaction", but with steam roller running over a million bags of tuned Doritos every time you play the hook. And I'm pretty sure that I know when the guy 5 doors down turns on his flourescent lights as well.

But the funniest thing was last night, while I was playing with it... I had the headphones on, since everyone was sleeping, and while I was tuning the trimmer, it suddenly came over the headphones that my favorite hockey team had lost again

Really not your run of the mill fuzz circuits. The PNP/NPN section really adds a lot of harsh nastiness to the sound. Just pure evil dirt. I wish there was a way to tame it, but I did find that an AC128 in the PNP spot seemed to make a bit of a difference. Somewhat less harsh. Hmmm....

The EDGE switch basically boosts, or lowers you treble. The DRIVE switch fattens the low end a bit with the EDGE on high, but with the EDGE off, and DRIVE off, there is a significant loss of volume. DRIVE on, EDGE off leans toward a nervous chihuahua Muffy type sound.

The ATTACK pot really is a killjoy at any setting other than wide open, pedal in the oil pan, max. Below that, the presence, sustain, and fuzzy fizziness, quickly fall by the wayside.

I would really like to know if anyone else tries this circuit, just to corroborate my findings. This thing is... umm... special. I'm surprised it hasn't shown up on a Black Keys album yet.

Many. many, many thanks to Grizzlytone for the pointers (and enduring my noobishness). Thanks to you, I go to bed less stupid at night. And that's where i'm headed now. Goodnight all!

[digi2t]

OK... Success (I think )

I went back to the resistors, keeping only the trimmer subbing for the 390 Ohm. On a whim, I swapped out all the NPN trannies for 2N5088's and kept the 2N3406 in the PNP. After several hours of poking and proding, and head scratching trying to understand the inter relationships between the trimmers, I FINALLY got all the voltages to within +/- 0.1 volt (for the major base voltages) of what you laid out on the schematic. And yes, tranny 5 is conducting, and sitting at 0.585v now. All the other voltages are very, very close.

This is one fizzy fuzzy buzzy crispy sizzly fuzz circuit. Almost a la Fuzz Factory, but crispier. Did I mention sizzly. Oh yeah, and it's real crispy too.

Was probably way to radical for anyone back in the day. Hell, sounds pretty radical to me now. It almost sounds like I'm playing my guitar through a Tesla coil amplifier. Think the Stones "Satisfaction", but with steam roller running over a million bags of tuned Doritos every time you play the hook. And I'm pretty sure that I know when the guy 5 doors down turns on his flourescent lights as well.

But the funniest thing was last night, while I was playing with it... I had the headphones on, since everyone was sleeping, and while I was tuning the trimmer, it suddenly came over the headphones that my favorite hockey team had lost again

Really not your run of the mill fuzz circuits. The PNP/NPN section really adds a lot of harsh nastiness to the sound. Just pure evil dirt. I wish there was a way to tame it, but I did find that an AC128 in the PNP spot seemed to make a bit of a difference. Somewhat less harsh. Hmmm....

The EDGE switch basically boosts, or lowers you treble. The DRIVE switch fattens the low end a bit with the EDGE on high, but with the EDGE off, and DRIVE off, there is a significant loss of volume. DRIVE on, EDGE off leans toward a nervous chihuahua Muffy type sound.

The ATTACK pot really is a killjoy at any setting other than wide open, pedal in the oil pan, max. Below that, the presence, sustain, and fuzzy fizziness, quickly fall by the wayside.

I would really like to know if anyone else tries this circuit, just to corroborate my findings. This thing is... umm... special. I'm surprised it hasn't shown up on a Black Keys album yet.

Many. many, many thanks to Grizzlytone for the pointers (and enduring my noobishness). Thanks to you, I go to bed less stupid at night. And that's where i'm headed now. Goodnight all!

Hahaha ! Great descriptions Dino, Congratulations on your success, Well done !!!
/M

[nightraven]

Was probably way to radical for anyone back in the day. Hell, sounds pretty radical to me now. It almost sounds like I'm playing my guitar through a Tesla coil amplifier. Think the Stones "Satisfaction", but with steam roller running over a million bags of tuned Doritos every time you play the hook. And I'm pretty sure that I know when the guy 5 doors down turns on his flourescent lights as well.

this is what i like to hear congratulations on getting it going!
any comparison to the Brian Eno recordings that have been posted earlier in this thread? myself and the other 70s prog fans are on the edges of our chairs in anticipation

[digi2t]

this is what i like to hear congratulations on getting it going!
any comparison to the Brian Eno recordings that have been posted earlier in this thread? myself and the other 70s prog fans are on the edges of our chairs in anticipation

Truthfully, I listened to the clip at the beginning of the week, but have been immersed in the breadboard since. Now that all the stars have more or less aligned, I'll have another listen on the weekend, and do some comparing. Albeit from my dim memory, with the drive off, and the edge on, you do get that thin buzzy fuzz that I heard in the clip.

Give me a couple of days, and I'll give you a proper analysis on that.

Have a great day all,
Dino

[RnFR]

I'm not able to see or download grizzly's attached schematics. anyone else having a problem, too?

[digi2t]

you make me really happy when you compare your breadboard to a Maestro FZ-1 check the little fuzz guitar melody approx halfway through this song and please give us a comparison!!! i have always assumed this was the same setup as in the infamous title-track but without all the ridiculous studio magic and layers. it does have a similar buzzy character as the Maestro but much much thicker with proper sustain.

Well, I listened again to the clip. and yes, I would say that it sounds pretty close to what I've got on the breadboard. It's always hard to say with certainty, but if it were me, I would say (maybe) Drive on, Edge on, and you should get pretty damn close to that tone with this circuit.

I did steal a little trick from the Axis Face, and inserted a 220pF silver mica cap between Base and Collector on tranny 5 (PNP). It helps to tame the really harsh high frequencies coming from that part of the circuit, and allows me to tune it much closer to the calculated voltages without getting into "constant buzz-tone" territory. Kind of smooths things out a bit, less fizzy.

I've got the main voltages set at 1.55 for Q3, 3.55 for Q5, and 5.12 for Q7. I seem to get the best fuzz/tone to noise ratio here, although the noise should drop when it's boxed, and the shielded wires are in the equation. What I'm running right now is 2N5088's, and a 2N3906. All in all... I like. Not for the faint of heart, but rewarding when you get everything reigned in

With out further ado, here is the promised video sound bytes;


Link


I hope someone's quest has been furfilled.

[nightraven]

wow very interesting! thanks!
it does have a Big Muff wall-of-fuzz thing going on but very detailed in the treble area. has a bit of that chainsaw peeling off your cheeks quality like a Burns Buzzaround. funnily enough it sounds like the fuzz on the Twink "Think Pink" album which i dismissed as the WEM ages ago very flexible unit and good to finally hear it! must find the original now

[RnFR]

thanks a lot for all of your hard work! it'll definitely make it alot easier when I finally get the time to do the work. sounds mean as hell!

[kaycee]

Sounds promising from the demo Digi, thanks for sharing your work on this, going onto the to-do list. Rudolph usually played humbuckers, so maybe a hotter input might fatten it up a bit?

Great stuff

[digi2t]

Sounds promising from the demo Digi, thanks for sharing your work on this, going onto the to-do list. Rudolph usually played humbuckers, so maybe a hotter input might fatten it up a bit?

Great stuff

Hmmm... I'll do another video in the next couple of days with a hb guitar. I have an Ibby with Motherbuckers on it, so it should be plenty hot enough

Stay tuned....