CMOS fuzzer for you all...

[gez]

Here’s a little trick you can do to get some asymmetry with a 4007.

The inclusion of a drain resistor for the n-channel creates a DC offset for asymmetrical clipping, asymmetric transfer function AND it reduces current consumption dramatically! (from 4.24mA to .62mA per inverter wired like this).  Not only is clipping asymmetrical but it’s softer than the pull-down resistor on input trick you sometimes see.  The end result is a smoother distortion (to my ears anyway).



Interestingly, wiring in a drain resistor for the p-channel and taking the output from the n-channel’s drain results in slightly harsher clipping.

Here’s a pic of one stage being driven with a ‘sine’ wave (the sine from my generator is a little on the pointy side).



4007s, to my ears, have a crunchier (Marshall?) vibe to them than other inverters and this circuit is no exception.  However, the intermodulation is slightly better than other 4007 circuits I’ve messed around with in the past (though I still think 4049s are better in this respect).

The above circuit could probably do with dividing down at the output and buffering, plus a tone stack but I have other fish to fry so I thought I’d post this in case anyone was interested.

[petemoore]

I'm not used to 4007 schematics but those look like Fets...
 Interesting experiment none the less, and I'm glad to hear that you're getting cool results !!!

[gez]

I'm not used to 4007 schematics but those look like Fets...

They're MOSFETs.  The 4007 can be looked upon as three inverters but with the connections 'undone', so to speak.  This allows you to wire the chip in various configurations.  The above circuit is just three inverters, I've just wired in a couple of drain resistors to two of the stages.

[brett]

Hi Gez.

Nice work.  

Couple of questions - any reason for the different sized caps between the first and second stages?  

Also, how "hard" does it sound without the 1nF caps to ground?

I'm thinking about doing something with 3 4049 or 4007 gain stages, but with a "mids-scooped" big muff tone section before the final stage.  It'd be a MOSFET chip circuit aimed more at hard rock.  (I once tried my 4049-based tube-sound fuzz with a really small input resistor to the first stage and it was kinda wild in a good way).

Keep up the good work.

[Marcos - Munky]

Thanks for share the schematic, Gez!!!

[toneman]

actually, they're (C)MOSFETs.
the "C" is 4 "complementry".
means a "P' channel & "N" channel both on the same substrate.
a little different from the (pwr)MOSFETs.
very nice schemo!!!  
3 discrete inverters(?) w/2Meg feedback.
very cool!!!    
can *TaP* INto the innards  :wink:

[jrc4558]

Ok, guys, I'm lost here... I'm loking at the datasheet for about 20 minutes now, tryin to assign the legs of the IC to the schematic. I don't get it.
Respectfull Gez, can you possibly throw in the pin numbers into your schem? Just for the technically challenged. :shock:
Thank you aplenty!

P.S. After reviewing the posting I realised that instead of DATASHEET I typed in DATASHEEP. Describes me perfectly :lol:

[Tim Escobedo]

Good work, Gez!

Pinout on 4007 always gets me. I always have to pull out the diagram to use one.

[gez]

any reason for the different sized caps between the first and second stages?

Yes, that's all I had left in the box!  I could have posted the values I calculated, but sometimes you never know how these things turn out in practise.  The 1n and 2n2 caps ARE intentional though (I only meant the coupling caps from stage to stage were all I had available).

"Also, how "hard" does it sound without the 1nF caps to ground?"

These caps partially de-couple AC feedback.  For frequencies above cut off the input signal 'sees' the first resistor only then 'passes' to ground via the cap, so this resistor sets input impedance (which is why the 2n2 cap is used for the first stage as the input impedance is less than half what it is for the following stages).  In a similar way the output signal only 'sees' the second resistor, so you can control gain by changing its value, thereby loading the output to a lesser or greater extent.

For frequencies below cutoff the input signal 'sees' a input impedance of 2M (just under 1M5 for the first stage as I didn't have a 1M pot handy) divided by the gain of that stage, which is an unknown quantity but usually around 20 - 30 at that voltage.  The lower impedance for these frequencies means that the bass gets cleaned up a little (less gain).

Without the caps to ground negative feedback will be more pronounced so the clipping would probably be slightly harsher, but gain would probably be lower (didn't try it).

Incidentally, don't bother driving the input with a JFET or op-amp, it sounds horrendous!

Here's a link to the data sheet:

http://www.philipslogic.com/products/hef/pdf/hef4007ub.pdf

I'll revise the schematic a little later on today and show the pin-out.  I didn't put the remaining pin numbers in because I haven't thought about layout for this thing and it's accademic which of the unlabeled devices gets used for each stage.

Hey Constantin, did you get my PM/email the other day?

[Ge_Whiz]

Hi Gez

I see I'm not the only person in the London area tinkering with 4007s... Something else to discuss on the 8th!

[gez]

Respectfull Gez, can you possibly throw in the pin numbers into your schem? Just for the technically challenged. :shock:

Done!  

[gez]

I see I'm not the only person in the London area tinkering with 4007s... Something else to discuss on the 8th!

Look forward to it!  

[gez]

PS  You can wire in source resistors for the last stage if you want to reduce the current consumption a little more.  This sounded a little woollier to my ears so I left them out.  However, if you do this (and you're so inclined) you can take outputs from both sources and feed them to Schottky diodes to get a octave up effect.

I did this last year sometime and it wasn't my cup of tea, but then again I'm no judge of octave circuits as I'm not a huge fan of them.

[jrc4558]

Hey, Gez! How are you?

Thank you for this cirquit. I modelled it last evening and I really like the way she distorts. Dynamic and soft, but not weak at all. Natural is the word I guess. However, noise is an issue here on my pad-per-hole. I get hiss, hum and crackle.

How was yours for noise?

[gez]

Hey, Gez! How are you?

Thank you for this cirquit. I modelled it last evening and I really like the way she distorts. Dynamic and soft, but not weak at all. Natural is the word I guess. However, noise is an issue here on my pad-per-hole. I get hiss, hum and crackle.

How was yours for noise?

I'm fine!  Glad you like it, but it's odd that you're getting all that noise.  Do you get it if you use a battery?  If not try some filtering on the supply (there's already a 100u cap so all you need is a small resistor in there).

Mine's really quiet (exceptionally so!).  Perhaps it's layout/wiring?  Could just be the chip you're using?  Try shielded wire for the input.

[jrc4558]

Oki-doki. Will try.