An overdrive/distortion.... A little different, this one.

[Ben N]

Thanks. I just printed out that article for my commuting-reading tonight.

[gez]

Also, w/re the question about paralleling inverters for better drive capability--would that not make sense ahead of the tone control?

Never tried it, but it makes sense to me!  Personally, I just DC connect a divider to earth and hook up its centre to the gate of a JFET follower.  So long as you divide down enough, you won't get clipping in the JFET stage.  Less current consumption this way (inverters are hogs in that respect)...

[brett]

Hi again
yes, inverters are current hogs, and you don't really want to drive heavy loads with this circuit.
But when you DO want to produce lots of ampage, inverters are quite good (I presume that its the very low on-resistance of MOSFETs compared with JFETs that does the trick).

Re: the last inverter.  100 ohm input and feedback resistors would be good. ( )

PS sorry about the name, Mark.  Gees, I must have stored your "FortyNiner" name in my sub-conscious.  I'll change it ASAP.  Maybe to the "Inverter" or "Smooth-o-matic"?

[gez]

Re: the last inverter.  100 ohm input and feedback resistors would be good. ( )

A 100 om input resistor is no different than a 100 ohm load to ground for the previous inverter.  Severe loading ensues.  If you use equal resistors for U1f it doesn't necessarily act as a unity gain buffer as inverters have high output impedance, so the output of U1a (if U1f's input resistor value is low enough) could be loaded and its gain reduced (less than unity).  That may, or may not, be desirable.  Loads of, for example, 10k will reduce the gain of a stage.  Sometimes that's good as it keeps overall gain throughout the circuit down and kills oscillation.

[Mark Hammer]

PS sorry about the name, Mark.  Gees, I must have stored your "FortyNiner" name in my sub-conscious.  I'll change it ASAP.  Maybe to the "Inverter" or "Smooth-o-matic"?

No problem.  I'm not offended and it's not trademarked.  As for "Forty-Niner", you probably got it from the same place I did: the song "My Darling Clementine" ("Was a miner, forty-niner...").  And THAT would appear to have entered my own subconscious via the old Huckleberry Hound cartoon show. 

The key thing is that every design that enters into the fray be identified in a unique way such that when person A refers to it by name, person B knows precisely what they're referring to, and when person C gets told to look for threads about it using the search function, that the search term work as planned.

[jrc4558]

Is there something one could do more easily with several parallel invertor sections on the output?  That could be as simple as being able to use a different value output pot or whatever.

I beleive the output impedance will drop, since they all will be in parallel.

[aloupos]

This looks great!  Has anyone started a pcb layout for it? 

[Ben N]

As for "Forty-Niner", you probably got it from the same place I did: the song "My Darling Clementine" ("Was a miner, forty-niner..."). 

No offense, Mark, but there is no way that anyone from south of the border would have gotten it from that place. 

[Mark Hammer]

Ah, but that's where you'd be wrong.  Remember, "My Darling Clementine" was Colonel Potter's favourite movie on "M.A.S.H."!

[aloupos]

How great is M.A.S.H by the way?   Still sad, hysterical and interesting 30 years later

[Barcode80]

has anyone actually built this yet?

[brett]

Hi
these might help simplify the building process for those interested. 
I'm calling it the "Smooth-o-matic" because the "Forty Niner" was used by Mark Hammer.  Ok, it's a weak name, but what do you want for nothing? 

Here's a PCB layout:

http://aronnelson.com/gallery/albums/BinOfBrett/Smooth_o_matic_PCB.sized.jpg

The overlay:

http://aronnelson.com/gallery/albums/BinOfBrett/Smooth_o_matic_overlay.sized.jpg

And for completeness, the schematic.  Although the last inverter isn't really used, I kept it in the schematic to be consistent with the layout above. 

http://aronnelson.com/gallery/albums/BinOfBrett/Smooth_o_matic_schematic.jpg

Please note that these diagrams include a couple of last minute changes that I "jumpered" on my board, so I haven't tested this exact layout yet.

All feedback is greatly appreciated.
cheers

[puretube]

sorry, folks,
but I don`t see an inverting buffer in U1f:
just a dead short between in & out;

I see two (shunting) FET-diodes going from the (center-biased) output of U1a to the respective rails. ("clipping"...)

maybe not "diodes", but "resistive loads" ? "constant current"?, "biasing devices"?
in Fig.2, here (N2/P2) an inverter is shown with in-/output shorted.

more theory, here...

[gez]

maybe not "diodes", but "resistive loads" ? "constant current"?, "biasing devices"?
in Fig.2, here (N2/P2) an inverter is shown with in-/output shorted.

more theory, here...

I couldn't get that PDF to open, but I think you're on to something here Ton.  If that last stage were non-inverting then yes, it would have no effect on the circuit.  However, it's inverting: its input is still being pulled high/low and it's output will try to swing in the opposite direction.  This probably has the effect of reducing the resistance seen at the output as the signal gets larger, in turn loading the previous stage.  In other words, I think it might have the effect of compressing the signal coming out of that 2nd inverter. 

Well, that's my guess...

[puretube]

that 1st pdf is from: US patent #4071777

yes, what I meant: there is nothing happening "from in to out" (horizontally) in brett`s last inverter,
there`s rather something (clipping, squashing or loading) going on
from the previous` inverter (centered) output towards either rail (vertically)...
(and it sure looks a bit like the inverted "S" transfer-curve, that is so typical for these inverters) .

[gez]

The acid test would be to simply A/B any CMOS fuzzer with that last stage tacked onto the output. 

[puretube]

any fuzzer (or even just booster...) whose amplitude comes close to the CMOS`s rails. 

(currently don`t have breadboardable inverters at hand...) 

[gez]

(currently don`t have breadboardable inverters at hand...) 

I've got an 4049 sine LFO on the board and one output is square.  Tomorrow, I'll feed it to a spare inverter wired up with a short from out to in and let you know what it does. 

[gez]

1kHz sine driving a single stage:




Now with the addition of Brett's 'Compressortron':




As predicted, there's limiting and further rounding (can't really see it here, but when I drove the first stage to the point of severe squaring, the 'buffer' rounded the corners out nicely). 

Well done Brett!

Start breadboarding chaps (first one to Ebay wins a cream cake)!

[brett]

Thanks for the pictures, Gez.  Very interesting. They match my scope and auditory results. ie. clipping happens quite progressively. 

At a low gain setting I found that I could throw a wide range of amplitudes of sine wave at it and get soft clipping (e.g. anything from a few to 1000mV).  Only at high settings of the gain pot (past 3 and 4 o'clock on a log pot) did I see anything squarish with typical inputs (10 to 200 mV).

RE: ebay.
How do you describe it?
It's got multiple, stage-filtered, feedback-stabilised, push-pull CMOS amplifiers.  Oh yeah, and parts LEFT OUT of the output buffer by some guy who was asleep at the design desk.