Understanding CMOS and CMOS distortion...

[Phorhas]

Hey...

Where Can I found article about CMOS and the way the work, apps in audio, why they sound so good and how to use and design crkt with them

:roll:

[Tim Escobedo]

http://www.fairchildsemi.com/an/AN/AN-88.pdf

ftp://nutsvolts.com/fet1-4.pdf

[Tim Escobedo]

http://www.fairchildsemi.com/an/AN/AN-88.pdf

ftp://nutsvolts.com/fet1-4.pdf

[aron]

From schematics one:

http://www.diystompboxes.com/pedals/jo4049.html

[Phorhas]

Thanx

[Tim Escobedo]

It is pretty interesting once you see the voltage transfer characteristics of linear biased inverters. They're pretty commonly displayed in the literature. It shows why CMOS inverters distort uniquely even at lower gains. The graphs also show why they tend to exhibit higher gain when run at lower voltages. C Anderton noted very briefly in his old book that they distort softer more like tubes, I presume because they are not very linear, both for low level signals and high level signals, having high gain, yet smoothing the peaks more than hard clipping might. The idea being that tubes behave similarly, though for different reasons. While the similarity with tubes may be debatable, it is clear that they do behave differently than the usual op amp or discrete designs.

There are other issues with things like output impedance, that are probably changed somewhat depending on the use of negative feedback. They also seem to have rather hot outputs. However, I can't definitively comment on either of these characteristics as I've never made a disciplined survey of these effects.

[Phorhas]

Sound very intresting... AND it sounds good.... But why aren't there more CMOS FX and amp, as there are (kazajillion times 3) op-amp,BJT,MOS,FET etc...  



is there a catch 22?

[bwanasonic]

How well do CMOS distortion circuits respond to dynamics, and things like turning down the guitar volume? I'm not suprised they can produce a pleasing *distortion* waveform, but I'm suspicious of the way they *feel* and respond to playing nuances.  How gracefully do they transition into clipping? Maybe I should look for some clips...

Kerry M

[Peter Snowberg]

How well do CMOS distortion circuits respond to dynamics, and things like turning down the guitar volume? I'm not suprised they can produce a pleasing *distortion* waveform, but I'm suspicious of the way they *feel* and respond to playing nuances.  How gracefully do they transition into clipping? Maybe I should look for some clips...

Kerry M

It all depends on the gain structure. Think of CMOS inverters as just another option like BJT, JFET, tube, etc.  It all depends on the circuit they're in. I think the CMOS "feel" comes from the non-linearity at BOTH ends of the input range. The upper and lower parts of the curves are very different, but both are nicely non-linear. Check them out in the datasheets.

With the right circuit and high impedance buffering on the in and out (important for crisp clean tones), you can get some really great clean tones at medium volume that nicely flow into smooth fuzz at you turn up to 10.

-Peter

[Tim Escobedo]

I find CMOS circuits can be pretty responsive to dynamics.

Responsiveness to something like a guitar volume is also possible, but I think there are at least a couple of different ways to accomplish this. The classic way I've found is to design the circuit with a low input impedance. This type of circuit does a more "classic" volume control cleaning up, similar to the Fuzz Face. But this is different that the type of dynamics I get with CMOS, which also responds to actual picking dynamics (amound of distorion depends on how hard you play).

Most of my recent efforts with CMOS has been with the desire to eliminate negative feedback, in order to get a clearer listen at the inherent distortion characteristics of CMOS inveters. A sound which I liken to "running hot", not necessarily distorted, but lively and sweet. And smoothly agressive at high gains.

[gez]

Where Can I found article about CMOS and the way the work, apps in audio

If your library has a copy of the Art of Electronics by Horowitz & Hill, there's a nice little chapter on using CMOS for linear applications.  Although this book assumes some prior knowledge and is not a begginer's book, I found this chapter pretty easy to understand when I was learning about this stuff, and it gave a good explanation as to how they work.

[gez]

Sound very intresting... AND it sounds good.... But why aren't there more CMOS FX and amp, as there are (kazajillion times 3) op-amp,BJT,MOS,FET etc...  



is there a catch 22?

Sort of, they tend to be current hogs and suck batteries dry pretty quickly.  If I recall, the original EH hot tubes had on board power to overcome this.  However, they are more efficient run at lower voltages and run at 5V (you can use a micropower regulator) or lower (down to 3V - there are various techniques), it's totally feasible to run them from a battery.  My current design uses three stages and it uses about 3.5mA (less than a third of what it would be at 9V!), and that includes a buffer!

Other advantages of running them at lower voltages are higher open loop gain and less headroom (more distortion).  Downsides are higher output resistance but they're easier to buffer as any output buffer can be run off the higher 9V supply meaning less chance of clipping.

Bandwidth is also reduced, but that's not a big deal as some of the high-end fizz will go.  Having said that they're pretty quiet.

Build one, you won't regret it!

[Mike Nichting]

www.runoffgroove.com     listen to the Double D. It has a 4049 in it.
I am in the process of building one now.

Mike