ggedamed wrote:marshmellow wrote:I don't understand why everybody always uses the inverters with those überhigh impedances. It's not like those chips would have problems driving 10k... But then use a compander to deal with the unnecessary noise they just produced
. Also, there are a couple of inverters doing nothing. Put them in parallel, cut the noise in half, and decrease the resistors even more because you have double the driving power.
Sadly, it doesn't work like this. There's a
nice explanation by PRR on DIYSB from which I quote:
Resistor hiss-noise is a non-issue: the MOS transistor self-noise is several/many microVolts, more than a 1Meg resistance.
Input impedance is often an issue. If you want gain of 30 and a 10K feedback resistor, you want a 330 ohm input resistor and have a nearly 330 ohm input impedance. That's awful low. In fact a CMOS amp won't drive 300 ohms well.
(Although, at 10V-15V supply, you hardly have open-loop gain of 30 so you can't really NFB-define a gain of 30 without cascading.)
OTOH with 1Meg feedback we get 33K input resistance, managable in many audio systems. For guitar we like over 100K, and 150K is popular. If we wanted gain of 30 (in this case maybe not) then the NFB computes to 4.5Meg. There's no realistic upper limit on the NFB resistor: CMOS gate current is VERY low, and in this case there's some gate current cancellation.
So, it's not the output impedance that's causing the resistor choice, it's the input impedance.
The inherent noise level of the CMOS inverters is really high. Sadly again, to make them clip/distort nice (and they can distort really nice) you have to starve them somewhat. The Bias control in a Emma Reezafratzitz will give you a good image of that. I fantasized since quite a lot ago about a "current compressor" to feed the inverters, so they have all they can eat at low levels (so lower noise), then when the signal goes large the inverters would be starved more and more, for the nice distortion they can make. Didn't find the time to do it yet, but who knows?
Also, in linear mode, the inverters eat gobbles of power. I can imagine using all six inverters leading to the overloading the smaller 9V/150mA wall-warts.
Bottom line, I think that the designer knew what he was doing and he played well enough within the constraints.
The designer missed several tricks: reducing the resistor values around the CMOS will improve the noise figure markedly, and won't affect any impedances "seen" by the guitar - that's why there's an op-amp buffer at the input. The use of the 571 as a compander will help a lot, but if the resistor (and capacitor) values are scaled, and the compressor is configured for a bit more output, the noise can be reduced to negligible amounts. It's also fun to make the output expander track the original guitar envelope (as a switchable option) so you can get "fuzz without sustain" which sounds bizarre and is very musically useful!
I don't like the fact that the CMOS switches are not electrically isolated from the output - I'd add either another op-amp buffer or a single transistor emitter-follower stage there for protection - CMOS switches are easily killed by static.
Further noise reduction can be achieved by making the CMOS stage gains smaller earlier on, and larger later - take a look at the Runoffgroove "Double D" for ideas there.
Finally - there's lots of noise (and consumption) variation in CMOS ICs. I've used some older 4069UBEs that I had lying around for years, and they're practically hiss-free. The 4049UB seems to be quieter (but more current hungry) than the 4069UB - it's always worth socketing the device and seeking the quietest one you can find from a batch! The "Emma" circuit deliberately "starves" the CMOS in an effort to reduce the noise (though they claim it's to allow the player to find the bias "sweet spot")....
If you want to get exactly the same kind of distortion with two orders of magnitude less noise (and no compander needed), just build a couple of FET µ-amps and feed one into the next. Drive them with a low noise op-amp buffer (sacrilege for the "golden eared" idiots) and increase (or decrease) the number of µ-amp stages to taste! The output can be buffered with the other half of your dual low-noise op-amp, and you can use a real switch (or relay) for bypass. This is almost exactly this Christmas's pedal given to a few friends. There's two footswitches - bypass and stage select - and three LEDs to indicate the number of µ-amps in the path. Levels are pre-set internally, and the thing sounds "gritty" (1 µ-amp), "overdriven" (2 µ-amps), or "screaming" (3 µ-amps)......
. More facts please,
