Fun with CMOS - part 2 - The Muff #49

[Chuck D. Bones]

Moving on from the CMOS Raincoat, I fiddled the gain on all four stages, deleted the FAT switch and put the TONE control between the last two stages. This is more like a BMP, except the limiting comes from overdriving the CMOS inverters rather than clipping diodes. More dynamic and less bottom-end than a BMP. The CMOS inverters clip slightly asymmetrically, for some even-order harmonics. The large feedback resistors (R6 & R8) allow the two middle stages to run pretty much wide open. Note the bright cap on the SUSTAIN pot. The last stage provides just enough gain to recover the losses in the tone network without clipping. Responds well to the guitar's Volume knob. Fort a fatter tone, increase C3 & C5. This time around, I used a Motorola MC14049UBE.

 

[lrgaraujo]

Still a long way to go before I'm capable of building beautiful looking boards, but it works! And it sounds VERY good, Chuck! As usual.
Tone stack sounds fuller in the mids (and that's what the tone stack calculator tells me as well).
PS: Belated happy birthday!

 

[Chuck D. Bones]

Thanks, Man.
That's right, no mid scoop. The freq response of the tone network is flat when the TONE knob is at about 4.5. It's easily retuned if you want a different freq response. Stage 3 is sensitive to loading, which means if we lower R9 or R10 it will reduce the treble content.

Which brand IC did you use?

 

[lrgaraujo]

It's an STMicroelectronics one (in theory at least, might be rebranded). It's quieter than I was expecting
I really like the tone network. This will probably replace the Big Muff I have boxed right now

edit: part number is HCF4049UBE

 

[Chuck D. Bones]

I believe that STM no longer produces CD4000-series. As I recall, the CD4049 in the Fat Sandwich I traced was made my STM. What is the Lot Date Code on yours?

 

[lrgaraujo]

What is the Lot Date Code on yours?

I have bought them at least 10 years ago, I reckon

 

[Chuck D. Bones]

My best guess is G4 is the factory, 1310 is the date code and TO is the production line(?).

If that's correct, then the part was packaged during the 10th week of 2013.

 

[lrgaraujo]

My best guess is G4 is the factory, 1310 is the date code and TO is the production line(?).

If that's correct, then the part was packaged during the 10th week of 2013.

That does make sense. I bought a bunch of 4049s when I built my Electric Mistress (early 2015, I once had the good habit of documenting my builds)

 

[Chuck D. Bones]

I tried it with a CD4069 (different pinout) and the gain was a little lower. Still sounded good. Then I changed R13 to 180Ω and put a C10K BIAS pot in series with it. With the BIAS turned up (min resistance) the sound is more open, less compressed, with a slightly stronger midrange. With the BIAS turned down, the tone is compressed and more dense (more harmonics).

 

[mzy12]

What do you do about the generally very high voltage noise of MOSFETs? I'm presuming in this case is just live with it, 'cos it's a muff and it's noisey anyways.

 

[Chuck D. Bones]

It's true that in general, MOSFETs are noisier than JFETs. Depending on the circuit impedances, JFETs or BJTs are the best choice for low noise.
MOSFET noise varies greatly from device to device. I've seen noisy BS170s and quiet BS170s. Some CMOS 4000-series are noisy and some are quiet. The CD4049 and CD4069 I have on my breadboards are quiet. The guitar's noise floor (mostly hum) completely masks any noise in the CMOS devices. The CD4049s and CD4069s currently in production (TI is the only source these days) are noisy. For whatever reason, their CD4007s are not noisy. Operating in the linear mode is not specified in the datasheets or tested on the production line. We get what we get and it can all change if the manufacturer changes the recipe. The NOS CD4049 or CD4069 parts from National, RCA, Motorola or Solid State Scientific that I have tested are quiet. There may be others, but I have not tested them.

 

[mzy12]

It's definitely annoying that MOSFETs can be so noisy, there are so many cool designs that use them!. But I guess you just gotta look big picture and realise that electric guitar signals are just noisy in general. There's a certain amount of voltage noise that's acceptable, but I really do like keeping it as low as possible as I run a pretty complex and big rig, so that stuff does add up.

 

[Chuck D. Bones]

It helps to cherry-pick them.