Hex inverter chips

[HamishR]

Lately I've been building Red Llamas with the CD4069 chip - because I was given a bunch of 4069s by a friend. It probably sounds exactly the same as a 4049 but at least the 4069 is two pins smaller! The two-knob version seems to sound better to me and it really does sound good with the 4069. Such a great boost, if a little noisy when used as a boost.

But it makes me wonder why these pedals use a hex inverter when they only use two stages - is there no such thing as a dual inverter chip?

 

[jwin615]

CMOS circuits are generally at least a dozen chips if not dozens so you rarely need one or two of something. Also, they're super cheap to produce anyway so no sense in making multiple varieties of the same thing.

 

[HamishR]

It rankles with the minimalist in me! I would rather that the pedal either used a dual inverter IC or used all of the stages! At least having six of exactly the same stages gives you options in your layout.

 

[Robert]

There are single inverters if you want to be super efficient and don't mind working with SMD.

https://www.mouser.com/c/semiconductors/logic-ics/inverters/?mounting%20style=SMD%2FSMT&number%20of%20circuits=1%20Circuit&output%20type=CMOS

 

[mckillacuddy]

Lately I've been building Red Llamas with the CD4069 chip - because I was given a bunch of 4069s by a friend. It probably sounds exactly the same as a 4049 but at least the 4069 is two pins smaller! The two-knob version seems to sound better to me and it really does sound good with the 4069. Such a great boost, if a little noisy when used as a boost.

But it makes me wonder why these pedals use a hex inverter when they only use two stages - is there no such thing as a dual inverter chip?

CD4007 is three. Quieter too

 

[MattG]

As Robert said, there are dual- and single-inverter chips, but they are all in SMD format. I've searched, and can't find any through-hole versions with fewer than six inverters. Edit: clearly I overlooked the CD4007!

If it makes you feel any better, the unused inverters, assuming they are wired correctly (input tied to V+ or GND, output floating), won't waste any power (maybe a few micro amps of leakage, but negligible even on battery).

I've never personally tried to use more than three inverters, but both RunOffGroove and @Chuck D. Bones have reported that three seems to be the upper limit before the noise is too much. See the RunOffGroove Double D, Chuck's take on the Double-D, and Chuck's CMOS Overdrive for good discussions on using these inverters in analog mode as gain stages.

An interesting experiment: the inverters can be paralleled for more current-driving ability. I wonder if running them in parallel has any effect on the noise? I used to build DACs from the old tda1387 chips, and running them in parallel was a common trick for improving noise characteristics.

 

[HamishR]

As Robert said, there are dual- and single-inverter chips, but they are all in SMD format. I've searched, and can't find any through-hole versions with fewer than six inverters. Edit: clearly I overlooked the CD4007!

If it makes you feel any better, the unused inverters, assuming they are wired correctly (input tied to V+ or GND, output floating), won't waste any power (maybe a few micro amps of leakage, but negligible even on battery).

I've never personally tried to use more than three inverters, but both RunOffGroove and @Chuck D. Bones have reported that three seems to be the upper limit before the noise is too much. See the RunOffGroove Double D, Chuck's take on the Double-D, and Chuck's CMOS Overdrive for good discussions on using these inverters in analog mode as gain stages.

An interesting experiment: the inverters can be paralleled for more current-driving ability. I wonder if running them in parallel has any effect on the noise? I used to build DACs from the old tda1387 chips, and running them in parallel was a common trick for improving noise characteristics.

I accidentally connected the unused inverter inputs to V+ in an early experiment and was surprised that it worked when I found my "mistake". Then I rebuilt it with the inputs connected to ground and I think it may have been a tiny bit quieter?

I'll have a look at the CD4007! (edit) Oh it's still 14 pins... Still, it might be worth an experiment.

 

[HamishR]

And now I read that a CD4007 may be quieter... Interested again.

 

[HamishR]

Ok, I've been looking at the CD4007 but I don't know enough to understand the pinout as it relates to a CD4049 or 4069. I'd love to give it a go but don't know which pins I'd use. With the 4069 I'm using pins 1 & 2 as one inverter and 12 & 13 for the other. But the 4007 is a different type of chip so I don't know which pins would correspond. Any ideas?

 

[mckillacuddy]

Bottom diagram. 3, 6,10 are your inputs

 

[Robert]

Build CMOS Logic Functions Using CD4007 Array - ADALM2000 [Analog Devices Wiki]

wiki.analog.com

 

[MattG]

Take a look at the schematic for Chuck's CMOS Overdrive. He has the CD4007 pin numbers on it. You can see how, for example, he's using pins 3, 6, and 10 as inputs (as @mckillacuddy suggested above); pins 2, 14, 11 go to supply voltage; etc. In terms of "converting" a CD4049/CD4069 design to CD4007... I'm not sure about the resistor between pins 1 and 5, and again between 13 and 8, but not between 12 and 9. But besides that, I'm not 100% sure, but I think you might be able to start with the same RC feedback values from the '49/'69 and apply them to the '07.

 

[Feral Feline]

MOAR RE-DINGDING: There are some "CMOS Workshop" guides that Parasit's Fredrik put out, as well.

I just had a look and couldn't find the link. Hit me up if you want them, I can send you the 4 PDFs I made.

 

[HamishR]

Wow - plenty to be getting on with. Thanks guys! Let's hope I can understand enough to make it work. I think it will be worthwhile - I have discovered that the Red lama works very well as a boost for me.