Strength of CD4000 substrate diodes

It's been a while, but I have done just that... "I won't sign off on it".

...Jim Thompson

Well, I opted not to push things this far. It'll be under 100uA now.

We can exclude all discontinued parts. The datasheets for part in production so far all seem to support 10mA. Which is actually quite good.

With 74C you have to be really careful. I wouldn't ever drive any of these into the substrate diodes.

Great! I'm proud of you ;-)

...Jim Thompson

Which reminds me of a circuit I designed recently:

tiny-logic gate

which upset some people, but I left it in because it's so cool.

John

Probably works for very large signals that are continuous... but I'm upset ;-)

...Jim Thompson

I can already hear the scoffing at the design review for the circuit I am doing right now.

Did you ever use one of those blazingly fast inverters with feedback for an analog job?

Amazing, working with a 30+ year old logic family!

Yep. A Fairchild sales guy told me umpteen years ago that this series was doomed and to be obsoleted soon. Didn't believe him. Sure enough, then everyone even came out with SO packages. CD4000 is here to stay, for a long time.

You can do amazing things with these. Plus you can run off batteries totally unregulated. That is a huge benefit when working on a design for a disposable. Same for very low power designs that must run off the mains rail.

Hello Mark,

A little P.S.: I am using a 2N3904 in this new design. That ought to be as old as Methusaleh. But it worketh just fine!

Naaah! A 2N3904 is a "staple" ;-)

Same with 2N3906.

...Jim Thompson

-- | James E.Thompson, P.E. | mens | | Analog Innovations, Inc. | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | Phoenix, Arizona Voice:(480)460-2350 | | | E-mail Address at Website Fax:(480)460-2142 | Brass Rat | |

No, is that what you're doing? Naughty boy.

My favorite non-approved application lately is using insanely fast dirt-cheap LVDS-to-TTL receivers as comparators or as pin drivers, or both. At least, that's one I can talk about.

We did find a really spiffy pin-fin heat sink we can epoxy to the top of one chip that we're using in a very unorthodox way, so if anybody does a less than exhaustive spy job, they'll think we have a custom asic.

John

Most companies require 40ma latch-up as a bare minimum, but simple logic chips should be more latch-up resistant, say 80ma. Of course, it is really bad engineering to test such limits, and my numbers are room temp.

My 1972 Motorola data book has the 2N3904 in it. The modern data sheet has the same graphs as the old data book, even same layout order.

Earlier this year I delivered a tiny board using 4000 series logic. Mighty fine logic series!

I'd probably still design some things with Ge transistors if the major mfgs would make them. When I was a kid I built all kinds of stuff with these. It was amazing to see a compressor amp still working on a single AA battery that already had begun leaking. Under 1V and it just wouldn't quit. Actually I was happily using it and only discovered the sub-1V condition because I saw something oozing onto the table.

Absolutely. I wish that they would offer a device similar to the CD4007 on the 74HC process.

On Fri, 08 Dec 2006 08:15:16 -0800, John Larkin wrote

I've done that, too, for a level translator. If the potential on the CMOS input "wanders off" it'll be clamped into range by the input diodes when the next transision occurs. Just keep that clock going!

But it sure looks weird.

robert

Yeah. The input here is from 3.3 to 100 volts p-p. It varies from kilohertz at the high voltage to as much as 20 MHz on the low end.

John