LTspice models...a decent 4049UB?

[dschwartz]

Hi all, lately i´ve been playing and learning this useful tool.. but seems that the default models included are pretty bad..
I found a 4049UB model on the ltspice group, but is no good for linear applications, since the input has to be biased to 2.1V @ 5 Vdd..AFAIK, inverters don´t need to be biased..

By the way..jfet models included are awful..i couldn´t get one to simply amplify..so y founda j201 model and worked fine..

any one knows where i can find (or hoy to make) a decent 4049/69 inverter model??

[snap]

Do yourself a favour and read through all or a couple of "cmos" threads (again).
And follow some links from there.
You`ll learn more of what you need to understand the inverters.
Bias comes to mind.
To my knowledge, there are no satisfying analog/linear models for them out there.

[R.G.]

To do a good job of this, you have to sketch up the internal schematic of the inverter, made from N and P MOSFETs, then go adjust the parameters of the MOSFETs to be as like the ones inside the IC as possible.

This is not easy.

Simulators tend to be either analog simulators or digital simulators. Even when a simulator touts itself as a "mixed mode" simulator, that often means that they simulate the analog stuff as analog and the digital stuff as simple on-off transitions and past the two together. If you read up on MOSFET semiconductor physics, you will eventually have enough info to get the N and P channel models right and get good results.

Someday I hope to get there... 

[puretube]

You could try to test a "Discrete Tubeless" in Spice:
just the 2 transistors...
(let Rp & Rl,  be zero Ohm, and regard Vdd, Vss, In & Out as the respective connections of an inverter).