AN: Basic OpAmp Output Stage Spice Model Posted to my Website

That's it, insures output currents are seen by the appropriate supply rail? If the output circuit comprised BJTs or FETs in a real-world fashion, this wouldn't be necessary.

...Jim Thompson

I'm trying to appropriately match real world devices... better BJT and MOS to come. ...Jim Thompson

I'm also seeking swing and current limits, independent for each rail.

Plus, ultimately, I will match datasheet curves, irrespective of actual chip implementation.

The ultimate object is to just insert information from a datasheet and spit out a model that matches every specification :-)

I already have the "core" portion of the OpAmp done... put information from datasheet into a spreadsheet and it spits out coefficients for a behavioral expression (only ~12 lines long :-) that matches...

DC Gain GBW Excess Phase Output Impedance Slew Rate + Slew Rate - Swing Limit + (and softness) Swing Limit - (and softness) All Noises versus frequency

And can handle OpAmps with built-in compensation, or external.

The most variability from device-to-device will be modeling the quiescent current behavior versus voltage and temperature... every designer does that differently. ...Jim Thompson

That sounds super useful Jim, we'll be eager to see it when you post to your website.

If it meets the rest of the criteria, w/o temp modeling, it'd still be super useful.

a few comments: * Output Impedance - this is vs freq and load? * Slew Rate - this is vs input error voltage?

Only if that data is provided on the datasheet, OR, I have access to an actual device-level model... only one of my modeling clients has agreed to provide true internal information :-(

Yes. ...Jim Thompson