6KG6, output stage & PSpice

[wfmali]

Tried to model the EAR859 output stage with pspice, using the duncan 6KG6 model. Does anybody know about the accuracy of this model? I really can't make the results to be close to what has been designed by Tim deParavicini. Certainly trust his real amp better than the simulation, but generally have good experiences with pspice and tube models.
The model gives me [email protected]=41V with his circuit, which can't be correct. Maybe it doesn't work for enhanced triode mode?
Thanks,
Marcus

 

[gingertube]

Before you model the entire amp set up a simple model with just the tube and power suppies and use step parameter function to generate the tube Ia vs Va curves (or Ia vs Vg2) for different grid voltages.

Then compare it to the data sheet. You will then know how good (or bad) the model is.

Make haste - slowly!

If you need some guidance on how to do this then the Norman Koren website is the go.

Hope this is of some help,
Cheers,
Ian

 

[wfmali]

That is pretty much what I did - modeled just the output stage. Found that on the Danielak Amp this model was used to derive the curves, so it should work. Will try if I get the same curves with my setup here.
Thanks for the input,
Marcus

 

[wa2ise]

What you could do is make a copy of the model, rename it, and start "hacking" it. That is, change some of the values of the parameters (one at a time and keep notes of what that change did). Eventually you'll figure out which parameter does what to the curves, and then you can tweak the parameters to more closely match the data sheets. That's what I've done to create new models for my 3f4 spice simulator.