THS4521 Pspice Model Error

This question is answered

Posted by daniel contreras

on Apr 23 2012 05:10 AM

We are trying to run simulations in PSpice with the model provided for this amplifier, but there is an error when launching the simulation. Here is a screenshot of the circuit:

And the errors encountered are related to the subcircuits in the model, several times:

**** EXPANSION OF SUBCIRCUIT X_U1.X_S2 ****

X_U1.X_S2.S_S2 X_U1.S1_2_DRAIN X_U1.N2663051 X_U1.N2662667

+ X_U1.VEE_BUF X_U1.X_S2._S2 X_U1.X_S2.RS_S2

X_U1.N2662667 X_U1.VEE_BUF 1G

.MODEL X_U1.X_S2._S2 VSWITCH Roff 1e6 Ron 1.0 VH

----------------------------------------------$

ERROR -- 'VH' is not a model parameter name

Is there anything I'm missing? I have seen that people have already made simulations with this model.

Regards

Daniel Contreras Solórzano

Grupo de Tecnología Electrónica

E.T.S.I. Sevilla

All Replies

Posted by John Miller - WEBENCH Design Center

on Apr 24 2012 08:40 AM

Expert4735 points

Daniel,

Thank you for the inquiry. I will look at the issue you have raised and should have a reply by the end of today.
One question: Which version of Pspice are you using?

Regards,

John

Posted by John Miller - WEBENCH Design Center

on Apr 24 2012 09:58 AM

Verified Answer

Verified by John Miller - WEBENCH Design Center

Expert4735 points

E2E_THS4521_042312.zip

Daniel,

I downloaded the THS4521 Pspice model from the Texas Instruments web page and quickly built up a circuit based on the one you sent.
It seemed to run fine in my version of Pspice (Cadence 16.2).
DC sweep and transient sims ran without any reported problems and gave the expected plots. The Pspice project has been uploaded to this page as a zip file.

A closer look at the six switch models used in the THS4521 macro found they use different parameter sets in the .MODEL statements.
Switches S5 and S6 appear to use classic Pspice parameters (Voff, Von), while the others use a different syntax (VH, VT) that is identified as SPICE3 in my references.
One of my colleagues said the VH/VT parameters might belong to ISSpice. At any rate, they don't seem to cause any obvious problems.

There may be some invisible or out-of-place characters embedded in the macro netlist. One way we have found to fix this is to open the netlist with the Pspice Model Editor and save it. It might also be worth downloading the macro netlist again from ti.com.

Please let me know if the problem persists. If it does, we will try to find a solution.

Regards,

John