12-02-2009 06:51 AM
Hope someone can help here.
I'm trying to create a simulation only component - a TI THS4271 amplifier. I've used the component wizard in the usual way and have used the SPICE model provided by TI. This was a .lib file - I changed the file extension to .cir to enable Multisim to pick it up. I've also used the .txt file provided and have the same problem.
On running the simulation I get numerous errors appearing on the error log of the form
"Error on line 72 in C:\................."
"Unable to find definition of model - default assumed"
Any idea what the problem can be? I've pasted the TI model below - many thanks in advance
* THS4271 High Speed amplifier "macromodel" subcircuit* updated using Model Editor release 9.2 on 11/05/02 at 13:11* The Model Editor is a PSpice product.** connections: non-inverting input* | inverting input* | | positive power supply* | | | negative power supply* | | | | output* | | | | |*$.subckt THS4271 1 2 3 4 5*Offset and CMRR Vos 1a 9 .005 Ios 1 2 .5u* upper Vic range limit drc1 16 17 dx drc2 16 18 dx Vcp 3 16 dc -0.4* input stage rc1 17 11 176.8 rc2 18 12 176.8 L- 2 2a .8n q1 11 2a 13 qx1 L+ 1 1a .8n q2 12 9 14 qx1 re1 13 10 159.07 re2 14 10 159.07 Cdif 1 1c 0.8p Rcdf 1c 2 50 Ccm 2 2b 0.4p Rccm 2b 99 50* lower Vic range limit d10 15 10 dx v10 15 4 dc 1.2 Iee 10 4 dc 3e-3 Icc 3 4 15m Rcc 3 4 2500* gain stage and dominant pole ga 21 99 value = {(limit(V(11,12),-.447,.447))*-35.6m} ra 21 99 158k ca 21 99 15.9E-12* GAIN STAGE SWING LIMIT DPC 21 23 dx VPC 3 23 1.7 DNC 24 21 dx VNC 24 4 1.74* zero ez 26 99 21 99 10 rz1 26 27 900 cz 26 27 .2p rz2 27 99 100* phase shift stage gps 25 99 27 99 -100.0E-6 rps 25 99 10.0E3 cps 25 99 10E-15 egnd 99 0 poly(2) 3 0 4 0 0 .5 .5 X_OP 25 99 3 4 5a THS4271_OP Ro 5a 5b .1 Lo 5b 5 .2n Rco 5c 99 10 Co 5 5c .8p.ends*$* Output stage* connections: non-inverting input* | inverting input* | | positive power supply* | | | negative power supply* | | | | output* | | | | |.subckt THS4271_OP 1 2 3 4 5* GAIN STAGE SWING LIMIT DOPC 1 38 dx VOPC 3 38 1. DONC 48 1 dx VONC 48 4 1.* UPPER DRIVE STAGE ROP 3 34 8.5 HLP2 34 33 VLP 30 VOP 33 32 0 HLP1 35 0 VOP 8 RLP 35 36 8 DLP 36 37 dx VLP 37 0 0 EOP 32 31 Poly(2) 2 1 3 4 -.8 1 .5 DOP 31 5 dx* LOWER DRIVE STAGE DON 5 41 dx EON 41 42 Poly(2) 1 2 3 4 -.8 1 .5 VON 42 43 0 HLN1 45 0 VON 15 RLN 45 46 8 DLN 46 47 dx VLN 47 0 0 HLN2 43 44 VLN 45 RON 44 4 12.ends*$*DIODE MODELS.model dx D(Is=800.00E-18)*$.model qx1 NPN(Is=800.00E-18 Bf=272.73 af=0 kf=9e-22)*.model qx1 NPN(Is=800.00E-18 Bf=272.73 af=2 kf=8e-8)*$
12-02-2009 09:44 AM
Hi there,
What you want to do is to create a component with the model. Here's a link on how to do that:
http://zone.ni.com/devzone/cda/tut/p/id/3173
In step 6, where you are asked to provide the simulation model, just copy and paste the SPICE model from the .lib file which have and you will be good to go.
Hope that helps.
12-02-2009 04:13 PM
I would also download the model again from the Texas Instruments website; it looks like the formatting of the model you posted has been corrupted.
12-03-2009 09:41 AM
Thanks much for your prompt responses.
I've used the procedure described in http://zone.ni.com/devzone/cda/tut/p/id/3173 successfully in the past and I'm not sure this is the problem.
Also the SPICE model as it appears in my original message is corrupted, but this occurred when I pasted it into the message, not when it was loaded into the component wizard.
I've successfully created models using the component wizard for Analog Devices components, but I've tried to create several Texas Instruments devices including the THS4271 / THS4275 but had no luck at all. I've also not had much luck with Nat Semi devices.
Has anybody noted any problems with TI SPICE model formats, as downloaded from their site, in Multisim? The SPICE model for the THS4275 can be found at
http://focus.ti.com/docs/prod/folders/print/ths4271.html#toolssoftware
if anyone wants to check this (its a nice well-behaved 1.4GHz opamp - very useful in HF work)
12-03-2009 11:02 AM
I suspect it could be your version of Multisim; these models should all simulate correctly in the latest version (10.1).
But just in case, I have attached the THS4271D component that I have tested in a simple test bench and appears to be simulating correctly. You will need Version 10 or newer to import this component.
Hope this information helps.
12-16-2009 04:02 PM
Thanks much again for the replies.
I can confirm that the THS4275D model sent through works in Multisim 10, as do several of the other component models I downloaded from Texas Instruments which failed to work in Multisim 9.
It seems that Multisim 10 provides improved support for models provided by manufacturers (at least from TI). If a model fails to run on an older version, its worthwhile checking to see if it runs on 10 by downloading an evaluation copy.
12-16-2009 04:24 PM
12-18-2009 03:22 PM
Its good to hear that you managed to get the model working.
Every version, we improve our model support capabilities. If you come across a model which is not supported by the latest version of Multisim, please let us know. We will do our best to add support for those models in future releases.
Thanks!