回應： Problem in spice model of AD8538...

[zenith]

Problem in spice model of AD8538...

Please see the attached file.

The spice model of AD8538 was copyed from Analog Device's website.

After simulation, it comes out the following warning message and the simulation is suspended.

error in C:\DOCUME~1\AAA\LOCALS~1\TEMP\02.CIR.cir(127):ez (145 0) (45 0) 1
Warning on line 127 in C:\DOCUME~1\AAA\LOCALS~1\TEMP\02.CIR.cir: ez:x8538 $(145:x8538 $0):x8538 $(45:x8538 $0):x8538 1
warning: unknown parameter 45:x8538  

Is there anything wrong in the configuration of the component AD8538?

[lacy]

There are a lot of spice models that will just not work in Multisim. This is apparently one of them. The way I see it if it wasn't in the database originially then nine times out of ten NI couldn't get it to work either. That is why it isn't there to begin with.

You may have to select one already in the database that is close to the parameters of the one your trying to use. I am not a Spice Model guru and I really don't know how to fix the model you have to make it work.

One request that I have for NI. When it comes to these models, we should not be having any problems having them to work. Multisim should be compatible with all Spice Models from all versions and types of Spice.

[zenith]

I see. Thx for your reply.

I've installed Multisim Analog Device Version.

However, there is still no AD8538 part. I tried to build it by component wizzard (in Multisim 10 Pro) but the simulation result is not good.

[OStep]

Hi Zenith,

In the model, replace

EZ (145 0) (45 0) 1

with

EZ 145 0 45 0 1

Future versions of Multisim will be more syntax friendly by supporting more variations. Better error reporting will also be addressed.

[zenith]

The simulation can execute now. But the result seems to be wrong.... @-@

[OStep]

If you're referring to the 5V output offset and if that is not a characterstic of the op-amp part, its a problem with the op-amp model, not the simulator.


Message Edited by MaxNI on 12-27-2007 02:30 PM

[zenith]

Hmmm...Thx!

I've asked ADI's SPICE Applications engineer to see if there are something wrong in AD8538's model.

[lacy]

Hi Max,

"Future versions of Multisim will be more syntax friendly by supporting more variations. Better error reporting will also be addressed"

I would like to thank you for the above. This would be a great help. Thank you very much for putting it on the list for future versions.

[zenith]

Finally.... I got correct simulation results in Multisim 10.1.1.

AD8538 is one of the analog components in the analog database now. 

Cheers!  ^___^