snvmb14a - ERROR(ORPSIM-15108): Subcircuit is undefined

Hello Ponsoby, 

Just to confirm, did you follow these steps: 

If so, could you share your schematic?

All the best,
Carolina

Carolina,

The link you have shared discusses adding third party components to a schematic. I've downloaded a TI PSpice file (snvmb14a). Are you able to download this file and test whether you're observing the same errors when running the simulation.

Hey Ponsonby, 

Understood my mistake, I thought you were having trouble importing LM6142.

I have redirected your ticket to the correct product line. 

All the best,
Carolina 

Thank you Carolina.

Hi Ponsonby,

Our group supports the LM5170.   We will ask our responsible engineer to take a look. Stay tuned.

Thanks,

Youhao Xi, Applications Engineering

Hi Ponsonby,

I can run the file download here with no errors on my end.

https://www.ti.com/lit/zip/snvmb14 

There are a few things you can try.

1. confirm it shows up in your simulation setting/ configuration files/ library and in your project tree, confirm the olb file shows up in Design Resources/Library, and the lib file shows up in PSpice Resources/model libraries

2. If step 1 is confirmed, try close and reopen all the OrCAD software and re-run the simulation.

3. Try open the model directly from the schematic, the LM5170 is on the top schematic, the LM6142B is in the Iset driver block. For each of them, you can check the model by right click on the symbol, and then select "Edit PSpice model", check if the model pops up.

Thanks,

Yinsong

Thanks Yinsong.

After saving this file in a different location and running the simulation, I can now simulate. However, I am getting the following message:

"INFO(ORCAP-15099): Detected an imported model containing transistors or diodes. For such models, PSpice for TI supports a minimum of one and maximum of three traces. Reduce the number of traces and simulate again."

(1) Which component is causing this so I can replace it with a TI part?

(2) I have a number of markers on the schematic and a total of three 'Mark Special' markers. Whenever I change which markers are defined as 'Mark Special', I get a different number of listed variables that I can plot. Sometimes it's 25, other times 20, other times it's less. Almost all the time the signals that I want to plot are not listed. Does this appear with your setup? When I 'Add Trace', how do I configure it so that it shows all the listed variables in the schematic?

Hi Ponsonby,

Good to see you get rid of the model errors.

1. That is weird because I don't see any imported models besides the LM5170 in the schematic. Please let me ask the PSpice team. Could you please let me know your PSpice for TI version number?

2. Could you share with me a screenshot of the mark special you did?

Thanks,

Yinsong

Yinsong.

1. 17.4-2020 S004 Windows SPB 64-bit Edition

2. Attached screenshots.

Buck no longer works (no gate switching) after removing the breakers from the original circuit and replacing with 20k resistors across BRKG and BRKS as per datasheet. Are you able to remove the breaker and simulate in this configuration?

Hi Ponsonby,

Just to make sure that we are on the same page, 

So are you able to run the simulation now? Or are you still having the errors?

Thanks,

Yinsong

Yinsong,

I'm able to run the simulation however, after removing the breaker circuitry, there's no gate signals from the LM5170. Can you remove the breaker circuitry and run the simulation and report back your findings?

I also attached images of the limited amount of signals I can plot. Is this what you observe as well?

Hello Ponsonby,

Please give us some days to finalize these simulations.

Hi Ponsonby,

Sorry for the delay. It is the nFault that disable the circuit. But I'll need to check the internal operation to figure out what is triggering the short detection failure.

For now, you can skip the detection process by connecting SYNCOUT to 10kohm resistor and that will enable the controller.

Thanks,

Yinsong

Can you kindly provide an update to my query?

Thank you in advance.

Are you able to provide an update with my query?

Hi Ponsonby,

This is under investigation. Please allow us some time.

Thanks,

Youhao

Hi Ponsonby,

Please see my last comment.

It is the nFault that disable the circuit. But I'll need to check the internal operation to figure out what is triggering the short detection failure.

For now, you can skip the detection process by connecting SYNCOUT to 10kohm resistor and that will enable the controller.

Please let me know if you have any question.

Thanks,

Yinsong