LM5155: LM5155 simulation very slowly and even not convergence

Bill Xu

Part Number: LM5155
Other Parts Discussed in Thread: PSPICE-FOR-TI,

Dear Product line engineer.

Nice to meet you! I’m Bill and an medical system engineer.

Recently, my customer ask me to supply a high voltage solution and show simulation result with PSPICE-For-TI for evaluate. Then I select LM5155 as controller to do a SEPIC design. the basic information as below.

Input voltage: 12V

Output voltage: 100V

Output current: 30mA

Frequency: 100KHz.

However, the simulation is very very-slowly and some-times even not convergence. I guess the Pspice model of LM5155 may has some issue since I just use this parts in the schematic. Could you please help find someone in your team to supporting and double-check the Pspice model of LM5155 has no issue? Attached all document and DSN file for reference. Thank you so much for great supporting!

regards,

Bill

HV-SEPIC.DSN

over 3 years ago

0 Bill Xu over 3 years ago

TI__Intellectual 1430 points

BTW, the simulation always paused but the status show 100% complete.

0 Stefan Schauer over 3 years ago in reply to Bill Xu

TI__Guru 60061 points

Hi Bill,

thanks for supporting the LM5155.

Let me have a look into this and come back till mid of next week.

Best regards,

Stefan

0 Stefan Schauer over 3 years ago in reply to Stefan Schauer

TI__Guru 60061 points

Hi Bill,
can you please also share the full PSpice project.
Thanks,

Stefan

0 Stefan Schauer over 3 years ago in reply to Stefan Schauer

TI__Guru 60061 points

HI Bill,

can you please check you settings for the Max step time and set it to 20ns:

Best regards,

Stefan

0 Bill Xu over 3 years ago in reply to Stefan Schauer

TI__Intellectual 1430 points

https://e2e.ti.com/cfs-file/__key/communityserver-discussions-components-files/196/SEPIC.7z

Dear Stefan

attached all the project to you. thank you so much for supporting!

regards,

Bill

0 Stefan Schauer over 3 years ago in reply to Bill Xu

TI__Guru 60061 points

Hi Bill,

please try to replace the power MOSFET with the standard Power MOSFET from the PSpice library.

Also set the simulation Max. Step Size as shown above.

Best regards,

Stefan

0 Bill Xu over 3 years ago in reply to Stefan Schauer

TI__Intellectual 1430 points

Hi, Stefan

it now convergence. but still very-very slowly and the simulation result is not correct as I expected. the Gate drive has no PWM waveform at all. this is no correct. is it correct for Pspice model of LM5155? thank you so much for supporting!

regards,

Bill

0 Stefan Schauer over 3 years ago in reply to Bill Xu

TI__Guru 60061 points

Hi Bill,

i have attached the version i have used - you may can check with that one.

Note: Spice simulation are very computation intensive, so will take a certain amount of time depending on the compute power of the PC.

https://e2e.ti.com/cfs-file/__key/communityserver-discussions-components-files/196/E2E_5F00_1129652.7z

Best regards,

Stefan

0 Bill Xu over 3 years ago in reply to Stefan Schauer

TI__Intellectual 1430 points

Hi, Stefan

yes, your are file runs good. but still need ~6 minute to finish the simulation. do we have any good method to reduce the simulation time? another issue is after 10ms, the output just arrive 100V. this is my expected. so I have to setup run time to 20ms for get more simulation result. the simulation time seems longer and stopped at ~11.5ms. not 20ms. could you please help to figure out the reason? thank you very much for supporting!

regards,

Bill

0 Stefan Schauer over 3 years ago in reply to Bill Xu

TI__Guru 60061 points

Hi Bill,

for acceleration of the simulation you can try to use check points. In a part of this tutorial they are demonstrated:

PSpice for TI: Transient Analysis | TI.com Video

Did you get the question on to much data? If yes, try the reduced data capture.

Best regards,

Stefan

0 Harry Hofner over 3 years ago in reply to Stefan Schauer

TI__Mastermind 19715 points

Hello Bill,

We have not heard from you for more than two weeks, so I am closing this thread now.
In case of additional questions you can re-open it or start a new thread if this one got locked.

Best regards,
Harry

Power management

Power management forum

LM5155: LM5155 simulation very slowly and even not convergence