Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

ERROR(ORPSIM-15138): Convergence problem in Transient Analysis

Status
Not open for further replies.

gypssysattva

Newbie
Joined
Oct 11, 2021
Messages
4
Helped
0
Reputation
0
Reaction score
0
Trophy points
1
Activity points
32
Hello, I'm trying to model ZVS boost converter in OrCad using cree mosfets spice model (C3M0120090D). Everything is fine when I connect Tj and Tc nodes to a voltage sources, but I want to leave it without connection. When I uconnect this pins, I get an error: Convergence problem in Transient Analysis. I have read some posts in this forum, and in the other ones. I tried to change the VNTOL, ABSTOL, ITL1, but it doesn't work. I'm pretty new in analog schematic modelling, so please help me deal with it.

There are my schematic and mosfet spice model and simulation log file in zip.
 

Attachments

  • boost.zip
    18.9 KB · Views: 72

Everything is fine when I connect Tj and Tc nodes to a voltage sources, but I want to leave it without connection.
Why? At least one node must be connected, otherwise you get infinite junction temperature. Or use a model without thermal modelling.
 

Why? At least one node must be connected, otherwise you get infinite junction temperature. Or use a model without thermal modelling.
I want to see power dissipation on mosfets, and when I connect these pins to voltage source, I get very strange results - about 14 kW on mosfet. When I left Tj unconnected and connect the Tc to the ground through resistor, I see about 7 W on mosfet which is looks fine, but I cant get results to the stop time because of error. Maybe I do something wrong.

And I didn't find this model without thermal pins.
 

It's clearly stated in the simulation model manual that at least one Tx node must be connected to a voltage source. Connection through resistor should work, if it corresponds to reasonable heat sink Rth.

I'd expect that the kW losses are somehow caused by your circuit.
 

It's clearly stated in the simulation model manual that at least one Tx node must be connected to a voltage source.
I saw it of cource, but as I told before it's work fine for everything except power. I was not sure that I was right usind resistor connected to case thermal pin.
Connection through resistor should work, if it corresponds to reasonable heat sink Rth.
could you give me a hint please? maybe some link with manual how to choose Rth?
 

Scaling is K/W, I'd expect 1-2 for real heat sink. The simulation can work of course with zero Rth as well.
--- Updated ---

I some faults in your simulation circuit.
- connecting both Tj and Tc makes no sense.
- V7 (the gate source for high side transistor) must not be grounded but refrenced to Q2 source

Also gate resistors are with 100 ohms unreasonably high.

I can't run your simulation because I don't have Pspice.
 
Last edited:

    gypssysattva

    Points: 2
    Helpful Answer Positive Rating
Thank you for checking the schematic, I've corrected it. But anyway I'm getting this error message.

I set Tstop = 10ms; max time step = 1e-8; VNTOL = 1e-3; ABSTOL = 1e-6; RELTOL = 1e-3; ITL4 = 1000.
 

Attachments

  • boost_sch.JPG
    boost_sch.JPG
    116.6 KB · Views: 111

Rth unreasonably high, Tmax will be exceeded with slightly above 1W power dissipation.
What's the purpose of L2-C4? Why is L1 so large?

Generally, convergence problem are often observed with MOSFET models, there's no simple solution for it.
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top