Designer’s Guide Community

‹ Previous topic | Next topic ›

Pages: 1

Reshaping transitions in veriloga voltage sources (Read 1683 times)

afisher

New Member

Offline

Posts: 2

Reshaping transitions in veriloga voltage sources
Jun 13^th, 2007, 5:04am

Hi all,

I refer my question to veriloga voltage sources, generally full-swing, but not only.

The default veriloga shapes of voltage signals at the transition points is something that looks like 1st order polynom. This might cause convergence problems in some cases, as well as large CPU resources due to non-continuoues derivative of the signal.

Does anyone has a "shaper" solution/equation for solving these "ugly" part of the voltage signals?

Thanks in advance,
afisher.

IP Logged

Geoffrey_Coram

Senior Fellow

Offline

Posts: 1999
Massachusetts, USA

Re: Reshaping transitions in veriloga voltage sour
Reply #1 - Jun 13^th, 2007, 6:04am

afisher wrote on Jun 13^th, 2007, 5:04am:

Hi all,

I refer my question to veriloga voltage sources, generally full-swing, but not only.

What is a "veriloga voltage source"?? Do you mean a voltage source from Annex E, which tells you how to instantiate voltage sources from a Spice simulator? In that case, it's your job to pick a well-behaved source. Eg, PWL is continuous but its derivatives are not, but that's not Verilog-A's fault.

Otherwise, you can write a voltage source directly in Verilog-A, and it's completely in your hands how you write the output voltage as a function of time.

-Geoffrey

If at first you do succeed, STOP, raise your standards, and stop wasting your time.

IP Logged

Pages: 1

‹ Previous topic | Next topic ›