I'm trying to run an AC Analysis on a 9-pole 3 kHz low pass filter using the LT6235 model. The simulation does not seem to converge, even after a half hour. When I replace all of the LT6235's with LT6238s, it runs in a few seconds. When I replace the LT6235 with AD8674, it runs virtually immediately. The filter was originally designed in Analog Filter Wizard. At one point I saw an error message relating to "DC operating point not found". I put in ".options cshunt = 1E-15" just in case.
The simulation seems to run fine with two other op amp models but not with LT6235. Is this a model issue or something a little more obscure?
I was told that if you add the following spice directives to your schematic the simulation will run immediately:
.options gminsteps = 0.options srcsteps = 0
I tried it and it did work.
There is also another technique of keeping the ESC key down while you invoke a simulation (running man) to do a transient analysis, and that also seems to work. I don't know what the effect of keeping / mashing the ESC key but it does seem to work.
I think this is very simply explained. I know that any type of analysis consists of successive stages. When running any type of analysis, the first step is always to search for a DC operating point, for example.
These steps can be canceled by pressing the ESC key and this is written in the LTspice status bar in the lower-left corner of the screen. If the analysis is very slow, you can easily read it.
I also experience difficulties when starting circuits with OP177, any analysis is slow and in the status bar I see that the process of finding solutions to equations and iterating through them is going on. I'm a little annoyed.