From: Marcin K. <mk...@po...> - 2005-08-14 15:43:46
|
I started using Qucs a few days ago. I'm using the 0.0.7 Windows = version. I'm actually working at my university on spiral inductors = simulation and modeling. Qucs core solver, without the GUI is what I was = looking for. Great job. Spice or other commerial solvers are much too = big for my needs.=20 =20 The distribiution contains qucsconv program to convert Spice netlists to = qucs netlist. How to use it? Using the command line I did get with -h = switch on both windows version and the one I compiled under cygwin, I'm = not getting any results. I used for test purposes *.cir files from = PSPICE. What could be the source of the problem? =20 Martin=20 =20 PS. Who prepared the polish translation of Qucs. I found the existing = sometimes weird and I would gladly help. |
From: Stefan J. <st...@gr...> - 2005-08-18 18:26:45
|
On So, 14.08.2005, 17:45, Marcin Kaluza wrote: Hello! > I started using Qucs a few days ago. I'm using the 0.0.7 Windows version. > I'm actually working at my university on spiral inductors simulation and > modeling. Qucs core solver, without the GUI is what I was looking for. > Great job. Spice or other commerial solvers are much too big for my needs. > > The distribiution contains qucsconv program to convert Spice netlists to > qucs netlist. How to use it? Using the command line I did get with -h > switch on both windows version and the one I compiled under cygwin, I'm > not getting any results. I used for test purposes *.cir files from PSPICE. > What could be the source of the problem? Of course you can use qucsconv as standalone. There is a manpage 'qucsconv.1'. QucsConv is used by Qucs when you try the SPICE component (see file data components). Try it... > PS. Who prepared the polish translation of Qucs. I found the existing > sometimes weird and I would gladly help. Darius did the polish translation (see CC). If you can help finishing the translation, this would be great :-) Cheers, Stefan. |
From: Stefan J. <st...@gr...> - 2005-08-22 07:17:27
|
On So, 21.08.2005, 10:30, Marcin Kaluza wrote: > Stefan, Hi! > hopefully, I was able to get rid of all mutual inductances in my netlists. > In the very particular case of the RLC networks I'm simulating, I was able > to erase them, adding their values or substracting from the inductances. > So > I'm can use qucs. > My goal is to calculate the 2x2 impedance matrix of my RLC networks, > calcualted for a single frequemcy point.. > I could do it the standard way, using the currents/voltages at the > entry/exit ports and having to change the elements connected to those > ports > to calculate the different elements of the impedance matrix which means in > fact running more than one simulation, to get the entire matrix. >>From my actual understanding of qucs, I could also use the S parameters > simulation and transform the resulting S parameters to Y parameters. ... to Z-parameters I think? > What I would like to achieve is to get from qucsator.exe as output only > and > only my 2x2 impedance matrix. I belive it is possible, but would it be > easy? > If I would like to modify the code for my purposes, where should I look > for? You can use the spice component, select the (2) ports you need and then place S-parameter ports at these ports, then select the frequency in the S-parameter simulation box. Also you need to place a ground at the reference port of the spice component. The add a equation to the schematic and add: Z=stoz(S) The simulation results will contain the Z-parameters then. Hope this helps, Stefan. |