-
-
Notifications
You must be signed in to change notification settings - Fork 109
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Ngspice netlist rework, fixes #463 #513
Conversation
Thanks for the contribution! I will test it in the next few days and post my feedback. I have seen also fix for #124 in this issue (export DC OP to the dataset). |
Yes, but for now it works if DC is not the only simulation. I have to figure out how to handle F8. |
I will remove checker error when the schematic contains DC simulation only. The F8 mode (DC bias) may be left unchanged. Qucsator doesn't export data to dataset using F8 mode too. |
I have found an issue with Nutmeg equations. Nutmeg equations are not attached to SPICE simulation now. This may lead to computation error and simulation failure. For example look at the attached schematic. The equation for |
I understood how new Nutmeg equation works. It is required to enter the simulation name in the |
A drop-down list wont allow entering partial simulation name that applies to all simulation of the same type, SP for example. One can have an equation 'db_s21=db(s_2_1)' that applies to all SP simulation and another equation that applies to a particular simulation SP2. A combo-box maybe, I have to look how SWeep does it. |
Just found an ussie, '.INCLUDE ngspice_mathfunc.inc' is the first line in the netlist. Above '* Qucs 24.1.0' line. I removed the '* Qucs 24.1.0' duplication and it was OK, but now I see that it was because there was no ngspice_mathfunc.inc file in the test build. Will fix it as soon as I have some spare time. |
0625e1e does it. |
I have inspected the
|
Yes, the DC simulation works as expected now. It makes sense to remove the |
I have added TODO list #518 |
I have tested this PR with my simulations and found no major issues. I am merging this. The improvement of the component properties dialog could be implemented later. @ivandi69 Please stay connected if some issues related to this PR will raise in the next weeks. And thanks for good job. |
I updated the ngspice 100 watt amp example to work with the latest nightly release. AC, Tran, Four and FFT in a single analysis. |
This fixes a regression in ra3xdh#513. If the only simulation is custom and it contains custom prefixed output, the custom prefix is not applied.
This fixes a regression in ra3xdh#513. If the only simulation is custom and it contains custom prefixed output, the custom prefix is not applied.
* Always apply custom prefix enclosed by '#' This fixes a regression in #513. If the only simulation is custom and it contains custom prefixed output, the custom prefix is not applied. * Parse nutmeg scripts to find used simulations This avoids prefixing when custom simulations don't conflict. * Relax regexs to allow variable names that include point Variable names like h2.out are valid so no reason to restrict them in custom simulation dialog and scalars print parser. --------- Co-authored-by: ivandi <ivandi@localhost>
This rework fixes #463 using the recently added custom prefix feature.
Prefixing is activated only if conflicting simulations are found, so the existing schematics are not affected.
Every simulation has its own output file named spice4qucs.<simulation_name>.extension, ex: spice4qucs.ac1.plot.
If any conflicting simulations are found, ex: AC and FFT, prefixing is activated and when the dataset is built the <simulation_name> prefix applies to all simulations. If a conflicting simulation is added to a schematic the existing diagrams have to be updated with the new prefixed variables.
Other changes:
Here are some tests:
463-tests.zip