My 3D rendering has an offset on almost all the throughhole resistors. Does anyone know why this is?
Is there a way to fix it? I’m using version 4.0.5 on Linux.
I really dig the functionality of the software, but some of the workflows are still clunky. But hey, it’s free software. And still does the job in most cases, though I think some features are missing or burried such as the SPICE simulator and autorouting and so on.
This can happen when the footprints are out of sync with the 3d models. You might need to download new 3d models from https://github.com/KiCad/kicad-library
It looks like the origin point of the 3D model is in the centre of the resistor, while the origin of the footprint (in line with KiCad Library Convention) is at pin 1 for THT components.
In KiCad main program, click on Preferences - Configure Paths. Please post a screenshot of what that shows.
In Pcbnew, move your mouse over one of the resistor footprints (are all the resistors using the same footprint?). Then press ‘E’ key to bring up Footprint Properties. Click on Footprint Editor button and check that the X and Y axes pass through pin 1 of the footprint.
While in Footprint Properties, select the 3D Settings tab and post a screenshot of that, please.
In kicad stable there is no way do update 3d models from within kicad. (The nighly builds have a 3dshapes downloader that does this for you.)
You need to download the files manually from github and either place them into kicads system directory.
(For windows somewhere under c: for linux: /usr/share/kicad/modules/packages3d/)
Or place the 3d files anywhere, as long as you set the KISYS3DMOD path variable accordingly. (kicad main window->preferences->configure paths)
The footprint libs are normally setup in such a way that they are automatically downloaded directly from github everytime kicad needs them. (This leads to problems like you have now where only a part of your lib is updated. In this case the footprints.)
The library is Resistors_THT these days.
The 3D model is preset with a 5.08mm offset, so @hackscribble is correct about the model using the wrong origin.