These issues are not typically associated with a worn or clogged nozzle. While they could be, such problems are usually linked to blobs or “zits” on the wall. A clogged or inadequate nozzle causes uneven filament flow at random. However, your examples show that this is happening at specific layers, suggesting that as the layers change, something is affecting either the filament flow or the Z-axis motion. As @Panamon_Creel mentioned, this could be due to step loss, pointing to a mechanical issue in either the rods or belts.
___________________________________________________
To test this theory, I recommend printing primitives in separate prints. Start with a cylinder primitive and scale it to simulate the volume of your test models. This will allow for a very consistent flow, as the slicer handles it, and eliminate any issues caused by the model geometry. Then, repeat the process using a cube primitive.
Let me provide an example. I was demonstrating a method to do a clear side-by-side comparison of two different filament manufacturers. These images show that changing the filament mid-print to another manufacturer proved that in the first example the formula was obvious different while the second set of two spools, the formula was identical, despite the different brand names and spools. This is a useful diagnostic trick that you can employ here to verify that the mechanical movements of your printer are sound.
Example of two primitives with filament change using pause command:
The difference is stark. Most importantly, you can see a smooth transition as well as a color and reflectivity change, proving that the model remained intact and the settings stayed the same, with the only change being the filament formula.
Examples of two primitives comparing filament of different brands but suspected of coming from the same factory
Now compare that with the example I used with two different makers. In the two on the left, you can see where I paused the filament and swapped. The seam is obvious, but the color and consistency between the two manufacturers proves there was no difference in formulation. The model on the right was my control: it printed using a single spool source with no pause or swap, just to illustrate that it wasn’t the model causing the seam.
FYI: This also illustrate poor pressure advance calibration in cube example. Those ripples should not be there in a well tuned profile but I deliberately used the PLA Basic Generic factory profile so that everyone else could easily reproduce my test.
______________________________________________________
BTW: The full post with examples of the original study can be found here. So you think your high priced fancy-schmancy filament is the the best?