So far i am very happy with my P1S+AMS. I printed several models from other people but now i need to print something i made on my own.
I made a simple box for a pcb i made. It is a rectangle box but a few holes in it. I made it in fusion360. But when it was done printing it seems to have created some layer shift? The walls are very ugly.
I tried it and it seems to make it a little better but only on 2 sides. The other sides I still keep getting the issue. Tried tin wall detect etc, lowerd the wall speed but does not seem to fix this.
Would making the walls thicker help ?
First. Thank you for posting the 3MF file and clear photos. This really takes the guesswork out of this question.
Cause
This has a straight forward answer. You are seeing changes in filament directly caused by structures in your model. It’s not your fault, it’s the laws of physics.
These features here are what is causing filament flow to change:
The change occurs at the layer height you are observing. The bottom of the hole changes filament flow.
Dry your filament. Weigh it before and after to confirm that moisture was present. This alone will save you a lot of time in pursuing the next step.s
Assuming your filament is dry calibrate your filament manually.
– Do not trust the factory profiles.
– In this case, I would look at reducing max flow rate and making sure you did a flow rate calibration. These are two different Orca tests.
If this is your model. Consider thickening the walls. 3 layers is too thin to get perfection, 5 or more layers is best.
Assuming you have access to the model and can increase the number of walls, do the following steps next:
– Increase wall loops. This will create more solid walls.
– Change the default print order from inner/outer to outer/inner. This will lay down the outside layer first giving the filament more time to set up before the next pass.
Thank you for your time and explanation. I appreciate you looking into the .3mf file to help.
Before you replied i did try the following:
Reduced printspeed of the outer wall to 60mm/s but has no effect
Increased the wall thickness in my model from 1.5mm to 2mm. Reason i did that is because i thought with an 0.4 nozzle and 2 walls per side i need 4*0.4–> 1.6mm minimum. I does not have much effect in the end. Think because if the Arachne wall generator.
I did already change print order from outer to inner. has some effect but not solve it all.
It seems with the changes above (exect print speed) i get decent result on the long side but not on the sort sides. See images below.
Well, you are on the right track. Situations like these rarely have a single cause. I often compare this hobby to baking rather than science because, like in baking, temperature, time, and humidity can have varying effects, even between events.
I suggest doing a simple test. To save time, change the layer height to 0.28 to speed up the process. Then, enable the “silent” mode in the slicer settings to slow down all print head movements. While it’s true that this adjustment can be made in the speed settings, it requires more time for fine-tuning, whereas silent mode is a single-click solution. Keep in mind that silent mode doubles the print time, which is why I recommended the 0.28 layer height as a tradeoff for this experiment.
You can also change this on the front panel which for me is actually preferable because I can ensure that the setting stuck.
Here’s the settings if you prefer to try it in the slicer and save it as a profile for future use. I’d recommend focusing on cutting the outer wall and inner wall speeds by 50-75% in two different print runs to see if either extreme changes the outcome.
I increased the walling but makes no diff.
It starts shifting in the height were i have the opening for an USB-B. But the side it has the big opening is better then the side that does not have the big opening.
So it would seem that the change in movement because of the opening is the reason. And i can understand the retraction argument but should it then not be on the side of the big opening ?
I will try in the meantime to the test you propose to see if there is a difference.
btw. I still am using Arachne and not the classic one.
Not necessarily. Part of the issue is that the filament has to have time and distance to reach it’s state where it begins to deposit molten filament. Where it might seem bewildering or even counterintuitive that the two side should behave similarly, they may not if the distance that the nozzle has to move compared between the two is dramatically different. This is why I lean towards a theory of speed being an issue although I’ll freely admit, this needs to be tested with your conditions.
One other test you can try is to eliminate draft or fan induced temp variations. This is really unlikely but it’s easy enough to test. Take your model and rotate it on the Z-axis 225 degrees so that the opposite corner is facing 45 degrees from the X-Y coordinate. See if that changes the outcome. If one side is being influenced by let’s say the chamber fan or the door being open, the change in orientation should also change the result.
I suspect that you are fighting moisture in filament, you can make a number of slicer changes to mitigate these and perhaps come up with a solution without drying but it’s just going to follow you to the next model etc…
I would quickly test this by loading the filament, heating the nozzle to printing temp, extrude it until its coming out of the nozzle normally and then inspecting, does filament slowly creep out of the nozzle when you aren’t extruding? Moisture if so.
The answer is straightforward, but most people lack the patience to follow what I am about to suggest. The Bambu community, by its nature, attracts customers who are not tinkerers and prefer a push-button solution. Based on your questions, I suspect you are not in that category, so I’ll provide an explanation.
If you want to micromanage speed at the layer level, the tool you need is the “Height Range Modifier” tool, which is cryptic and not well documented.
Right-click on your model and select “Height Range Modifier.”
I lowered the speed of the walling. And according to the speed analysis of the bambulab it is more even. Changed from inner/outer to outer/inner for the walls.
I only have some shift left on 1 side.
Sounds like you’ve made some excellent progress. Thanks for sharing.
With respect to the seam, if you haven’t already done so, you may want to try using the seam painter and manually painting the seam’s location. The seam painter isn’t perfect—sometimes it just won’t obey, so you may need to experiment with the seam’s placement. For a shape like this, I usually try to position the seam on a corner where it won’t be noticeable.
