I am attempting to perform a manual flow rate calibration utilizing the 9 printed test coupons. The surface finish on all the coupons is poor, and there seems to be an issue with the way the top (and underlying) layers are printed. I’ve uploaded some closeup photos to illustrate what I am seeing. It looks like there is a ‘gap’ between every other extruded line. So there are 2 extruded lines close together, then a gap, then 2 more extruded lines close together, etc.
Has anyone else encountered anything like this?
Looks like the wrong line width. Like printer thinking it has a 0.6 nozzle, but instead it has a 0.4 installed. There is a setting on printer to specify nozzle size and type.
Might also be a partially clogged nozzle. Could you share an image of the test piece with 20% more flow and also what filament settings you use?
1 Like
Clogs do not produce that kind of even extrusion. If it were a clog, one would see unevenness and sputtering of the filament.
I believe @maximit theory is the most likely culprit or some variant of filament flow misconfiguration. This uniform print looks like the slicer is anticipating a wider flow nozzle than what is actually being deposited on the build plate.
@rabrand1955 great job on posting clear photos BTW. I wish more of the members here would take such care in putting up a quality post like this. It makes analyzing a situation far easier when we can actually see what you are seeing. 
It totally can produce an even flow, I had some metal debris stuck in a nozzle, resulting in a pretty steady reduction of flow.
probably not relevant to this case, but the first time i tried calibrating i went from a print directly to calibration, figuring that loading the calibration would reset and overwrite all relevant settings. well. it didn’t and all the coupons came out identical. point and case. hitting the new project button matters, as does setting your filaments and nozzle profile before you load the test.
if you use the ams i recommend removing all filaments but the one you test in the filament list cause i found with the pressure advance calibration it only changed the profile on the first filament
Actually,. double posting, cause i picked up a coupon that was still on my desk here and yes… it does kinda have the same pattern in the lines. two paired strokes and a slight gap, in this case what youre actually seeing is the second to last layer. as the top layer is perpendicular to the lines.
also… i’m already coming to HATE calibrating translucent filaments…
I love a good mystery and this one had me going. It’s looking like @maximit theory is the one that can be proven out by empirical evidence. There may be other elements at play but I ran these tests using the same -15 tile that was baked into Orca slicer. Each test was started as a “new” project to ensure that no settings carried over from the previous test. The sole variable within the slicer was nozzle diameter. NOTE: The printer only has a 0.4 nozzle so we essentially lied to the slicer.
Here are those results. 0.8, 0.6 and finally the 0.4. The big surprise is that the slicer doubled the size of the tile when it was told it had a 0.8 nozzle. That was unexpected.
As you can see, the 0.6 nozzle produced wide gaps between the runs similar to @rabrand1955 above whereas in the native 0.04, it did not.
Now there may be a way to emulate this through a combination of layer height and filament width but this experiment at least proves that this can be replicated by tricking the slicer believing it has a larger nozzle than what is actually present.
I see that your printer is actually exhibiting the same behavior as mine is. That being that the extruded lines are definitely not equally spaced. That is exactly what I am seeing, only worse. It makes the two extruded lines going in opposite directions almost appear as one line. Belt tension? Gantry skew? I have never used anything but a .4 nozzle. It certainly looks mechanical in nature to me.
The test I conducted is far from conclusive but it’s at least a lead. I tried to reproduce this using just slicer filament width settings. I could not get it to respond the way. I’ll do some further investigations later this evening but in the meantime, here is what you might look at to ensure that it isn’t a slicer issue. You want to create a single layer model and look for similar gaps. If the gaps don’t show, it eliminates mechanical issues from the suspect list.
Here’s a quick recipe for making a single layer print.
- Using a generic or Bambu default filament setting
- Create a cube primitive and change the x and y to 50mm and make the Z thickness somewhere in the 1x of your layer height. 0.5mm will generally work.
- Set your top and bottom layers to zero.
- Make the infill 100%
- Slice and print
If the slicer complains that there are no printable layers, then increase the Z height of the model another 0.25 increments until the slice function is enabled.
When sliced, you should only see a single layer of filament. A further experiment would be to set the wall loops to zero, then all you have is a continuous flow of filament 1 layer high. This takes about 3 minutes to print the model so it’s a quick test. If you desire, expanding the x and z to 100 or even 200mm will just exaggerate the test.
As an aside, this test also reveals build plate issues such as contaminants and leveling issues when you print a 250x250 layer.
If you do still suspect mechanical issues such as stepper motor or belt issues, note that the stepper motors each move the print head in a 45 degree angle. You can witness this by powering down the printer and manually moving the print head and watching the belt movement. When you move it along a 45 degree angle, only one belt moves.
So to isolate which belt, stepper motor or pully may be at issue, change the 1 layer sparse angle to favor another orientation that is neither 45 degrees nor 90 degree and see how the filament layer behaves.
Here’s an extreme example of moving the infill direction to 343 degrees. You can see what happens to the direction of the filament. The point is, you want to move the direction of the nozzle into an orientation that will dramatically change how the filament is laid down from the default 45 degrees. The nice thing about this approach is that you can clone the model and change the direction for multiple samples and do it all in one print.
That had me confused for a bit, until i realized this was the X1 part of the forum.
A1 series printers are all X-Y linear movements. and seeing as i kinda have the same motion pattern of two paired lines, and the resolution on my coupons are fine in themselves on my A1 I’m inclined to cross step loss off the list of suspects.
this pattern is likely a slicer thing or part of the test, as for the really wide spacing of OP’s… could it be that you have some weird shrinkage factor on your filament settings?
what dimensions are your coupon?
for added reference mine are 4x2 cm without the tag, almost on the dot.
My calipers are reporting the same 20mm x 40mm dimension ±0.2mm as you measured. I included a ruler as part of my original photo for scale.
I don’t believe it is a mechanical thing either but only mentioned it since you brought it up.
My settings were all default settings. I have no X-Y compensation in the profile. In fact, I used the default Bambu BASIC PLSA even though I have Overture in the printer. This was again all for the purpose to rule out any deviation of settings. If you are using Orca with the default Bambu Profiles and your hitting “New Project” before you initiate the Orca Calibration routine, you should have identical slicer profiles as I have.
I underscore “new” and “before” because it’s all too easy to forget to properly save or incorrectly contaminate the profile on the next calibration test when going from one calibration test to the next. I only recently discovered this trick on a YouTube video where the person stated that he too made this mistake. It finally explained why I was having to recalibrate so often. I wasn’t always saving the parameters properly.
You’ve provided me with some great troubleshooting tips that I may not have thought of. I’m fairly new to 3d printing, only about 4 months in, however, I have an extensive background (34 years) in CNC machine tool operation, setup, programming and maintenance. It seems the printed line pattern should match the line pattern shown by the slicer after slicing. If you look at the upper right corner of the -15 coupon I posted the photo of, you will notice that the first few passes are nearly evenly spaced, then get progressively worse. Due to the way the lines are laid down, it makes it impossible to do a Manual Flow Calibration because NONE of the coupons look acceptable! I’ll try printing a single layer at various angles as you suggested and will post the results. Thanks for taking an interest in my issue, it’s much appreciated! BTW, the photos were taken with my Samsung S9+ phone. 