Issue with under extrusion

Hollzarr · December 9, 2023, 8:23pm

Hello,

This is my second post, unsure if this is a continued fault or a new fault.

Please see my first detailed post on my first issue; X1 Carbon Potential Flow Rate Issue

Continued;

I purchased a brand new 0.4 nozzle, and printed 5 prints all flawless, one after the other.
Then I started a 15-hour print and it failed at 4 hours. With strong signs of a nozzle clog.

I underwent the things you do when you have a clog, but I’m still having issues, but part of me is saying it’s not a clog anymore.

A little more info, I am starting from scratch, tuning all my filaments from nothing on top of this fault.

Here are a few images of me doing a pressure advance test on a dry Bambu Labs PLA Basic.
As you can see as soon as it speeds up on the second layer, little to no filament comes out.

Secondly, I’m unable to tell which option to pick as all corners are rounded which I think is part of the main problem I am having.

Hollzarr · December 9, 2023, 9:07pm

I just did a second pressure advance test and little to no filament was extruded, so I just did 2 cold pulls.

These were the results.
First

Second

EDIT!

Update:

Since I ran the cold pulls, this was the result of the pressure advance test.

I also did a Flow Rate calibration getting the result of 0.96

Now testing on a flow rate cube, using these settings and awaiting the results.

Layer height: 0.2mm
Wall Loops: 2
Top shell layers: 4
Bottom shell layers: 3
Infill: 15%
Gyroid.

Olias · December 9, 2023, 9:38pm

You’re taking all the right steps, just stick to it and you’ll get there. In particular, using your cell phone to zoom in on the image is a trick that most folks would have overlooked.

As far as your PA Pattern goes. None of those values are what I would allow if it were my filament profile. You’ll likely not find an exact perfect match but rather a goldilocks zone of 2 or 3 values that are close to each other. I noticed you already dialed in 3 digits but if I may suggest backing up a bit on the parameters first.

By that I mean, if you’re not getting the results you want between 0 and 3, try 3 - 13

You’ll recognize this as the default PA Pattern test.

Here’s a simplified version of the text:

“To improve results, I expanded the range by doubling the step size, which sped up the process. I also increased the Start PA and End PA ranges to make differences between values more pronounced. Once I had these larger ranges, it was easier to identify the ‘Goldilocks zone.’ I then repeated the process, focusing on the best 3 bars in sample 1, using the original PA Step of 0.005.”

Here are two examples of a crisp looking PA pattern. Albeit it’s often a judgement call, I felt that in this sample the third bar from the bottom for the black filament was best and I marked it with a paint pen. On the white one, it was the fourth one up from the bottom but one could have made the argument that I should have picked the second one. It’s a matter of opinion but the edges of number four looked better to me so I compromised.

Now having said that. I see something else going on with your filament that may or may not be related to the pressure advance.

If I read this correctly, I’m also seeing some stringing and some layer adhesion issues going on. Did you try running a temp tower? It almost looks like you’re at too high a temperature AND also not getting enough layer adhesion which is weird. This would be the point where I would reset my profile to a default bambu profile and try again.

BTW: Have you tried the PA Tower? Whereas I kind of moved away from that version once I learned how to identify the right values in the PA Pattern, the PA Tower is a lot easier to spot where your Pressure Advance Limits are. It’s just that it takes sooooo long and uses a lot of filament as you probably experienced. But… it is a more reliable method that is easier in my view.

An example of the Orca PA Tower tutorial which shows what I mean, the area of demarcation is really pronounced. And, if you dial in a narrower range(exact opposite of what I described above) you can super dial in the setting. But… that’s 42 minutes to print versus the PA Pattern at 11 minutes. Which is why I stopped using the tower.

Click here for recommendation for must-have tools for calibration

Must-have tools for filament inspection.

A jeweler’s loop is a must-have but I found these two on Amazon that really allow for some deep,deep inspection of layers. Note, you have to hold them right up to your eye and make contact with the piece under view but the results are amazing.

Be careful with this one buy only the 10X version. There’s a jackass on Amazon advertising a 20X version of this for 3X the cost. I purchased both so I would have two magnifications. They were identical, the 20x version is a scam.
Amazon.com

This is a 40X version and is legit. When I returned the scammer’s 20X, I purchased this in place. I paid $7 and it’s now $6. Note, it’s harder to use because it is almost like using a microscope but once you use
Amazon.com

I found both of these indispensable when trying to diagnose filament layer issues.

Hollzarr · December 9, 2023, 9:45pm

This is fantastic, thank you so much.

I’m gonna keep this on my screen and keep running a few tests, and like you said from using a bambu profile.

Again thank you, this is incredibly helpful info.

Hollzarr · December 9, 2023, 10:06pm

3rd update.

Cold Pull
Pressure Advance = 0.024
Flow Rate = 0.96

Did a flow rate block from bambu and this was the results, the top layer is confusing me a little on where I should go.

Bottom/First layer

Top Layer

Olias · December 9, 2023, 10:34pm

The first photo is about as good as one would expect. You’re not going to get a glass-like “molded” appearance without resorting to turning on ironing which is in itself another calibration effort unto itself.

However, the last one I see as having underlying infill issues. One way to prove or disprove this is to increase the top layer shells and run it again.

Although I can’t see you model, what your looking for is how solid the layer underneath is. Increasing the top layer will thicken up the plastic and make it less subject to what’s underneath.

Alternatively, you could increase the infill density and or pattern from the default grid to Gyroid or Adaptive Cubic. Both tend to make a more supporting underlayer for top surfaces to scaffold off of but more importantly, they are so vastly different from the default Monotonic Linear pattern that you won’t get that Moiré interference affect that occurs when you put two lines on top of one another which is what might be happening here.

Speaking of Moiré, you could also use different top surface pattern just to see if it brings out a different result.

If I’m going for a smooth top finish I generally revert to Octagram Spiral. It produces a very nice geometrically centric pattern. Looks great on symmetrical shapes, no so much on non-symmetrical geometries.

And then there is the aforementioned Ironing. It produces some very nice smoothness but at a really high cost of doubling print time. But… if you’re going to bed, what’s the difference if it takes the printer 3 hours vs 6? Usually, that’s the only time I’ll use ironing but when it works, it works really well. But you will have to experiment and not just give up after the first try.

I recommend starting with Top most surfaces only. This will save the ironing to the very last part of the print and save time. I would also use Concentric pattern for symmetrical shapes. It produces results that rival the results get when using a smooth plate for the bottom of the model. Expand this picture up to see what I mean just by how it looks in the slicer.