Unable to get PETG calibrated

I’m having the hardest time dialing in my PETG and PETG HF filaments.

I tried calibrating Pressure Advance for Bambu PETG HF Black and ended up with this mess:

Any idea at all what’s going on here? I have cleaned the plate multiple times, tried with and without glue stick, but no matter what I do it ends up like this. Tried different prints aside from this PA test as well, they all end up looking like this, all tore up.

For this filament I am using the default Bambu PETG HF profile (as it is automatically detected in the AMS).

Possibly related (in that it’s also PETG) I have dried a roll of PolyLite PETG Black for 24h at 70, but the prints from that filament do what appears to be the same thing: lots of stringing, shredding, etc.

I have seen several other threads on the forums regarding difficulties with PETG, but I didn’t want to hijack those threads, and wasn’t sure if they are the same causes.

Thanks for your help!

Update with additional info:

I have my filament in an air-tight cereal container with a hygrometer and desiccant, the hygrometer reads 20%. The dryer I am using is the Creality Space Pi Plus. Also, I live in Arizona where ambient humidity is relatively low. Sorry, I should have provided this info in my post.

1 Like

Try slow down to maximum of 100mm/s, turn off all fans

I’m not sure what the original objective was but the image suggests to things:

  1. You’re getting stiction problems with the filament on the build plate and you’ve tried to remedy it via glue as is evidenced by the glue surrounding the model.

  2. Your trying to calibrate PA advance(because that’s the PA advance test from Orca)

Stiction issues

If I were to guess, you have either the wrong plate temperature and/or the wrong filament temperature. Here are three things to try with PETG

  • Raise the plate temp first.
    – You should not ever need glue for a PA test. You have to solve this problem first. If the filament box says plate temps of 55-65C, then go 75 and above, be bold and don’t be concerned about even maxing out the plate temp, it’s only a test.
  • Raise the filament temp.
    – Same as with the plate temp. If the box says 55-65C, add 10-15C to test the upper limit.
    – Try running a temp tower to validate the filament is flowing correctly at the nozzle temps you’re using. Make sure you’re using Orca and are always doing CTRL-N for a new session between calibrations. Ignore the presets, they are not helpful as they just mislead newbies into thinking it’s their only choice. Instead, dial in the temp ranges manually. The settings only impact temps.
    image image
    – Verify that the tower is constructed properly using the temp drop-down. You will see the color gradient appear.

Pressure Advance

You actually nailed the PA test though whether you realize it or not. An test is actually a “vote” between best features. Few filaments get all of it right. CF filaments tend to perform best but they are expensive.

  • In your example. The best cornering was produced at the #1 spot in the photo. That would mean that your PA advance according to the photo if you were just looking for cornering, was 0.01
  • However, when we look at the tail end in example #2. That gives the best result there which would suggest a PA of 0.015 since it’s between 0.01 and 0.02.
  • But… sample #3 has three results that are a tie but since the one at the top of the photo agrees with sample #2, that’s the one that wins.

So in this test, your PA values that you want to try out are between 0.01 and 0.02 with a vote leaning closer to 0.015.

Now you could run this test again at finer resolution between those values if you wanted to verify the results but that takes time and in my experience, it does not produce much better results.

Tips for using the PA Advance pattern test.

Sometimes it’s damned near impossible to determine which pattern has the sharpest and cleanest test result. This is particularly true of filament colors that have sparkles, as in my next example or don’t show enough contrast for the background plate(Gold vs Black vs Smooth vs texture).

  • Tip #1 - Use your cell phone to do a closeup shot.
    – Pictures can be zoomed in and are far better than your eyes.
  • Tip #2 - Use an acrylic paint pen to color the area of the nubs that you feel produce the best results and then take another photo to see under magnification.

  • Tip #2 - Print a cube primitive after you’ve changed the PA advance and look at the cornering. You could use a test cube too but that wastes a lot of filament.
    – I use a cube primitive set to 25.6x10x15mm but you can make it taller if you want. Taller can make for more shadow affect and make it more obvious when cornering is not happening correctly.

    – Print multiple samples at and at different PA values making sure that you do a CTRL-N between prints to ensure that you’re always working from the default profile you’ve calibrated.

WRITE DIRECTLY ON THE MODEL, the PA value!!!


– The part of writing down values on the model cannot be understated. You will lose track of where your calibration was. Hitting CTRL-N between test prints ensures you do not mistakenly use the previous values. This is why I keep a supply of both black and white acrylic paint pens near my printer.

Don’t be afraid to test out some of your own models. Here is an example of a model that I drafted in CAD to test out a theory.


7 Likes

Thank you @Olias for your insights! I will give them a try today. I was feeling quite frustrated yesterday, your guide gives me hope :slight_smile:

2 Likes

So I took another stab at it today and everything worked like a charm. I double-checked to make sure moisture was 10% or less, reran the PA calibration test, and everything worked out great.

During this process I realized that I might have gotten in my own way by not selecting the correct filament during the prep step in the slicer, which defaulted to PLA in the AMS, then might have force selected PETG in the final confirmation window to start execution. I’m not certain, but that would explain why all the fans were running full blast when things were messed up, instead of only the chamber fan running now when everything worked. Thank you for bearing with me on this, and I am still following your tutorial @Olias, rest assured that your energy there was not wasted.

4 Likes

This topic was automatically closed 2 days after the last reply. New replies are no longer allowed.