Potential Solution For High Wall-Count issues

@SavvyPrints I think I’ve been having the same issue and have found a manner of fixing this; but I don’t know exactly which ‘fix’ was the part that worked.

I’ll give some pictures first: with a bit of explanation, and go through a detailed breakdown at the bottom.

Here’s some of my ‘Before’:

As you can see, inner walls start normal; and as my wall count goes from inner → outer, it all just goes to hell. This wouldn’t be an issue with 2 walls; but, as shown, it’s clearly not going to work with my 6 walls.


Mid-Process: I got this banding when doing final testing, due to over-tensioning the belts

After things were ‘Fixed’:




So, Here’s a breakdown of my printer/setup:

  • P1S w/ Vibration Dampening Feet (V2), Textured PEI sheet, Top class Vents (A 4-piece Print)
  • My Filament is a 5kg spool of eSun Black PLA+ (I had been using some from Amazon, and decided to use a larger roll; I’ve used 5kg spools for quite some time and have a special dual-bearing holder.)
  • I was using the Beta version of Bambu Studio and the latest Firmware

How did the problem start?

I originally was using .6mm nozzles, thinking that they would be faster than .4mm nozzles (larger size → more flow); But what I didn’t realize was that Max Volumetric Flow was a huge factor at these speeds. After investigating, I realized that: using eSun PLA+ with their profile settings (speeds predominately) would cut my print time down significantly; but I would need to use a .4mm nozzle (as that’s what they’ve tuned the profile for).

I made a custom .4mm profile for eSun PLA+ and had adjusted all of the appropriate values to that of which they indicate. Ran some test prints, and things seemed to go well (At this point in time, I was using the 1kg spools from Amazon.)

After using 6x 1kg Spools I decided to order the 5kg spool, and started using it, without changing any settings. Things seemed okay at first, until I increased my wall count to 6 walls.

As shown, the outer wall texture was terrible; and so, the journey began.

What did I do to Fix the issue
I’m not sure exactly which of these steps was the fix, so I’ll just list everything I had done, in sequence.

  1. Because I previously had both: run ASA and had tried to lubricate the pulleys (Bad choice); the belts seemed to be deteriorating and leaving some residue on the pulleys. I’m fairly certain this was mostly the attempt to lubricate the pulleys; So I took off the panels and cleaned all the belt pulleys. [This didn’t fix the issue.] I’m fairly certain that I didn’t tension the belts as intended. (I moved the extruder in circles, rather than moving the entire X gantry back and forth, as they do in the video. This might have no significance, but If you adhere to their instructions, moving the gantry will cause both X and Y motors to spin together: something I expect to be of importance.

  2. Next I considered a nozzle clog. The prints weren’t failing, as if heat-creep was the obvious issue; but it seemed worth clearing the nozzle. I skipped the “Unclog w/ pin” Method; and chose to use both the “Heated Hex” and “Cold Pull” methods. I feel that the most effective method was the heated hex-key. I turned off my printer and took the .4mm nozzle completely out. I then used a lighter to heat about 15mm of the long-arm end of the 1.5mm Hex key that came with the printer. After it was hot, I stuck it into the nozzle and held it in place for about 20s. The heat would melt any plastic, cool, and then solidify on the hex key. I was able to pull out portions of filament (2-3mm) each time (I did this process about 4 times: until the hex key was clearly hitting the metal at the bottom of the nozzle.)

    After completely removing any filament with the heated hex method, I did use their cold pull method. I reassembled the hotend, and took some of the green PLA that came with the printer. I cut the filament into 5 sections: of approximately 1ft each, and used this green filament as my cold-pull filament. Manually heating the nozzle to 220, I then extruded filament until it was clear that the green filament was being extruded. I then stopped the extrusion, set the temp to 100c and waited. After the nozzle was cooled to 100c, I reversed the extruder and pulled, moderately, on the filament. (You don’t want to pull to hard, as to rip the filament apart, but you do need to apply a reasonable tension so that the ‘Plug’: from the very end of the nozzle, is pulled into the extruder and pushed out.) [I did this process a total of 5 times. I don’t recall seeing any black filament within the green filament; but felt it was worth doing it multiple times, just to be sure.]

  3. Next I rolled back/uninstalled both the firmware and slicer software. I uninstalled Bambu Studio on my computer (Windows 11 - 64bit) and left that be for now. Then, using the Android version of Bambu Handy, I manually went into my printer setting and reverted the Firmware to an earlier version. The process to a few minutes.

    After several minutes, the P1S itself indicated that a firmware update was available. (I don’t recall if I had reinstalled Bambu studio by this time; but I would simply allow the update, via the P1S menu.) Before Continuing, I made sure that Bambu Studio was both installed and that I was logged into my account (since having an account will save your presets).

  4. Delete eSUN PLA+ custom filament profile, Slicing Profile, and Adjusting the P1S .4 Nozzle Printer Settings. I felt that: if the issue wasn’t within my printer, then my issues were almost certainly an issue with my slicing parameters. The Goal was to act as if both the printer and filament were completely new, and assume all values were incorrect.

    I first went and adjusted the Printer Profile: The only adjustment is a fairly important one, from what I can deduce. The MAX_LAYER_HEIGHT is going to default to something like .28mm; but, if eSUN’s PLA+ profile (Found on their website) is finely tuned, then this layer height might be way to high: given their print speeds. I set my value at .2mm, which is what they indicate (.1 to .2mm).

{ It would be important to know, if you don’t already: Using the Adaptive Layer Height feature does not limit the maximum layer height to whatever you have as your “Slice Profile”; It uses the PRINTER’S Max value. So, if you wanted to use adaptive layer heights and have something like .2mm set in the slice profile; the feature will still adapt the layers to range from .08mm for Quality and .28mm for speed. }

This “Issue” is shown below: my Printer has the .2mm MAX setting as above, but I set my slice layer_height to be .1mm; yet, when you use the adaptive layer height feature, it allows the max to go between the min ↔ max of the PRINTER.

  1. After doing the steps above, I redid my profiles and tested things: but realized my belt tension was wrong (From back when I cleaned the Pulleys; so I had to go back and redo that process, to fix the belts being too tight and causing ringing.)

    Regardless, I then proceeded to Calibrate the Filament: K-Value and then Flow Rate. [.022 and 1.029 seemed best, for my printer/filament]

    I then ran the Max_Volumetric_Flow calibration print (It can be found online). You have to change your filament settings to allow this to print as intended: MAX FLOW → 50, Disable the “slowing printing down for better layer cooling”. I use an Outer Brim 15mm wide.

The test should look like this:

You will have to manually stop the print, this is intended to start slowly and increase the speed until the max volumetric speed/flow rate is hit… at which point you will be able to see the print failing. I watch the slicer to see what layer it’s currently on, so that way I can go back and check the flow-rate for that layer height. You can measure it manually and use the formula given with the part file; but I feel like it’s just easier to determine the value via the slicer itself. (You still want to error down, so if it fails at 20, then you’d set the value for the filament a bit below that.)

This test was interesting for me, because the Amazon 1kg eSun PLA+ rolls were able to get to ~24mm^3/s, but the 5kg spool I have now was only able to safely get to about 18mm^3/s… Which likely was a key factor into my specific issues. I had been printing the 1kg spools and kept the same default 22mm^3/s value that eSun provides on their website; but the filament I have now is quite possibly a fairly old batch (5kg spools aren’t available on eSun’s site or Amazon; so, assuming it’s genuine, it’s likely bought in a large batch of 5kg spools.) It’s not a huge issue to me, because print time is dependent on many factors, beyond just linear speeds and Flow-Rate (Although they have a significant impact.)

Nevertheless, My issues may have been caused by the old, default flow-rate of 22 being too high; but the speeds and flow throughout the rest of the print were similar, so I’m not sure that was the only issue at hand.

Here’s a set of sliced images, using 22mm^3/s, of the print that was causing issues:


The print issues were showing prior to even getting to the outermost wall, so I don’t believe that the flow-rate alone was the fix; but I’m not fully certain.

  1. Finish applying the default eSun PLA+ parameters that they provide. One thing that I did adjust: which might have merit, is the Inner Wall Thickness. I dropped the value from .45 → .42 to match the others. [As that was the case in the previously shown flow-rate test. I’m unsure if this really had any affect on things; but it doesn’t seem to be detrimental.]

Here are images of my current settings, which haven’t be causing issues:

BLUE VALUES ARE SET SPECIFIC TO MY USE-CASE, YOU SHOULD IGNORE THEM OR CALIBRATE THEM IF YOU DESIRE






Hopefully this information is useful to help you guys resolve your issues. I realize that my setup might differ slightly, but it was clear that these problems only became apparent when I pushed up the wall count.

If you have additional questions: regarding what I had done or why I did certain things, just ask; I’ll try to explain my logic. Although, I am up and printing without issue, I can’t ascertain exactly which step, or set of steps, resolved my issues. All I know is, I’ve been able to print for the better part of a week without issues.

2 Likes

What a journey-

Aint 3d printing FUN?

Your thorough investigation and description is nice to see, AND it shows that no matter how good YOU are, there are just so many factors that make up the 3d printing salad

Yeah, There’s definitely a learning curve when it comes to print at speed…

My previous (and first) printer was a QIDI X-Plus v1: I couldn’t get anything to print faster than 60mm/s with decent quality. They really missed the boat on implementing many of the features that could have made that printer a much more respectable machine. (No Pressure Advance or Auto Bed Leveling.)

Now that I’ve got a P1S, It’s like I’ve gone from a rusty Honda Civic to a nice Porsche. [Yet I still drive like I have a crappy car, and can’t yet appreciate the full merit of what I currently have.]

It’s sure quite a unique change to be more concerned about the capabilities of filaments: rather than of the printer itself.

If only 3D printing was as simple as the naive people: who never have 3D printed anything, believe it to be…

But, Boy… It’s sure fun to watch this P1S extruder whip around frenetically.

1 Like

I’ll also add, I find changing from the default inner-outer to outer-inner does improve the outer wall quality.

Yeah,

That certainly is true, and would be a work around for this specific part: as it had a Fillet curving up towards the top of the main body.

Unfortunately, AFAIK, printing from Outer → Inner walls can reduce the efficacy of producing nice, shallow overhangs.

I would presume that: if tested, 45 degree overhangs would see minimal impact, but trying to print something closer to 70 degree overhangs: using outer → inner, would produce a less than favorable surface. (maybe excluding some, specific filaments)

But, given the context of this print, I would agree: using the outer-to-inner wall sequence would resolve the surface defects; but might lead to a propagating issue: inward, that would affect the top layers. I just know that the original person that posted, here: ( Higher wall count = z-banding - #19 by SavvyPrints ) was also struggling to print similarly ‘6-walled’ parts: noting that others didn’t seem to run into this kind of issue.

Hey! My reply is over half a year late lol but I had to say “thank you!”. I was dealing with life stuff that prevented me from continuing my 3D printing adventure, but I’m diving back in (I just ordered filament, I had to toss my previous supply). I look forward to thoroughly reviewing your post and seeing what I can apply to my printing, in hopes of a fix.

Thank you so much! :pray:

1 Like

Yea, No worries about the timing: Life happens.

I would certainly look through the info I provided; unfortunately, I didn’t test after each change (so I can’t tell exactly which of the numerous things had fixed the issue).

The newer firmware is great, it takes longer for calibration (26m) but allows for testing of the stepper motors: reducing the noise significantly.

Moreover, When you then add in the use of the new “Cross Hatch” infill, the print speeds are faster and far more quiet.

If I had to guess (briefly looking back on this issue):

Try to do the nozzle unclog procedure first (Especially the heated Hex Key section… that got deep into the tip of the nozzle, when I did it).

Next, I would take the little bit of time to re-tension the belts [following the wiki guide on doing such] (*Make sure to only move the gantry forward and backwards: as grabbing the extruder can cause the extruder to move along the gantry and won’t help balance out the X & Y axis evenly.

Lastly, A flow-rate test (as shown) is probably the final key to making sure that your printed walls don’t become wavy. [I’ve found that most of the eSun 5kg rolls I use can only manage around 16-18mm^3/sec: which is significantly lower than the ~22 value that is default].

For me, using eSun and their provided settings, I tend to stay between .1-.2mm for layer heights. I’ve had a similar ripple effect (albeit to a lesser degree) when printing at the .08mm layers.

As a generalization, this issue seems to stem from a speed vs adhesion issue: where pushing the print slightly too fast will create a feedback loop of worsening outer layers.

If you want/need, I have a custom 3mf file that runs the Flow-Rate test for a .4mm nozzle: but it’s adjusted so that the range is limited to a max of ~24mm^3/sec and have found it easiest to watch the print until you see issues begin, then glance at the current layer number. [ You can simply look back at the sliced file and see what flow-rate was running at layer ~75 (or wherever it starts to fail), and adjust down a little bit. ].

Hopefully following these steps and, if needed, the other instructions in the post, will help resolve your troubles.