Skip to content

5. 3D scanning and printing

alt text

This week’s group assignment was to test the design rules for our 3D printer(s).

Design Rules outlines the important considerations when designing for 3D printing, in order to get a good finish and reduce overall production time. The Design Rules published by ProtoLabs provides a good reference point for major printing processes.
alt text

Overview

Which 3D Printers we tested

We decided to characterise the design rules for these 3D printers;

Specs Anycubic Kobra2 Bambu Lab P1S Creality K1C
Slicer Anycubic Slicer Bambu Studio Creality Slicer
Build volume, mm 250 x 220 x 220 256 x 256 x 256 220 x 220 x 250
Layer height, mm 0.1-0.35 20%-70% of nozzle diameter 0.1-0.35
Positioning Accuracy (Precision), mm X/Y/Z 0.0125/0.0125 / 0.0025 around 0.1 100±0.1mm
Filament diameter, mm 1.75 1.75 1.75mm
Nozzle diameter, mm 0.4 0.4 0.4
Materials Supported PLA/ABS/PETG/TPU PLA, PETG, TPU, ABS, ASA, PVA and PET, and even PA PC, etc. ABS, PLA, PETG, PET, TPU, PA, ABS, ASA, PC, PLA-CF, PA-CF, PET-CF
Print speed, mm/s 200m max 500 300mm/s Max: ~600
Print bed temperature 110°C 55°C for most PLA ≤100℃
Extruder temperature 260°C 300°C ≤300℃
Input Formats STL, OBJ, AMF .3mf, .stl STL, OBJ, 3MF

Test model

After considering different tests, we decided to test with the All in one 3D Printer Test 2 and Clearance and tolerance test.

  1. All In One 3D Printer test (Made on November 20, 2017)
  2. Smaller, more effective version of above(Made on June 24, 2018)
  3. All in One 3D Printer test 2(Made on February 14, 2024)
  4. All-in-One Object from Kickstarter-Autodesk-3d(2018) - according to Nagano-san, this one takes 9 hours to print
  5. This one is not new but has 21 tests and prints in 1+ hour?
    https://fabacademy.org/2021/labs/waag/students/nadieh-bremer/blog/week-5/
  6. Comprehensive one made in 2025

Printing settings

We used these parameters when printing;

Parameter Description Value
Infil Density Expressed as a percentage, ranging from 0–100%. Higher infill percentage results in a denser, stronger print, while a lower percentage makes the object lighter but potentially less durable Value recommended by instructors; 15%
Support Use if over 45˚ None
Skirt Good for purging nozzle before printing model 2 rounds
Brim Good for increasing bed adhesion None
Raft Good for clean first layer of model None
Layer Height Smaller layer height produces a smoother surface finish but takes longer to print; a larger layer height prints faster but may have visible layer lines depending on application. 0.2mm
Print speed Depends on printer. To get more details, consider reducing speed. 230mm/s for Inner wall, 138 mm/s

For Printing Temperature, we kept the printer’s default value, which was around 220˚C (±10˚C for different printers, but we forgot to note down the exact values. PLA filament is usually between 205-225, adjusted depending on filament, printer, temperature on the day, etc. It is typically recommended to raise the temperature for faster speed printing.)

Summary of Results

These are the final prints.
alt text

And these are the test results, alongside ProtoLabs’ design rule recommendations for FDM printers.

Design Rule Description ProtoLabs’ recommendation Anycubic Kobra2 Bambu Lab P1S Creality K1C
Time to print Depends on object’s volume, complexity, and printer’s capabilities - 2h41m 2h41m 1h51m
Support & Overhangs The maximum angle a wall can be printed at without requiring support 45° 50° 60-70° 70°
Horizontal Bridges The span a technology can print without the need for support. 10mm 10mm 20mm 20mm
Diameter How accurately the printer can print cylinder diameters not provided (min:3mm) Almost perfect accuracy Account for -0.2mm error Account for -0.1mm error
Holes (Negative feature resolution) How accurately the printer can print holes not provided (min:Ø2 mm) Rectangles are acurate. For circles, account for -0.3~-0.5mm error Rectangles are acurate. For circles, account for -0.2~-0.4mm error Rectangles are acurate. For circles, account for -0.3~-0.5mm error
XYZ Tolerance Dimensional accuracy of a specific technology. ±0.3% (lower limit ±0.3 mm) All accurate Mostly accurate, except for y:+0.2mm for 10mm spire Mostly accurate, except for X:-0.1mm for 30mm spire
Connecting/Moving parts Recommended clearance between two moving / connecting parts (min space b/w each object), so they can move after being printed. 0.5 mm Could not test due to failed printing

1. Overhangs

alt text Ability of printer to accurately produce overhangs without supports.

What is the maximum angle the printer can print in good quality?
alt text

2. Horizontal Bridges

alt text Ability of printer to create horizontal bridges without support structures.

What is the maximum length the printer can print without drooping or sagging?
alt text

3. Dimensional Accuracy (Diameter/Clearance)

<br> Difference in diameter of printed filament versus acutal design.

What value to consider offsetting when designing?
alt text

4. Negative feature resolution

alt text
Ability of the printer to make holes, and their tolerance (how easily the holes can be easily released with your fingers) alt text

5. XYZ Resonance

How well the printer makes prints in the XY direction.
alt text

6. Clearance

Gap or space between two parts - awareness of this value ensures smooth fit. Especially critical for tight internal geometry and print-in-place objects.
At least how much apart should 2 components should be so that they can move after being printed?
alt text

7. Fine Flow Control

<spires>
How well the printer can make small spires. Measure the height of the tower that can be made, and whether there are any strings between towers.
alt text

Other

Bambu could not print engraved texts on flat surface (XY plane).
AnyCubic can print, but very roughly and not very legibly.
Creality had the best quality by far.

Files and Code