Square Mesh Sieves V2 - FINAL

FINAL UPDATE Version V2 - May 19, 2025

I have added a new print profile encompassing all (9) sieves ranging from 0.4 mm to 5.5 mm. Criteria for determining infill percentage are summarized in the following figure:

Furthermore, this new print profile incorporates a plate for printing a simplified sieve locking system, replacing screws with rubber bands. This enabled me to stack 9 sieves plus a base and cap (which would have required screws exceeding 20 cm!). While rubber bands are less robust than screws, they proved sufficiently effective (I used 4 rubber bands with a 40 mm diameter, 3 mm width, and 2 mm thickness).

FINAL CONCLUSIONS

In conclusion, the method of using modifiers to rapidly create a square-mesh “GRID” with reasonably precise mesh apertures appears reliable and reproducible. Achievable aperture sizes range from 0.4 mm (71% infill) to approximately 6 mm (11% infill); infill values (D) below 11% are poorly predictive of the actual printed mesh size, while values >71% result in distorted and/or occluded square meshes.

************************

Updated Version (V2) of my previous V1 model 

Based on the mathematical model I applied in version V1 (LM = a + b/D where “sparse infill Density” = D and Mesh Size = LM) I attempted printing new sieves of 0.4, 0.6, 2.75, and 4.75 mm using different PLA types (Bambu Lab Basic Grey and Silver, PLA Matte Caramel and Lilac Purple). All sieves were printed with a 0.4 mm nozzle.

Results: The 0.4 and 0.6 mm sieves visually appear superior to the 0.3 and 0.5 mm sieves from V1, exhibiting fewer mesh occlusions/distortions, particularly the 0.4 mm sieve, as shown in Photo 1; the dimensions of the new sieves are illustrated in the other photos.

The superior quality of the 0.4 mm sieve is undoubtedly attributable to the slightly larger mesh size, the mesh thickness (each individual mesh filament in V1 was set to 0.3 mm, while in V2 it was 0.4 mm), and perhaps also to the Bambu Lab Basic filament (my impression is that it presents fewer issues).

The 2.75 mm sieve (D = 18%) exhibited a measured mesh width of approximately 3 mm using a vernier caliper :-|, while the 4.75 mm sieve (D = 11%) yielded a measured mesh width of 6.2 mm! :-(

The problem likely stems from the fact that as the infill percentage decreases, the mathematical model adopted in V1 fails to provide reliable results, even due to limitations within Bambu Studio software (for example, to obtain LM = 4.75, the theoretical D value would be 11.1%, but BS only uses integers (11%), resulting in a 10% error immediately). Lacking insight into the criterion used for applying the “sparse infill Density”, I printed a sieve with D = 12% and obtained an LM = 5.5 mm, confirming that with low D values (high LM values) the V1 mathematical model yields unsatisfactory results.

At this point, I modified the mathematical model by performing a regression on the V1 dataset, to which I added the two real results (measured with the caliper):

D = 11% → LM = 6.2 mm

D = 12% → LM = 5.5 mm

I am aware that this is a questionable approach, but assuming that the LM values, both graphically calculated and measured with the caliper, are reliable estimates of the resulting LM values of the printed sieves, the statistical model (which may have physical significance) proved to be a Square root-Y reciprocal-X model: Y = (a + b/X)²

and the values estimated by the model are:

Comparing these “Predicted” data with those from V1, the LM (Y) values for D = 30% and D = 40% are lower than those calculated in V1, but in my opinion, were closer to reality (at least for an approximate measurement with the caliper).

Therefore, the updated V2 printing profile includes seven sieves (0.4, 0.6, 0.75, 1, 1.5, 2.75, 5.5 mm), retaining the D values from V1 for those from 0.4 mm to 1.5 mm; for 2.75 mm, the value predicted in model V1 (D = 18%), even if not perfect; and for 5.5 mm, a D value of 12%.

Conclusions: 

A) For sieves from 0.4 to 1.5 mm, adopt the mathematical model LM = a + b/D

B) For sieves from 3 mm to 6-7 mm, adopt the mathematical model LM = (a + b/D)²

C) For sieves with 1.5 < LM < 3 mm, adopt an intermediate value between the two models (… but I still need to verify this!)