search

dirtdigger / fleur_de_cali

Tool for calibrating dimensionality and skew of a 3D printer
GNU General Public License v3.0
61 stars 3 forks source link

Printing anomaly #11

Open boromyr opened 1 month ago

boromyr commented 1 month ago

Hi, I printed the 150x5 model, but there is an issue with the measurements: for the diagonals, D1 is greater than d1, while the difference is reversed for D2, D3, etc., and for the internal diagonals d2 is greater than D3 and so on; for the dimensions X and Y the same thing happens, X1 is smaller than Y1, but X2 and subsequent ones are greater than Y2, Y3, etc. Based on other prints, the correct values are the external ones, D1 d1 X1 Y1, while the internal ones are inconsistent with past prints. This behavior shouldn't be normal, I tried printing the file twice with the same result. What could be causing this anomaly?

dirtdigger commented 1 month ago

Hi boromyr,

It's hard to say for sure. It could be something mechanical, perhaps measuring the incorrect locations, print material interior shrinkage, ...

Couple of suggestions to try:

  1. Can you try with a different model, e.g. 150x3, to see if the problem persists?
  2. It could be that because the X2, X3, X4, X5 are interior (in the sense of they are not on the outermost perimeter of the print), this is related to interior horizontal expansion. Here's a post that may help on Prusa's forums link - they suggest printing perimeters first. Or, perhaps look into horizontal hole expansion in Cura.
  3. Can you confirm the measurement locations? Here's an updated map of the measurement locations that is more precise as to where you should be measuring compared to what's currently in the spreadsheet. meas_pts_5
  4. Can you provide detail as to the kind of printer? Creality printers (and perhaps others) have a known non-linearity which could account for this.
  5. What material are you using?
boromyr commented 1 month ago

I am printing in PLA with OrcaSlicer, Ender 3V3 Plus, which is a core xz, is one of the latest while the dimensional problem was about the first ender 3. I disabled the hole expansion in the second print and any type of dimensional modification, printing the outer perimeters first. I will try with another model, maybe a smaller one to see if there are any changes. calistar_150x5.3mf.txt

Here are the results, the last box is the difference with the theoretical one, as you can see all the internal measurements are very consistent with each other, but not with the external ones. If it were a problem of mechanical or systematic error, it should be constant in all directions. Could the shrinkage of the internal filling affect it, since it is not very isotropic in terms of shrinkage direction? <html xmlns:v="urn:schemas-microsoft-com:vml" xmlns:o="urn:schemas-microsoft-com:office:office" xmlns:x="urn:schemas-microsoft-com:office:excel" xmlns="http://www.w3.org/TR/REC-html40">

Diagonal (capital D) | 1 | 150,00 | 149,80 | 149,77 |   | 149,79 | -0,21 -- | -- | -- | -- | -- | -- | -- | --   | 2 | 120,00 | 119,73 | 119,7 |   | 119,72 | -0,28   | 3 | 90,00 | 89,72 | 89,73 |   | 89,73 | -0,28   | 4 | 60,00 | 59,80 | 59,78 |   | 59,79 | -0,21 diagonal (lowercase d) | 1 | 150,00 | 149,74 | 149,73 |   | 149,74 | -0,26   | 2 | 120,00 | 119,96 | 119,98 |   | 119,97 | -0,03   | 3 | 90,00 | 89,92 | 89,97 |   | 89,95 | -0,06   | 4 | 60,00 | 59,97 |   |   | 59,97 | -0,03

EDIT: with the 120x2 model rotated 180° same issue, I disabled PA and reduced the speed. I just have to try with PrusaSlicer instead of orca.

dirtdigger commented 1 month ago

Thanks for the update, bormyr. Let me know how the 120x2 works out with Prusaslicer.

I'm not super familiar with the Ender 3v3. I brought up mechanical problems not poking at your printer in particular. More, it was me wondering if there is something specific about the printer style that causes warping of sorts near the boundaries of the bed.

boromyr commented 1 month ago

Unfortunately, no changes even with prusa, but I did another test: image In the lower model, the internal and external diagonals are perfectly consistent with each other and the other prints, while in the upper model, only the diagonal marked in green is inconsistent, and I noticed that the thickness of those walls is greater than all the other walls, probably this is the problem that leads to an incorrect diagonal. It seems to occur only in the presence of holes.

boromyr commented 4 weeks ago

I also noticed that in the web version it is possible to iterate the calculation of the inclination, but not in the Excel sheet, I wanted to ask you how I can iterate the calculation in Excel, since I have not found any way to iterate the inclination calculation, I would be grateful to you.