6
$\begingroup$

If I'm working with standard PLA, and I want to print a box that I can stand on without any risk of it breaking, is there any good way to calculate the appropriate print settings?

I know that structural strength comes from the infill. Knowing this, and knowing the dimensions of the box, the weight of my body, the surface area of my shoes, and the material I'm working with, is there any good way to determine the minimum infill percentage I'd want to use in order to safely bear my weight?

$\endgroup$
3
  • $\begingroup$ different print settings, different PLA, and even different environmental conditions can all affect strength, making calculation VERY complicated on an exacting basis. $\endgroup$ – dandavis Jun 14 '18 at 18:49
  • $\begingroup$ this is a Structural engineering problem $\endgroup$ – esoterik Jun 18 '18 at 20:54
  • $\begingroup$ Related 3dprinting.stackexchange.com/questions/109 (but not quite a duplicate) $\endgroup$ – Sean Houlihane Sep 5 '18 at 11:12
2
$\begingroup$

Strictly speaking, it is difficult to do calculations on these materials, but not impossible (I've heard about a few commercial analysis tools that do that). The FDM process (Fused Deposition Modeling) creates a product based of fused slices of material causing an anisotropic material (this means that the properties of the material are different in different dimensions). Basically, your product will be quite strong and similar in the X and Y directions, but fragile in the Z direction (layering direction). You can imagine that every layer may be a seed for cracks to grow when you're pulling at the part.

When applying a compression load on a product like in your example, the walls need to be strong enough to hold the pressure (not all of the load as, based on the type of infill, the infill also can/should take part of the load!) and need to be of sufficiently high percentage, not only to take part of the load, but also support the walls to prevent buckling. I remember that stress calculations for buckling are difficult and require FEA (Finite Element Analysis) for more complex objects other than bars or beams.

I think it is difficult to determine or calculate the infill percentage based on the compression load beforehand as you do not know the exact material properties and the buckling behavior. You do know that a 100% infill will give you enough strength and support, you could try to print at a lower infill, e.g. 75%, and test if that works for you.

$\endgroup$
2
  • $\begingroup$ good answer, shall you include a dynamic aspect of load? Let say a man with 90kg steps on the stool with left foot, then raises other - will generate more pressure as the body will be moving to accommodate right foot movement and balance to stay on the stool.... $\endgroup$ – profesor79 Jun 13 '18 at 11:15
  • $\begingroup$ @profesor79 No I won't as that is not necessary. Actually you are describing a static load that is gradually applied. Also the safety factors in place on mechanically engineered parts is generally very high opposed to safety factors used on aeronautically engineered parts. The latter are usually subjected to complicated calculations to take fatigue into account. The strange thing is that fatigue life calculations originated from the railways (axles) but are nowadays not a subject for them (as of increased safety factors), but it plays a major role in the aircraft (engine) design. $\endgroup$ – 0scar Jun 13 '18 at 20:40
1
$\begingroup$

I would suggest doing some calibration runs - granted this'll use up a lot of time and filament. But an infill of even 30 to 40%, plus a reasonably thick set of walls and top/bottom layers, should have almost the same strength as a 100% infill. Look at the girders & beams on road bridges, for example. As Oscar wrote, modelling with FEA tools is unreliable, more so because every extrusion printer is a little different.
Try printing a test box, say only 10 by 10 cm, same height, and see if you can stand on that, before printing the full-size item.

$\endgroup$
6
  • $\begingroup$ you mean test box not text am I right? $\endgroup$ – profesor79 Jun 13 '18 at 16:04
  • $\begingroup$ @profesor79 ooops.... tho' it would be pretty funny as a text box. $\endgroup$ – Carl Witthoft Jun 13 '18 at 16:47
  • $\begingroup$ I just started my imagination here ... as a software developer :) $\endgroup$ – profesor79 Jun 13 '18 at 16:56
  • $\begingroup$ "Try printing a test box, say only 10 by 10 cm, same height, and see if you can stand on that, before printing the full-size item." But how valuable would that be, when it's well known that things get less strong as they get larger? (Square-cube law.) $\endgroup$ – Mason Wheeler Jun 13 '18 at 17:24
  • 1
    $\begingroup$ @CarlWitthoft To further build on your clear understanding of the drawbacks of these techniques (more so because every extrusion printer is a little different). Note that although the properties of the plastic may be known, due to these differences in printers or extruders it should not be taken for granted that the properties of the deposited plastic has similar properties. If you look closely there are local voids, under extrusions, separations, etc. This is especially a problem in metal powder laser printing. An even bigger problem is to certify components made by these techniques! $\endgroup$ – 0scar Jun 14 '18 at 11:11
-1
$\begingroup$

A fast way to do this is by using SolidWorks.

You can draw the box in it and run a simulation test with the max load expected.

Here is a link on how to make dynamic load simulations work in SolidWorks, How to apply dynamic load in solidworks simulation ?

The catch in the process is that SolidWorks takes cubes and most objects as complete solids, i.e. 100% infill in 3D printer terms.

You would have to actually design your infill pattern into the cube so as to get the best and most accurate result.

$\endgroup$
1
  • 1
    $\begingroup$ Just about every polymer has non linear mechanical properties, SolidWorks is notoriously bad at nonlinear FEA simulations, so not really a good way to calculate this for polymers $\endgroup$ – 0scar Jun 13 '18 at 10:42

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.