Strength Comparison of Infill Percentage and Patters

Has anybody done a comparison of different infill percentages and patterns, and the tensile yield/ultimate yield/bending strength of the parts? I'm very curious as to how much of a quantitative difference for example a 50% to 100% infill would make in a structurally loaded part, as well as the fill pattern from say honeycomb to rectilinear. I'm sure this data is out there somewhere and I just can't find it...

Hi Jonny

Ultimately this will be relative to how well/poorly each individual printer is operating. so 30% infill will have potentially lower strength than the same part printed on the same machine at say 50% or 70% infill, but the same part printed on a different machine will have different print conditions and quality that affect its strength for each ratio and type of infill used.

There will be little difference between 70% infill and 100% infill, but a bigger difference between say 10% and 40%. Temps and correct layer bonding play a large part in this, perhaps more so than a particular infil pattern, but also depends on the part and its use.

There will be more that others can add in relation to the strength of individual infill pattern styles. Ive heard the hexagon infill is very strong, however i have not personally done such testing but others have.

Hope this helps and gives you some insight/food for thought.

I'm not sure how much use detailed comparison tests would be. They would probably consist of rectangular bars put in tension or bending, but real parts (and the loads they see) are almost never that simple, and the performance of real parts will vary wildly depending on the specific geometry. That said, some general rules of thumb (like hex infill is typically stronger than rectilinear, etc) would certainly be helpful. Of course, I have no data myself, so I am utterly unhelpful in this regard.

If a printer prints too cold, the layers won't bond well. This may interfere with the results. As Mutley points out, it's such a difference between printers. I think a theoretical approach might be better in this case?

There are computer programs that can calculate stress/bend etc. I think Solidworks can also do this.

---

http://www.marinusdebeer.nl/