# DIY tensile strength testing

I want to tackle an experiment with the following goal:

Determine the correlations between printing parameters (temperatures, speeds, humidity, perimeters, infill, etc.) and tensile strength using a specific 3D printer, test specimen, and filament brand/model.

This goal calls for two parts then: a standardized test procedure and test specimen. For the test procedure, I've been asking myself:

What portable, measurable and roughly consistent tensile strength test does not require building a complicated machine, can be performed with ready-made tools or machines available at a large hardware store, and can be set up within 5 minutes?

I am thinking here about a procedure that lies somewhere between this hanging scale test and ISO 527. Definitely not using bare hands or pliers. Once the procedure is defined, this begs the question:

Is the ISO 3167 multipurpose test specimen an appropriate specimen for the test procedure outlined above or are there other specimens that are more suitable?

I was thinking that, since the usual filaments have an ultimate strength of around 40-60 MPa, perhaps the "recoil" would be too much and one needs to use a smaller, weaker specimen.

• If I can make a suggestion ... the yellow areas which are made by putting a > in front of the paragraph are commonly used to announce something you've copied over from somewhere else (or "quoting"). When you mark your questions with it, it becomes very strange to read. It is my suggestion, you bold the text of your questions instead ... I almost missed them while reading the body of your question. Aug 8 '18 at 23:50
• @Pᴀᴜʟsᴛᴇʀ2 Yeah, I was actually unsure about this and tried out different elements. Bold looked too much, though, which is why I settled for the blockquote, since it's also used for definitions and I actually have those questions pinned on my wall as definitions for the project.
– typo
Aug 9 '18 at 8:51
• I'm not sure this is really a 3D printing question. It appears better suited to Engineering.SE, since it's only asking about how to set up a reliable tensile strength meter (and I should warn you that good machines are very expensive) Aug 9 '18 at 14:00
• you can fill a water bucket and measure the water instead of trying to freeze a scale's heaviest reading. it's cheap and repeatable to. Aug 11 '18 at 7:40

For these kind of tests you could rely on the ASTM standards. They define test procedures and test specimen sizes for different types of tests. Or you can derive a specimen yourself based on these standards (e.g. for my bachelor's degree I used an alternative notch impact specimen as I was bound to the amount of available material of the turbine rotor blade the specimens were taken from). Considering the material, you could device up a contraption made from extrusion profiles or something.

Please do note that to get reasonable results, you would have to do a lot of tests as the spread in results is probably even more than in metals.

The company I work for does this, these material qualification programmes run for long times (years, as we also do fatigue and creep testing), and a lot of samples are tested to qualify for use in Aerospace applications.

• I suggest looking at CNC Kitchen, who did similar things. for example youtube.com/watch?v=Dbt4IMCZYPQ youtube.com/watch?v=mwS_2R2mIvo youtube.com/watch?v=mIv507btE08 and many more. Aug 9 '18 at 14:31
• Good idea about the extrusion profiles. Somehow I was only thinking about wood and steel. Also, I definitely don't want to do this professionally. I just want to have a device with which I can roughly check the tensile strength. And I will be focusing on one printer and one filament.
– typo
Aug 9 '18 at 19:40

A reliable and repeatable test can be performed by printing cylinders, tapping them to M4 thread and then testing the resulting bolts.

My Tech Fun

does something like that, but he prints the thread directly, which may be less repeatable.

Then you can perform pull and bend tests on the samples using simple attachments available at the hardware store.