The final installment of the HT filament test series—the combination heat-stress test and the final comparison of HT and PLA filament.

Previous Articles


Heat and Hammer Drop Test

Part one in this series discusses the basics of high-temperature filament and how to work with it. The third and final test performed in this article is a combination of the two tests conducted in part 2. This test is intended to test the true claim of the HT filament: that the filament performs at full strength at 100 degrees Celsius.

This test was significantly more difficult to perform than the last two. The test setup was the same as the hammer drop test with the only difference being that this time the material was heated in an oven and carried to the testing device. To ensure that we tested the material at the correct temperature, we used an infrared thermometer.

Unfortunately, this was the best testing scenario I could come up with since I don’t have access to a machine that can keep the material at a constant temperature while stress testing it. And while it is a decent approximation, my approach has two major drawbacks.

Firstly, the material cooled down quickly once it left the oven, so I heated up the material to a few degrees above the target temperature and dropped the hammer on it once it had cooled to the target temperature. This meant that the model had to be heated up and cooled down repeatedly for the test, which ended up being time consuming and potentially degrading to the model.

Secondly, the model was effectively undergoing two separate stresses simultaneously: the vertical stress of the hammer and the horizontal stress of the vice used to hold it in place.

Interestingly, the results of this test were very similar to that of the pure heat test. At high temperature, the HT filament became soft and folded on itself when placed in the vice. Unfortunately, this was a drawback of putting the model in the vice hot and letting it cool down to the right temperature; it was apparent that at even a few degrees above 100 degrees Celsuis, the filament retained none of its original strength. This picture was taken after the model was placed in the vice at 107 degrees Celsius.


Both models (PLA on the left and HT on the right) after being subjected to heat and hammer drops


I was not even clamping down on the model very powerfully; it was just enough to hold it in place. Despite this limitation, I was able to drop the hammer on the model at 100 degrees several times and it easily withstood the impact each time. This does lend credibility to the filament's advertised properties.



So what does that mean for the HT filament? It isn’t enough to look at only the heat resistance; one of the largest differences between the two filaments is how the strength drops off after they reach their highest recommended temperature. Normal PLA appeared to retain at the very least its rigidity at heat, even if it would not hold its full strength. In sharp contrast, once the HT filament reached its maximum recommended temperature, the overall strength of the model plummeted.

However, the HT filament vastly outperformed PLA in a regular strength test. I expect that it performed especially well due to one main reason: layer adhesion.

Looking at the way each the models split apart, the model in standard PLA split along its layers. This isn’t much of a surprise; anyone who has spent time printing models for strength knows the pain of carefully aligning their model so that the layers print in a specific orientation.

However, if we look at the way the HT model broke, it split almost perpendicular to the layer striations, meaning that the layer adhesion was just as strong as, or stronger than, the layers themselves. This is a major advancement for printing items that have to be strong across all axes and don't have the ability to relegate the inherent weakness of layer separation to one axis. Applications like action camera mounts come to mind.


The PLA model showing layer separation


Would I use HT filament in a high-temperature application? Yes, I would. But only if I could guarantee that the model wouldn’t heat up even the slightest bit over 100. In all honesty, in applications getting up to that temperature, it may be time to look for other solutions besides 3D printing.

Would I use HT filament for a high-stress application? It still depends. If the model had to bear a constant, regular force, a carbon fiber composite may still be a superior choice to HT. But in impact applications, HT didn’t just outperform other filaments in this area—it absolutely crushed it. Specifically, if layer adhesion is in question, I have never had a filament perform better. 

High-temperature filament does have some serious development ahead of it, and the 3D printing community will certainly find it useful once it performs well enough. However, as it is today it does not stand up to heat well enough to be useful in those types of environments. However, ColorFabb_HT provides a significant step in the right direction.

Soon 3D printers will be able to create parts that are as durable and heat resistant as traditional building materials.




  • InWonder 2016-09-16

    Although I haven’t looked up the name and character of the polymer that makes up HT, from your results, I’d guess that it’s a crystalline polymer, as opposed to an amorphous polymer.  Many of the higher temperature polymers are crystalline, but there are exceptions.  A key property of the crystalline polymers is their glass transition temperature, above which the polymer rapidly loses mechanical strength, as you observed.  But every polymer is different, and the plots of GTT vs. Temperature vary widely, with some showing not so drastic transition.  Some of the high temperature amorphics, such as Polyphenylsulfone or Polysulfone do not exhibit such drastic transitions.  You need to look at such properties in deciding these engineering questions.

  • Doktor Jones 2016-11-08

    Convenient way to test at 100˚C: test in boiling water. Assuming your vice can withstand being submerged, have the vice and ring inside a container of boiling water (with a heat source continuously applied)... any water that exceeds ~100˚C (give or take 2-3 degrees) will boil off, so the temperature of anything submerged will stay around there too. If you can set it up so the top of the ring is just barely submerged, you should still be able to do the hammer drop test while it’s in the water… of course you’ll want the appropriate safety gear, since dropping a hammer into a container of boiling water will likely cause splashing, and getting splashed with boiling water is just no fun.

    • Doktor Jones 2016-11-08

      To clarify, you’ll want the water at a steady simmer (where you can see small bubbles forming and rising continuously), rather than a rolling boil.

    • Michael Greer 2016-11-08

      That’s an interesting method, I hadn’t thought of that! I’d have to adjust the models because any air trapped inside of the print would heat up and potentially rupture the cavity, but that would certainly provide a more accurate test.