Compliant tube based headphone pad idea I had while riding a bicycle two days ago. I got the thickness and dimensions dialed for printing and bending, but I either need to setup my old KP3 kingroon with a longer 2040 Z extrusion or print this in 2 pieces. Either way, the joint connection needs more than just the overlap and glue. TPE would probably be better too, although I have no idea what shore hardness this $10 clearance spool of TPU has. The sound quality seems a little tinny and I have no way to tell how loud it is externally, but it is just my first iteration that I can put around my ear and test. TPU is so slow to print and the moisture levels impact the qualities drastically. I actually like the texture and properties of wet TPU more than dry, but it is hard to get it just right. With the design’s compliant bend, consistency is kinda important. Anyways, just another boring project. On the bright side, this seems cooler temperature wise when the TPU pad is against my ear.
I spent all day chasing custom logarithmic infill patterns that might incorporate a compliant bend but only learned about how not to do a thing like that in CAD.
Have you ever tried Creality’s HP-TPU? It only comes in transparent and white unfortunately, but I’ve printed it at 150mm/s and had no issues with bridging at all. I don’t have much experience with other TPU, but I’ve liked using that one. The base price isn’t very good, but sometimes they have really nice discounts on it.
I haven’t tried a lot of flex materials. I’ve only used them for things like a few seals. Compliant mechanisms have been a curiosity of mine for a long time, but I haven’t had the intuition to establish an entry point project worth trying on my own.
Like as a totally random aside, if this TPU is super dry like how the one test print that looks super crisp with sharp edges, it appears to be air tight. I see a lot of potential for building cheap pneumatic, cable, or passive force driven actuators while playing around with my thumbs sealing each end.
I intuit that this level of usefulness in mechanisms would be hampered by the low quality of the first bridging layers. Absolutely any moisture in the TPU causes random gaps to form as the steam escapes at the nozzle tip in small bursts. Any larger bridging is going to have some amount of dropped passes as a result. I don’t think this is a real issue if the TPU is very dry on a totally enclosed dry feed path to the extruder, but I don’t like the properties of this material when it is super dry. Overall, my design method in this case is likely oriented in the best way for the properties of TPU and the mechanical best case for compliant design. The layer deposition steps and top/bottom layer properties of FDM are not optimal for compliance in most cases. This particular design is capable of compliance both for the bending form to create the headphones pad, and as a pad against the ear after it is installed.
It is also ~$10-$15 for replacement headphone pads, so making and sharing such a design should be limited to materials most people already have on hand. I’m very tempted to try this with a 98A TPE, but it is just too expensive of a material to justify for this project when I’ve had 3 rolls of TPU banging around for years unused and only got them because they were dirt cheap clearance sale materials. I would do a lot more if I had an IDEX, but I don’t need that rabbit hole money bonfire.
If anything, a dirt cheap foaming TPU could be interesting if such a thing existed. It might be possible to create something functionally similar if TPU could somehow be exposed to a humidity controlled environment at a specific percentage, but I have no idea how moisture saturation works on a deep level, like if the saturation would remain regulated by the humidity percentage or if the exposure would allow the filament to always wick all available moisture where a much more complicated setup would be required to ensure consistent properties. Anyways, my point here is that the best properties for me are not from the super dried TPU needed for bridging and bridging is itself a poor mechanism with FDM. It is best if it can be avoided at the design level like I have done here.