Saturday, 25 February 2012

Replacing Small Parts by Rapid-Prototyping

Since getting back to the city I've been on a small spree of trying to find little things that I could replace or redesign using 3D printing, whenever there was a moment to spare.

Firstly, for a long time I'd been meaning to make replacement parts for a very humble purpose, the male buckle clips on a couple of Highlander rucksacks I have that had mostly lost their teeth after many years of fatigue when packing and unpacking heavy luggage, while the rest of the nylon construction remains as functional as the day I got it.
The buckles used here, with 'Rock Lockster' imprinted along the T-shaped guide, lasted for probably a few thousand clip-unclip cycles over the course of several years before teeth started to break off all the clips. The replacement I made is shown clipped in.

A good set of calipers came in handy in this situation, to measure the fit of these clips to make a compatible replacement. At first I made the base of the clipping teeth a little too stiff for my liking, after making it thicker than the original with the intention of extending its life, but one redesign fixed that.
My first and second versions respectively, in good quality save for the tiny blobs stuck to some sides as I hadn't tuned the extruder retraction quite perfectly.

This turned out to be a pretty good example for me of how well my reprap could handle creating something small and functional with quite complicated geometry to a workable standard of accuracy.

Later while trying to print out a Reprap Prusa-Mendel part that Adrian Bowyer modified, something sort-of cruicial broke on my own printer, making a slight mess of the part.
See if you can spot what was going wrong before I fixed it.
So it turned out that where the wires connected to the thermistor in the nozzle heating block, the use of a bootlace ferrule to crimp the tiny thermistor wire to the insulated wire meant that there was quite a sharp corner in the wire connection, and even though the set of wires going to the heating block didn't get tugged much at all during the x-carriage movement, the shaking about was still enough to eventually tear through that wire and break the connection.
Annoyingly, when the temperature-measuring circuit was open, it reported temperatures to the firmware program that bounced between about -20 and -8 Celcius, and instead of stopping the printer when the temp was negative, it carried on while constantly inputting heat to the nozzle. The result was that the overheated nozzle started dribbling plastic everywhere on movements through the air, while creating lots of fumes that stank the room out.
Thankfully I caught the problem before it was so bad as to ruin the print, and even saved it with a little bit of ingenuity; after switching the heater off for a while to let it cool back down to around a decent working temperature without the above effects, by observing how the open-circuit temperature was bouncing around I could see how to get it to self-regulate, so I set the target temperature to -12, thereby having it pulse the heat on-and-off again, managed to get it to finish the last bit of the print without any more trouble.
Bootlace ferrules and a little bit of heatshink, not good enough. In the background, you can see I had to take apart my x-carriage to get this extruder off, so have just printed new parts for it after I fixed this.
Of course, I can't just leave it at that for long unpredictable prints starting from cold, so I got a hand patching it up with some solder. Had I soldered the connections originally instead of using the ferrules that supplied me with, it might have lasted a bit longer, but I still think that joint would have given way eventually.
If anyone else is using an extruder that has a threaded heater block, with a 100KΩ thermistor that has teeny-tiny short and thin wires, I'd like to know how you connect it up, especially if you've found a way to protect the connection from being tugged while still easily being able to unscrew the heater block for maintenance. While terminal blocks can be good for quickly connecting/releasing a set of wires like that, the wires of my thermistor aren't long enough to fit one without it melting against the hot-end, not to mention that how thin they are would make it difficult to hold them in, so I'm a little bit stumped on this one.

Some other things I was working on prototyping included a replacement shower-dial for a friend where their original injection-moulded ABS part was slipping off the single shaft used for combined water/heat control, and a 'sheila maid'-style laundry drying rack. Not exactly spectacular or groundbreaking, but pretty useful at home.

Incidentally I made an observation of an unexpected plant use recently, when I pulled out the last of a few leeks I had bought quite a while ago, the last one that had been lying around for a while was starting to dry out slightly and wrinkle up. My other half decided that the outer leaves were too thin and 'papery' for her liking, even though it was being chucked into soup. Fiddling with the discarded outer layers, their consistency while there was still some moisture in them reminded me of thin sheets of expanded/inflated LDPE that I've encountered as a first (inner) layer of the original packaging for a flatscreen monitor. A moist material can't be used to package electronics of course, and I wanted to know how they would turn out once dried, so left a couple of leaves (when cut straight up the tube they are usefully rectangular) to dry over the back of a chair near a radiator.
Scribbling on the rougher side of very 'papery' dried leek leaves with biro.

After some messing about with the dried specimens, I've found that while one side is slightly ridged with the leaf veins, the opposite side is slightly smoother, and it makes a very good makeshift notepaper. A biro pen runs right over it with ease, although a 2H pencil that I had to hand quickly punctured it.
If supply lines break down somewhere currently oil-dependent and you don't have access to a paper-mill, this could be quite useful as it requires next to no preparation compared to paper made from minced-up plant fibre such as hemp or nettle - although I just chucked it over the back of a chair to dry, it might turn out less wrinkled if held taut in a frame. Of course with the natural veins, the material is anisotropic, in that it's quite strong along the grain and easily torn into strips by pulling across the grain; that property could have some use, but I haven't thought of one yet.

1 comment:

  1. With regards to the anisotropy of the dried leek: You might obtain a pretty strong material if you get two "sheets" and layer them so that the grains are set in a right angle, and then gluing or pressing them together. A bit like papyrus was made in ancient egypt: layer several thin sheets across each other to make a single sheet that's tear-resistant in all directions.