cheap low profile fasteners for joining multiple parts

I was thinking about how to join two cylinders without adhesives, welding, or mechanically screwing them together, and I remembered reading about an ancient stone joining technique that cast molten copper/bronze/iron in "I profile" notches in the stone. The metal cooled and cinched the two stones together.

The idea is to do the same thing with plastic. Basically, design into the work piece flow paths for molten plastic and then extrude into a fill hole to fill the gaps. In this way you can put things inside an enclosed case without having to design in screw holes or threading. Plus its low profile, so you can build it into the walls of the structure.

Also, the technique isn't dependent on the properties of the plastic, so you could join dissimilar plastics, metals, ceramics, etc without too much problems.

Problem is that its a non-reversible process, but in some special cases it could be very useful when weight or clearance are issues.

... you can use Fields-matal melting at 71 Centigrades or the less toxic Roses-metal melting at 98 Centigrades too

Its not the process, its the pattern of thinking that is important.

Any material that exhibits shrinkage on controllable parameters, i.e. temperature, solvent evaporation, curing times, will work.

... shrink tubing is an interesting methode - should be possible with I-shapes or other geometries too.

Simply heat and stretch a part from a 'thermo-elastomer'-material, so the molecules are straight aligned, then let it cool down.

When heated again, the internal stress releases and the part is reversed in it's normal shape ...

---

Viktor
--------
Aufruf zum Projekt "Müll-freie Meere" - [reprap.org] -- Deutsche Facebook-Gruppe - [www.facebook.com]

Call for the project "garbage-free seas" - [reprap.org]

I like it!

Only problem is for precision parts I wouldn't trust a friction grip, but for run of the mill tasks, you could design in barbs to catch as the shrink wrap pulls in.


For what I am doing, I am going to use a syringe filled with a viscus acetone and ABS blend to inject into the holes. Making sure I include overflow blow holes and decent anchor points. I know from experience that as the acetone evaporates, the ABS will shrink and pull together.

Ok, I think this idea is important enough to design up and release a test file.

Here is the thingiverse link.
[www.thingiverse.com]

Here is a picture of the working untested prototype.


and here are some notes.

This is the first version of the injection coupling that I’m happy with. The way it works, is that you connect two cylinders together by putting together the ends and filling every other hole along the seam. This should produce three filled holes with non filled holes in between. This should block air passages and allow the second filling to push material throughout the entire structure, hopefully without too much back pressure pushing material out. If this is the case, then the first filling should be left to sit for an hour or so or can be baked to boil off the solvent. Once all the holes are filled and the piece has been set to cure, it should give a satisfactory bond.

Things to do. Get someone to test the idea. Find optimal acetone to plastic percentage. Try method with detached extruder barrel for hot extrusion. Try method with Roses Metal. Look for potential improvements.

Known issues:
The inner ring teardrop cutouts are not as pretty as they could be. This is partly a limitation of openSCAD, as I can’t rotationally extrude solids. It could be fixed by figuring out exactly how the openSCAD teardrop cutout works, but I’m leaving this as an exercise for the motivated student.

Can you do a rotated cut of, not the solid, but the 2D teardrop profile?

thats what I'm doing. Its just that the size of the union square and triangle are off. They fit mostly-ish. I could have figured it out by looking up the point of the circle which comes off at a 45 degree angle, the length of that chord and its distance from the radius, but got tired and lazy. They also do something similar with another openSCAD object which is the union of a cylinder and a cube.

I fixed the problem. It was easy. the latest version is on thingiverse

I know something MUCH easier than doing this. I have spent a long time designing a milling machine with a frame made out of 1/16 inch plastic sheet, and to connect the thin parts at right angles, I use zip ties. Basically, there is a slot that one piece fits into on the other piece, kind of like a mortise and tenon like seen on the cupcake printer. Only I use zip ties to hold these joints together. I have attached a few pictures. I got two bags of a thousand zip ties from ebay>>
[cgi.ebay.com]

seller is "goldmouse66" I got 2000 4" zip ties for 10 bucks! That should last me a lifetime of projects. My milling machine I have designed only uses around 200.

I have also attached a picture of one of the bags of zip ties.

Also, this is not just experimental. I have actually milled out a test piece for this, and it is extremely sturdy, maybe even better than a nut and bolt type connection, since the nuts can slide out of their slots. Plus it is very quick to put together, and it won't come loose! It is also more compact than the nut and bolt + slot method. Too bad I lost the little test piece. I will try to find it.

2000 zip ties/ 10 dollars= 0.5 cents each. Nuts and bolts are around 2 cents each at the least, coming to 4 cents per joint. Just something to think about.

Although interesting, it has some problems. Those zip ties are going to stretch over time if they are under load, so you can't use them for anything that you want to remain precisely in place.

It might be useful to hold pieces together for glue to set. So, design in zip tie holds for all your pieces, construct them all and then set them with epoxy.

also, 1/16th inch plastic sheet? for milling? It could work, but it will be very difficult to design.

I have attached a picture of my design for the milling machine.
By the way, nuts and bolts loosen much more than zip ties, just from the machine moving and vibrating, believe me. I have a home built milling machine and the nuts have to be tightened every couple months. Just because the zip ties stretch doesn't mean they can't be used in high accuracy situations. They just need to be tightened when they get loose. And they don't get loose that fast. I have had that little test piece for a long time, and I have stretched it and pulled it to see if it would loosen, but it doesn't. I think it is because the zip ties I got from ebay are very stiff. They are rated for 50 lbs, and they don't stretch much before they break. I really think they are much less work and money than bolts and nuts. Plus for every one bolt and nut, you can put on 8 zip ties for the same price. With enough of them, loosening and stretching would be negligible.

Galaxyman7, you should post your design and its rational in its own thread! Its interesting and you've put a lot of work into it.

For loosening bolts I would look at loctite, or for a super cheap version, superglue with a very light dusting of baking soda will also work. The baking soda activates the superglue without having to be exposed to air.

Your right that there will always be an easy pre-made solution, but what I was trying to demonstrate was a method of joining plastics that can be built with just a reprap. The method still requires a solvent, epoxy, molten plastic, liquid metal, etc... Yet given that large variety of options, it is likely that you'll have something on hand.

I think something that my design doesn't show, but yours illustrates in an off hand way, is the power of flexible plastic mechanical fasteners. Literally snap together pieces printed from your reprap. Thats what zip ties really are anyways.

I'll design up some zip ties maybe this week, and maybe I'll get someone to print them and test them!

Actually I was thinking you could make the pieces snap together, like a belt buckle. The two ends spread apart and snap over another piece. This way it could be very easy to put together and take apart.

I second the call for you to start a thread about your milling machine. Love it.

What are you drawing it with?

I am drawing it with solid edge. It is meant to be a milling machine that can mill itself out, and also can add onto itself and make itself larger. Right now it takes around 5- 2' X 1'sheets of polypropylene for all the parts, plus around 200 zip ties. It uses 1 inch conduit pipe for the rails, and polypropylene wear pads instead of linear bearings. It uses $5 geared hobby motors for each axis, using 1/4 threaded rod. It is all controlled through a $3 parallax sx microcontroller, which reads the step and direction signals and makes the motors move accordingly. The aim of the entire project was to make a $50 milling machine including electronics (not the dremel though). So far it is around $60. The original thread is on cnczone.com>> here
[cnczone.com]
I am pretty much done with the design, and I have the polypropylene sheet. The next thing to do is design the electronics, which require a couple hbridge ICs and some protection circuitry. I already have an mdf cnc, so I will make all the parts plus the pc board with that. If you want I will make a thread on reprap.

Any ways, lets get back to joining two cylinders. Do you mean a motor coupling or just to hold the structural rods together? Are the rods plastic or metal?If they are metal, how will you actually mechanically attach the two metal rods without drilling into them? You said that you couldn't use a friction grip.

The way I have done it is with small set screws from fastenal. They have a bag of 100 for $3 or so. Of course I know you don't want to use outside sources. The only other way is a friction grip or by putting some kind of pin through a hole in the rod.

The initial intent was to enable making fully enclosed areas and to join lengths of plastic pipe without using bulky fasteners and to build strong permanent bonds.

The initial need was that I needed to encapsulate a strong magnet inside a plastic rod for use in my magnet ball bearings.

However, it also allowed me to overcome a difficulty I was having with the lengths of tube I was printing being too long.

This brought up a whole new concept, for me at least, of being able to print small pieces and put them together without having to worry about compatible cements and weak bonds.

Mostly, its an interesting construction technique. Like a paint brush for detail work. Useful when you need it, and useless for broad strokes.

In other news, I designed up a plastic cable tie. The ratchet ones you have are better, but I was surprised no one had designed one yet.

I think there actually was someone on thingiverse that made one, but I dont remember where I saw it. It was pretty thick and large though.

Anyways, I see what you are talking about now. You want to connect multiple plastic cylinders together into one. I really think this can be done without the molten plastic or epoxy, using small plastic clips like the ones to hold down parallel ports on a pc board. I have attached a picture.

They wouldn't be too hard to print and use. Just print 4 or more around the circumference of the cylinder, and they all snap into the next cylinder.

I will create a seperate thread for the milling machine

Edited 1 time(s). Last edit at 03/22/2010 07:11PM by galaxyman7.

This is a nice idea, but sounds a little like using a sledge hammer to crack a nut.

Joining two plastic parts is done regularly in industry. The most common method is vibrational welding. The parts are vibrated together and generate heat at the interface and weld together. Its a variation of friction welding. Its perhaps a little complex for hobby applications but could be adapted. If you look under the hood of most modern cars you will see this technique used on many of the plastic mouldings, two shells welded together. Check out the brake reservoir, inlet manifold etc. (I design such stuff for my main job)

Stir welding is also a possibility. It is a patented process for metal, but could be adopted for plastic. You basically use a blunt rotating spindle to melt the two parts together.I visited the Welding institute in Cambridge UK and saw this process. Its very neat and is used by people such as NASA for rocket fuel tanks. Perhaps a Dremmel with a blunt steel tool might be worth a try to stir the two parts together.

You could also use a hot knife or soldering iron to locally join two parts.

To solve the plastic to metal problem, design two plastic parts that when welded together hold the metal piece in place. For example on the round part design a retaining ring that welds to the top edge to fix the metal part.

The image attached shows a disposable knife. I have seen these where the blade is retained using a "spot weld" in the plastic. For most applications its just a case of designing the part to provide easy weld points, but using a mechanical pocket to give accurate location.