You probably read my blog and know I am working on a design for robot legs inspired by the legs of Theo Jansen's Strandbeest.
Below is a picture of a plywood prototype on its way to become version 6.
The three layers of ply are connected in the pivot points by means of paperclip wire through 1mm holes. This is by far not strong and durable enough for the application I have in mind. This leg can easily carry a 2 kg load. But I need to improve the pivots first. As it is, the holes wear out fast and become very loose.
You can see the black glass head of a pin sticking out in the center. This pivot hinges from the robot's chassis. I call it point €. At the far right is the point that is to be driven by the crank shaft. All the other pivots connect to leg parts only.
The pivot design I am looking for must be: * tight enough to keep the flat parts snugly together * without causing too much friction between the plates * positioned with precision * allow the rear side to be mounted flush to the flat robot body (so no screw heads sticking out)
I am considering some metal reinforcements of the holes through the plywood. Like hollow rivets or washers and bushes. I am also considering using a different material, like plastic or metal. But I am afraid that those materials will also wear out very rapidly.
I appreciate your inputs.
Update 16 feb 2009: For completeness, I include Chris' video on brass cutting. Thanks dude!
This is tough with the plywood only 1/4" thick. Also you have some at 2 layers, some at 3 layers and some with the vert. pieces… Hmm. Just off hand, I would say 1/16" I.D. brass tube with 1/16" piano wire through the inside. The brass tube will be the sleeve while the wire is the hinge pin. On the inside the wire can be bent to a 90 and pressed into the wood .And on the outside a coller (with an allen key set-screw) can keep the wire in place. The company that sells the little collars is called "Dubro" and all model airplane shops carry it.
Here is a tail wheel I made with this tube/wire thing. In this case, the brass tube was soldered to the little copper bracket you see, but this kind of configuration should work for you.
My calc shows 3.6 / 25.4 as appr. 1/7th of an inch.
And 1/16th inch is almost 1.6 mm (inner diameter). That’s a decent, strong pin I guess. The collar would help tuning the tightness of the fixing (and the pressure between the plates).
I guess I have a visit to a model builder’s shop in my future.
I made this practice piece. Each corner has a different pivot.
This one has a piece of 4mm Ø, thick walled, brass tube running through bare holes in the plywood. Actually it is 5/32 inch. Dutch model builders stores buy their stuff in Chicago apparently.
This is "5 mm" Ø, thin walled, brass.
Cutting brass at these sizes is not very easy. I bought a tubing cutter, but it compresses the tube while cutting. The deformation makes it hard to fit the 4 mm into the 5 mm tube. Which is exactly why I bought three different sizes to begin with. Here is a 4 mm piece sleeved through two 5 mm bushes. I found that the outer tube does not deform (much) when I put a piece of the thinner tube inside it while cutting. But that will eventually lead to damaging the inner tube, making it impossible to slide off the tiny brass ring from the inner tube.
Cutting brass with my jig saw worked, but it costs me a blade per cut. Maybe I am not using the right blades....
Hard to judge now. None of the brass solutions I made hold the pieces together, yet. I did not fit a screw or bolt through them. The combination of small and wider tube as a pin though bushes works very well. The pivot turns very smoothly. But the fabrication is hard work: six cuts per pivot.
Perhaps if I used the inner tube as a split or "bifurcated" pin, I could use that to hold the plates together. I suppose I could complicate fabrication further by cutting into the end of a 4 mm tube with a small hack saw or a wire cutter maybe.
The cheap plastic rivets really are not made for this task. I had to weld the ring on with the hot glue gun, or it would slide off slowly with each movement of the pivot. Also the head and ring of these rivets are too big to sink into the 4 mm plywood.
Man I could have saved you some time… Here’s how you cut that brass tube…
Use a brand-new razor blade - the kind you would use in a retractable utility knife. Place the blade on top of your tube on a nice, flat table. Now roll the tube back and forth (with the blade) with gentle pressure on the blade. This is the same idea as the tubing cutter but without the rollers pinching the other side of the tube. Simple, -razor blade and roll it back and forth on a table. Be sure to “stay in the groove” to make sure you are not cutting a spiral. This will make a perfect cut and it won’t “mushroom” into the center! --Also those metal cutting blades must be leaving burrs.
A trip to my armory resulted in a plethora of blades that could be regarded as utility blades or even as razor blades. Notice from my appearance (elsewhere, some other day preferably) that I have not used the first one in this picture for quite a while.
Top to bottom, loosely translated from "Dutch common names": "scraper" (sold as actual razor: Gilette Sensor, has its blades upsidedown in this photo, that's how long it's been), "Stanley knife" with spare blade (scary utility knife, retractable version), "small breakaway knife" (hobby knife also by Stanley), "big breakaway knife" with spare blade(s) (bigger hobby knife, Stanley ripoff), "pocket knife" (better not carry this in your pocket in crowded places like airports, more of a camping knife), "Jokari" (big ass wire stripper: the wire has a big ass, not the stripper).
Something in your Walter post tells me you are probably talking about the second tool...
I speculated about the blade and went with the scary type. Here is a result.
Notice: 1) no stain, yeah! 2) spiraling, booh! 3) slight burring, oh well... 4) no mushrooming, yeah!
Some lessons learned: 1) this still is a job that will drain all the power from your fingers and fore arms; 2) do not subject your good tabletop to this method (use a pizzabox desk protector); 3) move the tube, hold the knife still - I started out by "riding the knife on the tube" like a clown would walk on a barrel; 4) there is no shame in starting a good groove with the original cutting tool; 5) although this may not be necessary any longer, now that I know to hold the knife still; 6) some brass saw dust will collect in your work area - mind your electronics! 7) a lathe might replace the (strained) left arm, maybe a cordless drill on low speed?
Definately a better looking one than the one I ended up with.
The original store tried to sell me a K&S cutter, but could not find it in their cluttered display cabinet. And the second store sold me the one with a metal roller. Which looks so much more coarse than the one you’re ponting out:
Bigger pipes May have used the one you have, on 3/8 (9.65 mm) copper tubing and even 1" (25.4 mm) galvanized pipe. I think they only cut well down to a certain diameter. Little ones like the K&S probably work better on the small stuff. I may need to get one soon, if I can find one. The K&S bins around here seem to be only metal, no tools, though there is a hobby shop or 2 that might have one.
Wow, I just had one of those “I cannot believe that has been my problem all along” moments. Very reminisent of the “Ahah-firmware-erlebnis”.
Here’s the pictures that will save my all that typing:
I had a trip to the hardware stores today and bought all this and more. Notice: two types of blades, two different brands. Let's focus on the black-and-yellow brand.
That's right: each length of blade also has a different thickness of blade! Being the pseudo scientist that I am, I tried them both. I also took a close up of the cutting edges.
The picture does not do the sharpness any justice. But I can now report that the shorter, narrower blade is indeed much sharper. I think the photo shows that is has a very differently sharpened edge. Its edge is so much more smooth and refined.
And it cuts through brass tubing! YEAH BABY! Carpenter style!
Controlling the blade takes some skill though. I found it was easier to do on a stable worktop (unlinke my wonky bench). Also, clamping the blade in its intended knife helps me a lot to prevent the cut to "spiral out off control".
No video just yet. I am way too tired and, speaking of "razors" due to get cut.
And I am getting the hang of this macro photography. Look at my prototype for a pivot bolt.
The base is thick walled 4 mm tubing. Which I threaded with an M4 thread cutter. I really like how pretty these little tiny grooves look! It awakens the crow in me. Shiny!!! CHRAWWH!
From left to right I added: a M4 washer and a bit of 5 mm tubing and soldered them onto the inner tube. I added the 5 mm tube, so that the washer would have something to sit on. The hole in the washer is not exactly a snug fit. The 5 mm bush will be "press fitted" inside a 5 mm hole in the first piece of plywood.
Then comes a loose washer, a loose 5 mm bush and another loose washer. These will turn freely around the inner pin and will fit the second piece of plywood.
The lock nut will be tightened just tight enough to give some pressure against the ply. Overall length is 48 mm.
Or that's the theory at least. First order will be to make one that is actually to size with the other materials. This prototype only serves to prove that fabrication is doable. Now I need to prove that I can make it much shorter: .5 + 4 + .5 + 4 + .5 + 5 + 1 = 15.5 mm overall. And that it actually works!
Still I am wondering if I just reinvented a nail, or a bolt, or a threaded nail?
I bent this one though. The threads just look ugly! CHRAWWHfull!
And the bush won't slide over the nail's shaft all the way to the head, because there is a tool mark there from shaping the wire into a nail in the factory.
And no, I am not interested to read what "she" had to say about that!
