Wednesday, October 12, 2016

D-Walker build part 4 - Arm

Judo chop

That's my dining room table and I didn't clean it for this photo.

D-Walker's arm didn't turn out exactly as I wanted, but it functioned well enough for another improvised, one-day build.  Again it is made of 6mm Celtec and some assorted hardware.  The arm attaches to the rotating shoulder joint on the body, giving it a little under 180° of motion.  It also utilizes a steel cable attached to the arm controls to pivot at the elbow.  

Steel cable is represented by paracord in this picture.  The spacer at the elbow bolt point lets the forearm pivot.

The above photo is pretty much all of the planning I did for the arm.  A steel cable anchored to the forearm and bent around the elbow could be retracted through some guides in the shoulder, straightening the elbow.  This would make D-Walker do a chopping motion, but I was not able to get very much range of motion out of it.  I tried a few different variations of the geometry, but they all had roughly the same results.  The limiting factor was the rubber-sheathed steel cable, which only had about one inch of travel.  The cable was from another non-essential stock part of my motorcycle (the exhaust flapper system) that was sitting in my junk bin.

Arm control components built by Yamaha Motorsports.

The above photo show the arm controls in the cockpit.  The rubber throttle grip rotates, pulling on the steel cable.  The zip ties are a temporary way to anchor the cable to the grip, something I forgot to do before I got to the con.  The whole control armature can be moved in and out to rotate the shoulder, as demonstrated in the frame section of this blog.

In order to transfer the motion of the steel cable through the 90° bend at the shoulder, I used more of that static kevlar cordage in place of cable.  It threads through a U-bolt in the upper arm and attaches to the end of the steel cable that protrudes from the shoulder.  This makes attaching/detaching the arm easier too, because the kevlar cord is easier to handle than steel cable.

 Not sure why I didn't countersink the bolts.  It might be because once I test-fit the arm together, I never took it apart.

This photo shows the elbow "piston" / return mechanism in a dismantled state.  A spring (later changed to a shorter spring than shown here) is attached to a wooden dowel that connects to a pivoting point on the forearm.  This keeps the elbow in position, but also lets it bounce around a little under its own weight.  The sheath that covers the spring - made from the cores of register paper rolls that I collected at my old bank job - is moved away to show the spring in this photo.  Normally it is zip tied to the upper end of the piston.  The same register paper cores also make up the columns between the two plates that make up the upper arm.

The assembled piston.  The wooden dowel is wrapped in foil tape.  It's not perfect but looks alright considering it's actually functional, and made of random scrap.

Here is the hand.  The fingers are made of paracord with the interior nylon strands removed and replaced with 8 gauge solid copper wire from my old solar job.  They are pliable enough to position but rigid enough to hold light objects.  The finger tips are rubber caps that were leftover from a variety pack which I bought for something on an older motorcycle, I don't remember what.  The hand pieces bolt together through the eye hook at the wrist, which allows it to be manually rotated.  Also visible in the above photo are the nuts that hold the arm to the shoulder.  These "strut nuts" are specifically designed for the aluminum channel that the shoulder is made out of, again both from my solar job.

I should have made the fingers out of those flexible BBQ lighters, but I didn't think of it until much later and I would have had to buy them.  

 Didn't need to countersink the bolts anyway.  This picture is before I fixed the corner near the elbow to match the foam.

Foam panels cover each part of the arm.  These are contact cemented on the borders only, so that if I ever have to take it apart, it won't be completely destroyed.  I would have preferred to make them removable with magnets, but I decided against it for the sake of simplicity and time.  The edges of all the Celtec parts are visible in the final assembly, so they were all painted ahead of time.

The bolt heads on the outer side of the foam panels are all fake, and I foolishly forgot to plan it out so that the fake bolts matched up to the real bolts.  You can see in the below photo that the piston anchor in the upper arm is slightly misaligned with the fake bolt.  Each of the real bolts were positioned based on mechanical necessity and a vague recollection of the source material.  Luckily it ended up being pretty close.

The hand was supposed to be white but for some reason I decided not to paint it.

The shoulder cover is a couple pieces of foam with a magnet bolted to it (wide fender washers keep it from pulling through the foam).  It mates up to another magnet on the arm so that it is easy to get to the bolts that hold the arm on.  Those leftover magnets were used to hold the magazine in place on my MIDA Multitools.  

The C-channel aluminum strut I use comes in two varieties - regular and deep.  I used the deep variety in the shoulder, but it turns out I should have used the regular, more shallow type.  This would have let the arm rest a little closer to the body.   I didn't notice that until final assembly, aka too-late-to-change-anything-cause-the-con-has-already-started time.

In the end, I didn't get enough range of motion out of the arm to pick people up and smash them into the ground.  I would have needed a more complicated, heavy, gear-driven shoulder, and it wasn't worth it.  I think people enjoyed shaking D-Walker's hand more anyway.

No comments:

Post a Comment