Colibri Hummingbird Dynamic Sculpture - WIP

[kvom]

Today I wanted to make the 7 remaining parts that are 1/8" thick.  However, apparently Mic6 doesn't come in 1/8" sheets, so my idea was to use 1/4" plate, machine down 1/8", and then face off the bottom.  Unfortunately the best piece of material I found had room for only 6 of the parts, so I included the ones that are needed soonest in the assembly.  Here's the result:



The 4 on the left are spacers for the cam wheels, and the two on the right are the thin shoulder pieces.

Unfortunately the face off plan didn't work as clamping the sheet in the mill vise bowed it, and the face mill screeched when trying to cut it.  So I decided to cut all of them out individually.  The shoulder pieces could be clamped in the vise firmly enough for the facing op, and the round parts were faced on the lathe.  For the odd shaped cross I'll probably need to make soft jaws for the vise.



My day's production:

[Dave Otto]

Hi Kirk

Why not thin them up while everything was still in one piece and on the vacuum chuck?
You could do a pocket routine on each part so you are not cutting the entire piece of stock?

Dave

[kvom]

Once they were milled the chuck couldn't maintain vacuum.  A better idea in retrospect was to facemill the stock to thickness as the first op.   

[Dave Otto]

Yes that is kind of what I was thinking, only just pocket each part down to 1/8" thickness then do the profile. The other option is, they do make 6061 T6 rolled aluminum plate in 1/8" thickness.
You are making good progress on this one.

Dave

[Art K]

Kirk,
Dito on what Dave said. Seems you could have done the 1/8 thickness with a face mill then did the profile. Hind site being 20/20 and all.
Art

[kvom]

6061 would have been OK for the spacers since they're no very visible .

[kvom]

It seemed to me that the remaining spacer could be milled to thickness on the Bridgeport if clamped fairly lightly and light cuts used.

So it proved to be:





Then I moved on to making the last remaining pieces from 1/2" plate.  One of the problems I foresaw was drilling small, deep holes.  I used peck drilling, and also kept the bristles of a brush around the drill bits while in operation.  This had the effect of clearing the flutes on each retract.  After several hours I had this:



Another hour of deburring plus drilling out the remaining .01" in all of the holes:



The list of parts needed to be cut from plate is pretty small at this point.  There's a group that can be done together from some 3/8", plus some more spacers from round stock.  A number of parts need lateral (xy plane) holes drilled and/or tapped, so I'll start to polish these off pretty soon as well. I don't have material big enough for the base as drawn, but I may make a temporary one for use in the assembly.

[kvom]

I took a parts inventory today and discovered two 1/2" parts are missing plus I screwed one up yesterday.  Seems in adjusting a parts location on the stock I moves the profile but no the holes.  You can see the bottom part on the photo has the holes off center.

I also found four 3/8" parts left to do, and I started to machine them on the vacuum chuck.  I machined a pocket just slightly too deep, and the flashing cracked and slit the film underneath.  So I lost vacuum.   Luckily the stock didn't move enough to cause a problem before I shut it down.  So I reduced the vacuum area to exclude the leaky pocket, and continued on.  But now the machining force apparently was enough to move the stock on the table.  So I'm going to finish this when I remount the vise tomorrow.

There are also two 1/4" parts that will also be done on the vise.

[kvom]

Today I finished what should be the last 3/8" parts to be machined on the mill.  As stated on the previous post, I had to move the work to the vise, which meant swapping the vacuum check for the vise, tramming, etc.  Anyway, worked through it all and ended up with this assortment.



The long pointed part is the bill, a section of which I'll either paint or dye black.

Now with the vise installed I could make the center section of the flower from some 1.5" hex  brass.  I turned one end down  on the lathe so that I could find the center before milling.



Then back to the lathe to part it off.  Here's a trial fit with one of the petals and the stem.



I'll have to shorten the screws of course.

[crueby]

You mention dyeing a part - with what?

[kvom]

I have some chemical blackener that works on aluminum.

[kvom]

Spent the day doing subassemblies.  Most of the work for these involved turning spacers and parting tube at the lathe.  But the first job was to fix up a temporary base from a piece of 10x10" aluminum and attach the support frames to it.  I used the dimensions from the plans for the holes.  But since the frames are made from 1/4" plate vs. 3/8" wood, the space between the frames is 1/8" wider than per the design.  That shouldn't make a difference as long as the main frame is shifted right 1/16.  And even that isn't a big deal.  The position of the main frame within the support frames is decided by a spacer on the drive shaft that passes through the lower pinion gear.



You can just see a bearing mounted inside the right support frame.  The bottom cam rides on this and controls the up and down motion of the entire mechanism.

The 4 assemblies show below form part of the mechanism that drives the flexing of the wings.  Each pair is connected by a strong magnet that allows them to rotate relative to the magnet axis.  The magnets were chosen in the design because of very little space for any other type of rotational fastener that would be easy to fabricate.



The 3 assemblies shown below are the "followers" for the other 3 cams.  Two have needle bearings that ride on the cam, and the smaller one on the top rides the the small nub.  I'm wondering if this could cause a galling problem on the aluminum cam.  Similarly, the round spacers will contact the faces of other spacers on the frame.  These can be separated by brass shims if there is a problem.

[crueby]

Ooohh! Now we are getting close to seeing how all those shapes come together! 




 

[b.lindsey]

This is all looking really great Kirk!! Now the fun part of seeing it all come together and working!

Bill

[kvom]

Today I attacked 6 more sub assemblies.  The first is the "hummer body right" which combines the right side of the body (red acrylic) with the cam link that drives it.  In Colibri, each of the 3 cams (tail, body, and head) have a "follower" that pivots as the cam turns.  The followers have a slot in which bearings attached to a cam link travels.  The combined motions of the follower and link result in the motions of the three main parts.



The left side of the  body will be added to the other side at a later stage of the assembly.

Here are the two spinner assemblies.  They will eventually be connected by a shaft and activate the wings.  Since the small gear turns 8 times for each full rotation of the cames, the wings will "beat" 8 times.



The most time consuming job of the day was mating the cams with their spacers and pins.  Having drilled the holes for the pins with a 1/16" drill, I found them too tight to insert through the cams and spacers.  By enlarging the holes with an .063" drill #52 I was able to get a moderate press fit.



The pins ride in the circular openings in the hypocycloid gears while the 1/8" shafts from the eccentrics ride in the centers of the cams.

The number of unattached parts is getting smaller day by day.