An ILM R2 Story

Ah - so I was wrong, then. The bars looked to me like they were supposed to be folded on the blueprints.
Well, blueprints are great and all, but things often change when it comes time to actually build things (if that was what the blueprint really intended).

I imagine the builders found it was a heck of a lot easier to do the edge folds on the back plate and then lay in the solid bars, rather than do a crazy number of back and forth bends to make all the "bars" as ridges on a single piece of sheet metal.

That'd be, like 20 bends on a little 4.5" piece of metal!
And there are 4 vents per robot!
And they made 6 robots for the first film!
That's almost 500 bends! Just for the vents! Oi!
 
I love the sheet metal construction on these parts. Just the way it should be done. And the trapezoid shape on the bars is news to me. What a great detail.
 
Well, blueprints are great and all, but things often change when it comes time to actually build things (if that was what the blueprint really intended).

ANH R2 is a bit special in that he was not built by prop guys. He was built by engineers.

In the past folks have tried to make light of the body blueprints, but have not yet been able to produce an incorrect dimension.

One thing we have to remember that is special about Dr. David Watling and Peteric Engineering is that these guys were PHD engineers and had WW2 era machinists working for them. David states that they build the body to the 2nd issue body blueprint.

One thing to note about the body blueprint is that it only dimensions the outer surfaces of things. Its does not call out any specific manufacturing technique. In fact when the blueprints were first drawn up, it was still thought that the bodies would be made out of fiberglass.

To NKGs comment, the dimensions should all be correct but the manufacturing method is not captured.
 
Well, blueprints are great and all, but things often change when it comes time to actually build things (if that was what the blueprint really intended).

It will be interesting as the project progresses to see if anything was changed. For the vents, showing that the trapezoidal details are actually on the real props, seems to be the first bit of public reinforcement that David did faithfully use the blueprints.
 
I've been sorta studying techniques for building homebuilt aircraft for a long time. (More finding and buying books, magazines, and articles than studying them in the depth like I should.) But when some of the reasonably detailed pics of the original R2s started appearing a few years ago with rivets holding the skin layers together, it made me think that some of the construction could probably or maybe was done using tradition metal aircraft fabrication techniques. You can build your own airplane at home if you have the space, the cash, a lot of time and persistence. And as I recall, Spitfires were more handbuilt than strict assembly line like in the US. So maybe some of the old metal aircraft builders, or at least a few of their techniques, were being used for some of R2... Just an idea I had. I've seen some AMAZING homebuilt aircraft, and I can imagine some of those metal bending, curving, shaping and riveting techniques being used to build an R2.

Even though CNC routed and laser cut parts are more affordable than even a few years ago, I'd still kinda like to try to build an R2 mostly all out of metal. Without having to wait for "someone to do a run on ______" or sending out an order to some jobber. Seems like it'd be quite a bit cheaper. Probably take waaay longer, too....

I should get some ambition together at some point....
 
Just another comment about blueprints when it comes to Star Wars; and I can't say if this was true for the original trilogy, but the blueprints made during production of the prequels were drawn after the models were built. Initial designs were sketchy, so there was a lot of creative design that was figured out on-the-fly during the process of making the models - so it makes sense that to draw them after the fact due to the design not being complete until the model was done.

I wasn't privy to the set construction blueprints, but I imagine those WERE drawn before being built just because of the scale of the construction. Minor details might be tweaked; but for the most part, a set crew needed a clear set of plans to go on.

Of course the original aluminum R2's are a whole different animal, since they were farmed out to a vendor that was used to working from detailed plans, and thus couldn't (or wouldn't) work from the loose sketches that a production shop is used to.
 
I've been so happy with how the vents came out, that I'm thinking of offering them for sale. Maybe I should start an interest thread.

In the mean time, I wanted to get on to the lighting in R2's dome.
There are two types of lighting gags in the dome: the logic displays and the PSI flip-flop lights.

The flip-flop lights (as I like to call them) switch from one color to another and back again. I've heard that originally they had some sort of sliding mechanism to slide the colored gels back and forth to change the color. I can neither confirm nor deny whether that is how they were done; but by the time I got to work on the R2s, they were instead using a divided cup with two bulbs behind color gels in it that switched back and forth to change the light color.

The front flip-flop light is red/blue and the rear flip-flop light is green/orange. I used 12v LED auto turn signal bulbs on mine because they run cooler, last longer and use less power than incandescent bulbs. You can kinda see one under the orange lighting gel.

21.4 Red Blue PSI.jpg
21.4 Green Orange PSI.JPG



I poked around online and found a how-to on making a 555 flip-flop circuit to control the back and forth flashing of the lights.
It was super cheap to make and worked great! It even included a potentiometer to adjust the speed of the flip-flopping.

Exhibit B:

20.4 Flip Flop 4.JPG


The video color doesn't look as saturated as it does in real life, but here we go!

ezgif.com-video-to-gif.gif
 
Last edited:
Yes, the research we’ve done (especially Robert), and interviews and conversations that people have had with two surviving principals involved in making the original aluminium droid bodies, indicate that they did rely on the blueprints for much of the work. In fact, Peteric started work on the droids when the first batch of blueprints arrived, and then had to throw the work out and start again when the revised blueprints were shipped over.

The legs and feet were partial exceptions - as I recall they used models built at EMI Elstree to draw up interim drawings (now lost) for the dimensions, though of course those parts were drawn in the blueprints. The dome dimensions were also partly based on the plaster cast made at EMI of the original lamp housing from Lee Studios.

The ANH blueprints were reused for the pattern making of the Empire droids, and the difference between blueprint revisions seems to account for many of the differences between the droid generations, particularly the panel positions. :)
 
For all of R2's personality, he only really has a couple lighting gags - the flip-flop lights that I showed above, and the tiny flickering lights of what has come to be called the logic displays. ( I love how the fans have come up with names and functions for all the parts of Star Wars paraphernalia.)

One of the upgrades that was done on the R2s when renovating them for the prequels was to convert them all to LED lighting. They still used the original fiber-optic bundles; but for the new films, Grant Imahara made a "puck" of LEDs that faded and flickered on and off to simulate the original color wheel lighting gag that made up the original logic displays. This reduced the power consumption considerably from the original incandescent light and motorized wheel set-up. It also was a lot smaller (making it less cramped in the dome) and had less of a chance of breaking since it was solid-state. The new "LED pucks" were also silent, which the sound guys were thrilled about, because the color wheel motors made a constant whirring noise during filming.


But,,,,

I didn't do that for mine....

I decided to go "Old School".

My R2 is using color wheels!
I like the organic way the lights move around with the older system. The lights don't just fade on and off - they move.

Since I had access, I was able to get the same fiber optic strands as the originals, and I could make the bundles the same as the originals.
Unfortunately I didn't have any documentation of the old color wheel mechanism. So if any of you are trying to make a super accurate R2, don't view my design as gospel - it's my own design based on motors and lights that I decided to use.

...and here it is!

21.5 Color wheel 1.JPG
 
Last edited:
I started with the fiber optic bundles. Their size determined the size of the color wheel disc.

20.4 Logic Fibers Lg 2.JPG



Then I found small 12 volt gearmotors with a slow speed of about 3 rpm. I ended up getting some adjustable speed controllers for them in order to really dial in the right speed.

20.4 Color Wheels 1.JPG


I machined some hubs to mount the clear discs to the motors.

20.4 Color Wheel Hub 2.JPG

20.4 Color Wheel Hub 3.JPG


Then I used styrene sheet to build up the housing to hold it all inside.

20.4 Color Wheels 12.JPG


They came out looking like little Millennium Falcons - kinda' appropriate:p. You can see the end of the LED turn signal bulb that I'm using as the light source. I then used color gels taped to the clear disc to create the color patterns. I'll probably keep playing with the color gels till I'm happy with the way they move.
 
Nifty stuff! Hardly anyone has successfully managed to emulate the swirling, moving, pattern look of the original droids, which is a shame, as it looks great. Flashing LEDs are fine, but without complex programming they, well, just flash. A lot (like the Imahara version) change brightness but not colour. I'm sure he would have implemented that design differently if colour-change dot addressable LEDs had existed when he worked on the film!

Look forward to seeing your video. :)
 
I confess I'm curious to see how the colour wheel will compare to the more complex LED solutions like the "R-Series" offered over on Astromech. Those make use of RGB LEDs to pulse not only brightness but colour - but I suspect they're actually too random to really compare to a colour wheel.
 

Your message may be considered spam for the following reasons:

If you wish to reply despite these issues, check the box below before replying.
Be aware that malicious compliance may result in more severe penalties.
Back
Top