By Christopher O'Connor
I've always loved R2D2 and though I considered tackling a cardboard build of the little astromech once... that idea didn't last long as the build wasn't looking too good. Now I have access to 3D printing and the very generous works of Mr Baddeley (look for his patreon if you want the files... you can get the files without joining... hence "very generous").
I had planned to keep this as a project to do "a little further down the track"... but for some reason I decided to do a "test print" of part of the dome and once that was done I kind of figured "well that's done... I might as well keep going".
I'm not exactly happy with the quality of my build... but I'm forging ahead. I'm working on the notion that I can always redo the dome another time... in so doing I would certainly endeavour to make sure all greebles etc fit well... as my current build stands it looks like some parts are kind of popping out a bit... I'm going with the "claim" that it's "battle damage"... not the most plausible... but what the hell.
On the positive side of things, I found a lens that looks just about perfect for my projector... well... except that it's about twice the size I was hoping for. I have ordered a possible replacement lens that is actually supposed to be just right for size... but I will have to see how it performs optically and may end up having to design a mount for the larger lens.
On the projector side of things... my first test of the ultrasonic mist maker was successful... I have since designed and printed an attachment to facilitate laminar flow of the mist to hopefully provide a decent screen to project on (though I may need to expand the length of it for the final build)... I haven't tested it yet but it's ready to go.
So I'm currently waiting for some mosfets, chips and other bits to go together to create the circuitry for the Logic lights and I've started printing the Upper Ring (part of the system that when combined with a lazy susan will provide the rotational movement of R2's dome.)
The starting place... if I can't get this right then there won't be any point doing the rest. Fortunately, following Mr Baddeley's plans... it's actually not too hard. The dome model is split in to two versions, one for large build plates and one for small. Due to my current build plate limitations I had to use the small version but it's designed in such a way that each part matches it's neighbour and bolt holes have been incorporated (as well as holes for small pieces of filament to be used to help alignment of the top (pie) portion. The larger portions took around 6 hours to print at the recommended .3 layer height and Mr Baddeley was quite right in that he stated he designed it to not loose too much detail at that layer height, but it does shave a lot of the print time. Having said that... a lot of sanding, filling, sanding, filling and so on was required to get a finish I was happy with (well at least willing to accept).
Painting is another issue. I ended up using an aluminium colour with Rub'n buff silver leaf on top... it turned out ok but part of me prefers and earlier version that used the same colours but due to having some of the black plastic exposed and a touch of the orangey/brown filler ended up looking kind of like brushed steel.
With the main part of the dome printed and prepped... it was time to move on to the panels and greebles :D
I have been told it's tough to sand PLA... but like many things you can't just be told that... you have to learn it for yourself first hand. Now I should clarify here... it's not that it's "hard" to sand... it's more that it's hard to get it to a point at which adding paint over the top won't show the layer lines etc.
I managed to get many pieces to a point where they felt quite smooth to touch (typically using sandpaper from around 120 to 240, 400, 1200 and even 2000)... but inevitably it would still show at least a faint hint of the layer lines beneath when paint was applied. I even went down the path of using spray filler... but found that it was not overly effective... it really is more for small... and I mean Small imperfections. Putty does do a good job but can end up costing quite a lot (ergo always buy the "economy" size tubs.).
To that end (ie I have a very tight budget) I thought I'd try some multi-purpose filler I had left over from a roof repair job I did last year. Now it is meant to be used on plaster... but I wondered if it could be used on 3D prints.
It turns out yes it can. Now I would add a caveat to that, namely that the finish seems a bit more grainy (though to be fair I haven't tried the wet sandpaper finish that I did on the dome on the pieces that I've tried so far... I may give that a go and see how it comes out). I also don't know if durability is likely to be the same... so they are issues to consider.
As of typing this I am planning on at least doing a test print of one of the panels at either .15 or maybe .2 layer height to see if that might improve the result sufficiently to swap all the pieces I've already done... well that and changing the print orientation... so I will probably be "back tracking" somewhat... but I do find myself getting very picky with the results on this build. Either way... sanding is something you just have to get used to... but at least the plaster intended filler sands very easily... I mean really really easily.... so I do suspect that it's too good to be true because it took a lot less time to get the pieces to a reasonable level of finish than it has with any other approach.
So while I wait to print/reprint some parts... I have started ordering some bits and pieces to give R2 life... that and my dad actually sent me an Arduino Uno, a servo motor and a few other bits and pieces... turns out the servo is the exact same one mentioned in Mr Baddeley's plans for use with the holoprojector so I should be able to test that mechanism once it's printed :D
I had ordered most of my bits and pieces from Aliexpress so it will take a while to get here (hence dad sending me the advance kit)... but I recently sold a review set of discs on ebay and figured I was entitled to use that money for more supplies... but this time I made sure it was from Australian suppliers.
So I should have a bunch of LED's, some prototype double sided copper circuit boards (they call them breadboards... are they technically?) I figure I may be able to use these for the LED arrays in some of R2's front displays.
I also have my ultrasonic mist maker and a little fan plus food storage container for my mist making machine... I'm still waiting on my mini MP4 player to arrive to take apart and see if I can convert it into a mini projector... the theory is I may be able to make a "mist screen" and project onto it... it's a big if and I've only seen other people mention the idea but haven't come across anyone who has pulled it off yet... so fingers crossed I can manage it.
Originally I was thinking of making a small LED array to provide the switching red/blue light on the front of R2... in mentioning it to my dad (and fellow Hackerspace members) it was suggested that a Neopixel ring might work well. Dad then promptly sent me one and some Arduino Sketches to see it in action. That was great but I found it a bit confusing commenting out parts of the sketch trying to get the results I wanted (or more correctly, trying to comment parts out to observe cause and effect). So a bit of google searching later I found a page that covered the basics of how to run Neopixels... a little bit of trial and error later and voila! I had not only the right colours activated but arguably looking a lot like that little colour changing part on the front of R2 (if memory serves someone said the original basically used a disc that would slide in front of the lights to swap it from red to blue).
The diffuser I used was simply a diffuser sheet from an old LCD panel I had lying around... turns out that does a great job too.
I'm now printing the body for R2 and have hit a slight snag in that a couple of my parts suffered a layer shift... there's a pretty reasonable chance this is due to me fiddling with the heatbed cabling, so after attempting to put it back the way it was, we shall see if I've fixed that... it only took a 5 hour plus print to notice the error.
Oh and I'm also taking the opportunity to print a direct drive upgrade for the Sunhokey Prusa i3 clone... when I got this one (again very gracious thank you for it!!!) part of the extruder portion was broken... what I should have done is printed the replacement part and fixed it... but being impatient I instead ziptied it.... which arguably has been working very well... but it does tend to let the fan/sensor droop a little in the back and that can be a bit too close to the print surface.
So we shall see if I end up installing the upgrade or, if the upgrade prints without issue the more likely future for the printer is it simply goes back into factory mode of churning out R2 parts :D
I am still considering getting my redesign of my DIY printer (currently Bitsabot) done too... but that might be a bit hard to bring online simply because of the cost of parts (mostly due to the requirement of a whole bunch of aluminium extrusion (probably 2040) for the frame... most other parts I should have ready to salvage from Bitsabot to make the new printer.