So a couple of days ago, after two and a half months longer than originally anticipated, six months to the day since the original post here, the battle rifle master was complete! The master is the finished original build that doesn’t get used, but gets moulded so that the final strong, lightweight plastic resin versions can be cast from the moulds.
The delay was simply a lack of experience on my part- it turns out that if you want to finish something to a professionally high standard, all the layers of priming, sanding from 220 grit on up to 2000 grit and clearcoat applications take just as long as the build itself. Of course, I was entirely unaware of that going in.
If you’re interested in the details of that finishing work, I produced a video guide on it while working:
That was, admittedly, a not-insignificant portion of the delay in finishing, since 30-minute videos are non-trivial to put together even when conceptually simple, but since I hadn’t really seen any information on it prior to starting I thought it was important to share at least my initial experience with others in that position. My techniques may change over time, I don’t know. If they do I’ll probably do another updated video.
I think that’s about it, so let’s get to the photos!
This took about three weeks and innumerable individual actions and solutions, so I’m going to keep it fairly abridged and streamlined.
I now have a picdump thread up on the 405th where almost all my pics will be going, and you can see these at a little higher resolution. The disadvantage there is that the embeds aren’t 600px wide like on WordPress, so you have to click through the lightboxes to see clearly.
Here, I’m going to stick to a general overview with a handful of more interesting events. Let’s get started!
First I did the cylindrical parts because they were the easiest:
The anchor at the front of the frame snaps into this part very satisfyingly. I made it so that the finished piece can be moulded and shipped in pieces which should theoretically glue together nicely with minimal need for careful fit and alignment.
Yeah, more of a grip slash trigger assembly, but I’m labelling the build more by traditional rifle geography than by actual mechanical purpose.
I’m also not doing a step-by-step, day by day build. I more or less did that on the make-it-up-as-I-went-along Scifi Pistol build, and ultimately other than some new toys, the process is largely the same. Here I just have dimensional plans to stick to.
So with the frame done, I worked on the trigger assembly section first, because I was procrastinating on the grip, which looked horrible in terms of symmetrical compound curves to carve freehand.
This is where we left off in the last post…
And now I can fast forward through the boring stick-cut-sand-stick-cut-sand process, with the magic of technology:
There isn’t a whole lot to tell, it’s fairly self explanatory. Having a set of various size files and rasps is invaluable, though the winner at everything ever for all time is the table sander. Not even necessarily the belt sander part on top, since that doesn’t have much of a precision guide on it and results aren’t a whole lot better than a hand held belt sander. Just the disc sander. 1/3HP, 6″, 80 grit. Magical! If you’re yet to splash the cash on a bench sander, I’d recommend ignoring the combos and just going for a separate pair of disc sanders, one small and low powered like mine, and then a bigger 10-12″, 1-1.5HP one.
Oh, and a band saw would be crazy useful, but I don’t have one… yet. 😀
Anyway, the single-sided contact cemented paper was peeled off and I started creating some semblance of an assembly, and it looked rather nice:
At this point I couldn’t really ignore the grip any more, so I started wrapping my head around its geometry. I could scratch the centre line in around it with the calipers, and then from there I added spot measurements and joined them together with the pen.
The first part to do was the heel of the grip, where it has an indent for the heel of your hand. It looks bullet-shaped on the bottom, and then it has a fairly established simple curvature on the face, and then up the back it ‘s convex as it gets ready to morph into the wider grip curves.
Like the rest of this process, it was a combination of careful bench sanding, resin bonded sandpaper and filing. Because the forms were already drawn on as guides, symmetry only really came into it at the very end to finish up.
I added the side scales to bring it out to around 26mm thick. That’s about 4mm short of its actual thickness, but having to use imperial sized materials whilst working in metric occasionally brings some limitations… And I don’t think being a little easier to handle is a bad thing so I’m not rushing out to find 2mm craft foam!
The back side of the grip I did all in one single go, so there are no progress pics there. I think, unless a model is huge, it’s best to do organic shapes in a single session… Kind of a “mindset” thing to do with the visualisation, it’s hard to describe.
The end result was fairly promising, next to the on-screen model. The right angle gap at the heel will be filled with Bondo to form the curve rather than cutting the scales oversized and sculpting them down. That should give me a more defined edge at the bottom, and let me blend it better at the top.
I considered chopping the bottom off and creating a DIY-Pepakura replacement using UV unwraps, but that didn’t work so well.
Next up was the “rear grip”, the section behind the thumb which blends into the stock and meets the mag well.
I like to notate my templates so I don’t have to keep going back upstairs and referencing each part while I’m building.
It came out quite nicely, considering I shaped it entirely separately from the grip without any checking. Only very minor filling should be needed:
Then, once it was all put together, the grip fell off! The contact cement hadn’t cured for some reason and the superglue hadn’t seeped in to grip as much as I’d like- I must need less viscous superglue.
This wasn’t a bad thing though, since I’d previously decided I wanted to remove the grip in order to get at the trigger mechanism area that I’ll have to cut into it, in order to install a switch and spring and all that good stuff. common sense would have indicated that I’d have cut the recess out of the central slab before attaching the scales, but live and learn.
A couple of additions and details and cleanups here and there, and it’s finally ready for Bondo and sanding.
And, adding it back into the original structure gives us this:
Next up, the upper reciever. That’s what I’m calling the middle black bit below the carry rail, encompassing the barrel cooling area and all that good stuff.
Not having easy access to a large format printer = Tetris.
Another brief update. I’m trying to make these updates not end up as monstrous essays like I tend toward. Anyway, the skeleton is done! It took 3 days, though that probably only equates to around 10-15 hours of constant work, mostly at the weekend.
I haven’t drilled any more lightening holes, since I remembered that some of the parts are actually on external display and don’t want to have to do a load of filling later.
I haven’t gone out and got the DAP Weldwood, which I’m using instead of Barge for at least this project because it’s more convenient to get hold of and I’m not doing much, if anything, with the EVA foam.
So I’m using what’s around already, and that turns out to be Loctite’s version of Super 77 aerosol contact cement, which works nicely on flat planes, and some Loctite gel super glue which I’m reinforcing smaller areas of the super 77 with around the seams. The wife claims not to be a great fan of Loctite, but she bought ’em. *shrugs* They’re working well on fibrous materials so far.
Next up is the grip, trigger and lower receiver area (is it still an LR on a bullpup?), which is relatively straightforward and should ease me into the rest of the body fairly nicely. As the skeleton might imply, I’m building it up from the inside out, starting with the deeper, black (blued) steel parts and finishing with the grey painted body shell plates.
The middle slice of the grip is also done, since that was just a plain 1/2″ slab. It had to be modified slightly to fit around the LR section of skeleton, which it’ll probably directly attach to. If I do alternate grip sizes, I’m planning on solid casting them and then grinding down from there.
Note: If you’re going to sand a lot, particularly MDF and polystyrene but probably anything really, invest in a shopvac. Even a small one will make your life much easier and avoid killing your regular vacuum cleaner!
Once I had my blueprints more or less put together, I needed to figure out the scaling (hence the use of vectors and not just tracing).
After a full two hours of research on the lengths that other people use for their props, the appropriate scaling to move from SPARTAN (6′ 10″, plus 4″ in armour) to human (let’s say 5′ 10″, I’m somewhere around there), the lengths of real life battle rifles like the G36 and M16, use of BR55HBs by humans in-game, etc… I decided not to bother scaling at all. 991mm for a scoped medium-range service rifle doesn’t seem to be unusual at all, and the pictures of scaled versions always seem a little small to me. Of course, the asset itself is variably scaled in-game and in cutscenes, depending on scenario and user, so some research materials may dictate scaling for some people.
That’s the theory. What about practice? My only concern in practice was getting my hand around the grip. It seems rather fat compared to real-life assault rifles, which seem to have skinny, straight-box grips. The only way to be sure is to print it at actual size and try it!
My concerns weren’t without merit. While I could get my hand around the grip, and even relax enough to wrap the imagined full 3cm width of the 3D grip, it wasn’t exactly what I’d call comfortable. So here I have my first decision. Do I scale the gun? Do I just scale the grip, risking it looking disproportionate? Or do I leave it as-is, and relegate it primarily to “showpiece” status?
Since I can actually get my hand around it to begin with, I’m inclined to go with the latter solution- do nothing. Given that the project is fairly complex to begin with, I don’t want to add complexity with partial scaling as well as the risk of the result looking strange. It’s not intended to hike around a con for hours anyway; probably just a photoshoot or two and spend most of its time on static display.
So it’s settled… I’ll just build it as-is, full-size. If I cast it (in multiple parts), I may break off at the grip to allow smaller ones to be used for comfort.
Now, onto the chassis. I’m not building the gun like a typical prop is made; starting with a full outline in ~1/2″ MDF and build detail outwards and inwards from there. Due to the fact that the shape of the gun, with its long chunky carry-handle and complex barrel cooling geometry, is more like two separate barrels and it doesn’t have a consistent “center thickness”, I want to build a skeleton frame in various thicknesses of MDF, appropriate to the location on the gun. I’m typically putting internal seam lines at the same point as external ones, so that the plates being added onto the outside are getting glued to a single flat surface.
This overall process is a little more time-consuming in the blueprinting, but should make actually designing and attaching parts relatively straightforward, and also allow me to have parts mounted at various angles- not just laterally. I don’t know if this method is used in the industry, but it makes sense to me and the way I think in 3D space.
Some things have to be taken into account when designing the skeleton, like leaving space for cylindrical sections, which will likely be some form of tubing (or some Delrin bar I have lying around, if it’s the right diameter… I don’t yet have a lathe) that I can stick straight in place, and leaving space for the magazine to fit. This isn’t visible in the external model, but I’ve separated out many of the parts of the OBJ and I’ll just use its vertices’ coordinates to work out the spar for the top of the stock.
This is what I’ve come up with, it’s just a case of transferring to appropriate stock and putting it all together.It’ll need some care in handling until it gets more built up, of course. I’m going to clear off a dedicated assembly area to avoid moving it too much.
I identified the innermost vertices with reasonably consistent distances from the centre plane as each other (+/-0.5mm, say) and then took those sections from the vector blueprint. There were several around 6.5mm out, 11mm and 14.5mm were also popular. I just took the parts that were the least visible from the outside, and then extended those parts some way around the visible area to make up an internal structure.
The resulting skeleton should allow me to pick and choose between traditional rib-and-spar model construction and current prop replica sheet-sandwich construction as appropriate, over the whole body. This should allow some use of alternate materials and fillers that displace the heavy MDF, like polystyrene (EPS/insulation foam), which I found a couple sheets of yesterday, or expanding foam. I can also drill lightening holes along the beams without sacrificing too much rigidity.
Obviously, it also significantly reduces material usage, helpful on a limited budget, and perhaps more importantly, it eliminates all that time cutting around tricky silhouette details. Lots of straight cuts and a bit of sanding and gluing. Should be a snap.