It has been my goal to put together a top notch replica of the side mechs for my car. Regular readers of the blog will know that my pal Tim Neill in canada modelled the side mechs based on extensive research and reference, and they are absolutely amazing. Tim offers them for free to download on his Batberry blog. Here's a link directly to the files, should you be so inclined.
A friend printed these files for me, and I set about the task of cleaning them up. My goal is to produce them in resin and/or fiberglass. I have kicked around the idea of getting them plated, but honestly I think painting them will be good enough.
For reference, here's what they look like on a screen used Returns car:
The process of getting this right has been incredibly time consuming. Cleaning up the 3d printed parts is super labor intensive, and also intersects poorly with my anal-retentiveness when it comes to perfection. I have invested easily a couple hundred hours in puttying, sanding, primering... repeating until these things are at a point of quality that I am happy with. Cleaning 3d parts is not only an art, but it's also a tremendous hassle. The filament sands differently than putty and primer. It's very resilient. A real pain.
I finished cleaning up the part that is often referred to as "The donut" almost a year ago, and finally just got around to molding it. In this post, I will walk you through the process I used to mold. This will be a fairly technical discussion, so if you don't really care about materials and techniques, you can skip this. I found a great video by my pal Bill Doran that showed his technique for doing a brush on, jacketed mold, and this proved to be very valuable to my process. Here's a link if you'd like to check it out.
Prop: Shop - Molding & Casting 101: Brush on Molds for Helmets & Masks
I started by bulding a box for the donut, with about 1/2" of space around it. I used 1/4" foam core, which you can get at Michaels or any art supply store worth its salt. I cut slits into a strip of foam core at about every 1/4" so that I could shape it into a circle. Building a square or rectangular box for this project would have wasted a lot of rubber. I used Smooth-On's Rebound 25 for this mold, which is an expensive platinum cure silicone that is resilient and really takes the shape nicely. It's billed as a product that sticks to vertical surfaces. While true, it's not quite... accurate. More on that in a moment. I brushed on two coats of the stuff using a 1/2" chip brush from home depot. Oh, the base is made out of a scrap piece of MDF. I glued the donut to the MDF using super glue gel. No accelerant needed. Here's a pic of the donut with the first two coats of Rebound 25 on it.
The real takeaway from this photo is that the stuff isn't thickening up in all the right places. The horizontal surfaces are acquiring a nice, even and thick coat, but the vertical surfaces are not. Though thick, the rubber still tends to succumb to gravity, and only leaves behind the thinnest of layers on vertical surfaces.
And that's where a product named "Thi-Vex" comes in. It's what's known as a thixotropic agent, which basically means it thickens the stuff up while retaining all of the rubbers original properties. Ie, it doesn't effect the cure time, the pot live, the shelf life, etc. I mixed up another batch of rebound 25 and eyeballed an amount of ThiVex and went to town. You basically just add it in a little bit at a time until you get it to a consistency that you are happy with. I was going for a cake frosting level of thickness. I then applied it to the donut using a "senior" tongue depressor. Those are the big ones, mind you. Not like typical Popsicle sticks. So complicated!!! I tried to make it as smooth as possible, though I was definitely going for function over form. Here's a pic of the donut after one attack of thickened rebound.
I ended up doing another pass with thickened rebound, then a final pass of non thickened rebound. I then pulled off the box walls. I used a piece of copper tubing to bore holes in the outer lip of the mold. These will serve as keys for the rigid jacket, and hold the flexible part in place.
I brushed a ton of wax paste on to the MDF base, as it is very pourous and anything I put on top of it will probably attach to it pretty firmly. Now it's time to make the jacket. I chose Smooth-On's Plasti-Paste, which is also what Bill uses in his video. It's super easy to work with, fairly non-messy, and is really quite rigid even at thinner applications. I mixed up a batch and as with the rubber applied it with a tongue depressor. I did two coats, which was probably not necessary.
I left that to cure, and in a few hours I pulled it off the MDF base using a spatula and all was good in the world. Here's a pic of the underside of the piece.
At this point it occurred to me that a mistake had been made. In order to save time, I decided to forego making a seam down the middle, which would have allowed for easy removal of the jacket. The donut proved impossible to remove, so I ended up cutting a seam line down the middle of the jacket using a dremel tool and a big cutoff wheel. No damage to the mold or donut, but it added some complexity to the final steps, as I now have to manually align the two halves of the jacket and keep them stable while casting up a donut. Not a big deal, but if I had it all to do over again, I would have done the jacket as two separate parts.
Success was still within my grasp, and as a test I used Smooth-On's 65D to rotocast a donut in two pours. It came out EXCELLENT and with a little clean up it will be ready.
The other side mech pieces are also coming along. Here is the first pull out of the mold of one of the attachment pieces. I don't even know what the name of this piece is but who cares it looks great.
That's it for now. Thanks for reading. I'm really happy to be making progress on this build again. Maybe some day I will actually finish it.
No comments:
Post a Comment