Today I took a crack at casting a prototype window sill corbel. Why a prototype? Because I've never ever ever formed up a small mold and attempted to cast a small object, and I'm confident that it'll take me at least 3 attempts before I get a result I can live with.
Backing up a hair, I'm going to address some of the more egregious issues on the house's front wall this year. The big ticket item will be the windows and while I'm confident in my ability to build the windows, replacing the precast sills and sill corbels will be a brand new experience.
That said, I've poured concrete, I've formed concrete, I've built plenty of things, and in theory, I know how the concrete casting process is supposed to work. But knowing how to do it and having actually done it are two very different things.
I started out by trying to figure out the dimensions of the curvy front of the existing corbels. They're so caked with paint and mortar that actually measuring anything was out of the question so I traced around all 4, took a pic of the tracing, dropped the photo into SketchUp, scaled it appropriately, drew the shape on top of the tracing, and reverse engineered the dimensions from that.
Building the form, aside from the curved portion was pretty simple, but the curvy portion took some work. Making a long story short, I grabbed a scrap 2x4, cut it into 4 short, manageable pieces, jointed them, planed them, roughed out the curved shape on all 4 on the bandsaw, glued the 4 blocks togethers, let the glue dry, spindle sanded the 4-block assembly to its final shape, and then cut the assembly to the appropriate height.
The rest of the form was just melamine-skinned particle board, quick and dirty. When the entire form was assembled, it looked like this:
Then I mixed up some concrete, using whatever seemed like the most appropriate concrete variety available at the big box stores. Normally I'd go to the local concrete supply place a couple miles down the road and get something based on advice from people who know what's what, but they're only open M-F and I didn't get around to buying materials until today (Saturday), after what's becoming a weekly Saturday morning excursion, taking Freckles to Tower Grove Park.
Anyhow...I mixed up concrete. I don't think I picked a great product for this application, it seems like I need something with smaller aggregate to get a good finish. I'll find out tomorrow when I strip the forms.
The rest of the job was straightforward: put concrete in the form, vibrate everything to remove voids and (hopefully) liquify the concrete a bit such that the sides will all be nice and smooth (vs. real aggregatey), screed it, then let it sit overnight.
I wasn't real concerned with tooling the top; 1, this is just a prototype and 2, no part of the top will never be visible. Today's exercise was primarily to run through the process of building the form and figuring out if my concrete mix will work.
I made a couple mistakes in the construction, and sorta painted myself into some corners when building the curved portion - very small pieces of wood take a lot of tools, like the table saw, planer, and jointer, out of the equation - but it was a good learning experience.
Tomorrow I'll strip the forms and see how she turned out. With any luck I'll only have a couple things to correct, but there's also a chance it'll turn out halfway decent. Either way, once I get the corbel casting figured out dealing with the sills ought to be pretty easy.