Hereís a bit of a mixed bag with this update. First up I wanted to show some images from the studioís workbench to illustrate what Iím doing to improve and streamline my mould making process. Iím sure thereís a few of you whoíve been reading my recent articles outlining my progress working with Servitor Solus and wondering Ďwhat the heck is taking so long to get casting?!í and I wanted to take a moment to show what Iíve been doing thatís delated me a bit. This is an effort to reduce the labour of producing casting moulds to improve their production in the future, while also aiming for a top quality outcome that will ensure the best possible casts for the customer.
Iíve mentioned it a few times over the years, but Iíll reiterate, I hate
Ďsplit mouldsí which are very common in casting resin models; with a split mould you place the object in the mould box and pour one complete block of rubber to completely encase the object, and then cut into the rubber down to the master object to free it from the mould rubber. While this is a method that saves labour during the mold making process it comes with problems that I simply despise. Not only can you damage the casting master as you cut it free, if the object is small and/or thin it can be very difficult if not impossible to cut the item free in a clean manner; even if you can cut the item free it generally creates an uneven mould line that you have very little control over. As such, they tend to Ďslipí rather easily producing really obvious mould lines at best and horrible detail ruining mould slips at worse. Ever see a really nice resin cast model with a big ugly mould line/slip in a rather odd/obvious location? You can thank quickly made split moulds used/run as quickly as possible for that.
My studio will Never
use split moulds. By producing a two-part mould extra time and labour is
needed but you can have complete control over where the mould line goes and you can produce a mould that literally resists mould slipping and therefore creates almost invisible mould lines almost every time. Done right, this can also help the mould open and close making it easier to extract the components without damaging the mould and/or badly warping the part. Given the cost of RTV rubber, putting some extra labour into making a mould that can be in use for years to produce dozens of copies of an object seems like a shrewd investment; since it also means that it noticeably improves the quality of the components produced, this is simply a no-brainer choice for the studio to standby.
A quick first point, the RTV rubber is not
binding to the surface of the 3D prints made by Solus as I make the moulds. This small detail is a huge
positive for the studio. Past 3D prints made with the PolyJet process (what Shapeways uses) creates components that have a porous surface that needs to be properly sealed before making moulds, or the rubber binds with the object causing all sorts of problems. Servitor Solus makes parts with such a refined and smooth surfaces this issue simply doesnít happen; to say this is a good outcome would be a huge understatement. Itís not something most people might even concern themselves with, but know that Iím the kind of perfectionist that does it on your behalf.
Encouraged by the accuracy that Solus has been achieving Iíve been working out the tolerances to produce Ďinsertsí that fit into the negative space of parts during the mould making process.
If an object youíre casting has a nice flat back itís a simple process of laying it on the casting clay, adding the vents/gates, and pouring RTV rubber over the item; once the first half is cured flip the mould, remove the casting clay, and pour the second half. Now, if the object has all-around details with no obvious flat side and/or has obvious overhangs and/or holes that will lock the item in the rubber, you need to find some way to back-fill them to block the rubber in the first half of the mould. Up to this point I did this by hand using the casting clay, sharp blades, and sculpting tools to fill in these kinds of locations. This is where labour in creating a two-part mould can add up, and every time the mould wears out the process needs to be repeated. Iíll do it if I have to, but I wanted to find a better way.
With Servitor Solus completely at my disposal, opposed to outsourcing my 3D printing, I was able to do the trial-and-error necessary to get the tolerances as tight as I could manage to create standardized inserts for the components that will benefit from them. The example above is quite simple, but even with more complex objects creating a seat to occupy the negative area is reasonably straightforward in Solidworks. So, in most cases where itís needed I should be able to create an insert for a component to simplify the process; place the object on the insert to make one side flat, place them on the clay, add the vents/gates, pour the rubber, cure the first half and flip, remove the clay, pull out the inserts, and pour the second half of the mould. Also, note how doing it this way will have the mould line follow the inside corner/edge perfectly making it really easy to clean up.
The last hurdle Iím trying to figure out with this process is how to better seal the paper-thin gap between the insert and the component so it resists the RTV rubber from seeping in between the two objects. If itís not sealed the pressure curing process I use on my moulds creates a film of rubber that needs to be cleaned up before pouring the second half of the mould. Not a huge deal but it would be nice to avoid it to further streamline the process. Iím researching if thereís a readymade product that can do the job but it might just be as simple as adding a bit of petroleum gel in the gap and cleaning the edge. So, a few more tests are still ongoing to see if I can solve this, but the overall idea of making precision fit inserts in general seems to be viable thanks to how well Solus works.
All that said, the first moulds for the Pintle Weapon Kit (Certamen Mk.1, 2, 3) will be starting to finish over the next few days and the first casts and test assemblies will follow shortly after; I canít wait
to see the parts in grey resin. With limited equipment right now mould making is a bit of a catch-22, since I need to use my casting chambers to create my moulds the process unfortunately stops me from being able to cast, and vice-versa. Naturally, more chambers are planned, but for now itís an annoying reality.
Ok, with all of that word salad dished out, letís have last look (for now) at the final successful iteration of the Dragoon/Ironstrider bits to see how they turned out, shall we?
Since the red color tends to mute the contrast in the parts, hereís a screenshot of the last bits I did for the Dragoon/Ironstrider in Solidworks.
The first print I did of the Phosphor Serpenta arm was good but it didnít quite Ďfeelí right to me and seemed a bit heavy; after it was commented that it seemed a bit off balance I figured it wouldnít hurt to tweak the 3D model a bit. In the above image the back end of the weapon has been slimmed down a little to remove a bit of bulk while still keeping the same form, opposed to the Serpenta in the photograph below. Itís a subtle change but I think it suits the arm better now. These bits were a bit of a distraction within a distraction, but Iím so pleased with how all the components turned out Iím glad I took the time to make the extra bits. The components only change, what, 5-10% of the model? But it really does give the model a distinct look while not diverging too much.
After the second iteration that missed the mark it dawned on me what to do to quickly zero in on the fit I was trying to achieve.
The first parts were done Ďblindí by taking measurements of the existing components and model which will work to get the general shape but will struggle to get a really exact match. Once a printed component can be placed on the model there is something to provide solid reference points to work with. I simply sketched the shapes of the model details I was trying to conform to (the round cap of the hip and the oval area of the pipe connections) how they appeared out of alignment, then lock the sketch so it couldnít move, and altered the 3D model to conform to the sketchÖ And Iíll be damned if it didnít work as well as I could have hoped. Again, itís a little technique thatís actually a large insight that will be very useful for many future projects where Iíll need to zero in on the fit of a curved and/or complex component. It was good to figure it out on something small like this so it can save me time and materials on larger projects. *Subtle glances over at his half assembled Knight and mutters, ďSoon.Ē Under his breath*
Next up, I need to get the other two Dragoons cleaned up and assembled, so this project will go dark for a bit before returning once Iíve got the group closer to being ready for primer. Naturally, if I do anything else of note for the project along the way Iíll try to remember to document the process. Iím trying hard to force myself to photograph what Iím doing more often, as I keep finishing things and thinking I should have documented the process.
Thanks as always for reading and following along. Ok, now Iím all worded out.
*Subtle shuffles off to find something to jam into his food hole*