Composite Rudder Project
See Van's Airforce Forum discussion thread about the rudder project HERE
3/16/19 - Mold Prep Cont. & Rudder Skin Layups
Came back home this past week from school for spring break and made some good progress. I will be leaving in May to go back to Scaled Composites as an intern again this summer, so I really wanted to get my skin layups done so I didn't have to take the big mold back out to Mojave, CA with me. Here's the update on this week's progress:
I started the week out with another test piece since I have quite a few I'm planning on making, and to confirm my skills were up to the challenge of laying up and vacuum bagging the rudder skins, before I got around to using big pieces of material. This piece went smoothly and I worked out all the problems I had previously mentioned with my vacuum system.
I continued where I left off prepping the rudder skin mold. I had finished two cycles of sanding and resin wiping both sides of the mold (one mold on each side of the tooling block as previously talked about) over Christmas break. I continued by wet sanding in steps down to 800 grit. After that, the surface was thoroughly cleaned and I added joggle tapes to both the leading and trailing edges of the mold. After all of the substructure is bonded to the inside of one skin, the two skins will be bonded together with a blind closeout. Once the rudder is all assembled and bonded together, the excess skin will be cut down to the final dimensions of the rudder and the two skins will be additionally secured by using fiberglass "tapes" along the leading and trailing edges. For that reason, an equivalent thickness of fancy joggle tape (duct tape) layers were placed, topped with a release tape, in the respective locations so that the fiberglass tapes are recessed and conform to the intended outer surface of the part. The tip and bottom of the rudder will be cut off later to make fairings (much like the RV-8 rudder fairings) to accommodate a counterbalance and rudder cable attach bracket. This will allow me to apply the fiberglass tapes on the interior surfaces of these parts, hence no joggle tapes in these areas. After the joggle tapes, I applied and buffed four coats of wax until I was happy with the mold surface (shown by the third picture).
Next up, I shot PVA onto the mold surface, a liquid release agent that dries into a thin barrier film, separating the layup from the mold and allowing for easier release of the cured part. As suggested by Eric Stewart, I used a cheap Harbor Freight spray gun to apply a mist coat, followed by three heavier coats. Eric has given me a lot of advice regarding this project, and is a contributing author for Kitplanes magazine as well as the designer and builder of the SR-1 Project, a composite airplane designed to break speed records in its class.
I brought out my CAD model to confirm dimensions of the foam core, and then cut the core pieces from .25" Divinycell H45. I then aligned the edges of the foam on the edge of a table and beveled the core to a tape line.
Next came cutting out the rest of the materials. I used a sheet pressed down to conform to my mold, and traced an inch outside of the part, to make a template for cutting out my materials. I cut out appropriate amounts of vacuum bagging material, breather/bleeder, perf, peel ply, and carbon fiber at varying appropriate offsets to my template. The carbon plies were all cut at a 45 degree bias relative to the spar position. This allows the skin (primarily loaded in torsion) to most efficiently handle the air loads it will see in flight.
With everything ready and in order, I started the layup. First I mixed an appropriate amount of resin in a few cups, followed by a wet out of the mold surface and plies of carbon. My rudder skin design consists of a foam core sandwiched in between plies of carbon fiber, which greatly increases stiffness of the end part with little weight penalty. The core had a thin layer of dry-ish micro applied to fill in the porous surface of the foam, before being inserted into the layup.
Next went on the last ply of carbon, followed by peel ply, perf, breather/bleeder, and then the bag. Vacuum held nicely for both skin layups with no major leaks to fix. The first skin (left skin) that I did ended up with a couple small voids at the concave area near the bottom of the rudder horn, due to bridging of the vacuum bag over that area. On the right skin, I made sure to put some extra breather material in that area so that the bag had plenty of material to push down on in that area. This fixed the problem for the second skin, and the layup was sucked tightly against the mold surface.
My parents' basement sits at a cool 65 degrees, no good if you don't have a year to wait on resin to cure! I put a space heater under the table and made an enclosure out of cardboard to raise the temperature of the layup, greatly speeding up the cure time. I had to be careful not to get the layup too hot, or the resin may drop too much in viscosity, allowing too much of it to bleed out of the part and into the breather/bleeder. The enclosure was adjusted a bit, and in the end, kept the part evenly right around 90 degrees. I was pretty happy with this temperature, and in the end, both skins ended up with an expected resin content.
After I went through all of these processes for both rudder skins, I de-bagged and released the part from the mold. Both parts came out with no problem at all thanks to hours or sanding, waxing, and applying PVA to the molds. After the skins were taken off the mold, I rinsed the parts with warm water, which dissolves the thin layer of PVA that ends up on the outer surface of the carbon fiber part.
The finish on the outside of the skins was good, but not perfect. I learned that every tiny imperfection on the mold surface really does show up on the part. I got a little too excited to see the nice and glossy carbon fiber finish and opted not to put a layer of peel ply on the outer surface of the skin. Looking back, I would do this differently if I make additional rudder(s) and I would put a layer of peel ply on the outer surface. No, the cool glossy finish won't be waiting for me when I take the part off of the mold, but in the end it really is a better way to go. The cool and sleek carbon fiber surface will have to be painfully painted over for heat (resin Tg) considerations anyway, and the peel ply would make for a much easier process when applying the fiberglass tapes into the joggles and then painting. A layer of peel ply would also aid in hiding any of the small surface imperfections on the mold.
Overall, I was really happy with how the skins turned out and I'm sure they'll make a great end product. I learned a lot just to get to this point in my composite rudder project from the initial design work, to CNC programming and cutting for the mold, and applying my knowledge of composites to make structurally efficient parts and fabricate them myself. I still have a lot more to learn as I test structural samples, finalize the interior structure design, and fabricate the spar and ribs before the final closeout and finishing processes. Followed eventually by flight test. More progress will have to wait until May, after I finish out this school year at Purdue and head back out to Mojave.
1/9/19 - Mold Prep and Structural Samples
I've been slowly chipping away at some work on this project between school and break and whatever other time I can get. I had some vacuum bagging equipment at school with me so I could lay up some carbon fiber C-channel pieces. My goal with these is to experiment with some different fabrication methods and learn how to best construct carbon fiber parts for my rudder. The goal is to be as light and strong as possible. I will eventually be breaking these pieces to evaluate their strength to weight ratios, which will help me choose construction methods for parts, such as the spar, for my rudder.
I 3D printed some simple plastic forms to vacuum bag the carbon fiber to. The idea behind vacuum bagging is that it pulls the plies of carbon fiber tight against each other, while also helping to achieve an optimum fiber to resin ratio. In this first picture you can see the small 3D printed mold, which was covered with a release tape that will help the part break free from the mold. I also used some mold release wax to help with that too. Before the layup is started, it's important to have all your materials cut out and ready. Once the resin and hardener are mixed, there is a certain amount of time available to get the process completed - though not quite as important on such a small part like this. This first part was more of a practice run, and was a simple layup of 4 carbon fiber bidirectional plies, cut on a bias at a 45 degree angle. Carbon fiber is considered an anisotropic material, which means it has different mechanical properties with different fiber orientations. This is opposed to materials like metal, where essentially the same strength is observed no matter where you push or pull from. These fibers are cut to run at a 45 degree angle in respect to the beam's longitudinal axis to increase the amount of shear strength in the part.. whole textbooks are written on the complex topic of fiber orientation and the resulting mechanical strength properties.
The first picture below shows the vacuum bagging setup. Not shown is the vacuum pump. The yellow line is a sealant tape the provides a leak proof seal between the clear bag material and the table. The white fabric seen is called a breather/bleeder ply, which absorbs excess resin pulled out of the part by the vacuum pressure. As you can see, quite a bit of resin is saturating this ply, especially on the left side. Ideally, the bleeder ply should have a bunch of dots like it does near the middle of the part. between the bleeder ply and the part is a perforated film which prevents the bleeder from getting stuck to the part, but allowing resin to pass through. I had issues with the vacuum relief valve on the setup, which resulted in too much vacuum pressure being applied, hence the saturated bleeder ply. The bleeder ply was also not as dense as would be appropriate for a 4 ply layup. This was confirmed by the weight of the final part, which was significantly lighter than what I had calculated. This means that there wasn't enough resin in the final part, which will likely decrease the strength of the part. The part was then trimmed to its final shape which will eventually be tested. I plan on fabricating a variety of test samples using different forms of carbon fiber such as pultrusions, unidirectional fabric, and carbon tow in the top and bottom caps of the beam.
The Mold:
Over break I made some progress prepping the mold for the rudder skins. I designed the outer shape of the rudder first since the dimensions were fairly constrained. It needed to be a drop-in replacement for the current rudder, so the shape of the trailing edge was the primary change. Here are a couple screen shots of the computer model.
After I designed the rudder skin shape I made a computer model of the mold. I was lucky enough to get a hold of scrap pieces of high density tooling foam that were big enough to adhere together to be CNC milled to the shape of the mold. Since this was all I had to work with, one side of the foam was cut for the left skin and the opposite side of the foam was cut for the right skin. This ended up working well, the only downside is I can only work with one side at a time.
The large piece was scrap material due to quite a few deep cuts on one side of the foam. This was easily taken care of by making some filler from resin and the white material seen below to form a mixture called "micro". The white stuff is a bunch of very tiny glass bubbles called microspheres, which are finer than grains of sand. This thickens the epoxy to about the consistency of peanut butter, which can by pushed into the slits and then be sanded flat.
All of the slits and divots were filled and the mold should work just fine. The mold also had some imperfections and ridges from machining, along with the surface being porous. Each side was sanded and then I applied a thin coat of resin. Then it was sanded again with a finer grain, followed by another coat of resin. Hopefully I will be able to sand this to a very smooth finish followed by a coat of wax and a thin film of a release agent called PVA before finally being ready for the vacuum bagged carbon fiber layup.
9/23/18 - Status Update / Composite Rudder Project
It's been a while since I've had a chance to work on airplane stuff. School has been keeping me relatively busy, but hopefully I'll get a chance to go home for a weekend soon and give a good start to the horizontal stabilizer..
I have also been working on a custom carbon fiber rudder for my airplane that I have high hopes for. It has some influence from the Game Composites GB1 Gamebird - quite an eye-catching aircraft. I think the swept rudder design will give a whole new look to the side profile of my plane. It will definitely be different and stand out from the rest. I plan on posting some of my progress with that project on here as well. As of right now, the profile of the rudder has been designed, and molds for the skins have been CNC'd. I have also done some preliminary internal structure calculations and design. I plan on fabricating and breaking some test ribs/spars and refining the structural design from there. The rib molds will be 3D printed, and I've got nearly all the supplies I need to build the entire rudder. Not a cheap project, but I see it as an investment in my education as well. I plan on programming, designing, and building my own composite post-curing oven so I end up with a quality part. The more I get into the project, the more I realize there is to learn and do. Not quite as simple as I first imagined..
Below is a sneak peek of the skin mold. One skin mold on either side of the tooling foam. The mold is made from high density tooling foam and I'm very happy with how it came out. It has yet to be prepped before the first layups.
This was a rough preliminary sketch, but the tail of my plane will turn out to look a bit more like this, as opposed to the rudder on the plane at the top of the project status page.
