In 2017, we were required to make our own nosecone. To maintain consistency with the rest of the rocket, we decided to use a fiberglass/epoxy composite. To manufacture this, we needed a mold to hold the shape of the part while it cured; we decided on a female mold because it would make it easier to remove the part after curing.
 |
|---|
A first-attempt mold, laid up on a small nosecone. The photo shows the part before it was released from the master. |
The first attempt at a nosecone mold was to layup fiberglass on top of an existing nose cone. However, the mold tended to flex during the layup and cure because it was so thin.
| ??? |  |  | ??? |
|---|---|---|---|
The CAD model for the mold, with all the details. On the left is the nose cone section, and on the right is the coupler section. | The block of MDF being glued together before machining. | The mold being machined on a CNC router. | The mold after being coated with gelcoat, and sanding. On the right, an accessory tool for nosecone layups. |
The second attempt at a nosecone mold was to cut it out of a large block of MDF.
- We glued together 4 sheets of 1" MDF from Boulter Plywood with Titebond II; we intentionally made the bottom sheet slightly wider than the rest to create a ledge for handling and for fixturing to the router.
- We cut the mold on a CNC router. We screwed the bottom ledge to the router table, and then started by drilling several holes for alignment pins. Then, we cut slots to allow the molds halves to be pried apart easily. Then, we did a rough pass of the shape with a 0.5" straight router bit. Finally, we did a finish pass of the shape with a 0.5" round endmill, using the constant scallop feature of MasterCAM.
- We coated the mold in polyester gelcoat, and sanded down that surface until relatively smooth.
- We accidentally cut the coupler section of the mold too large, so we glued on a sheet of thick mylar to make the part smaller. The mylar was just glued on with 3M super 77, and released the part cleanly later on.
- We bolted the coupler section onto the nosecone section, and used wooden alignment pins to keep the mold parts correctly positioned.
The actual nose cone layup was very challenging and tended to create suboptimal parts; however, with enough layers, they were deemed sufficient.
- We cut out many rectangles of fiberglass, in several sizes. We probably cut approximately 30 each of 3"x4", 4"x8", and 10"x12", as well as 6 strips of 3"x48".
- We laid out the mold halves next to each other, mold side up and pointing the same way.
- We coated the mold halves in many, many layers of mold release.
- We placed a strip on the right side of each mold, and wetted it with epoxy.
- We placed a strip of wax paper on the left side of each mold, covering about 1-2" of the mold surface.
- We placed fiberglass rectangles on the mold, and squeegeed epoxy onto the fabric.
- We repeated step 2 until we had completed 5 layers near the tip, 7 layers near the base and 9 layers in the coupler.
- We folded over the glass on the left side of the mold to make space, and removed the wax paper.
- We put together the mold halves, taking care to insert the free fiberglass strip under the folded-over fiberglass so that they overlapped. A useful trick here was to use scrap aluminum as a shim, to keep the mold halves apart while we reached our hands into the parting line.
- We used the stick to fold back the folded-over fiberglass, such that the layers overlapped.
- We used the stick to apply 4 more strips of fabric on the inside of the nosecone; 2 on either side.
- We moved the nosecone to the floor, with the cone pointing downwards.
- We used two very large clamps to clamp together the coupler section, and some weights to push the rest of the mold together.
- We used two balloons to apply pressure to the composite as it cured; the balloons were inflated with the vacuum pump because we were all wearing respirators.
- We left the part to cure overnight, and then removed it.
Of course, there were several recurring issues arising in the parts:
- Large divots near the parting line: because the cavity was so large, it was very difficult to make the overlap neat, so some areas didn't overlap correctly.
- Stray tufts of fiberglass inside the nosecone: because we weren't able to reach the end of the nosecone, the tip of the nosecone was blocked off by stray tufts of fiberglass that weren't smoothed out.