eJar0k
Member
Hey y'all.
I have a couple of wild projects lined up for this year, including a spicy flight at BALLS that I'm working up to over the next few months. I'd like to have basically every part on these builds be custom composites, and, since I'll be cracking Mach 3 on some flights, I'd like to seriously improve my manufacturing skills. Supersonic re-kitting and remote fence pole deployment are... not fun... so I want input from the composites gurus here.
I already have solid experience working with braided carbon and FG sleeves from Soller Composites, but getting a good surface finish, consistent outer mold line, and controlled thickness with them can be challenging on nose cones. Rocketry might be 80% sanding, but I'm naïve enough to think I can reduce it with a traditional split mold. After doing some research, including looking at @watheyak's excellent nose cone molding thread/L record build and a lot of Easy Composites videos, I think I have a decent plan for producing good composite tooling and nose cones without a CNC mill or lathe. Here's my thinking:
First, limits. No CNC mill, no CNC lathe, no turning abrasives on a manual lathe. I don't have my own lathe, and any lathes I could use are off-limits for FG/carbon. I could order parts from services like PCBWay/JLCPCB, but that's pushing my next limit: I want to keep the cost per mold reasonably low so I can iterate quickly and cringe less at the motors I'll be buying. Next, I'd like to avoid prepreg if possible, both due to cost/MOQ and not having a temp-controlled cure oven. If I can make this work with sleeve+wet layup, the technique gets much easier and more accessible. Vacuum bagging is a maybe, but that adds to the equipment cost and consumables; this is a budget and accessibility question. Lastly, I want to leverage FDM 3D printing as much as possible to get decent dimensional accuracy and quick iteration. This means oven-curing the mold itself might be impossible.
Next, requirements. The nose outer mold line, wall thickness, and shoulder ID need to be well-constrained, and the surface finish should be as smooth as possible coming out of the mold. My builds will be mostly non-standard diameters/thicknesses (submin, i.e. matching case diameter) and going very fast, so these parts need to mate well to my "airframes". I want minimal post-processing and no lathe ops. Next, the parts have to be strong... duh. My part's resin should be able to tank aero heating for Mach 3 flights, and the fiber reinforcement should be optimized for flight loads (thinking about recent nose failures at BALLS). Lastly, I want to be able to apply this to a number of sizes and profiles, e.g. 38mm-98mm, VK, conical, 4:1-7:1, integrated straight sections.
With that said, here is my first draft. Brace yourself for some MS Paint, because I don't feel like opening SolidWorks yet.
Tooling mold cross section:
1. [Red/Pink] An exact profile of the nose half is printed in PLA/ABS/ASA with mirrored indexing features and sanded to mostly knock down layer lines. Print axis is perpendicular to the split plane, i.e. up in the drawing.
2. [Light grey] The surface is coated with a thin layer of laminating resin or a similarly thin, tough epoxy, left to gel, and recoated once or twice again. This is then sanded nearly back to the print , going up to around 1000 grit. Next, the surface is polished with a cutting compound, something like NW1. I have to take care not to round out the inner corners or indexing features here. A few coats of mold release are then applied, either something like Partall Paste #2/Hi-Temp Wax or Frekote 700NC.
3. [Dark grey/Black] A layer of gelcoat, probably Proline 4500 (used in its intended purpose!), is applied to the whole surface. Carbon tows are wet out with gelcoat and packed into the mold corners for reinforcement. After the first coat gels, the surface is recoated a few times, making sure large radii are built up on the indexing features and corners. I'm not sure whether to apply the tows in the first or subsequent coats here.
4. [Light blue] Once the gelcoat.. gels.. a layer of fiberglass or carbon is laid up onto the mold. I think I would first apply a layer or two of light glass, 1-2oz, then several layers of heavier glass, 6-12oz. 6oz or heavier carbon is also an option for better stiffness here. After gelling, the edges are trimmed square.
5. The tool is demolded and cleaned up as needed, and the process is repeated to make the other half.
Done well, I should end up with two mold halves that fit together perfectly.
Next, the plug. This is far simpler and takes directly from my prior experience.
1. [Red/Pink] An exact profile of the nose interior is printed, i.e. the nose profile minus the intended wall thickness. A hole runs down the center for a threaded rod, dark grey.
2. [Light grey] The surface is coated with laminating resin, sanded, polished, and release coated identically to the tooling mold. This also could be gelcoated, but I'm not sure.
Next, layup. One big issue with braided sleeve is that it has a limited range of diameters it can expand/contract to, meaning that a nose's base and tip diameters can't be too dissimilar. You can coax the fibers beyond the listed minimum diameter by aligning them more towards the nose tip, but this dramatically increases its thickness. If you want more consistent thickness, you have to manufacture a pretty huge nose tip, which can get very long and heavy for high LD profiles. My solution is to use multiple sizes of sleeve and have them overlap in alternating layers, giving a staggered concentric seam. Once compacted in the mold, the seams should flatten out and give a clean, strong transition between the two sleeve diameters. The challenge here is not having big gaps filled by resin in the joint. More MS Paint!
The hardest part will probably laying up the seam and minimizing gaps between each pair of sleeves without bunching up fibers. Inserting the plug into the mold without distorting the weave will be the next hardest step. Once laid up and inserted into the mold halves, I have a few options:
1. Curing under no compression. Probably the worst option, but also the easiest. Take a mold half, drop in the plug, attach the other half, secure together, then jam the plug down to consolidate a little and let cure.
2. Curing under bolted compression. Not much harder, but could be messed up. Assemble as usual, but then use the threaded rod running down the plug to compress the plug against the mold axially. By extending the rod out the top end of the mold and tightening a nut down, the layup gets some mechanical compression. This requires the mold haves to be bolted together or similar.
3. Curing under vacuum. More equipment, more consumables, but probably the best result. Lay up, assemble, bolt halves together, run breather to each end, bag, suck. However, I have no idea how a printed plug with partial infill would handle under vacuum, given the trapped air inside.
Once cured, I demold, trim, clean up, and post-cure if needed. Overall, I think I could make a full mold set in about a week, then pump out lots of consistent tip-less noses from them. I'm left with lots of questions before starting, however, and I want to save myself a few botched attempts.
- Does the workflow make sense overall to those who have made split molds before?
- Has anyone tried doing similar parts with sleeve?
- What mold releases are best? I'm leaning towards Frekote 700NC, but I have had good results with Partall Paste #2 before.
- Carbon tow reinforcement in the first or subsequent layers of gel coat? Are they even needed with something like Proline?
- Polishing compound recommendations for epoxy? NW1 is just what I've seen in Easy Composites videos.
- What reinforcement should the mold get? Glass, carbon? How many layers? What weights? Should I use lighter weights to conform to the surface, then build with heavier weights? Can a single weight suffice?
- How do I design the indexing features? My current thought is a frustum with a pretty sharp draft angle (30-45 deg?), rounded edges, and a match-drilled bolt hole.
- Can this be vacuum bagged with a printed plug? Any good references for bagging nose molds?
- Should the plug be coated with gelcoat or laminating resin before sanding/polishing/release coating?
- Is the printed plug a good idea, or are there better methods? Prepreg + tubular vac bag is less labor intensive, but that obviously defeats the point of using sleeve. Making a composite plug, meanwhile, would double the amount of molds I need per part.
- Any resin picks for extreme flights? I'm leaning towards HTR-212, which I've had good experiences with but haven't pushed too fast yet. Cotronics stuff is the next pick, but pricey.
- Should I stop procrastinating on a filament winder instead?
If you made it to the end of this and have input to give, thanks! I would also appreciate links to any RC forums or similar, since I hear their parts are on a whole different level. Basically, sanity check me before I waste my time.
I have a couple of wild projects lined up for this year, including a spicy flight at BALLS that I'm working up to over the next few months. I'd like to have basically every part on these builds be custom composites, and, since I'll be cracking Mach 3 on some flights, I'd like to seriously improve my manufacturing skills. Supersonic re-kitting and remote fence pole deployment are... not fun... so I want input from the composites gurus here.
I already have solid experience working with braided carbon and FG sleeves from Soller Composites, but getting a good surface finish, consistent outer mold line, and controlled thickness with them can be challenging on nose cones. Rocketry might be 80% sanding, but I'm naïve enough to think I can reduce it with a traditional split mold. After doing some research, including looking at @watheyak's excellent nose cone molding thread/L record build and a lot of Easy Composites videos, I think I have a decent plan for producing good composite tooling and nose cones without a CNC mill or lathe. Here's my thinking:
First, limits. No CNC mill, no CNC lathe, no turning abrasives on a manual lathe. I don't have my own lathe, and any lathes I could use are off-limits for FG/carbon. I could order parts from services like PCBWay/JLCPCB, but that's pushing my next limit: I want to keep the cost per mold reasonably low so I can iterate quickly and cringe less at the motors I'll be buying. Next, I'd like to avoid prepreg if possible, both due to cost/MOQ and not having a temp-controlled cure oven. If I can make this work with sleeve+wet layup, the technique gets much easier and more accessible. Vacuum bagging is a maybe, but that adds to the equipment cost and consumables; this is a budget and accessibility question. Lastly, I want to leverage FDM 3D printing as much as possible to get decent dimensional accuracy and quick iteration. This means oven-curing the mold itself might be impossible.
Next, requirements. The nose outer mold line, wall thickness, and shoulder ID need to be well-constrained, and the surface finish should be as smooth as possible coming out of the mold. My builds will be mostly non-standard diameters/thicknesses (submin, i.e. matching case diameter) and going very fast, so these parts need to mate well to my "airframes". I want minimal post-processing and no lathe ops. Next, the parts have to be strong... duh. My part's resin should be able to tank aero heating for Mach 3 flights, and the fiber reinforcement should be optimized for flight loads (thinking about recent nose failures at BALLS). Lastly, I want to be able to apply this to a number of sizes and profiles, e.g. 38mm-98mm, VK, conical, 4:1-7:1, integrated straight sections.
With that said, here is my first draft. Brace yourself for some MS Paint, because I don't feel like opening SolidWorks yet.
Tooling mold cross section:
1. [Red/Pink] An exact profile of the nose half is printed in PLA/ABS/ASA with mirrored indexing features and sanded to mostly knock down layer lines. Print axis is perpendicular to the split plane, i.e. up in the drawing.
2. [Light grey] The surface is coated with a thin layer of laminating resin or a similarly thin, tough epoxy, left to gel, and recoated once or twice again. This is then sanded nearly back to the print , going up to around 1000 grit. Next, the surface is polished with a cutting compound, something like NW1. I have to take care not to round out the inner corners or indexing features here. A few coats of mold release are then applied, either something like Partall Paste #2/Hi-Temp Wax or Frekote 700NC.
3. [Dark grey/Black] A layer of gelcoat, probably Proline 4500 (used in its intended purpose!), is applied to the whole surface. Carbon tows are wet out with gelcoat and packed into the mold corners for reinforcement. After the first coat gels, the surface is recoated a few times, making sure large radii are built up on the indexing features and corners. I'm not sure whether to apply the tows in the first or subsequent coats here.
4. [Light blue] Once the gelcoat.. gels.. a layer of fiberglass or carbon is laid up onto the mold. I think I would first apply a layer or two of light glass, 1-2oz, then several layers of heavier glass, 6-12oz. 6oz or heavier carbon is also an option for better stiffness here. After gelling, the edges are trimmed square.
5. The tool is demolded and cleaned up as needed, and the process is repeated to make the other half.
Done well, I should end up with two mold halves that fit together perfectly.
Next, the plug. This is far simpler and takes directly from my prior experience.
1. [Red/Pink] An exact profile of the nose interior is printed, i.e. the nose profile minus the intended wall thickness. A hole runs down the center for a threaded rod, dark grey.
2. [Light grey] The surface is coated with laminating resin, sanded, polished, and release coated identically to the tooling mold. This also could be gelcoated, but I'm not sure.
Next, layup. One big issue with braided sleeve is that it has a limited range of diameters it can expand/contract to, meaning that a nose's base and tip diameters can't be too dissimilar. You can coax the fibers beyond the listed minimum diameter by aligning them more towards the nose tip, but this dramatically increases its thickness. If you want more consistent thickness, you have to manufacture a pretty huge nose tip, which can get very long and heavy for high LD profiles. My solution is to use multiple sizes of sleeve and have them overlap in alternating layers, giving a staggered concentric seam. Once compacted in the mold, the seams should flatten out and give a clean, strong transition between the two sleeve diameters. The challenge here is not having big gaps filled by resin in the joint. More MS Paint!
The hardest part will probably laying up the seam and minimizing gaps between each pair of sleeves without bunching up fibers. Inserting the plug into the mold without distorting the weave will be the next hardest step. Once laid up and inserted into the mold halves, I have a few options:
1. Curing under no compression. Probably the worst option, but also the easiest. Take a mold half, drop in the plug, attach the other half, secure together, then jam the plug down to consolidate a little and let cure.
2. Curing under bolted compression. Not much harder, but could be messed up. Assemble as usual, but then use the threaded rod running down the plug to compress the plug against the mold axially. By extending the rod out the top end of the mold and tightening a nut down, the layup gets some mechanical compression. This requires the mold haves to be bolted together or similar.
3. Curing under vacuum. More equipment, more consumables, but probably the best result. Lay up, assemble, bolt halves together, run breather to each end, bag, suck. However, I have no idea how a printed plug with partial infill would handle under vacuum, given the trapped air inside.
Once cured, I demold, trim, clean up, and post-cure if needed. Overall, I think I could make a full mold set in about a week, then pump out lots of consistent tip-less noses from them. I'm left with lots of questions before starting, however, and I want to save myself a few botched attempts.
- Does the workflow make sense overall to those who have made split molds before?
- Has anyone tried doing similar parts with sleeve?
- What mold releases are best? I'm leaning towards Frekote 700NC, but I have had good results with Partall Paste #2 before.
- Carbon tow reinforcement in the first or subsequent layers of gel coat? Are they even needed with something like Proline?
- Polishing compound recommendations for epoxy? NW1 is just what I've seen in Easy Composites videos.
- What reinforcement should the mold get? Glass, carbon? How many layers? What weights? Should I use lighter weights to conform to the surface, then build with heavier weights? Can a single weight suffice?
- How do I design the indexing features? My current thought is a frustum with a pretty sharp draft angle (30-45 deg?), rounded edges, and a match-drilled bolt hole.
- Can this be vacuum bagged with a printed plug? Any good references for bagging nose molds?
- Should the plug be coated with gelcoat or laminating resin before sanding/polishing/release coating?
- Is the printed plug a good idea, or are there better methods? Prepreg + tubular vac bag is less labor intensive, but that obviously defeats the point of using sleeve. Making a composite plug, meanwhile, would double the amount of molds I need per part.
- Any resin picks for extreme flights? I'm leaning towards HTR-212, which I've had good experiences with but haven't pushed too fast yet. Cotronics stuff is the next pick, but pricey.
If you made it to the end of this and have input to give, thanks! I would also appreciate links to any RC forums or similar, since I hear their parts are on a whole different level. Basically, sanity check me before I waste my time.
