# Easier mould-making for a ultralight Low Aspect Ratio

### Help Support Homebuilt Aircraft & Kit Plane Forum:

#### nestofdragons

##### Well-Known Member
Just trying to find a way to make a wing in composites for a motorized, ultralight low aspect ratio.
My goal of this discussion here: finding a way to make the composite wing with as easy to make moulds. I had very bad experiences with endless sanding of moulds that start with hotwired or CNC routered foam.

Lets start with why i choose to make the wing with three panels at each side. Easy to explain. The round wing would demand that technique of CNC routering and placing gelcoats and so and i really really would like to avoid that way. It is not easy to do for everybody and it demands tiiiiiiiiiimeeeee. And if you look at the wing. It nearly has the same shape as the three panelwing.
You can see my Low Aspect Ratio ideas here. Low aspect ratio sport plane ideas
The more lineair editions can be seen at Low aspect ratio sport plane ideas
That project is still being developped.

Now lets make a mould for that kind of wing.

I would make wooden jigs combined with alu plates or some other plate material. Three sections need to be made and assemble them afterwards. Here you see how the three parts would look together. Sorry, i drew a single panel where the jigs are placed together to be quicker.

Each section can get its mould panel easily. Just slide in a plate of alu or copper. Push it onto the cut out shape. If you want you can use silicone to keep it a bit in place while you draw a line at the part which sticks out. So, no need to draw a complex unfold template. Just lay down. Fix it a bit (but keep it removable). Draw the line so you can later cut of the part that sticks out. Remove plate, cut of part you don't need. Glue the plate back onto its place. This time it needs to stay there.

Do that three times. And place the sections together. There might be need for some filler and some sanding at the joints, but ... i guess it is a doable task to just sand those joints.
You now have this.

Ok, you can now start doing your composite skin. I would prefer some carbon, thin foam and another carbon layer. You might go at the same time for your carbon for the spar caps.

You can install D-tube rib noses now.

The ribs don(t all have the be full ribs. Halfway you can place a trapezium foam part over which you place some composite to create a tough fake rib. Just like my team did in the DragonWing. It really is easy to do. You make one foam part and you can use it for all fake ribs.

Now i need your opinion. How would i easily make a spar vertical so the spar doesn't buckle. Just insert a thick foam plate which has carbon layer at both sides? Or do i place L-angles just next to the spar caps inside edge so i can glue a thinner foam with carbon layers onto it onto that L-profiles? I could repeat the same L-profile and plate technique to close the D-tube. If any fear of these verticals buckling one might add some trapezium reinforcements like mentioned above onto the vertical thinner foams at the "inside of the spar". You can do that after you glued the first vertical. For the closing vertical, you need to do that in advance before glueing the vertical to the L-profiles.

#### Attachments

• 110587-ca2f2e02d2695880787b149ab13bb8f9.jpg
40.8 KB · Views: 0
• 113376-2faa2f7de8808c6d463ac3f3fabc03dd.png
248.1 KB · Views: 0
Last edited:

#### Victor Bravo

##### Well-Known Member
Supporting Member
Just trying to find a way to make a wing in composites for a motorized, ultralight low aspect ratio.
My goal of this discussion here: finding a way to make the composite wing with as easy to make moulds.

Hop on the airliner and get to Los Angeles on Friday. I'll pick you up at the airport, you can stay at our house, and Saturday at dawn we will fly to Tehachapi for the Experimental Soaring Association event. There, you will be introduced to BobK, who has done significant research in this specific area.

Or... you can ask BobK for his advice here on this forum... but you will miss out on the good BBQ lunch we have planned in Tehachapi

#### nestofdragons

##### Well-Known Member
Hop on the airliner and get to Los Angeles on Friday. I'll pick you up at the airport, you can stay at our house, and Saturday at dawn we will fly to Tehachapi for the Experimental Soaring Association event. There, you will be introduced to BobK, who has done significant research in this specific area.

Or... you can ask BobK for his advice here on this forum... but you will miss out on the good BBQ lunch we have planned in Tehachapi
Victor Bravo, i reaaaaally appriciate your offer. But ... i will not be able to travel to USA the following years. Financial choices to make.

#### wsimpso1

##### Super Moderator
Staff member
All that wood! How about if you invest in a large light duty 3 axis router. Carve skin tools directly from foam. Or carve massive cores directly from foam - do one side, laminate it, turn it over on fixtures, carve the other side, and laminate that side.

#### rotax618

##### Well-Known Member
For ruled surfaces like the above design, the surfaces of the foam (if XPS) can be cut with a hot wire.

#### nestofdragons

##### Well-Known Member
All that wood! How about if you invest in a large light duty 3 axis router. Carve skin tools directly from foam. Or carve massive cores directly from foam - do one side, laminate it, turn it over on fixtures, carve the other side, and laminate that side.
To me this solution sounds like the endless noise of sanding. And ...i really want to avoid that. Maximum sanding i like to do is fixing the joins between the sections a bit.
Also ... getting myself a CNCrouter sounds like big money. Having a large piece like this CNC cut is still big money. I would only go to foam if i would hotwire it. But ... i still dislike the sanding, so ... no thanks. I am a bit stubborn about the "no sanding if possible" rule here.

##### Well-Known Member
Supporting Member
There's probably someone with a large CNC hot wire cutter within easy shipping distance of you; the materials are cheap and the machines are fast enough to be super cheap to run. Surface finish is pretty darn great.

CNC routing of foam is slower (so more expensive) and leaves a worse surface finish, but shapes are unconstrained. If you look at automotive startups near you, they probably have a foam cutting machine of some sort (even if just a Kuka wielding a big bit) for clay models and other prototypes.

#### geraldmorrissey

##### Well-Known Member
I would build my molds out of plaster. Hydrocal B11. An industry standard for 50 years. Cheap, durable, haul it to the dump when your done. Seal it with shellac. Reinforce it with glass or hemp. You tube has videos to help. Before CNC it was how OML was defined on nearly every aircraft. Mahogany was the other popular material. Once saw a picture of a mahogany P-38 fuselage mockup. Beautiful.

Last edited:

#### nestofdragons

##### Well-Known Member
I would build my molds out of plaster. Hydrocal B11. An industry standard for 50 years. Cheap, durable, haul it to the dump when your done. Seal it with shellac. Reinforce it with glass or hemp. You tube has videos to help. Before CNC it was how OML was defined on nearly every aircraft. Mahogany was the other popular material. Once saw a picture of a mahogany P-38 fuselage mockup. Beautiful.
I totally forgot that technique. Thanks for mentioning it. But i think it still requires the need for a smooth positive shape. And ... i fear that that sounds like a lot of sanding. I will try to look up the magony technique. Never heard about that one. Might be i know it with other name.

#### JamesRussell

##### New Member
Hi all:

What geraldmorrissey said.

Make eggcrate templates of your wing shape, fix them to a backing structure, add plaster w/ hemp for lower level reinforcement, then sweep plaster only using the templates to guide the flexible sweep to generate the OML surfaces.

Heavy, messy - you bet! Fast, cheap - yes. The plaster will fully cure in a day, then you lightly sand (if needed), seal with shellac. Release with wax & PVA, lay up female tool as normal...

I was lucky enough to work at Bell Helicopter and helped/learned how to do this from the old heads.

Then I built Cobra inlet breakout tools using plaster/hemp.

No faster, cheaper way I know to generate large surfaces...

The US Navy training booklets, "Templates and Layout, Mock-Up, Parts 1 & 2" cover this process in exhaustive detail and are for sale on eBay at very cheap prices.

See attached.

Regards,
James

#### Attachments

• USN Bk No. 17 Pg. 0.jpeg
1 MB · Views: 2

#### Jay Kempf

##### Curmudgeon in Training (CIT)
Supporting Member
That wood and metal solution is actually pretty cost effective. When making a deep composite part like that out of foam and using either hot wire or CNC you are going to waste a lot of more expensive materials than a couple sheets of MDF or equiv. My 3" foam board that I CNC Into molds is $60 for 4 x 8 sheet. That adds up quick at full scale. The hard part with the foam is getting the bag to seal. With the metal sheets you only need to get the sheet metal edges to seal to each other (silicone caulk) and then you can seal directly to the metal. I think that is why Boku headed that way. If you put a whole foam block inside the bag you then need to worry about how much vacuum and crushing. For the spar web I would make a C channel spar in a simple table top mold and then bond to the skins. #### wsimpso1 ##### Super Moderator Staff member Y⁶ To me this solution sounds like the endless noise of sanding. And ...i really want to avoid that. Maximum sanding i like to do is fixing the joins between the sections a bit. Also ... getting myself a CNCrouter sounds like big money. Having a large piece like this CNC cut is still big money. I would only go to foam if i would hotwire it. But ... i still dislike the sanding, so ... no thanks. I am a bit stubborn about the "no sanding if possible" rule here. Perhaps we misunderstood your parts. Are you doing a compound curve set of wings, or are you doing three prismatic pieces joined with hard chines? If you are doing a compound curved wing, that still sounds like a lot of mold shaping between forms. If you are doing linear parts with hard joins, won't that be, um, inelegant and inefficient? CNC 3 axis router kits are out there, scaleable in size, and not awful compared to a lot of other schemes. Probably some used ones too... Other options include having someone else cut your molds. As to sanding, use a ball end cutter, make the final shape passes a lot closer together than the cutter radii, and it won't take but a little finish sanding on the mold. Surfacing is done by putting bag film over the mold, then vacuum bag your part. All of these will result in a female formed composite wing skin. These can be pretty good, but most of us amateurs will still either accept a rough looking wing or do a fill and fair process. Your choice really. Look at my build at: Sanding for finish is covered in posts 25 and 40. Profiling my flaps and ailerons were about an hour per side. Belly and lower sides worked out to about 19 hours. Smaller amounts of sanding followed each coat of that white epoxy primer, as it shows all the evil remaining. If you hate sanding as much as you seem too, I suggest a close review of each of the construction types for the amount of sanding needed and perhaps re-evaluate your chosen build methods. For the rest of us, ALL build methods have a bunch of sanding before you are done. RV's might have the least, but you still have the cowling, root fairings, canopy, tips of wings and tailplanes, and landing gear fairings. Get to all composite and everything is filled and faired. Fabric covered can be only as big for sanding as an RV (no sanding of fabric and stop at silver OR go Oratex) or as big as a composite ship if you want a finish that stops folks in their tracks. Billski #### geraldmorrissey ##### Well-Known Member If your sanding on a plaster mold you are destroying your contour. They are built to net. Sanding, other then buffing, will ruin a plaster mold. Parts pulled from plaster molds that are bagged, should need nothing more the a light buffing with 600 wet to knock off the odd inperfection. James got it right and the books he suggests will explain all. It's caveman but served the aerospace industry well from the teens through the 1980's. Last edited: #### rotax618 ##### Well-Known Member The ruled surfaces are no more inefficient than the facets on the Facetmobile, they certainly make it easier to build. The design really should have a more rigid, say light ply covering the leading edges (to resist deflection of the LE due to air pressure) amd fabric (Oratex) covering over the remainder. #### geraldmorrissey ##### Well-Known Member Whatever method you use, keep at it. It's an interesting design. Last edited: #### nestofdragons ##### Well-Known Member That wood and metal solution is actually pretty cost effective. When making a deep composite part like that out of foam and using either hot wire or CNC you are going to waste a lot of more expensive materials than a couple sheets of MDF or equiv. My 3" foam board that I CNC Into molds is$60 for 4 x 8 sheet. That adds up quick at full scale.

The hard part with the foam is getting the bag to seal. With the metal sheets you only need to get the sheet metal edges to seal to each other (silicone caulk) and then you can seal directly to the metal. I think that is why Boku headed that way. If you put a whole foam block inside the bag you then need to worry about how much vacuum and crushing.

For the spar web I would make a C channel spar in a simple table top mold and then bond to the skins.
Can you make a rough draft of what you means with C channel spar.

#### wsimpso1

##### Super Moderator
Staff member
The hard part with the foam is getting the bag to seal.

yep, most likely way is wrap the whole thing in bag film. Any hard spots in your parts and tools then punctures the bag film. It requires meticulous attention to de-horning the stuff inside…
With the metal sheets you only need to get the sheet metal edges to seal to each other (silicone caulk) and then you can seal directly to the metal. I think that is why Boku headed that way. If you put a whole foam block inside the bag you then need to worry about how much vacuum and crushing.

I see folks mention foam crushing, yet the closest I have ever come to seeing any is rounding sharp corners on 1.5 to 2 pcf styrene foam. Let’s calibrate ourselves a little. 20“ Hg is about 10 psi or 70 kPa. Even the flimsiest white coffee cup foam has a compression strength of about 40 psi - 4x what my pump can deliver.

Some epoxies have more solvent effect than others, and this can be checked by just cutting samples into little cups of resin and hardener. If your foam melts in your resin and/or your hardener, maybe you need to use a system with less solvents. West and Proset will round sharp corners on styrene foam, but I have never lost any thickness on styrene cores, and Divinycel stays sharp.

For the spar web I would make a C channel spar in a simple table top mold and then bond to the skins.

Channels (why do some folks call it a C channel? Channel describes it fully) are easily tooled. Straight ones are just done on a work table for open wet layups, or on an impermeable table for vacuum bagging. You can fixture right on the table or do foam cutouts and vacuum bag over or within them.

For a spar with distinct kinks, you hotwire white foam with templates at each break, align the pieces and glue them together, release tape it all with packing tape, and either do an open wet layup on the male shape, or make wet layup mold with extra length and flanges, then vacuum bag in the mold. There are about 8 different ways to explore these options…

My spars are all channels. I have talked about them before…

Billski's Fiberglass Bird

see post 16’s photos to see how some of them came out.

Billski

#### nestofdragons

##### Well-Known Member
My spars are all channels. I have talked about them before…

Billski's Fiberglass Bird

see post 16’s photos to see how some of them came out.

Billski
I am really impressed with your project. Yes, you do composites as a master. Respect, respect.
Here in my own project, i try to avoid moulds as much as possible. Making C-channels (or just channels as you call them) demands another mould to be made. If i could minimize the need for moulds in my project to just the mould for the left and right D-tube and a flat plate for all the 2D parts, i would reach my goal.

At this moment i think that metal plates will help me to create "moulds" on the construction. I think i need to explain that a bit.

Example: between the D-tube and the rear spar (on which all control-surfaces will hinge) i will place several ribs. Those ribs will be made on that flat table (probably a glass plate). But they need caps to give them resistance against bending and more glue surface for the fabric (I really hope to use Oratex). So i intend to use metal plates on the mounted ribs to create a "mould" for these caps. How do i do that? Well, just bend a metal plate over the topside of two ribs. Use some weight to keep it down. Use silicone to "glue" the rib-side (which you will not do) to the plate. Once that is hard just remove the weight and turn wing upside down to place some layers into the corner you created at both ribs. Let it cure. Remove the silicone you used to glue the plate and remove the plate. Cut away the composites that were placed too much.

I guess this technique demands two days for each section. One for hardening out the silicones, one for adding and curing the composites. If some trick could be used to speed up the fixing of the plate onto the ribs, you can save a day. It might look like time-consuming. That might be right. But ... to me ... it sounds like suuuuper easy to do and only needs garage tooling.

For a wing that has very small amount of ribs, i guess it is a doable task.

Extra info: all my ribs will be different in shape, so the caps will have a different angle left and right of the rib. I guess one can add the wax to the metal plate while the silicones are hardening (the weights keep plate down). A removable paper tape keeps the wax from the rib surface.

#### nestofdragons

##### Well-Known Member
I was wondering. If i would not wax the metal plate, but fix peel ply to the plate. Would that not speed up the process? I need to wait for the wax to dry before placing another layer of wax. And repeat that three times.
Now i would just place tape, which sticks on both sides, to the metal plate. Place a strip of peel ply over it. Done. Ready to place the layers in that corner.

##### Well-Known Member
Supporting Member
Just trying to find a way to make a wing in composites for a motorized, ultralight low aspect ratio.
My goal of this discussion here: finding a way to make the composite wing with as easy to make moulds. I had very bad experiences with endless sanding of moulds that start with hotwired or CNC routered foam.

Lets start with why i choose to make the wing with three panels at each side. Easy to explain. The round wing would demand that technique of CNC routering and placing gelcoats and so and i really really would like to avoid that way. It is not easy to do for everybody and it demands tiiiiiiiiiimeeeee. And if you look at the wing. It nearly has the same shape as the three panelwing.
You can see my Low Aspect Ratio ideas here. Low aspect ratio sport plane ideas
The more lineair editions can be seen at Low aspect ratio sport plane ideas
That project is still being developped.

Now lets make a mould for that kind of wing.

I would make wooden jigs combined with alu plates or some other plate material. Three sections need to be made and assemble them afterwards. Here you see how the three parts would look together. Sorry, i drew a single panel where the jigs are placed together to be quicker.
View attachment 129389
View attachment 129390
Each section can get its mould panel easily. Just slide in a plate of alu or copper. Push it onto the cut out shape. If you want you can use silicone to keep it a bit in place while you draw a line at the part which sticks out. So, no need to draw a complex unfold template. Just lay down. Fix it a bit (but keep it removable). Draw the line so you can later cut of the part that sticks out. Remove plate, cut of part you don't need. Glue the plate back onto its place. This time it needs to stay there.

Do that three times. And place the sections together. There might be need for some filler and some sanding at the joints, but ... i guess it is a doable task to just sand those joints.
You now have this.
View attachment 129391
View attachment 129392

Ok, you can now start doing your composite skin. I would prefer some carbon, thin foam and another carbon layer. You might go at the same time for your carbon for the spar caps.
View attachment 129393
You can install D-tube rib noses now.
View attachment 129394
The ribs don(t all have the be full ribs. Halfway you can place a trapezium foam part over which you place some composite to create a tough fake rib. Just like my team did in the DragonWing. It really is easy to do. You make one foam part and you can use it for all fake ribs.
View attachment 129395

Now i need your opinion. How would i easily make a spar vertical so the spar doesn't buckle. Just insert a thick foam plate which has carbon layer at both sides? Or do i place L-angles just next to the spar caps inside edge so i can glue a thinner foam with carbon layers onto it onto that L-profiles? I could repeat the same L-profile and plate technique to close the D-tube. If any fear of these verticals buckling one might add some trapezium reinforcements like mentioned above onto the vertical thinner foams at the "inside of the spar". You can do that after you glued the first vertical. For the closing vertical, you need to do that in advance before glueing the vertical to the L-profiles.
Great, great, great post! This is darn-near identical to what I am fixing to do with my molded composite build... stand by for private message so we can discuss further!