Molded tails with expanding foam core - RC Groups
Shop our Airplanes Products Drone Products Sales
Thread Tools
Feb 23, 2012, 09:55 PM
Registered User
tzyoung's Avatar
Build Log

Molded tails with expanding foam core

I have been secretly working on a new project over the last few months, and now that I have built a few usable parts, I am ready to document and share my progress. I was looking for a way to build molded tails that does not rely on CNC machining. The parts are built by laying up skins in each mold half. After the skins have cured, a core is formed by pouring in a castable expanding polyurethane foam that bonds the two halves together. I have attached some pictures of finished parts. The next posts will show more detail about the molds and parts.

Sign up now
to remove ads between posts
Feb 23, 2012, 10:09 PM
Registered User
tzyoung's Avatar
The first step was to build a master of the vertical tail. I blanked out the shape from balsa and carved it to shape with a razor plane and sandpaper. Once I was happy with it, I applied a layer of 3oz glass cloth on each side. The cloth weave was filled with two-part auto body glazing compound followed by high build primer with plenty of block sanding. A coat of rattle can black lacquer was used a the topcoat.
Feb 23, 2012, 10:11 PM
ground penetration specialist
Nathan Schmoekel's Avatar
That's a neat idea!

Tail looks to be on the large I'll guess 9+ grams unless that foam is really light.... am I close?
Feb 23, 2012, 10:28 PM
Registered User
tzyoung's Avatar
The next step is to build composite molds from the plug. I used a piece of cheap melamine faced shelving from the home center as a parting board. I used a router to cut out the relief area for the plug to sit into the parting tray as well as the channel that will form the mold key. The plug was potted into the parting tray using polyester body filler. To do this, coat the downward facing side in wax and pva and press the plug into the wet Bondo. Once it cures, the plug is popped out and the potting material is blocked down even with the melamine. If I do it again, I will use epoxy splooge in the place of Bondo as it is easier to work with in this application.

The whole works is reassembled and a depron strip coated in tape is glued to the parting tray as a buck to form a resin trap at the trailing edge. A border of coroplast finished off the parting tray assembly. The whole thing was treated with wax and PVA before molding
Feb 23, 2012, 10:37 PM
Registered User
tzyoung's Avatar
Now it is time to mold the first side. I had read vague references around here about using portland cement as a mix in for epoxy to create a surface coat. In all of my searches I couldn't find any examples. So I tried it here. It actually works quite well and makes a tough dark green mold surface. I recommend sifting the cement before mixing to avoid clumping. These molds are sand cored to hedge my bets against containing the pressure from clamping to keep the foam from pushing the molds apart.
Feb 23, 2012, 10:49 PM
Registered User
tzyoung's Avatar
The rest of the mold building is rinse and repeat on the other side, but I will highlight a few problems that I had. The parting board stuck on a chunk of the mold key and ripped it out where I had some bridging in the glass next to the surface coat. I rescued the chunk and CA'ed it back into place.

Here I also had to make arrangements for the foam overun to escape when I poured in the core. This was accomplished with clay risers that bridge the resin trp buck with the mold edge. I erred on the side of caution on sizing and placement with these. I might move this feature to the LE where you would cut anyway to attach the tail to the boom in order to tighten up the trailing edge.
Feb 23, 2012, 11:07 PM
Registered User
tzyoung's Avatar
Finally on to building parts. The skins were layed up from .75 oz glass cloth on a bias with a uni carbon spar cap. I tried to cast the core against wet epoxy on the first attempt. This did not work that well causing some of the fabric to shift and seemed to negatively affect cure time. The next set worked out much better when cast against cured skins.

I still have some issues to iron out with this process. The foam needs to be at least 80 degress F in order to reach full expansion. I am hoping that once I build a hot box I will have more control over expansion ratio which will help to get the weight down.

I have also had some quality issues stemming from the casting technique. I pour foam into open molds and then close and clamp them. So far I have brushed the foam into the leading edge of one mold half and let the mixture rise up through the trailing edge. I get the best surface quality at the initial pour site and the end up with blisters right at the transition from the pour zone and where the foam starts expanding freely. I will try at some point to fully coat both skins and let them expand toward the middle in hopes that I can push the bigger bubbles toward the center of the core.

I have also tried another experiment in these molds where I vacuumed bagged a piece of 3mm depron against the skin layup in each side of the mold. After it all cured the foam that remained proud of the surface was removed with a hot-wire bow and the two sides were epoxied together clamped in the mold. Danstrider has all the pictures from building that part.
Feb 23, 2012, 11:11 PM
Registered User
tzyoung's Avatar
Good guess on the weight. The fin has about the same area as the Swindell tails everyone has been raving about. The urethane foam cored one is about 9.5g. The depron cored version that we unmolded today weighs in at ~8.5g. I think I can beat that with expanding foam if I can get it to a reasonable temp for full expansion.
Feb 23, 2012, 11:32 PM
Registered User
I've been thinking about this for a while now, but don't have the $ to experiment with it right now. I'll be following this thread with great interest.

Feb 23, 2012, 11:51 PM
Registered User
That is soo cool. I hope this works out for you. The only thing stopping me from doing moulded tails is the cores.

I wonder if this foam that uses a catalist [spelling?] would cure any better with the wet skins but I guess you would still have the issue of the fabric shifting.

Anyways. here is another option
Feb 24, 2012, 11:28 AM
Father of Fr3aK, DLG Pilot
tom43004's Avatar
Feb 24, 2012, 12:06 PM
Registered User
Hi Trent

I've been making rocket bodies using this technique since the late nineties but not for a few years now and even looked at making wings. The problem in both was that 2 part polyurathane foams available to me had set times [pot life] of only a couple minutes, full cure in 10-20. What would happen on the larger molds was that the expanding surface would set up hard as the underlaying part was still expanding. The body molds were 22" x 3.5" with the expansion jambed at 1/3 up. That was solved but it would be sooo sweet to find a resin that had a longer pot life. Ergo; what resin are you using and what is its pot life?

I was a mason by trade and tried cement molds but that was too much like work Never occurred to use it as a filler for epoxy tooling. I therefore have a natural affinity for the way you made your molds.

Feb 24, 2012, 12:31 PM
Aurora Builder
The quality of the first parts out of these molds is very good, at least as good as the best bagged tails, having had the opportunity to see parts first hand! So the method is very viable, just requires some special mold construction, which Trent did a wonderful job on. 9.5g isn't bad for a first pull on such a large v.t, 8g would be a good goal.

I am very excited to see the pictures of the depron tail work. My immediate 2 questions are:

1) Did you use a shear web sliced into the foam? I think Depron may be a little weak for a shear web all by its lonesome, but I've never flown depron on a DLG so perhaps it will work?

2) Did you block off the resin channels to make it easier to drag the hotwire bow across the surface?

Thanks for posting the pictures, very exciting stuff!
Feb 24, 2012, 03:50 PM
I'm All Thumbs
AirBornOne's Avatar
Hello TZ,
Interesting work.I think everyone has considered some method of using expanding foams for model a/c.
Have you considered foaming the two halves separately? Curing the foam in open molds,hotwiring to be flush with the molds and then joining the molds.A little Gorilla & water might be adequate.
You could even add a layer of glass if it didn't get to heavy.
Feb 24, 2012, 04:21 PM
Twisted and Confused
flyonline's Avatar

have a chat with Timbuck here on RCG - he's been doing this for some time now with DS wings. I can't find the link to his thread from a quick search, but I'm sure he'd be more than happy to help you out.

He uses the method Jay mentions above.


Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Build Log Building molded DLG tails tom43004 Hand Launch 25 Mar 01, 2013 11:32 PM
Help! CNC Foam Cutting for Foam Core Wings 1stfoot CAD/CAM 6 Jan 13, 2012 06:00 AM
Discussion Disser with no backing on foam core? DKNguyen Composites Fabrication 32 Mar 20, 2011 11:45 PM
Discussion Expanding Foam Traian Hand Launch 8 Mar 06, 2011 09:49 PM
For Sale 3 x custom TabooGT models (std, disser + molded tails, narrow wing + foam tails) olgol Aircraft - Sailplanes (FS/W) 6 Mar 05, 2010 08:04 PM