[RC Wing Sailboat Project] Third Prototype

Designing and building the shrink wrap covered balsa ribbed wing

Nikolas Osvalds
10 min readAug 18, 2020

Originally Published Spring 2011 as part of my capstone engineering design project at Brown University

Now it was time to build the wing in a more accessible way, since CNC hot wire foam cutters cost in the $1000’s. The method that first came to mind and that is used on many other rigid wing sail boats uses many vertically spaced airfoil shaped “ribs” which are then covered by a film to produce the shape in 3D. Examples of this style of construction can be seen in many of the pictures in the background section. Model Airplanes builders employ the same sort of construction and are on the same scale as the wing I planned to build. Therefore I used model airplane building sites as references to design and build the wing.

Following these resources for advice I decided to build the wing out of balsa wood due to its light weight and cheap cost. The design for the main element of the wing includes 23 3/32in thick balsa ribs spaced 5cm apart along the span of the wing. The ribs are given strength and support by “spars” that pass vertically through all the ribs. My design includes a 1/2in diameter dowel that makes up the leading edge. Another 1/2in diameter dowel passes through the ribs at the wing’s pivot point which is placed 4.5cm back from the leading edge at the maximum thickness of the airfoil (30% of its chord) . A spar that is 3/32in thick and 1in wide is slotted into trailing edge of the airfoil. Finally two 3/8in x 1/4in rectangular spars run along either side of the wing at the maximum thickness.

The flap has a very similar deisgn, except for it does not have the internal dowel since it does not need a pivot point. The dimensions are as follows. The leading edge dowel is 1/4in diameter. The trailing edge spar is 1in by 1/16in. The rectangular side spars are 3/16in x 1/8in.

To cut the 23 balsa ribs the airfoil shapes were first plotted in excel using an equation for plotting NACA airfoils obtained here. These data points were then imported into Pro Engineer. The spars were then added to each of the rib’s CAD models. Then a cross section was printed out that could be used as a template for the ribs. Here are the PDF files of the rib plans. These files also include spar dimensions and placement for reference. MainElementRibPlan.pdf , FlapRibPlan.pdf . Screen shots of rib plans below (not to scale).

To cut the ribs quickly two templates were made out of thicker plywood. These templates were first taped to a stack of uniformly sized blanks. Holes were drilled through the reference holes of the template into the blanks. The blanks could then be bolted on either side of the stack of rectangular blanks of 3/32in balsa using two reference holes to align them. The reference holes could be of any size. I used 6X32 bolts since that is what I had available. The stack of 12 pieces of balsa could then be uniformly shaped at one time using a hack saw, files and sandpaper. This method was used because it made uniform ribs more quickly than tracing the pattern 23 times onto a sheet of balsa and using an xacto knife to cut them out. A picture of one such stack and a set of main element ribs can be seen below.

Finished set of ribs
Templates bolted to stack
Stack after some shaping

Laser Cutting

To make this process easier and less labor intensive for anyone to replicate I looked into CNC laser cutting. This is definitely the way to go for this sort of repetition. With a little internet research I decided on . I created a 2D AutoCAD file of the ribs to their specifications and sent it in for a quote. To laser cut the 46 required ribs to build the wing it would cost $11.56 plus shipping. This is a great deal considering it took me at least 5 hours to make each set (main and flap) of ribs by hand. If anyone wanted to have the ribs made, it is now as easy as emailing the file I made to AK models. Picture below and Link to download the file here: WingsailRibCut.dxf


Once all the main ribs were cut out, they were placed onto the center spar and spaced at 5cm apart. Each rib had to be sanded a bit to get them on the spar, you want a snug fit. Two ribs were used at the top and bottom of the wing since they will support the flap when it is attached. This could also be accomplished by using a thicker balsa rib or other wood. In my case I was one rib short so I used one half of the plywood rib template for the top piece.

The leading edge dowel spar was marked for spacing and fitted into the ribs. It was held in place using a combination of rubber bands and zip ties. The trailing edge strip was also marked for spacing and fit into the ribs.

The process is basically the same for the flap, however since there is no center dowel spar, the zip ties or rubber bands are placed around the leading edge and trailing edge to hold the flaps for gluing. I used a level to make sure the leading edge dowel was held straight.

After the balsa pieces were dry fitted and checked to be in good alignment, They were glued using fast drying thin CA glue or superglue. The glue worked well in that it set up very fast and held a tight bond. However because it is so thin it does not fill gaps very well. This can be a problem because if the spars don’t fit perfectly flush with the ribs,which can be the case with balsa wood construction, you don’t get a very strong bond. If more time was available for drying, wood glue would be a good option to fill the gaps. In my case the wing seemed to be quite strong with just the CA glue so I didn’t used any wood glue.

Once the spars were glued and dried, the two side spars could be added. To do this each slot in the ribs had to be carefully filed and sanded so the spar would sit flush in the groove. Once the spar was fitting well it could be glued and hand held until dry. This process is identical for the main element and flap.

After all the side spars were glued in place they were block sanded to make sure they sat flush to the rib. The wing skeletons are now ready to be prepared for their heat shrink coating.

Before the heat shrink was put on I added some 6X32 hardware to the flap and main element for the flap control system. I conveniently used the reference holes of the flap to place a bolt through. On the main element I measured to 85% of the chord and drilled a hole. This is the point the flap will pivot about. I had to trim some of the trailing edge material to get the bolt in.

Heat Shrink Covering

To prepare the wing skeletons for covering I followed the directions on the heat shrink package. I block sanded the edges that would be touching the film with 280 then 320 grit sandpaper. I spent some time making sure there was a smooth transition between the ribs and spars, especially at the trailing edge. Then the wing was vacuumed and wiped to make sure all the dust was off of it. This ensures that the plastic covering will adhere nicely to the wood.

I used Hangar 9 brand ULTRACOTE heat shrink covering for the wing. This is a commercially available product made specifically for covering model airplane wings. There are many types of heat shrink film available and they all have slightly different properties. The ULTRACOTE has an adhesive built in, so that at around 220 deg F it will begin to stick to the wood. Then starting at 300 deg the film begins to shrink, with the maximum shrinkage at 350deg. The directions recommend using a small hobby iron to attach the heat shrink to the skeleton. They are actually very cheap and I would recommend getting one. I only used a heat gun, and it was pretty difficult to control the area where the heat was being applied. The temperature range was also not very accurate, luckily I had a nice thermocouple to set the correct temperatures. I didn’t get the best results because this is somewhat of an art and I really didn’t have the time to master it.

The process I went through is as follows. I first cut a piece large enough the cover the balsa skeleton I was trying to cover. Then I started by adhering one end of the heat shrink film to one side of the trailing edge, beginning at one end and moving down the whole span of the wing. I then slowly worked up and down the span and around the whole wing. It is pretty difficult to get the film to lay perfect and wrinkle free, especially when only using the heat gun. It may have also been easier to cover one side of the wing at a time so I didn’t have to work with such a large piece of heat shrink, however I was worried that I might get a bad gap at the leading edge. The good thing about the heat shrink is that once you have it fully adhered to the skeleton, you can use the higher temperature to take many of the wrinkles out and tighten the skin up. You have to be careful not to get too close to the edges however because they will start to peel up when the skin shrinks. Some picture of my process are shown below.

Heat gun
Starting with one side of trailing edge
What happens when you shrink to close to an edge
Some of the better sections of finished covering
Some of the better sections of finished covering
Some of the better sections of finished covering

If you follow the directions and take your time there should be no problem in getting a nice smooth surface covering for the wing. Using a hobby iron would also make it easier than my process probably.

Flap Control System

Once the main element and flap were covered I could finish the job of connecting them and designing a flap control system. I used two rectangular section acrylic pieces to attach the flap to the main element. A few washers were needed to space the flap correctly due to the fact that it wasn’t exactly the same length as the main element. I used nuts to attach the acrylic to the flap and wing nuts at the pivot point so the flap angle could be easily adjusted. I then drew up some angle guides and glued them to an extra piece of balsa. This piece of balsa was then bolted and screwed in line with the flap so the angle could be accurately measured. See pictures below.

The last step in building the balsa wing was to drill a hole in the center dowel and screw in a length of threaded rod. This acts as the attachment point to the testing cart.

The total cost of my building materials was around $50 dollars. If you add in laser cutting of the ribs and buy a hobby Iron it will be more like $75. Considering that the wildcat model costs around $100 itself, I think this is a reasonable price for anyone pursuing such a project. A list of all the materials I used and some notes about them can be found here. MaterialsListFinal.pdf

Originally published at http://engin1000.pbworks.com.



Nikolas Osvalds

I’m passionate about doing good, giving back, and helping to tackle the climate crisis with my working life, ideally with code.