The Corsair82: 82% F4U-1A Corsair

In June of 2002 I attempted to make the first wing skin. This attempt was to build the top skin for the left wing center section. Well, everything that could have gone wrong that day did. Even though I had done all of my preparation work well in advance of my lay up, I could not complete the lay up before the resin started to gel. I ended up with a mess of bubbles and dry spots. Anyone who knows anything about composites knows that a part like that gets thrown away. I found that one guy couldn't do that size lay-up alone.

I needed to find a way to build my large fiberglass parts, so I investigated my alternatives. I went to a few fiberglass fabricators. When I told them I needed help building airplane parts they either laughed or slammed the door in my face. Then I approached a few airplane kit manufacturers. Even with me supplying the materials and molds, their prices were unbelievable. At some point I remembered a process a long lost friend had once mentioned. He called it "Vacuum Infusion".

One day I did a word search for "Vacuum Infusion" on the Internet. I found a site that sold a DVD training course for the process. I purchased the DVD and became an overnight expert. Right! I still had a long way to go before infusion became the answer to my problem.

So what is Vacuum Infusion? Simply put, Vacuum Infusion is a method for wetting out composite materials whereby vacuum draws the resin through the "structural matrix" (that means the fiberglass and core or the "composite sandwich"). However, just like everything else involved in working with composites, you must learn the proper procedures and techniques to do quality work. In the case of infusion, you have either spectacular results, or spectacular failures.

I'm going to oversimplify the process to better explain it. To infuse a part with resin, you first put all of your composite materials, the fiberglass and the core, into the mold. This is called "dry stacking". You then place a sheet of plastic (the vacuum bag) over the materials and seal it all the way around the mold, just like you would if you were going to vacuum bag a hand lay up. You insert and seal a vacuum hose through the plastic at a predetermined location. At another location, usually on the opposite end of the mold, you insert and seal a feed hose that goes to your resin reservoir. The feed hose must be closed off with a valve or clamp at this point. It is also a good idea to use a resin trap between the mold and the vacuum pump. This keeps your pump from being ruined if the resin makes its way into the vacuum line. Now you allow the vacuum pump to evacuate the air trapped between the mold and the vacuum bag. This compresses the dry stacked materials to some extent. When everything is under control, the resin is mixed in the reservoir and the valve is opened. The resin then flows through the materials. There is no mess and almost no smell. The only surface area of resin exposed to the air in your shop is the surface of the resin in your reservoir. That means less than a square foot of area venting fumes, as opposed to the entire surface area of a hand lay up. With Vinyl Ester or Polyester resins the styrene smell is cut down to almost nothing.

Now comes the bad news. My first few attempts at infusion were flops. I tested the process on a small lay up of about two square feet. The first problem I discovered was that the vacuum bag was too tight against the mold surface at the point I was feeding the resin. This stopped the resin flow as soon as the resin arrived at the mold end of the feed line. I tried different ways of feeding the resin and found that there needs to be a flow enhancing material (called a flow media) between the mold and vacuum bag and also touching the laminate. Trying this improved the initial flow of resin, but the resin stopped at the laminate. After a lot of testing I found that woven fiberglass (like 7781) would not allow the resin to flow. I tried some "knitted" materials and found them to flow well. A knitted fiberglass is basically multiple layers of unidirectional fiberglass loosely stitched together with the required fiber orientation.

I still was not getting the flow rate I was looking for. I learned of a few different core materials that were engineered for infusion. I was given some samples of these materials and found a vast improvement on my flow rate. The last step in this process was to get some resin designed for infusion. A few companies were making reduced viscosity epoxies and vinyl ester resins engineered for infusion. I was supplied with some samples of these new resins to test. This was the turning point. I was now getting spectacular results. I could infuse a 30 square foot part consisting of six layers of fiberglass and a core, in about 15 minutes. One guy can't wet out 30 square feet by hand, place a vacuum bag, and draw a vacuum in 15 minutes.

The boat building industry developed Vacuum Infusion, for the most part. The materials I was experimenting with were designed for boats. The fiberglass was very heavy and the core materials were very thick, and of high density. With the help of a few manufacturers and their sales and engineering staffs, I was able to develop and test material combinations that were better suited for airplanes. Between November of 2002 and late 2003 I spent the majority of my time building and testing samples for the engineers to test (read that destroy). I now have a catalogue of fiberglass, resin, and core combinations to use in the construction of my replica Corsair.

The year I spent learning Vacuum Infusion and testing material combinations will allow me to do most of my work in a clean environment with little or no help. It will save time and materials. All of that pales to the superiority of a vacuum infused part. There is no comparison to a hand lay up part. I found a consistent 65/35 ratio of material to resin. With infusion there are no dry spots or voids. This means a consistent quality through the whole part.

The following series of pictures show the setup and operation of Vacuum Infusion. The diagrams are a depiction of the setup. The first series of pictures is a test of the procedure. The second series is the actual wing skin for my left center section top skin.

figure1.jpg
figure2.jpg
vi-1.jpg
Setting up. I am using a sheet of glass as a mold surface. This is the sealant tape for the vacuum bag. It is two sided. The protective paper is left on until the bag is placed.
vi-2.jpg
The glass surface has been treated with mold release and covered with a layer of tri-axial, knitted fiberglass.
vi-3.jpg
This is the core material. It is engineered for infusion. Another layer the fiberglass will cover the core.
vi-4.jpg
The resin feed and vacuum tubes are in place and the vacuum bag is taped on two sides.
vi-5.jpg
A vacuum has been drawn and the flow of resin can bee seen. The flow media will fill first, then the resin will flow through the materials.
vi-6.jpg
vi-7.jpg
vi-8.jpg
vi-9.jpg
vi-10.jpg
The wing skin infusion sequence follows. It consists of two layers of tri-axial fiberglas on each side of the engineered core material.
vi-11.jpg
vi-12.jpg
vi-13.jpg
vi-14.jpg
vi-15.jpg
vi-16.jpg
vi-17.jpg
vi-18.jpg
vi-19.jpg
vi-20.jpg
vi-21.jpg
The finished wing skin will remain in the mold until it is time to use it. Notice the flow media has been removed.


Return to Building the Corsair
Return to The Corsair82 Page