Click here to back to the Niwotian home page

Hovercraft Construction

The pallet full of parts

The pallet came a week or so after we placed the order. It included everything except the skirt material and the epoxy. It was neatly packed and arranged and arrived undamaged.

The first layer is layed out.

The first of the two basic layers of 2" styrofoam for the hull is laid out to ensure there are no gaps before assembly.

The plastic sheeting here is laid out so we can piece it together and cut it to size. it will serve as a vacuum bag to apply uniform pressure to the styrofoam boards so that a strong bond will form with the epoxy.

Vacuum bagging in progress.

This is what vacuum bagging looks like. Epoxy is spread over all joints (edges and surfaces that adjoin one another). All of the pieces are placed together, then a layer of plastic is added to the top so that some blankets can be laid down without sticking to the epoxy. Sandwiched in the blankets are some perforated PVC pipes connected to a vacuum cleaner. The whole thing is wrapped in plastics, the edges taped, then the vacuum turned on. It really works! See the picture below to see how much pressure is applied.

Vacuum bag pressure.

This is about 240 pounds over maybe 100 square inched - or about 2.4 lb/square inch. That's 5 cups of water per square inch (a column of water six feet high!). That's a lot of pressure from a vacuum cleaner.

Cutting the hull to shape.

That's Craig cutting the hull to shape using a basic carpenter's saw. The foam is very easy to cut.

The lift and thrust ducts.

Meanwhile, down in the workshop....

...we've been piecing together the 1/8" plywood into strips that get wrapped around some forms to make ducts for the hovercraft. The smaller one is the "lift duct" - it goes through a hole cut in the deck. a fan is placed in the duct to force air down under the hovercraft to provide lift.

The larger duct is the thrust duct. It goes around the rear propeller to provide an efficiency gain of about 10-15%. To get this gain in efficiency, the duct must be shaped like and airfoil. This is accomplished through the use of more styrofoam...

Styrofoam cut to form the airfoil.

This photo shows two pieces of styrofoam cut and spliced together to form a single piece large enough to wrap around the thrust duct in the rear of the picture. The shiny surface is fiber glassed. It will become the outside of the airfoil. The other side of the styrofoam will be cut to an airfoil profile using the hot wire technique described in a few more photos.

Just in case it isn't clear, the thrust propeller and duct are mounted on the rear of the hovercraft. They provide the thrust to propel the hovercraft forward.

the thrust engine stand.

This contraption is the thrust engine mount. The 10hp engine mounts on the top light colored pine board that has clamps on it. The shaft is parallel to the bottom board and extends towards the two 'winglets' on the mount at the end nearest the camera.

The thrust duct mounts between the upright portion of the mount and the two winglets. The two winglets support the pivot point for two parallel rudders that will enable the pilot to steer the hovercraft.

Landing skids.

These four 2x2s were cut and shaped by Chris. They serve as protection for the craft when the lift engine is cut and the craft settles to the ground, landing in these four skids. The skids will get a piece of aluminum screwed to them and then they will mounted to the underside of the craft.

The wooden edges are attached to the hull.

And now back up to the garage....

Here you can see 1/2" wooden strips that have been epoxied to the shaped edges of the hull. I needed ratcheted straps, screws, and duct tape to hold the wood against the hull with enough pressure to form a good bond. Notice that the edging is much thinner than the hull - a fact more obvious in the next photo.

This shows the material that must be removed.

This photo shows that there is some material to be removed. The top portion of the hull is to be rounded on the edges. The is both to remove excess structure and weight that is not needed and to make the craft look a little more sleek. The bottom potion of the hull will be bevelled. This will also save weight. As the wooden strip serves as the outer attach point for the hovercraft's skirt, the beveled portion of the hull will not be visible on the finished craft.

This is the first hot wire 'saw' we built.

This is a rather poor photo of the tool I first used to cut the foam. The wire is connected to an automobile battery charger. One lead is connected to a legal sized hanging file holder - it is stiff and has an opening of about 15". A hot wire (tungsten?) is connected to the far and of the wire holder and stretched taught back to the wooden handle where is connects to the other wire lead. Turn on the battery charger and the wire is hot enough to melt styrofoam. I then clamped the handle to a board at the correct angle and slid the board along a line in order to keep the rounded edge parallel to the center line.

In later versions, I just put straight edges down on the board where I wanted to cut and moved the bare wire along those - much easier.

The hull after two passes with the hot wire.

This is the hull after two passes with the hot wire. The rest of the rounding I'll get with sanding. The chalk line shows me where not to sand - that line is at exactly the right height. I will need to sand the two ridges on either side of it down to form one continuous curved surface.

The final shape of the rounded edge.

This is the result of sanding. The edge is now rounded and ready for a layer of 1/8" marine plywood. Because the styrofoam is so soft, the sanding went very quickly.

A lightweight piece of pine is imbedded in the styrofaom

Prior to placing the plywood on the top surface of the hull, we embedded four pieces of pine into the styrofoam so that we would have something substantial into which we could screw the marine cleats. This got epoxied in place in the same step we put the plywood on.. With only about 30 minutes to work on the hull before the epoxy started to cure, it can get hectic - thus the written instructions.

Compressing the plywood onto the top of the hull

This is vacuum bagging step number two: the plywood top. You can see the plywood poking out around the edges of the hull. You can see the bathroom scale under the clear plastic at the far end. I hope the neighbors didn't mind the vacuum running for many hours.

Nick routes a hole for the lift duct

After Nick practiced using the router by drawing a whale on the waste portion of the decking, he routes an oval hole in the plywood so that the lift duct can be placed through the hull at a 20 degree angle. Once the slot was cut, we used a drywall saw, set at the correct angle, to cut all the way through the foam. Nick similarly cut an oval in the partial plywood coverage on the bottom to complete the hole effort.

Hull bottom

This shot lets you see the additional layer (strip?) of styrofoam and plywood we placed on the bottom. This also gives you a good view of the bevel on the underside of the hull. A 1" x 3/4" wood strip goes along the plywood and serves as the inner skirt attachment point. The skirt, therefore, looks like a tube of sorts all around the underside edge of the craft. In the distance, you can see the lift duct temporarily set in place. It is actually supposed to protrude from the top side, not the bottom.

Fit testing the lift duct

This lets you see the front end of the hull bottom - and perhaps you can see why just running the hot wire along the two strips of wood can be much easier than the contraption above. This is also the fit test for the lift duct. We used geometry to calculate the exact size of the oval needed to fit the duct at a 20 degree angle. As you can see, the fit is just right. It is nice when theory and practice agree.

Glassing the edge is easier with the hull on its side.

In order to fiberglass the edge of the hovercraft, we built a couple of simple stands to hold the craft on its edge. This made it fairly easy to get good coverage without a lot of dripping epoxy.

Installing the cockpit walls

Yes, we admit. It is now more than a father/son project. The whole family can now claim to have participated in its construction. This picture shows both the landing skids that Chris built and mounted and the scaffolding we used to make sure we did a good job with the fiberglass on the very edge (top). We put two layers of the heavy cloth on the edges, wince we expect it to get the most abuse. We'll likely put a last layer of finer cloth on it to improve the finish.

Installing the cockpit walls

OK. We skipped a bit, but now we're finished with the basic hull and are starting to build up the cockpit. We've attached a .75" x .75" triangular strip to the hull to define the shape of the cockpit, then we nailed and epoxied the cockpit side panels to the triangular strip. In the rear of the craft, you can see the thrust engine mount (pictured somewhere above). Note that the side panels had to be built from two pieces of 1/8" plywood butt-jointed together with fiberglass. Nick is the guy that's been driving the pneumatic nailing gun.

Building up the styrofaom around the lift duct

Here you can see we have built up several layers of 2" styrofoam around the lift duct. The gaps have been filled with great stuff. Once hardened, this will help the 1/8" plywood duct retain its shape after the internal form is removed. This will also help prevent flying propeller parts from penetrating anyone standing nearby.

The firewall and dashboard help shape the cockpit top.

The curved top of the firewall and dashboard shape the top of the cockpit. The firewall, not called for in the plans, helps contain anything that the propeller might throw.

The cockpit top is installed and foam has been inserted into the gaps around the firewall.

This shot shows the cockpit top installed, trimmed, and fiberglassed in place. We planned on nailing the top to the cross supports and another triangular 'furring' strip, but the nails wouldn't hold. It turns out duct tape worked great for this. It was done in two pieces: the upper piece went from the "thin point" on each side of the duct up to the dash board. The lower piece went from the thin point down to the nose. This approach is much easier than one full sheet (and less wasteful of the limited plywood). We left a gap between the duct and the cockpit top and filled it with great stuff (a bit too much, perhaps). Turns out it trims nicely with the same hot wire we use on styrofoam. The only problem is it really stinks when it burns, so we have to do it with the garage door open.

Fit testing the lift duct

Meanwhile, down in the shop...

We continue to work on the thrust duct. The big arc of styrofoam shown earlier has now been shaped and 'kerfed' so it can bend around the thrust duct form. The next step, using epoxy to bind the foam to the plywood, needs more epoxy than we currently have, so it awaits a new shipment of epoxy before moving forward to the next step.

The spray foam and Styrofoam have been carved to shape and covered with one layer of fiberglass.

Our work with the hot wire is evident here. We carved both the Styrofoam and the spray foam (great stuff) to an aerodynamic shape using the hot wire. The great stuff is much softer - meaning spongy - which made it a little difficult to sand to the final shape. Once we were happy with the shape of the foam, we coated it with epoxy and overlapping sheets of fiberglass. We only put one layer on before dropping the two lift engine support boards in place. The discussion below will give you more insight.

The lift motor mounting boards are in place and the lift fan hub has been screwed to the duct shaping disk assembly.

Lots going on here. First of all, we measured, cut, and placed two 1x4 boards (for mounting the lift engine) into the lift duct and tacked them in place with a finishing nailer. We first rounded the edges of the board with a 1/4" roundover bit to both improve airflow and so it is easier to wrap fiberglass over the boards. We then put two layers of fiberglass over the boards and where the board joins the duct. We will have six layers of fiberglass on that joint by the time we are done. Before we do any more fiberglassing, we want to remove the disk in the middle. As it marks the center of the duct, though, we want to get the motor mount built before the disk comes out.

After the fiberglass was dry, we mounted to 1" aluminum angle brackets to the boards. These are made of 1/8" aluminum - which we hope will hold the weight and vibration. I used a 1/8" x 1" flat piece on the outside mainly for looks - I think washers would have been fine. We then put an axle through the two disks, slid the lift fan hub assembly down on it, and screwed it to the disk. We then removed the axle and set the engine, with another part of the mounting bracket, into the hub. The next photos might explain it better.

A mounting bracket for the lift motor was fabricated from 1/8 inch alimunum sheet, angle brackets, and flat strapping.

The part of the mounting bracket that attaches to the lift engine is shown in this photo. A flat sheet of 1/8" aluminum was cut to the desired shape so it would fit flat to the engine bottom. JB weld was then used to weld it to a pair of short angle brackets and a couple of pieces of flat aluminum for rigidity. Two longer cross members were then bolted and JB welded in place. The result is a bracket that should drop into the frame in the lift duct.

lift engine in place in the lift duct.

And this is the end result. The lift engine is now perfectly centered over the lift duct. The next step is to remove the mounting bracket, the two disks in the lift duct, and fiberglass the duct foam with a couple of layers and the engine support board joint with another four layers of fiberglass. I think we're going to need more epoxy.

The original form used to shape the lift duct is removed.

The engine and bracket were removed and Nick helped to knock out the original chipboard form used to create the lift duct. We've added a couple of layers of fiberglass to the opening. We need one more layer on the joint between the duct and the 1x4 motor supports, then the last layer of fine fiberglass so we don't leave any pinholes. You can see the skirt attach strip on the bottom of the craft, along with the skirt filling air deflector, through the list duct. Feels like we're getting close now.

lift engine in place in the lift duct.

Meanwhile, back in the shop (again)...

We have finally assembled the huge piece of foam together with the plywood thrust duct assembly (plywood and chipboard form). This has been taking up room in the shop for a couple of months and it's great to see it finally assembled. This took all three of us. Nick and I spread the epoxy while Chris controlled the rotation of the duct. Once it was epoxied (pretty thick coat), I fed the foam onto the duct form. Nick taped the leading edge while Chris guided the foam into place by watching a mark on the trailing edge (and taped that edge as required). The tape was temporary, as once it was in place, we wrapped the whole thing with three ratcheted nylon straps to ensure good contact between the foam and the plywood (squeezing out any excess epoxy - still waiting to see if any gets out). We put a small piece of plywood under the ratchets so we didn't dent the foam.

Thrust duct dry fitted in place.

After sanding the leading edge of the duct to shape (turned out to be easy), we put two layers of fiberglass on the leading edge and one on the trailing edge of the styrofoam so it had a firm binding to the plywood duct form (the ring, not the disk). We cut off the a flat spot to mount to the deck and we notched the leading edge so it fit around the two sides of the motor mount vertical panels. We then dry fit the pieces together. It looks good from this perspective...

Thrust engine mount is about a half inch too low.

...but this perspective shows the motor shaft is about one half an inch too low to be in the center of the thrust duct. This was an easy fix: just put a 1/2 inch piece of poplar under the motor and epoxy it in place.

The thrust duct mounted, foamed, and fiberglassed in place.

With the 1/2" piece of wood, the shaft now aligns with the duct. In order to ensure the alignment as we mounted the duct, we again put the bushings on the shaft and on the center of the duct. Once the duct was mounted on the shaft and appropriately squared to the hovercraft deck (both vertically and laterally), we foamed the duct in place. We used two 2-inch strips of 1/8 inch plywood to contain the foam on the sides of the duct - they were simply held in place with a brick on the outside and the foam pressure on the inside. Once cured, we trimmed the foam to what you see here (basically vertically from the deck both in the front and back of the duct). We then glassed over the foam and added the duct braces. We raked the forward edge of the braces a little bit to look more sleek and added a 1/4-inch thick by 1.5 inch wide strip to leading edge to stiffen the flimsy plywood. All of it was sanded round and the leading edge glassed. Now to mount the rudder supports.

This brace supports the rudder support bar.

The last thing to do before pulling the thrust duct form is to mount the upper support bracket for the rudders. To do that, we needed to cut and shape two pieces of wood for the braces and mount them in the duct. About 22 degrees worked for the angle to support the cross brace. We glued the wood in with gorilla glue then put two layers of fiberglass over it.

lift engine in place in the lift duct.

We've dry-fit the rudder assembly onto the rear of the craft. This lets us mark and cut the brace. Once mounted the upper support is mounted, the disks can come out of the duct. Note that we've used the easy way to make the rudders - a 1.5x2 foot piece of aluminum sheeting sandwiched between two pieces of 1/4x2 inch pine - with the leading edge shaped and fiberglassed. We've the material to build the rudders out of styrofoam, but will make those only of the performance of these rudders is inadequate. We do want to get this flying soon.

The thrust duct has the disk forms removed.

The thrust duct is now mounted and free of all forms. I am surprised at how stiff the final duct is. All that's left now is sanding everything, coating it all with resin, sanding it again, priming it, sanding it, and painting it. Anyone out there like to do bodywork? Then come the engine controls and propellers, which, by the way, are fiberglassed but not yet completely balanced. More on those steps below.

All exposed wood is coated with epoxy.

We finally have put the last coat of epoxy on the craft. The wood grain is really quite beautiful - it doesn't come out well in the photo, however. While the epoxy is great for bonding fiberglass to anything but plastic, it still breaks down under prolonged UV (sunlight) exposure, so we can't keep the craft this way (bummer). As a result, we have to sand the surface to rough it up so that an oil based primer will stick to it. Here Nick is sanding the side of the cockpit with a power sander.

The first coat of primer is applied.

Primer coat one complete. We used a roller for the primer coat. We decided not to spray it due to the mess and excessive tarps we'd need to seal off the garage around the craft. If it looks like an orange peel finish in the end, we'll call it "The Lemon Slice".

The coat of paint is on.

This is the final (base) color. The skirt is a matching color. For a paint job done with a brush and roller, it turned out better than expected. I did find a few pinholes, but fewer than we expected. We used a 4" foam roller for this and it worked quite well. As this is a bang-around vessel rather than a sports car, we took the advice of a couple of builders to use an oil based paint for both the two primer and two top coats. I've been walking around on the paint for a few days while mounting engines and controls and it is holding up quite well.

The engines and cleats are mounted, controls started.

This is just a quick update photo showing some intermediate steps. Here you can see the cleats are mounted - just over the landing skids, in fact. Both engines are mounted. We've fired up the thrust engine - it's rather loud in the enclosed garage. The boys are wondering if earplugs will be required for riding in this. Not visible is the rudder steering mechanism (not yet finished - but started). You can see a pair of beach chairs to be used in the cockpit. We'll be doing some center of balance measurements to determine where to mount them. We plan adjustable mounts for the seats so that we can shift weight forward or back to trim out the craft.

Profile view of the entire craft with the engines mounted.

Here's a profile view. In this perspective it looks long and sleek. Land speeder, anyone? The tools on the garage wall kind of interfere with seeing what's mounted on the craft, so we'll probably replace this photo when we get it outside again and take another photo. We also began mounting the engine controls at this point (photo to be provided).

The inner part of the skirt is attached to the hull.

This is an in progress view of attaching the skirt to the hovercraft. K&C marked, measured, and cut out the pieces. Dad glued it together then Kristin reinforced it with Kevlar thread (it wont rip now). We attached the inner edge of the skirt all the way around before screwing on the outer skirt. I had to pause and think about how to align the inner and outer parts of the skirt to ensure proper (even) inflation around the craft. In the end, it was fairly simple to mark the skirt and the craft to make this an error free operation. The concept is to mark the skirt outer edge first with the halfway mark, then mark the midpoint of the center mark and the ends, and repeat marking the midpoints until you have marks about every 10 inches. Then mark the inner edge of the skirt by transferring the marks from the outer edge so that the pairs are directly across from one another (perpendicular to the edge). Mark the corresponding marks on the outer edge of the hull, then use a square to transfer the marks to the inner rail while holding the square perpendicular to the outer edge of the hull. It might be hard to put into English, but it really made things simple.

The craft is hovering for the first time.

AT LAST! Here is the craft hovering during a quick check of the skirt alignment. All things looked well aligned, so after this photo was taken, we inserted more screws along the outside edge. We flew it (without rear engine) several times after that as well. Whew. At least we got it flying before the snow flew. These are some neighborhood kids that stopped by to help with the hard part. They're keeping it from drifting downhill into the cars. Nick is providing the ballast in this photo.