Follow us on Twitter Follow us on FaceBook Kitplanes Videos Get RSS Feed

Builder Spotlight

June 2019 Issue

Making Progress on the GlaStar Project

Fixing the tail, installing the engine, and hanging the wings.

It takes some extra help to muscle the wing into position on the fuselage. Luckily, we have plenty of other airplane builders at Chino (California) Airport to lend a hand. I gladly treated everyone to lunch afterward.

As I write this, it has now been a year since the first article in this series appeared. (See “Taking Over Someone Else’s Project,” December 2017, and “Getting Started on the GlaStar Project,” August 2018.) Progress has not been what I expected, but I always expect too much, which is becoming more evident as I proceed through the golden years. When my good friend Ed Zaleski and I built the Sportsman, we routinely worked 9- or 10-hour days when we could. Nothing close to that has happened this time, resulting in a much more modest rate of progress. What exactly is golden about the golden years? Anyway, work does proceed, and fortunately most of the work done previously has been useful.

Correcting Old Work

When I said most of the old work has been useful, it is worth noting that most is not the same as all. Upon closer scrutiny, some things that initially appeared to be useful proved to be deficient. I knew that the gap between the vertical stabilizer and the rudder was a little on the fat side, but the more I looked at it, the more I felt the need to correct it. The problem was that the gap only needed to be narrowed by about 1/8 inch. Ed came up with a great idea to form a consistent gap. He cut out a piece of 1/8-inch acrylic to fit the gap and then filled in the top of the fiberglass vertical stabilizer to meet it. Since the epoxy resin we used won’t stick to the acrylic, it was easy to break it loose when the resin set. This left a perfectly even gap between the rudder and the tail. A little sanding and filling will easily finish it off.

We have been using West epoxy for most of the fiberglass repair and alteration work on the GlaStar even though it was originally fabricated with vinylester resin. We have had good luck with it in the past, and it is easy to use. Instead of weighing resin and catalyst as one does with vinylester resin, we simply pump out the resin and catalyst we need.

A note to builders: Epoxy resin will mechanically bond to fully cured vinylester resin without any problem, but air-cured vinylester parts will require additional preparation before epoxy resin will stick to them. This is because air-cured parts are never fully cured, so the uncured surface resin must be removed (sanded off) before epoxy will adhere to it. The alternative is to stick with vinylester resin. In that case only normal surface preparation will be required to get a good bond.

With the rudder gap issue under control, the other fiberglass problem is that all of the screws that attach the fuselage to the cage were covered over with filler only and no fiberglass cloth. I didn’t notice this problem during the pre-buy inspection, but closer examination revealed that this work was not properly done. Now I need to grind off every spot that was filled and first place a small piece of fiberglass cloth over the screw head. The cloth will keep the filler from cracking later as the fuselage flexes in relation to the steel cage. The option was to leave the screw heads showing, but once the previous builder began the process, we really had to just keep going on the same course.

The steel cage bolts to the fiberglass fuselage in a number of places using countersunk screws. These can be left exposed or they can be covered up. However, if they are to be covered, it is important to place a patch of fiberglass cloth over each screw head to prevent later cracking. This was not done by the previous builder, so all of this work must be redone.

Aside from fiberglass work, the biggest repair so far is to the shear plates in the tail of the plane. These aluminum plates bolt to flanges in the tail of the fuselage to provide structural support to the empennage. They were drilled but not installed during the pre-buy. It turns out that they were not properly drilled and had to be remade. It seems likely that the door lock mechanism will suffer a similar fate, but we are not at that point yet. Some things look good until you actually try to use them. Then the poor workmanship reveals itself. This is just a hazard of taking over someone else’s project.

Installing the Engine

There is something very gratifying about installing a new engine on an airplane project. It seems like such a big step. In truth, it is just one of a thousand important steps, but it is what we in construction called “show work.” It’s like when you frame up the walls of a new house. It makes the house seem to just spring up out of the ground—it shows. Hanging the engine is like that. It is very noticeable. When someone drops by your hangar they go, “Wow, you have an engine now.” Of course, you really got the engine a month ago. It just now got moved from the floor to the engine mount.

Laying out and drilling holes in the firewall for some items before installing the engine made access a little easier. Here I use a step drill to make a hole in the stainless steel in preparation for using a chassis punch to make the 2-inch hole for the cabin heat valve.

Before we could install the engine, we had to finish and install the firewall. The previous builder had cut out and Clecoed together the stainless steel firewall, but it still needed to be riveted together. Since it was supposed to withstand 2000 F, we used Monel rivets instead of aluminum. These take a bit more force to drive, but otherwise presented no challenge. Luckily, it fit well into the fuselage opening. Next the firewall was bolted to the cage and riveted to the fiberglass fuselage. Since it was being attached to fiberglass, we used aluminum rivets for that. All firewall bolts used all-metal AN363 nuts.

With the firewall wide open, it made sense to install a few items that would be harder to reach later, so the eyeball fittings for the throttle and mixture went in as did the cabin heat valve. The last thing to do before the engine goes on is to seal the firewall with 2000-degree caulking. We used 3M Fire Barrier 2000. Please note that red RTV sealant is not acceptable for this since it is only rated for 500-degree continuous exposure.

When it came time to install the engine, we tried something different. We attached the engine mount to the engine first and then attached the assembly to the plane. In the past we had proceeded with the more traditional process of installing the engine mount to the plane and then attaching the engine to the mount. It seems like it might have been a little easier to install the mount to the engine first, but the difference was small. Here is how the process went.

With the engine out of its crate and sitting on the bench, we attached the two top mounts first. This went quite smoothly. Just as a reminder: When working with a Dynafocal mount, the gold or, in this case, thinner mount bushings go toward the plane, and the silver or thick bushings go toward the engine for the top mounts. I add thick and thin because the Van’s bushings we were using were not color coded in the traditional way. The top bolts should be snug but not fully tight to allow a little movement to get the lower bushings in place. We installed the lower right bushing next, which also went in relatively easily. Remember, on the bottom bushings the gold or thin bushing goes toward the engine and the silver or thick bushing goes toward the firewall—just the opposite of the top bushings.

A top engine mount bushing in place. Note that the thin bushing goes away from the engine on the top mounts. Installing these with the engine out of the plane seemed a bit easier than the more conventional method.

The best strategy for the lower left (last) bushing seems to be to slip the engine-side bushing into place and then tighten the bolts for the other three bushings. This will pull everything more or less into position. With luck, the last bolt will go in with only minor difficulty. A large drift punch or a bolt with the threads ground smooth can be used as a guide to get everything to line up. Finally, tighten all the bolts and insert the cotter pins.

The fixed-pitch Catto prop went on quite easily. We chose this to save weight and cost, and because Catto has such a good reputation for getting fixed props just right. The 78-inch prop should give good takeoff performance and reasonable cruise speeds. We thought about going with an 80-inch prop with less pitch, but cruise would have suffered in the trade-off. I’ll just have to see how it performs when we get it flying. It most likely won’t match the performance of the Hartzell Trailblazer prop that I had been eyeing, but the Catto comes in about 30 pounds and $10,000 lighter.

With the engine in place, it was time to install other firewall-forward items such as the master and starter relays, the battery holder, the gascolator, and the oil cooler. The oil cooler actually is on hold until I decide how I am going to get air to it, but the other work is proceeding.

Yours truly torques the bolts on the prop extension needed to fit the Catto prop to the GlaStar cowl. It was very gratifying to get the engine and prop installed. It really makes the project look like a real airplane.

Hanging the Wings

Attaching the wings to the fuselage is definitely “show work.” It is another very visible and important step in the assembly process. After lowering the wings from their perch on the mezzanine railing, we attached the strut to each wing first. Then we rounded up the gang to actually attach each wing to the fuselage mounting points. As we were going over the hardware, we found that we had two AN8 bolts where we should have had AN7 bolts. I had ordered the hardware off the Sportsman parts list, assuming that it was the same as for the GlaStar (most things are the same), but it turns out that this was one of those exceptions. One of our friends said he thought he might have two AN7-12 bolts in his hangar, something I seriously doubted, but I’ll be darned if he didn’t have just what we needed to save the day. With all the hardware in hand the wings went on without much drama. It was nice to have a lot of help.

We needed to attach the wing strut before hanging the wing on the fuselage. This is mostly blind work in tight spaces. My longtime partner in crime, Ed Zaleski (left), is assisted by RV-8 builder Eddie Rohwedder. Eddie’s plane should be flying before you read this.

With the wings in place it was time to check the dihedral and the angle of incidence of each wing. Fortunately, the dihedral was acceptable because the struts had already been cut and drilled by the previous builder. An error here would be expensive to correct because it would have required buying new struts.

We used a SmartTool digital level to measure the angle of incidence of each wing. We made some spacer blocks for the top of the level to allow for the curvature of the bottom of the wing. With a standard thick washer top and bottom on the starboard side hinge, we got both wings within less than 0.1-degree difference.

The angle of incidence is set by placing washers in the appropriate places at the starboard wing pivot hinge. A smart level and a little effort took care of this rather easily. This is a critical step because any variance in angle of incidence will make the plane fly wing low on one side or the other and tend to make it drop a wing as it stalls, two things we did not want. Our acceptable tolerance is a difference of not more than 0.1 degree.

By stretching a string from wing tip to wing tip, we can measure the drop to the root of the wing and calculate the dihedral angle. Ours turned out to be just a hair shy of 1.5 degrees, the design dihedral angle, but we’ll live with it.

What’s Next?

We have a lot of fiberglass work ahead of us. Fortunately, Ed likes fiberglass work. I can do it, but I would be hard pressed to say I like it. In any case, we have plenty of it coming soon. The wing fold hatches must be remade to fit the larger Sportsman flaps. We originally thought that this would be a serious project, but now with the wings on, it doesn’t look like it will be too hard.

The really big fiberglass project will be fitting and reworking the cowl. Besides making the cowl fit, it must be widened and have an RV intake scoop added to work with the RV intake system. The original GlaStar cowl was designed to fit the Lycoming O-320 engine. The O-360 engine we are using is just a bit too wide to fit in the cowl. This means we will need to split the cowl open from the front and add about inch of width to each side. The challenge will be reshaping the intakes to look right after widening them. The RV scoop addition is pretty straightforward, so we don’t expect that to give us any grief. It is just more fiberglass work to do. All of this promises to add a thick layer of fiberglass dust to everything in the hangar for the next month or two.

When we aren’t getting fiberglass everywhere, we have rigging to do and fuel lines to run. We expect the rigging to be extra difficult because the wings are closed up. Normally the aileron and flap cables are fitted before the top wing skins go into place, but we will have to work inside closed wings. Maybe it won’t be so bad, but we have been warned to expect the worst. We will just have to see.

With the plane in level flight position, it has great clearance under the wing due to the big tires and extended gear legs, even for your 6-foot-3-inch writer. There is still much to do, but with the engine, prop, and wings on, it is really starting to shape up.

The Panel

The instrument panel will be the last big project and expense. We are thinking about Garmin G3X touch, but the blank instrument panel we got from the previous builder won’t work with that equipment. We can either secure another larger panel or modify the one we have. No decision on that yet or on the exact equipment we will install. Although we don’t really use it, we are looking to get IFR capability because it adds considerably to the value later. Decision time is coming soon if we want to finish this year. I’ll let you know where we end up in the next article.

Download File

Homebuilder's Portal by KITPLANES
Photo by Richard VanderMeulen
N958PD first flightSubSonex Serial #21 (N958PD) made its first flight out of the Dayton Valley Airpark in Nevada on the 24th of March, 2019. Winter kept trying to hang on in the Sierra Nevada mountains, but the weather smiled on us, I had a great crew that made themselves available on short notice, and we got "Loki" …
RANS S-6 N646GTN646GT started with six members in an LLC, and currently has four left, 10 years after we bought the kit. One of our team is Richard Hogan, the designer of the new Commutercraft. Our RANS S-6 started construction at a hangar at the now defunct Mathis Airport in Suwannee, Georgia, then moved to another hangar …
Phillabaum-HelicycleFirst flight of my Helicycle helicopter was on January 31, 2019 after two years of part-time building. Previously, before I retired, I built an RV-8 and a Zenith 750. The Helicycle was my first post- retirement project. I discovered I had more building time when I was working. Now trips, lake time, skiing, etc. interfere …

lycoming-thunderbolt-o-360-redAt Sun 'n Fun 2018 we interviewed Jeff Schans, Lycoming's Sales manager and we asked about the popular Thunderbolt Engines. Watch the short Product Minute video as Jeff describes the benefits of the Thunderbolt product line and what it means for the aircraft homebuilder.
Dynon Avionics' latest-generation SkyView integrated avionics called the SkyView HDX has a newly designed bezel and user controls for easier use while flying in turbulence, plus brighter displays and a reworked touch interface. Larry Anglisano takes a product tour of the HDX with Dynon's Michael Schoefield in this video.
Video Archive


All mailed correspondence, including subscription invoices, renewals, and gift notices, will bear our address:
PO BOX 8535
Big Sandy, TX 75755

Third parties claiming to be selling KITPLANES subscriptions are not legitimate.