Upper Fuselage Joint – Part 2

The last part of the fuselage joint, at least the inside work, involved the hatch area up in the nose.  This panel will eventually be cut out and turned into a removable access hatch to get to the electro-mechanical equipment that will be installed up there.  Or a jacket.

The process here was the same as before.  First a single ply across the joint, then some foam core to fill the joint, then 2 plies over the whole mess.  The only difference in this area was the use of 1/4″ high density core instead of the 3/4″ core I had be using.  I like this 1/4″ stuff better anyway since it’s easier to shape.

After the basic layup was complete, there was an extra step to build and attach 2 stiffening ribs to the inside of the hatch.  Since it will be removable, these will allow the hatch to keep its shape.  The stiffeners were built out of the same 1/4″ high density foam core, then bonded in place.  The edges were rounded off, and the joint radiused with an epoxy micro mixture, then 2 plies of S2 glass were added over the top.

The S2 glass is a special weave that allows it to form nicely over tight areas and compound curves.

Overall, I think the installation came out pretty good.  In the below photos, the fuselage is upside-down for the first 3 photos, and upright for the last one.

Upper Fuselage Joint – Part 1

Joining the fuselage halves at the upper joint was identical to the lower joining process, so I won’t bore you with all the details… I’ll just put a few photos at the end.

But before I could actually get started, I decided it would be a good thing to invert the fuselage first.  I guess I could have done the job over-my-head, but that would have been messy.  And insane.

So in enlisted my buddy Mike Wagnon to help me invert it.  While not particularly heavy, it would have been difficult for me to do myself, given its size.

Once inverted, I build a little saw horse cradle to support the nose so that it wouldn’t tip over while I crawled around inside.

So then it was back inside the fuselage for me (oh, my aching back), to repeat the process of ply – core – plies.  Again, the most time consuming part of the joining process was constructing the foam core sections so that they nestled just perfectly into the joint valley.

Lower Fuselage Joint – Part 3

With the lower fuselage joint nearly complete, the last step was to add 2 plies of EBX-1200 (if I recall correctly) fiberglass cloth over the top of the joint.   With my ever-increasing fiberglassing skills, this task wasn’t particularly challenging.

For smaller layups, I found that wetting the fiberglass cloth with epoxy resin was easier when accomplished on the work bench.  But for larger pieces like these, I prefer to do the layup in place.

The first ply of EBX-1200 cloth was laid up over the exposed foam core, after filling the open cells with a wet micro mix, of course.  A 1 – 2 inch overlap on all sides was the standard.  Then another ply on top of that one, with anothere 1 -2 inch overlap.

Apply a final peel ply, and let the whole thing cure.  Voila! Lower joint complete.

A side note here for those of you with bad backs: don’t build this airplane.  I don’t have a bad back, but after several hours hunched over in the fuselage working on this joint, I sure felt like I did.  And I was only half way done.

Also, get yourself a set of knee pads.   You’ll thank me later.

Lower Fuselage Joint – Part 2

Now that the lower fuselage joint was essentially bonded together, the next step was to strengthen the joint.  This is achieved by filling the “valley” with rigid foam core, then glassing over the top.  For this project, the Builders Manual called for medium density 3/4 inch core.  Although lower density than the 1/4 inch stuff I was using to make bulkheads, I found this foam more difficult to work with.  It was difficult to sand, although responded fairly well to rasp files.

Trying to fabricate one long piece of foam to fill the entire ~10 foot long valley would be have been ridiculous, so I attacked it 16 inches at a time.  Each 16 inch (approximately) section of core was carefully filed, shaped and sanded so that it fit exactly into the valley.  This was a very iterative and time consuming process of shape-test-shape-test-shape.  Once I was satisfied with the fit, the foam section was bonded in place with a wet mix of epoxy micro, with some dead weight to keep it in place.  I prefer Play Sand for this job.


 Then the process was just repeated over and over until the entire lower valley was filled up to the area of the NACA inlet.

Lower Fuselage Joint – Part 1

With all the bulkheads now complete, I could move away from the workbench and onto the airplane as I began Section 2, Chapter 9 of the Builder’s Manual.  As the title suggests, this is the point where I permanently bonded the 2 fuselage halves together.

Let me preface the chapter by saying this was a major task, and took way, way, way longer than I imagined.  I partially attribute this to the weather turning colder, and my generally lower enthusiasm for working in a cold garage.  Regardless of my own desires for comfort, working with epoxy resin requires an environment of at least 70°, so a space heater was now a necessity if I was to continue working through the winter.  But you can read about that in my Building in the Winter entry.

Before actually bonding anything together, the normal first step would have been to align the 2 fuselage halves. Recall, however, that I took care of that a few months earlier when I was still awaiting the arrival of my epoxy (see Aligning the Fuselage Halves) so I could jump right into the joint (sounds like a blues song).

The basic concept of the fuselage joint was pretty straight forward;  a single ply of fiberglass is laid up across the joint valley, then some foam core is bonded in, and another ply of fiberglass covers the whole mess.  After that, I could cut off the flange (mohawk) and fill the remaining gap flush.  Sounded simple enough, but it sure took a long time – highly attributable to my meticulous manner I suppose.

Typical of composite construction, when two pieces are to be joined together, the core is tapered down to zero so that a solid laminate joint can be achieved.  I referred to this area as the “valley.”

I started by sanding the valley area by hand to remove any imperfections and high spots.  Next, some light weight filler (epoxy/microballoon mixture (“micro”)) was applied to any small voids or low spots so that I wouldn’t have any trapped voids in the joint.  After a full cure, the filler was sanded smooth and flush.

Then comes the messy part: a heavy ply of fiberglass, about 12″ wide, is laid up across the joint (is it “laid up” or “layed up?”) and allowed to cure with a top layer of peel ply.  After cure, the peel ply was removed.  I started at the nose and worked my way aft over the course of a few days.

Layup over the lower fuselage joint.

Nose Gear Plates

The Nose Gear Plates, or NG-2 Bulkheads, are pretty much like the other bulkheads with a few additions. Construction began the same way: cut out the high-density .25″ thick foam core and sand it to match the template I fabricated a few weeks earlier.  Actually, in this case I built 2 cores, because both left and right parts are required, and they need to be perfect mirrors of each other.   “Why?” you may be asking.  As the name implies, the Nose Gear Plates serve as the attach points for the nose landing gear and its retraction devices.  If the parts are not alike, the gear installation will be skewed which could lead to bindin

g during retraction (or worse–during extension).

Adding solid hardpoints.

The inboard surface of each part was glassed with 2 plies of EBX-1800 cloth, cured, trimmed, and sanded to match the template.  On each part, there are 4 areas (bolt/attach points) that need to be strengthened by boring out the core and building up the area with glass plies instead.  This operation is a bit tricky because the hole saw needs to be positively controlled: cut too shallow and you don’t get all the core out.  Go too deep, and the hole saw cuts into the inboard plies that I had just layed up.  Using a drill press, and setting the cut depth using some scrap pieces of core helped a lot.  Then came the tedious job of cutting a bunch of circular pieces of glass.  Thousands of them.  Maybe tens of thousands.  Well, maybe I’m exaggerating a bit.  But I’m sure it was at least 100.  The circular pieces were then layed up into the holes, and the outboard sides of the parts were glassed.

But wait, there’s more.  On the outboard side of each part, 4 aluminum crush plates are bonded, faired around the edges with an epoxy/flox slurry, and glassed with some Bi-Direction (BID) cloth.  On the first crush plate, I made the mistake of sloppily applying the flox and allowing it to cure before adding the overlay BID ply, thinking it would just take a few minutes to sand it down to a smooth radius.   Wrong.  As I mentioned before, cured flox is some tough stuff.  I’d say it took about an hour to grind/file/sand that stuff down to a nice radius.  Needless to say, I didn’t repeat that mistake when installing the other 7 crush plates.

Original NG8 plate.

But wait, there’s still more.  On the inboard side, yet another aluminum crush plate (NG-8 for those of you taking serious notes) gets installed at the nose gear pivot point.  Unlike the other crush plates, this one is mechanically fastened rather than bonded.  The NG-8 plates that were supplied with the kit were in

New NG8 plate installed.

pretty bad shape: scratched, holes out-of-round, lousy countersinks, and generally not the type of parts I wanted on my airplane.  I faxed a drawing of the parts to my Dad, and he built me 2 new ones at his workshop which looked – and worked – great.
Once all the parts were installed, the last step was to mate up the 2 bulkheads, and open full size holes through both parts simultaneously.  The drill press strikes again.  That’s about it.

Nose gear plates complete.


The Firewall construction was not unlike the other bulkheads I had recently completed: the core was cut to match a template then fiberglassed on each side.  The main difference here was that I used a .25″ thick birch plywood sheet as the core instead of high-density foam sheet.  This made the part substantially heavier than the other bulkheads, but since this is what the engine ultimately mounts to, weight = strength = good.

None of my tools were suited to accurately cut a highly curved piece of plywood, so I borrowed a jigsaw from a friend which worked quite nicely.  Plywood doesn’t sand as easily as foam (duh), so I cut the piece as close to the template line as possible to minimize the amount of edge sanding required.  Glassing was as before, except these were the largest pieces of fiberglass I had used yet – roughly 3’x4′. The other significant difference between this part and previous bulkheads was the inclusion of integral hardware.  (Hey, I finally got to use my cool Ingersoll-Rand pneumatic drill.)  If you look closely at the photo, you’ll notice 4 silver dollar sized circles in the “corners” which are the engine mount attach points.  They are actually .25″ thick aluminum plugs which were bonded into the plywood before the fiberglass was applied. Firewall screws Additionally, after the first side was glassed I bonded in 6 (.190″) countersunk screws (3 on each side) which serve as attach points for the rudder cable pulleys.  (Hey, I finally got to use that 100 degree countersink I bought at Boeing Surplus.)

The hardware was bonded to the plywood using a slurry of epoxy and cotton flox (which is finely ground cotton–almost like a powder–for those of you not “in the know”).  I had some doubts about the structural integrity of the screws bonded in with this mixture, and their ability to withstand a decent torque.  But once it cured, I was a believer: cured flox is some tough stuff.  The heads were flush on the other side, and glassed over, so unavailable for a screwdriver.

That was about it.  The completed part was set aside for installation sometime in the future when it will get fire-proofing materials applied and a metal face sheet.

FS-37 Bulkhead Part 2

As mentioned in an earlier post, the FS-37 Bulkhead – one of the more complicated layups – was my first casualty.  After the initial trim and final cure, I was sanding down the edges to meet the final contour.  The top of the part is straight, so I figured I’d sand that with the belt sander that my Dad gave me (surplus from his shop).  That damn tool was way too agressive for this composite part.   It sanded right through the edge in no time, and made a neat concave shape.   By the time I finally had a straight line, I had trimmed away much more material than planned.  Now what to do?  I had several hours (not to mention materials) invested in the part – I hated to throw it away and start all over again.  It may still be okay.  I could wait until installation and see how it fits, right?   Well, probably.  But I had just enough material to make another part, so I convinced myself to scrap this one and start all over again.  After all, this part takes 90% of the canard lifting loads and transfers them into the fuselage structure, so it’s not a trivial part.

It was a bit of a setback, but the second part came out even nicer than the first.

– – – – –

Hindsight note: After installing the canard (much later), the original FS-37 bulkhead with its slightly concave top surface would have been just fine.  A little filler between the canard and the top of the bulkhead would have been fine, provided the edge distance of the crush plate inserts was still adequate.

Forward Fuselage Bulkheads

Finally, armed with a fresh shipment of epoxy resin, I could actually start building parts.

This first part I built was the FS-16 Bulkhead.  (FS-16 represents Fuselage Station 16, meaning it will be installed 16 inches aft of aircraft datum.)   The process for building all the bulkheads is essentially the same, but the FS-16 bulkhead one is the most straight forward, which is why it’s a good starting point.  First, I obtained a piece of high-density, .25″ thick foam and trimed it roughly to the shape of the FS-16 template that I built a few weeks earlier.  The foam was then sanded/filed until it exactly matched the contour of the template.  Hey look! I made a bulkhead core.

Next, I carefully weighed out 3 parts of epoxy resin and 1 part of hardner, and mixed in some microballoons to form a light-weight slurry.  This slurry was then spread into the surface of the foam to fill any open cells, which in turn cuts down on the total amount of epoxy required for the layup, and ultimately leads to a lighter finished part.  For subsequent parts I used the Clearstream (Hudco) 8040 filler in lieu of the microballoon slurry.  Had I not taken that Aircraft Composites Course a few month earlier, I would not have learned that trick.  The SQ2000 Builder’s Manual doesn’t go into detail like that.  It specifies which materials to use, ply orientations, etc., and covers some techniques, but it isn’t an all encompassing guide to aircraft construction.  But I digress…

While the slurry was still wet (tacky), the appropriate type of fiberglass cloth was cut slightly larger than the core and layed in place.  A new batch of epoxy was mixed up (no microballoons this time) and spread over the dry fiberglass with a brush until nicely wetted out.  A small texture roller can be of assistance here to force the resin into all areas of the fiberglass cloth.  Then, while the epoxy is still wet, a layer of peel ply was layed over the fiberglass and wetted out.  The peel ply is a thin, nylon cloth which epoxy has a hard time sticking to.  After the epoxy has cured, the peel ply is removed (“peeled” away–hence the name), leaving behind a nice, even texture which is perfect for subsequence epoxy bonds.  Even if no further bonding is necessary, the use of peel ply is encouraged because it leaves such a nice finish.

I did the layup on a 1″ thick piece of particle board with Formica facesheets.  This is a nice, smooth, flat working surface.  During the over-night cure, I sandwiched the bulkhead between 2 of these boards, and placed some additional weight on top to make sure it would cure flat.  Warped bulkheads can only lead to trouble down the line.  Since it was the middle of summer, I didn’t need to worry about heat during the cure: the garage never got below 70 degrees.

The only tricky part about this process is the trim.  The manual recommended that I let the epoxy cure for a few hours until it’s tacky, then knife-trim the excess fiberglass to the shape of the foam core.  This didn’t strike me as a very good construction technique, and it limited me to glassing one side only.  If I glassed both sides, then allowed a full cure, I could just cut/sand/file the excess glass.  So that’s what I did.  Turns out that cutting the cured glass to the exact shape of the foam, without damaging the foam was harder than I thought.

For the next bulkhead, I followed the manual’s advice and knife-trimmed the excess.  Timing becomes an important factor here: if you wait too long after the epoxy has been applied, the composite becomes too hard to cut with a knife.  If you don’t wait long enough, it’s too sticky and gums up the blade.  And the ambient temperature will affect this window of opportunity, so if you’re heading off to your girlfriend’s house for dinner on some warm summer night after a layup, be sure to warn her that you have to leave in a couple hours to “trim your bulkhead.”  I’m sure she’ll understand.

And so it goes for the other bulkheads.  The shapes and layups become more complex, but the process is the same.  The FS-37 Bulkhead has numerous plies and 2 cores, and took quite some time to build, so it was the most challenging for a new builder such as myself.  As luck would have it, this part was also my first casualty, but you can read about that here.

Aligning the Fuselage Halves

The SQ2000 fuselage comes in two pieces – left and right halves – which makes perfect sense since it’s built in a mold.   The Velocity fuselage is also a two-piece affair, but I believe theirs is split horizontally, forming an upper and lower half.  Does one design have an advantage over the other?  Hard to say for sure, but it seems to me that the highest stresses will be travelling down the centerline of the fuselage (butt line 0) so that would give the edge to Velocity.   I’ll have to be very cautious doing the layup of this complex and important joint to make sure everything comes out perfect.

The fuselage halves arrived from the factory bonded together with structural adhesive.  The manual is very clear to point out two things:

  1. It is critical that the two fuselage halves be in perfect alignment.
  2. The factory doesn’t do this before they bond the halves together.

To the casual eye, the two halves looks pretty well aligned, and indeed they are.  But they weren’t perfect: off by about 0.125″ I’d say.  It’s too bad the factory didn’t take a few extra minutes to do the alignment, because it would save the builder several hours.

The first step was to break the blobs of structural adhesive (dark grey circles visible on the flange (mohawk)).  A hammer and chisel worked pretty well, although I did delaminate a few plies here and there.  Not to worry though, the flange gets cut off eventually.

Once all the bonds were broken, I had a little help to remove one half of the fuselage (it’s not heavy, but kind of big for one person to handle) and set it aside.  You can see the grey adhesive blobs pretty well in Photo 2.

Next I took my trusty angled grinder and ground off the remaining adhesive.  This was pretty tough stuff.  I also sanded away any loose plies that my chisel victimized.  This exercise was repeated for the other fuselage half, of course.

Side note: this was about the point my neighbor came down and said “What the hell is all that noise about?”

With a little help again, the halves were once again mated up, and carefully aligned.  There are several lines molded into the fuselage to assist with the fore/aft alignment, but nothing for vertical alignment… you just have to go by sight and feel.  Once I was satisfied, I drilled holes through the flange, installed fasteners, and just be safe, hot-glued the flanges together.  And no, I don’t wear suspenders with my belt.

By the way, special thanks to my Mom for taking the photos (featuring me) and helping me move around the fuselage pieces.   She was passing through Seattle that weekend on her way from California to Montana and I didn’t hesitate putting her to work (payback for all the dishes she made me wash as a kid)