Ch 20 - Winglet and rudder - Part 3

Top winglets (61.2 hrs)

With the issue of the plans' error behind me, and the foam cores already glued together, it was finally time to get dirty again. 

First though, I would bond the cores to the table to prevent any twist from happening during the curing process. I used a couple of foam blocks and a straight 2x4 screwed to my table to construct a pretty sturdy foundation.

Using the foam the winglet was cut from as a base for glassing

Foam duct-taped over up for the front, and a straight 2x4 to support the trailing edge.

Weighing everything down while the 5 minute epoxy cures

The fiberglass layup schedule for the winglet is pretty straight forward. Two UNI plies corner to corner crossing each other, plus an additional BID ply in a 14” x 18” area where it will later connect to the main wing.

Layup schedule for both sides of the winglets

I started with the right winglet first.

I accidentally cut from the wrong roll (BID vs UNI) so all this glass was later chucked

After starting over, I laid the two crossing plies of UNI over the winglet.

Made sure the final BID ply didn't cover the custom sized rudder

Peel-plied the whole thing

The next morning it was time for cleaning up, which as usual took all day.

The plans have you trim the glass right at the leading edge apex

 

I like to roll the glass over the other side an inch or so even though cleanup is harder

I like having 4 plies instead of 2 at the very leading edge should I ever hit anything

I also decided to cut the winglet down to size at 49” along the foam trailing edge. 

Winglet sanded

The 49" is still one inch long, but I'll trim that after mounting it to the wing.

Used the Japanese saw for this job

Easy peasy

You might remember from my last post that the winglet is only 48” along the foam, plus a one inch urethane foam cap.

Original winglet plans

So, why did I do it this way?

You see, step #2 of the plans has you add the 1” urethane foam cap, and step #3 has you measure the distance from the aileron inner forward corner (WPRP) to the rear tip of the winglet (measurement C in the picture below).

Measurement "C" (118.35") is taken from the top of the urethane winglet's cap (48" + 1")

Since the 1” urethane foam cap gets installed before the C measurement is taken, and because I would like to do a backward sweeping top cap to the winglet, I chose to cut the winglet one inch long, as if the regular urethane top cap was already installed. This is only for the measuring purposes during winglet installation of course. Once the winglets are mounted to the wings, I will cut the top one inch, and install a more artistic 1” inch tall top cap.

James Redmond's beautiful Berkut

Next, I took care of adding micro to the trailing edge. This is normally done while fiberglassing, but I decided to peel-ply it instead, and take care of it at a later time, when the trailing edge is stiffer and more able to resist my pressing on it.

Trailing edge taped with pure epoxy applied 

 

Adding dry-micro to the trailing edge's juggle

    

With peel-ply over the dry-micro

A day later I ripped the peel-ply off and sanded the micro smooth. That’s when I noticed the gap.

This got me stumped

I couldn’t understand this since I thought the addition of this micro was meant to create a smooth transition back there. 

Did I screw up fiberglassing?

I checked the original shape of the left winglet trailing edge foam, and found the same problem.

Checking the foam on the left winglet

This one has a gap too

This was getting weird. 

Did Eureka CNC screw up hot wiring the foam?

I dug up the plans for a closer look, and this is what I found…

I'll be damned! The plans have a gap as well.

I guess nobody screwed up.

I felt much relief that I hadn’t caused the problem, but now I totally questioned the rationale behind the technique. Putting micro on the trailing edge during glassing (or afterward in my case) does not even come close to filling the gap in the trailing edge area, and so it doesn't save any future finishing work, rather it adds another step to an already long day.

I resolved to skip the micro trick from now on (it is not actually called for in the plans anyway), and with that in mind I moved on to the left winglet.

Reused the same 2x4 and foam blocks to support the other winglet 

Filling in all the major foam imperfection with dry-micro

Using UNI right off the bat this time 😄

Ply #2 down. Didn't catch a photo of the smaller BID ply

 

Unfortunately I ran out of the sheet peel-ply, and made do with a roll of the smaller 3" (8 cm) kind.

Sometimes the lack of proper planning creates the need to improvise

This worked well, but the next day it proved to be worse at cleanup time. Ripping the peel-ply off left a bunch of strands on the surface, and there was a small ridge of epoxy where any two plies overlapped, requiring quite a bit of sanding.

This should teach me to plan ahead next time

Transitioning the 2 plies to the bare foam smoothly takes time

With cleanup completed, I cut the left winglet to 49” also.

Trying the vibration cutter this time

The outer surface of both winglets was now done.

Looking sharp

Better finish these guys before the foam gets any more dents in it

I moved to the inner curved surfaces next. First order of the day was removing the fishtail and the peel-ply embedded in the trailing edge, including the staples that held it to the foam during glassing.

Cutting off the excess foam (aka fishtail)

You can see the staples sticking out of the foam

Ripping the trailing edge peel ply

"Now, that's a clean trailing edge!"

Like with the canard, the winglets are used to house antennas internally, where they do not contribute to the overall drag of the plane. So, while the canard was perfect for hiding the horizontally polarized NAV antennas, the winglets are ideal for the vertically polarized COM antennas, once again leveraging the radio transparency of fiberglass structures to our aerodynamic advantage.

First though, I had to make them.

How the Com antennas are made

Antennas bits and tools

RG400 coax cable stripped and ready for use

Coax soldered to antenna dipole

Baluns prevent reflected transmission energy from traveling back to the radio

Using hot glue to fix the baluns position on the coax cable

These antennas are very delicate, and easy to damage as you’ll see shortly. The good news is they can be repaired almost as easily.

Antenna location plan

Cutting a trench for the coax cable

Made the trench narrower than the cable so that the foam would hold the coax down

Coax exit point

All parts in place

Left enough loose coax to go through the entire wing and emerge at the center section spar

I modified both winglets in this way in order to have two COM radios down the road, same as JT now has.

Com#1 and Com #2 antennas are in place

With the antennas mounted, and continuity verified, I moved to glassing the left winglet.

Once again I wrapped a long straightedge with duct tape, and used it to support the trailing edge, just in case. I also raised the entire winglet off the table and over the edge in order to be able to glass the leading edge overlap more easily.

Not taking any chances with the straightness of the trailing edge

Hanging the leading edge over the table edge to ease glassing the overlap

Before any glassing could be done though, I had to fill various holes, and the coax cable trench I cut earlier. Dry-micro always worked well for small holes, but filling the trench used up a lot of it, and although I worked very hard to reestablish the winglet contour, I was never able do it to my satisfaction. 

This means that the fiberglass would turn out slightly flat over the trench, and the proper rounded contour would have to be achieved only at finishing time. Quite alright, but not ideal.

Dry-micro over the big depressions

Micro-slurry over the foam

First ply on. Not too happy about the dry micro sag in the coax trench.

Another shot to show the coax exit pocket unevenness

Ply #2 going on

BID patch added where the winglet will attach to the wing

Got more peel-ply sheet in

I spent the whole next day cleaning up.

This looks a lot better than when I used the 3" wide peel-ply

The outer facing side of the winglet is going to need some cleanup at both ends

This overlap is going to take time to blend in

First though, we need to cut the extra material

The trailing edge needs to be reestablished using the straight edge and a cutting tool

Excess fiberglass trimmed off the rear of the winglet

That's a good looking trailing edge

Wouldn't want to run into this edge

Glad to see the trailing edge coming out this straight

The left winglet difficulties I encountered inspired me to try something else on the right winglet. I purchased pour-foam to  fill the coax trench and any other depression, instead of having to rely on dry-micro again.

I have never used pour-foam before

The two part foam expands rather quickly to thirty times its original volume, so a little goes a long way. This stuff is also super tacky, and it sticks to everything, to the point that mixing sticks, cups, and anything else used cannot be reused. It also dries quickly, and subsequent pours can be made after just ten minutes.

To make sure I wouldn’t run into any issues, I went above and beyond to protect the winglet from this stuff.

Setting up the right winglet. Note the straight edge under the trailing edge.

Creating a dam for the X-30 foam test pour

My first time using it did not go so well. I should have mixed a little more than I did, making the foam a little denser, instead of trying to spread the little amount I had over a large space. I resolved to use more next time, and fill the voids later with a bit of dry-micro.

I must have overestimated the pour-foam expansion

Unfortunately, while removing the tape that was holding it down, I broke the antenna.

Removing the winglet protective layers ended up ripping the fragile antenna foil 

Luckily, fixing it was a quick and easy job.

After trimming the pour-foam down, and getting the roll of copper foil back out.

New copper foil spliced in and tested

Detail of the soldered splice

Another continuity check revealed no further problems, and I decided to give the pour-foam another shot at a couple more low spots.

Trying the pour-foam again

Got a nice build up this time

Both spots cut and sanded back to proper height and shape

This worked quite well at reestablishing the curved contour, so I decided to give it another try in another area that was still slightly low.

I am getting hooked on pour-foam now

"This is fun!"

"Loaf of bread anyone?!"

It cut and sanded perfectly, and the proper airfoil profile was easily reestablished.

This technique was easier to use and worked much better than the dry-micro

The only thing left to do now was glassing, and for the first time in a long time I had a helper.

I don't always accept apprentices, but when I do they must have impeccable resumé. 

Now that Chris owns a flying LongEZ, he could benefit from some hands-on experience fiberglassing airplane parts. So this ended up working well for everyone, Gina caught a break, I got some help, and Chris practiced fiberglassing on a much larger scale than one gets at an EAA workshop.

Getting started with the UNI

I'm gonna make a builder out of this fly-boy

The session went well, and was a lot shorter than my previous ones. Chris was satisfied enough that we might see him slinging epoxy again on this blog. 

Unfortunately he was nowhere to be seen the next two days, when the cleanup took place 😂.

Blending the overlap to the rest of the winglet

As slow as I am when building, cleanup is an even slower process for me, and I just couldn’t ask anyone to endure something I myself wish I didn’t have to do. More importantly though, it is relatively easy to damage the underlying structure if one is not careful with the precision sanding required at times, and since I am the guy who is going to be in the front seat of this plane one day, I feel it is my responsibility to perform this important and delicate task.

Blending in progress

Both winglets completed and going into storage