This blog is for entertainment purposes only, and is not meant to teach you how to build anything. The author is not responsible for any accident, injury, or loss that occurs as a result of reading this blog. Read this blog at your own risk.

Friday, April 18, 2014

Nose and nose gear - part 3

Glassing NG-30 inside out (12.8 hrs)

Have you ever had one of those moments when you thought you were really clever, only to find out you had it backward all along?

While busy congratulating myself on solving the interference issue between NG-30 and the gear motor, I didn’t even notice that I was applying the fix to the wrong side (outside) of NG-30!!!


Initial recognition of the problem on the inner surface of NG-30.

Let me explain a little further. 

The glassing Wade and I did in March (see Nose and nose gear - part 1) was on the inside of NG-30 (per plans), even though my brain kept looking at it as the outside of this symmetrical (at the time) piece. 

Happily glassing the inside of both NG-30s, while I'm still thinking it was the outside.

Since the gear motor goes between the NG-30s, once the inside was glassed, it should have been no longer possible for me to relieve the foam structure on NG-30s' inner sides

Note how the outer skin is glassed here, contrary to the real NG-30.

But this didn’t stop me from thinking that I could, and going for it.

Had this been NG-30's inside surface, it would have been a perfect picture, sadly it wasn't.

Theoretically, in order to be able to relieve and glass the proper side (inside)I should have been thinking about it before the initial fiberglassing session even began, but that was before I knew I had any fitting issues, so by the time I figured it out it was already too late.

Still with me? 

Ok! So, what’s the big deal? If the parts are symmetrical, why don’t I just swap them out and be done?

Good question. You see, the reason the inside is glassed first is to ensure there is a perfectly flat surface for the hardware to mount on later (I didn’t appreciate this finer point until now)

Filling up the “moon craters” with the 15 fiberglass disks called for by the plans leaves a small depression that, while not an issue on the outside, it would be troublesome if it were on the inside.

Short nose NG-30 showing the normal depressions left on the outer skin after glassing

This caused a lengthy “analysis paralysis” period while I contemplated my options. 

1- I could cut the triangle relief out completely as others have done, but I wanted to preserve the small structure, and prevent any further strength reduction in this area.

Note how NG-30 has been cut here to allow the head of the motor to fit properly

2- I could fill the depression back up with micro and foam, glass per plans, cut a new depression (on the inside this time), and skin it. The drawbacks with this approach were many, and it would have also taken longer than remaking the parts.

3- I could start over with new NG-30s, wasting away expensive foam.

4- I could swap the NG-30s, figure out how many additional layers of fiberglass it would take to fill up the craters completely (thereby leveling the inner surface once again), and glass away, fat, dumb, and happy.

Wait a minute... I resemble that remark!

I decided to try to salvage my previous work, and swap NG-30s around (option 4), and in case that didn't work out, I could always start over again (option 3).

I calculated that it would take 23 layers of BID (as opposed to the original 15) to completely fill in the craters, and hoping to be correct I finally got on with it.

Task number one was cutting the 184 fiberglass disks to fill the depressions up. I chose to make the disks incrementally wider (in 4 steps) to try to match the crater profile. This sounded like a good idea at the time, but it unnecessarily increased the level of complexity of the task, and I wouldn’t recommend it.

The big problem now was how to make precise patterns to use for marking 8 different circle sizes on the dry fiberglass.

One can always overcomplicate things, so I decided to use the most complex technology I had available to solve one of the simplest problem... drawing circles. What the hell, having just finished (for the most part) my CNC mill conversion, I decided to take it for a spin.

Now, you may notice in the following video that I mention 160 fiberglass disks, that was before I took more precise measurements, and came up with the new total of 184 disks. I also made the patterns slightly wider to ensure the edges of the NG-8 aluminum disks would rest on the hard points, and not accidentally end up on the bare foam-backed skins.

Complex, and a lot more fun solution to a simple problem.

So, I started cutting the disks...

Adding a few layers of BID circles to reach 23 for each hole

... and the new inner skins.

Since orientation is optional, I'm trying to get the most out of my BID.

For either one of the structures I prepared 5 BID and one kevlar skins (to be buried within the glass ones).

These took longer to cut than all 10 BID NG-30s combined

I started working on the inside surface by “pre-pregging” all the disks.

Circles stacked by sizes

Pre-pregged circles

I chose to use flox instead of micro for the fillets, because I was concerned I’d end up with micro between the skin and the glass disks.

Floxing the outer edges

When I laid the staggered disks in the craters, I was glad to see that 23 disks were indeed enough to come up level with the foam.

23 circular BID pieces precisely filling the craters

I repeated the process for the larger front craters.

Pre-pregged circle stacks ready to deploy

Top BID perfectly even with the foam surface

Hard points completed

Next, I “microed” the foam surface, making sure no micro-slurry touched the fiberglass disks...

Adding micro to the foam before receiving the inner skin

... and worked on the flat layup.

Inner skin finally being applied

Kevlar ply inserted between BID plies

Additional BID ply applied over the kevlar one

After curing overnight, I trimmed the first NG-30, then sanded away some of the strange ridges of epoxy I still get on the fiberglass surface when using large peel-ply sheets (as seen in Ch. 5 - Fuselage sides RH - Part 2).

Trimming time

Inner NG-30 surface sanded smooth

Later, I repeated the process for the left NG-30 (formerly right).

Pure epoxy to lead the way

Flox for the edges

Hard points ready

Glassing the left NG-30

Left NG-30 glassed and peel-plied

Once again I left it in the shop to cure overnight at 80℉ (26.5℃).

Note where the peel-ply sheet caused a few epoxy ridges to form on the skin surface

Both NG-30s side to side as they'll fit in the nose

I’ll trim and sand the left NG-30 when I get back home, then I'll figure out where to go from here.

No comments:

Post a Comment