Archive for June, 2009

I have gained a new title on the project: Master of Imperfections. It’s my job to insert some little stuff-ups into the fabrication of the window test unit.

This is actually a valid test method used in the aerospace industry. When you are testing the full lifetime of an aircraft structure, you need to include everything that the structure might see in its lifetime. This would include accidents and repairs. The goal is to certify that the airplane will be safe, even if it does have a few bumps and scrapes on it. If the testers only tested an absolutely pristine structure, then they could only certify that it would withstand the tested forces in perfect condition. As soon as a real plane got a scratch, then it would be operating outside the tested range and would be deemed to be un-airworthy.

Another way of putting it is to say, “Engineering is the Science of Failure.” We are actually looking forward to blowing stuff up and having failures. If we don’t break something then we never find out what it takes to break it. If we don’t know what breaks then we can’t fully understand what actually keeps it from breaking.

Today we are bonding in the first window test article. This window is larger than any window planned for the actual plane and it has intentional defects in the bonding. We still use the exact process that we will be using to build the actual plane but we are being “less careful” about the workmanship. If we created a “defect” like skipping a step of cleaning, then that doesn’t count as a proper test: the cleaning solvent may make the final part weaker than the one we tested. The types of defect we are testing are things like putting a fingerprint onto the bond surface after it is cleaned.

Once the window has passed (or failed) the initial test pressure and temperature, we will add more defects such as an accidental scratch right across the middle of the window. Then we test pressure and cold again until it breaks.

The first plug was given its first coat of epoxy today. The glistening covering makes it look like a beached whale, complete with dorsal fin and even a blowhole, thanks to the single remaining alignment hole.

The epoxy will cure overnight and then we can do more sanding.

Windward Performance and the Perlan Project have been featured in the June edition of Machine Design. Click on the cover image to go straight to the online version of the article.

The first fuselage plug is now all bonded (glued) together. The first two blocks of foam were levelled to within a hundredth of an inch – less than the thickness of a piece of paper. This makes a perfectly flat foundation for the other blocks. This is also the centreline of the fuselage so getting this line straight is necessary so that the other parts of the aircraft will fit and it will fly straight.

The other blocks were then aligned onto the first pair, using alignment rods and careful measuring before the epoxy resin was poured.

The next step is to fill in the holes from the alignment rods, since they will never be used again, and sand the surface to smooth out the grooves made by the CNC machine. It won’t be long until we start laying up carbon fibre onto the plug.

The design for the Perlan 2 relies on polycarbonate as the window material.  (You may know it better by the trademarked name “Lexan”.)  Polycarbonate is known to provide good impact resistance and low-temperature strength.  However none of the published data goes below -55°C (-65°F)and most of the data specifies -40°C (-40°F) as the minimum.  So we had to buy a sheet and do the testing ourselves.

We tried three tests at room temperature and at dry-ice temperature (-68°C / -110°F.)

1. Bending to 90° over the edge of a bench
2. Hammering a nail through the polycarbonate
3. Bending with a deep scratch across the bend line
    

   Room temperature on the left, cold temperature on the right

    

   The results were as follows:

    

1. The cold test sample was much stronger (more difficult to bend) and had less plastic (permanent) deformation
2. The cold test sample was so stiff that the nails bent and could not be hammered through
3. We were finally able to break the cold sample but only after it was bent past 50° with a deep v-notch cut all the way across with a chisel

Light aircraft windows are usually made of acrylic, primarily because it is easier to form into complex shapes and doesn’t go yellow under UV light.  We tried test #2 on a cold piece of acrylic.  As soon as the hammer touched the nail, the sample split into two and sprang away from the impact so violently that the two pieces slid off the opposite ends of the work bench.