Question Gravity
Tandem AirBike Builder's Log

February, 2003 -- Fuselage finishing and misc.

Page 1


Pack a lunch, my friends. Lots of things to report in this installment. I tried to keep the verbosity down, but failed once again ;)   I broke it into three pages in an attempt to make it more user-friendly.


I offer this photo and observation relating to our discussion in Tandem Talk about heat embrittlement around welds. To remove the improperly-drilled lift strut brackets from the fuselage, I ground through the welds as deep as I dared with an abrasive cut-off wheel. Then it only took a few light blows with a plastic mallet to break the bracket off. The arrow is pointing to one of the embrittled areas I found adjacent to the MIG welds. Here the metal broke easily (instead of bending as normalized 4130 would), and the material is visibly very coarse and grainy. A better word might be crystalline.

This illustrates the concern I had that made me research the heat-relieving or stress-relieving of welded joints. I found two different schools of thought on the topic: One saying that electric welds produce a very small heat affected zone (HAZ) and therefore inconsequential embrittlement. The other says that all (critical) welds should be heated to a dull red and slowly cooled to relieve the stresses induced around the HAZ. Actually, even as I write this, I'm still a bit confused and wondering if stress relieving has much effect on the type of embrittlement illustrated above. I remain open to being further educated on the topic, but meanwhile I did borrow a rosebud torch and stress-relieve the most critical welds on the fuselage (including the newly-attached and properly-drilled lift strut bracket replacements Wayne sent me).

Budd Davison has an excellent article on the EAA site about welding, where he offers lots of wisdom on both sides of the stress-relieving topic and lets the reader decide for themself. His bottom line seems to be that in the short term, electric airplane welds (MIG and TIG) seem to be OK without stress-relieving, but nobody knows how they'll hold up long-term. Given the embrittlement I found around the MIG welds, I feel that the AirBike's most critical welds warrant a little caution and extra attention.


I wanted shoe hooks on my rudder pedals to resist the natural tendency of your foot to rotate off the pedal, and since I don't have access to welding equipment I came up with this weldless solution. Mild steel was what I had on hand in the appropriate thickness (.100-.125"), and as shown above it was bent, drilled and tapped to be held in the end of the upper pedal tube with countersink head machine screws and Locktite.


I made a lot of progress on my second wing panel since the last update, and took a moment to document the glue-starving concern I warned about in May 2000. Here one of the yellow pine (I think that's the species Ison uses) anti-drag brace corner blocks is coated with T-88 in preparation for bonding into the corner shown at right. The grain is parallel to the long edge of the block, so the two glued surfaces are at 45° to the grain. This means that these edges are mid-way between end-grain and side-grain, and therefore not as likely to suck up the glue as true end-grain.

Danger, Will Robinson. Five minutes later, most of the glue has disappeared into the block. Had this block been simply coated and installed, the result would have been a weak, glue-starved joint. Now is the time to re-coat the block with more glue and then observe it a while longer before assembly. Be sure the absorption has stopped before assembling end-grain or cross-grain pieces like this. Some wood samples will require several re-coats, and the longer you have the joint open, the thicker the resin will become which will further reduce its tendency to soak into the wood.


Here's another wing assembly topic I'd like to revisit with a visual aid. As I said in August 2000, once your ribs are glued to the spars, there is not much you can do to correct a wavy trailing edge. If you want straight trailing edges, take a few minutes to get the ribs properly aligned before you glue them to the spars.

Two years after building the first wing panel, my table was not quite as flat and straight as it used to be and I was having a hard time verifying the alignment of the rib TEs. So I cheated and set up the music wire and zither pin trick again (see the May 2000 log). Here I've raised the unglued wing slightly by placing 1/2" thick blocks under the spars at multiple locations. The front spar is clamped to the table's straight edge with the wedge clamps (also shown in the May 2000 log). The rear spar is biased toward the front spar with wedge clamps to give the desired rib pressure against the front spar. The ribs' trailing edges have already been carefully trimmed to the same length, so fore and aft discrepancies shouldn't be an issue now. The music wire is stretched tight just under the rib TEs. You'll probably find one or more ribs that don't tow the line with the rest of the ribs, and now is the time to correct that. The offending ribs may be loose enough on the rear spar to allow you to move their TE up or down. If not, you may be able to slip some sandpaper between the rear spar and the rib and remove a small amount of material to bring the rib into alignment. It shouldn't take much, as the motion is multiplied at the TE. Do avoid removing so much material from the rib cap strip that you risk weakening it, though.

Now, with the TEs as straight with the wire as you can reasonably get them, glue the ribs to the spars as described in the assembly manual, but with the wing still on the 1/2" blocks, clamped to the table's straightedge, and with the wire still under the ribs so you can align their TEs as you nail them to the rear spar. To complete your beautifully straight TE, all you have to do now (after the spar/rib assembly is cured) is glue the TE stock onto the ribs with reference to the ribs' bottom edge (see Aug. 2000). The ribs will tell the TE stock where to go (whether you use my methods or not), so you don't need to nail it down to the table for installation.


The tips of my elevators were out of alignment with the tips of the horizontal stab by about 1/4", so I sandblasted the tips and added Bondo to get them to match. The fabric covering will never tell.


I chose to sandblast my metal parts for ultimate paint adhesion. Here's the forward fuselage half in the sandblasting booth, supported by the rotisserie that was also used during priming and painting. The black pin on the right locks the assembly in any desired 45° increment of rotation. I know other builders somehow get by without this fixture, but I honestly don't know how. It is difficult enough to reach all the tubing surfaces WITH the help of this device...


Here's my little hell away from home for several weeks. In this temporary, filtered cross-flow spray booth I painted all the metal parts. The left (upstream) end features four paper filters and a simple flap door that seals with the fan's suction.

The downstream (exhaust) end features a window fan taped to the tent with an air filter on the inside. Although the MSDS says these waterborne paints are combustible, I didn't buy it and had no problem with the non-spark-free fan.

I was about to make a PVC frame for the tent like I did when I painted my Ultra-Piet, when I realized I had this portable awning frame that would be perfect for the project. The plastic sheeting I had on hand was some heavy polyethylene for the roof and some thin Visqueen for the walls. Tape didn't stick worth a darn to the Visqueen, even duct tape, so I ended up gluing the sheets together with 3M 77 spray adhesive (which saved the day).

Continue to Page 2 >


Page 1 | Page 2 | Page 3