N254MR

Yes, I've been aware of the leaky tanks problem for well over a year and a half, but I'm still researching [dithering about] all the options for dealing with it. After an astonishing [Ed. appalling] amount of cogitation, both on and off the Rebel Builders list, I've finally reached some almost possibly definite conclusions about the pros and cons of different schemes and methods for the job.

The bottom line question is which approach should be taken first, and which will be done next or last as backup, if need be. A solution to that riddle isn't clear unless and until one makes certain assumptions. If one assumes "the whole thing" (integrity of a fair amount of existing sealant) is suspect, and "the whole thing" really can't be completely cleaned and redone properly through the limited reach and angles available via bottom port access alone (most would agree there) then it makes sense to go in from the top first -- not through bottom access ports. Removal of the top tank skin allows difficult but certain elimination of all the original sealant (except what's captured between riveted rib, stringer, and spar flanges) followed by best practices prep of the surface, and the top-down "bathtub" sealing technique, done right, followed further by the best top skin reinstall that can be done, even though, as noted by many, the top skin is frequently the most leak prone area, given the difficulties of getting it down and properly sealed without benefit of its own bathtub treatment from below. [The common use of the bathtub metaphor is meant to signify that hyper-gooey Pro-Seal application works best when gravity isn't pulling it away from where it's supposed to stay, before it firms up. This is important even though it's well known that Pro-Seal is the only material in the universe, only a small thimble-full of which will, inevitably, completely cover your entire body, permanently.]

Further, if a person is confident new bottom ports can be sealed up well, if need be, at whatever point -- as I am, pretty much -- then it makes sense to redo "the whole thing" from the top first, because if the reinstalled top skin doesn't seal fully (even though it's usually impossible to determine a leak's precise source) only then do you have to cut new bottom ports, to better seal the top.

Problem is, if you flip just one or two of those assumptions, the logic plays out differently. But not to worry, because....

Eureka !! The ultimate procedure, the one most likely to result in a secure, leak proof tank repair -- though certainly not the easiest -- has finally become clear, from A-Z, thus:

A) cut bottom ports,
B) install and seal the doublers [if flush mounting] but
C) just cleco the covers in place temporarily,
D) flip the wing over, right side up,
E) remove & clean the top skin,
F) remove all the original sealant with PolyGone [not around new doublers],
G) fully prep* and bathtub seal everything anew from the top, including
H) top coat the doubler-to-skin seams,
I) reinstall top skin "real good",
J) flip over again, bottom up,
K) through bottom cover holes, complete an overcoat, bathtub seal of the top skin seams, then
L) flip right side up again [last flip],
M) seal & rivet the bottom port covers one at a time, from the bottom,
N) double coat & overcoat the port cover seams & rivets from inside through adjacent bay ports, and
O) do the outboard cover under the filler neck last in order to
P) top coat the last cover, top down through the filler neck too,
Q) cure [and medicate anxiety] slightly less than forever,
R) test,
S) final clean,
T) check finger screen & sump fittings for leftover crud,
U) dry for minimum of one week in each of nine (9) Multi-Polar Attitudes®, and [ta-da!]
V) fill 'er up!
(W-Y) - [Reserved]
Z) Detox before flight.

The [only?] virtue of this procedure is that everything gets a newly-prepped, top down, bathtub-ish, overcoated seal. YAY!

Hmmm. Multiply by two wings = epochs of work. Might be flying again by 2013. Or not. 

In addition to possibly not even survivng so much work,   the main risks in this procedure are potential damage to the top skin, or worse, a rib, requiring fabrication of a new one, and, with lower probability and consequence, incomplete sealing of a bottom access port cover.

As important as it is, and as complex as it seems, is it possible to over think even this problem? As my good friend, fearlessly inventive engineer, Rebel owner, and indefatigable tinkerer Garry Wright would surely reply,  "I don't know, but I'm not sure."

__________________________________

[* Whether new construction or repair, proper surface prep is the most critical task. Essential details are available elsewhere, including comments by Wayne O'Shea and others in the Rebel Builders mailng list archives.]

Working on the layout of the tank access ports, drawn onto the bottom of the left wing. Each wing tank will have three of these ports, one in each bay. The ports will be 8" x 6", with either a 2" or 3" corner radius. The 3" radius probably looks a little better. No, an even smaller radius would be better. Maybe.

A longer perspective view of the size and location of a port. (The black-circled rivets are leakers and will be replaced.)

Without a doubt the most surprising thing about the right tank so far is the significant quantity of fuel that was still inside, even after the fuel inlet had been left open for almost two months! Obviously, all fuel and vapors have to be removed before metalwork begins, due to spark and explosion hazards. To dry it out, I decided to blow warm air from my Steinel variable heat gun (150 F on the digital readout) through a 5 ft. piece of 1" urethane tube. The tube keeps warm surfaces of the heat gun, and possible sparks from the heat gun fan motor, well away from any concentration of fuel vapors. The air cools quite a bit in the tubing, before reaching the tank. With the top inlet around the tubing sealed with duct tape, the air flows through the tank and out the blue fuel outlet fitting at the lower right.

After about four hours of this, the vapors at the output seem to have diminished, and the tank is no longer visibly wet inside. I will still have to rinse it out, just to be sure I have fully diluted any fuel that could still be remaining somewhere in the tank, but the warm air setup will aid drying again, after rinsing.

While waiting for the tank to dry out, I looked more closely at its condition as visible on the outside. Among other things, as seen in the next photo, I found a blue avgas fuel stain around a hole (photo lower left) used to mount the bottom wing root fairing. That stain was underneath the clipnut into which the bottom wing root fairing fastener goes. You can also see a bit of abrasion or nick in the gray Pro-Seal sealant next to that hole, no doubt also caused by the edge of the clipnut. Although I'll certainly want to protect the sealant there in the future, all I can tell from this so far is that leaking fuel was trapped under the clipnut long enough to stain the surface. I doubt it leaked there, because the sealant at the ding isn't stained. There's also blue staining on the flange under the fuel outlet fitting, however, the bottom of the outlet fitting itself is not noticeably stained and there's no direct evidence of leakage at the finger strainer or outlet fittings themselves.

As was apparent even while the wing was still on, there was leakage flowing out the stringer hole seen at the far right. The stringer is outside the tank proper, just aft and downslope of the tank's rear wall. Because leaking fuel can show up in a wide variety of places far from the actual leak, none of these observations yield anything truly definitive about the source point of the leaks. Instead, the above observations just suggest possibilities.

Here's a closeup of the pinholes in the Pro-Seal near the fuel outlet fitting, some of which are centered on rivet heads. Pinholes can be caused by several things, including, for example, suboptimal mixing, or thinning of the sealant before it's applied. It will be interesting to find out if there are many pinholes like this on the inside. So far, I've not been able to see anything like this inside, using a borescope.

[UPDATE 1/13/11 - Closer inspection with a borescope does show quite a few pinholes, for example, along the bottom edges of the first rib inboard of the filler neck. Also some bottom rivets at that junction may not be fully coated with sealant. These observations will have to be confirmed when I open up the tank.]

Looking in the wing from the outboard end in the shot below, there is an area of blue fuel stain on the outside of the outboard tank rib, as represented by the drawn blue line. Couldn't figure out how to get it to show up in a photo, giving the lighting challenges (might take a color filter of some sort) but it's there, and probably indicates leakage coming from somewhere around the top skin of the tank.

It's theoretically possible that leakage from this area could show up on the outside bottom of the wing where fuel seepage has appeared around some of the aft, bottom rivets, and it could even account for some of the fuel leakage flowing out the stringer hole, 3-4 feet away at the inboard end of the wing, as seen in the second photo above. Hard to tell for sure.

Before undertaking a tank-repairing surgical operation, you need an operating table. A 4x8 piece of plywood covered with carpet pad will suffice.

With all the tips and flaperons removed, all the fuel and vent plumbing, and fuel gauge wiring and torque tubes disconnected, the wings finally came off -- for the first time in about four years. Below, the right wing is loaded up on the operating table, ready for pre-op tank rinsing. The left wing is on the reassembled, carpet sling wing stand in the upper right, which has been lurking in the back of a shed for four years, but it went back together in about 15 minutes. Yay!

Before painting, quite a few things have to get done. In addition to tank repairs, skylights, etc., I decided about two months ago to change the engine mount. The original mount, built long ago by Jabiru, doesn't have the correct geometry. (They've since corrected the design.) The engine thrust line hangs 1.5 deg. below horizontal now, even with maximum shimming, and does not have the Murphy-recommended 1 deg. right offset to counteract the well known left turning tendencies that most small single engine airplanes have.

Of course, I've known of this issue for a long time -- well before cowl construction -- but was persuaded that it wasn't significant enough to get a new mount. As the mount rubbers wear, it's gotten a bit worse, and I've decided to just bite the bullet and get it right. Unfortunately, changing the engine mount and thrust line will require rebuilding the back half of the cowl in order to keep the nose bowl centered on the propellor shaft. Although several parts can be reused, it's still going to take some work. There's no point in painting an interim cowl.

For the last week or two, I've been disassembling all the wiring and plumbing going to engine and/or mounted to the mount tubing, as well as other engine accessories like heat muff and muffler. I've taken lots of photos so I won't forget how to put it all back together, like this one:

And of course, breaking it all down makes a mess.

Yes, it does feel weird to be taking so much of this apart but... it's gotta be done. At this point I think I'll be lucky to have the plane flying again in May.

Because the rudder mod required removing the stabilizers, it presented an opportune moment to complete installation of the VG's on the bottom rear of the horizontal stabilizers.

As expected, the addition of stablizer VG's provides a noticeable and welcome improvement to elevator control authority.

As the unfortunate consequence of a firm landing on the Oshkosh trip (at Ellensburg, WA, the last stop before reaching home!) the tailspring flexed enough that the top of the tailwheel impacted the bottom of the rudder, causing a dent. My previous conclusion that the original clearance (~2-5/8") was proven wrong, even though vigorous bouncing of the tailcone had not put the tailwheel even close to the rudder. Oh well.

Both the bottom rib and edges of the skin required repair. After consideration of possible effects on rudder performance, balance, and potential for flutter, I decided to raise the line of the bottom rudder rib to provide more clearance from the tailwheel. The actual reduction in weight would be minimal (only the reduced skin area), and, given that the airflow in this aera is already "dirty" after passing around the stabilzer struts, rudder cables, etc., the effect on aerodynamics would be very slight, if any. A line for the new shape was drawn on the rudder skin.

Before cutting the skin, some layout lines were drawn to calculate the change in run-to-rise ratio along the new shape. Then the damaged rib was removed, and the skin trimmed to the new line, as seen in the next photo.

To construct the new rib, a plywood form was cut to conform to the new run-to-rise ratio. As seen below, it was then used to bend up the edge flanges of the new rib, drawn on a piece of aluminum stock.

The completed rudder repair and mod looked like this:

The major bend in the tailspring was also closed an additional 5-6 degrees, resulting in the geometry shown below. The tailwheeel king post, which previously had a slight forward lean, now has a slight backward lean with the plane unloaded, and is very close to vertical when the plane is loaded, as desired.

Doing this repair/mod took some time, but was not difficult, and resulted in an overall improvement of the clearance geometry. Unloaded, the new clearance is 4-5/8".

[Ed. Previous blog posts covering the July trip to Oshkosh (OSH) have been consolidated in a story, a work still in progress but already containing more photos, and still growing.]

Faithful readers, be assured that although I've been off the air for quite awhile, you've not been abandoned completely. The causes of the recent hiatus are many, including a surgery and recovery which took your correspondent out of action for more than a bit. After that, there have been travels, including a long overdue, two week road trip to see friends in California & Oregon. I will also confess to a certain ennui, the result not only of physical ailments, but some not inconsiderable dread of having to come to grips with the ongoing tank leaks, and the prospect of a heck of a lot of other work that must be done before painting, and the painting to come. All this has kept me from doing much, or writing much, or flying much at all -- in fact only once in the air since mid-September. You'd think I would have learned by now not to be bogged down by such a task list, but no. The weight of what needs doing is still daunting.

The ailments and travels are done, at least hopefully, for awhile, and work is once again underway in the hangar. So do visit again soon, because whether in the shop or in the air, the chronicles of 254R are never done, and the story will be told.

Many thanks to all those who have prodded and cajoled me to continue with this, and more, over the last few months of a difficult autumn.