08 May 2014

Empennage: Horizontal stabilizer. Factory error on skin? Must redo entire stab.

Update 13-Dec-14:  The reason why the spar was mis-punched has been identified.  Scroll to the bottom or click here.

So again, Van's has been very reasonable with me on this issue.  They continue to demonstrate their focus on the customer.  They're fantastic to work with and knowing that I will have one of their airplanes gives me great comfort.  Now to the story...

My original horizontal stabilizer had what I (not Van's) determined to be a factory error on the rear spar.  You can read about it here.  In summary, there were four places on the spar with 0.05" too great of spacing between holes, so the spar couldn't line up with the skins.  It took a lot of time to diagnose but, no problem:  Van's kindly sent me a new rear spar without issue after I explained the issue.

I couldn't test-fit that new rear spar until it was countersunk in order for it to fit the skin dimples.  So, after fully prepping the new spar, prior to priming it, I cleco'd it up into the stab assembly and it fit perfectly on the right side, top and bottom.  It fit perfectly on the left bottom side.  But on the left top side...it wouldn't fit!  And where did the alignment problem begin?  At one the exact same places as before:  Between the 26th and 27th hole from outboard on the top of the left skin (location "B" below).  So, this was definitive evidence that old spar was indeed mis-punched, as my previous measurements showed as the new spar fit in places where the old one didn't.  But why wouldn't the new spar fit on the top of the left skin?

Dumfounded, I measured the hole spacing on my new rear spar with my caliper, afraid of another mis-punched set of holes.  Spacing was fine:  Exactly 1.125".  I then measured the hole spacing on the skin.  It too looked fine to me.  But clearly, there was a problem either on the spar or the skin (see first image below).  So, to ensure I had an accurate measurement, I seated unset rivets into the spar when it wasn't on the stab assembly.  Separately, I did the same on the skin when it didn't have the spar in it.  This way I could use the dimples in the AN426AD3 rivets as a good reference.  Turns out, the skin had two holes with spacings 0.02" too great!



Even though 0.02" isn't a big deal, two of them naturally add up to 0.04".  This creates a skewing of holes outboard.  Simply drilling these out up to a #30 wouldn't solve the problem because the dimples still wouldn't fit in the countersink for all holes outboard of the misalignment.

As another sanity check, a fellow -14 builder who completed a -6 and a tech counselor A&P who worked for United and the original Frontier and has also built several kit planes, to include a -12, both came to look at my stab to see if I was missing something.  They agreed with my assessment, though the tech counselor felt I could just drill out the holes to a #30 and use NAS1097 rivets in their place (my other tech counselor advised against this approach when the original problem surfaced).  I didn't believe that was feasible given that the dimples didn't seat into the countersinks due to the misalignment.  It would also greatly increase the risk of the dimples cracking due to the knife-edge cut from the drill.

I cut off the implicated pieces (spar and skin sections) from my original stab and sent them to Van's for examination (the image below may not be suitable for all audiences).  When they received my parts they told me the next day that they couldn't find anything wrong with them (click here for an update on this outcome).  They stated that they found the spacing to be correct and that my spar countersinks were a bit under-countersunk.  They felt the latter may account for the less than optimal layup.  To their credit, they involved several relevant people internally to investigate possibilities including looking at their CAD files, checking inventory parts and even cleco'd up a few stab mock-ups.  They took this issue very seriously even though I'm the only -14 builder to report this issue with these parts to-date.  They sent me images showing the skin portion riveted to the spar portion.  For sure, the mis-punched skin and spar portions would fit fine, but this is because they were both mis-punched in such a way that the two misaligned parts align fairly well.  However, that skin won't fit properly on a good spar and neither will the bad spar portions fit good skins (I wonder if they've misunderstood the problem).

Their assessment perplexes me.  Obviously I disagree with their conclusion, as precise measurements illustrate incorrect hole spacings on my original rear spar and left top skin and the fact that the new spar fit in 3 of the four places the old one didn't.  Even slightly deep countersinks couldn't account for these observations.  At this point it's moot.  Despite the different conclusions that we've both reached, they've agreed to send me parts to rebuild a new stab.  That represents a significant cost on their part and that is highly appreciated on my part.

My original horizontal stab was probably one of the best pieces in the build:  Only one very minor dent on the skin that no one would have known was there.  I hope that can be repeated on stab #2.

I guess I now have some hangar art in the form of a grounded, handicapped horizontal stabilizer.

Update 13-Dec-14:  Another builder ran into the same problem as I, posting on VAF.  In his communications with Van's, they found that the spar is too large to chuck up once, so it has to be rechucked in the machine at about the locus of where mine and his mis-punches occurred.  In case other builders run into the same issue, Van's is manufacturing several rear stab spars without holes punched so they can simply match-drill the unpunched spar to their skins directly.  I'm not sure why a simple replacement, properly punched, spar isn't the better approach.  But at least they identified the problem and I don't have to be the lone builder relegated to Crazy Town.

No comments:

Post a Comment