A Rivet Issue

I am fortunate enough that my former hangar mate and Sonex builder/pilot Mike Singleton is also one of our EAA Chapter's (774) Technical Counselors.  As it was approaching time to attach the bottom skin to the mid-fuselage section, I knew I had to get Mike to come inspect my work prior to closing up some of the fuselage structure.

Mike came by on October 25, 2012 and inspected my work.  He was complimentary of my project and asked a lot of good questions about my tools and my paint priming process.  During the inspection of the mid-fuselage, Mike noticed that the shop end of the top row of AN426AD4-8 and AN470AD4-8 rivets on the aft bulkhead, F-1204D, did not appear to protrude far enough through the F-1204 center section channel.  I had not noticed this until now.  A quick check with the rivet gauge suggested he was correct.

The picture below shows two rows of rivets with the shop end of the squeezed solid rivets exposed.  Mike referred to the bottom row of rivets as "overdriven".  The diameter of the shop head is within spec of the go-no-go rivet gauge for a number 4 rivet.  It is the thickness of the driven head that is lacking.  Note in the following 2 pictures that the fuselage is currently inverted as I am preparing to attach the bottom skin.

 The next picture is of the manufactured heads of the same AN426 and AN470 solid rivets. 

The head-driven thickness of the rivets is about .040 inches as measured in several places along the line of rivets with a micrometer.

The morning after Mike's inspection, I called Vans to get further information.  During the course of that conversation, the tech support rep never did address the possibility that the designated rivets might have been too short.  Instead, he acknowledged that drilling out those rivets would be a challenge, especially now, since the baggage ribs and floor skins were already in place, which significantly limits access with a drill.  His suggestion was to use a right angle air drill with a number 40 bit, followed by a number 30 bit if I elected to drill out the rivets.  He went on to cite a table on page 104 of the sixth edition of the Standard Aircraft Handbook which calls for a minimum upset height of 3/64 inches (.047) for a 1/8 inch solid rivet.  As noted above, the rivets are about .007 inches short by this standard.

As I further researched the issue on www.Vansairforce.com internet portal, I found a reference to the Mil-R-47196A, published September 6, 1974, which is the military's specification bible for all things structural.  In Table III, under the heading of Rivets, Buck Type, Preparation for and Installation Of, Standard Flat Driven,  Head Dimensions, the 1/8 inch rivet lists a Driven Head Thickness Minimum of 0.050 inches.

Finally, a search of the archives on Vansairforce yielded 2 relevant threads published by other RV-12 builders who had the same result in their F-1204D aft bulkhead rivets.  One thread in particular referenced a response from Ken Krueger, former Chief Engineer of Vans Aircraft with substantial knowledge of the RV-12 aircraft.   In that response Ken simply cited the Note in section 5D, entitled, "Riveting" of the builder's manual.  The note reads as follows:

"Note: There are times when the correct rivet length is not available.  Depending on the application a shorter rivet can be used or a longer rivet cut to the proper length.  Using a longer rivet, as is, can result in the shank being bent over like a nail. We have chosen to use a rivet that may seem too short in some places, but will do the job adequately."

It was also noted in this thread that the parts in question (F-1204D riveted to F-1204) are experiencing shear loads, which would either eliminate or minimize any tensile loads, in this area.  Tensile loads are normally the worst case for testing the head strength of a rivet (manufactured or shop).

Another reference also found in section 5D of the builder's manual pertaining to the subject of drilling out rivets is based on an excerpt from the Alcoa Aluminum Rivet Book, dated 1984. Without quoting the entire paragraph, the upshot of this reference is summarized in the following 3 points:

    1. Poorly set and cracked rivet heads tested in tension were still within 5% of the strongest properly set rivets indicating that minor deviations (.007 inches) are not cause for concern or replacement.

    2. A second rivet that is driven in any one hole is likely to be more defective than the first...and will be more likely to buckle and form an imperfect head.

    3. Tests have shown that very small rivet heads are sufficient to develop the strength of the rivet shank, even when the rivets are subject to a straight tensile pull.

It is my conclusion that the risk to reward of drilling out and replacing 73 rivets, (a combination of AN470 and AN426 rivets) in the center section structure is absolutely not merited.  There is sufficient evidence both from within and without the RV-12 community to support this position.

The previously cited 2 threads of relevant discussion on this subject can be found here:
    1. http://www.vansairforce.com/community/showthread.php?t=44303&highlight=20-04
    2. http://www.vansairforce.com/community/showthread.php?t=39106&highlight=20-04

Forward Lower Fuselage Assembly

One of the first steps of the forward lower fuselage assembly is to dimple and then rivet nutplates along the sloped edge of the left and right tunnel ribs.

Next, the first of two pair of engine mount brackets are installed on the tunnel ribs.

Now, the tunnel ribs are riveted to the forward fuselage (think floor) skin.

It is now time for the firewall bottom panel/skin to be attached to the tunnel rib/forward fuselage skin.

And finally, the left and right forward fuselage corner skins (curved sections) are riveted to the forward fuselage skins and firewall bottom.

The entire section is, of course, cleco'd together prior to riveting.  In this picture you will see a series of colored dots.  These are to mark locations that are NOT yet ready to receive rivets - a no rivet zone!

Mid-Fuselage Parts Assembly, Part 2

The mid-fuselage assembly continues with the final assembly of the two control column mount assemblies.  Cleco'd together and then rivet.

Next is a shot of the F-1203A bulkhead and the F-1203C and D wing spar stub receptacle components. The front wing spar stubs slip into the pocket created by these parts.  This bulkhead resides at the forward end of the seat ribs.

The "flip side" of the same bulkhead with its interior color coat and doubler, which also ties into the previously mentioned spar stub receptacle.  It has a combination of solid and blind rivets which secure the assembly to the bulkhead.  The entire doubler/spar stub receptacle assembly jigged together with clecos

With the first solid rivets of the assembly set, it is time to move on to some of the blind rivets.

All of the solid and blind rivets set.  Remaining clecos are there to remind me to NOT rivet in those holes until later.

Back on the primer side of the bulkhead, it is time to rivet the control column mounts and another rectangular doubler.

Just to the right of the control column mount is the other rectangular doubler previously mentioned.

The finished bulkhead now awaits its turn to be attached to the mid-fuselage assembly.

 

 
 The left-hand seat assembly is now attached to the mid-fuselage assembly.

The seat assemblies attach to the center section via the floor skins and the ribs.

Picture time!

Next step is to attach the bulkhead caps and rollers (not pictured) to the side assemblies.  The caps are cleco'd and then riveted to the side assemblies.

 

It is now time to attach the forward bulkhead to the front of the seat ribs.  As usual, jig in place with clecos and follow with rivets.

 

 

***********************Need to add forward floor ribs and skin pictures here.

Next comes the corner fuselage skins as they are cleco'd into place prior to riveting.

The fuselage bottom skin has had its edges formed or "broken" with a slight bend as well as the 20 nutplates attached to the perimeter of the inspection holes.

The inverted mid-fuselage section is supported in two places in preparation of the bottom skin being attached.

In light of service bulletin, SB 12-09-26, I am considering making the F-1275G landing gear cover plates removable via nutplates.  This would facilitate easier access to the landing gear attach fittings where improperly torqued bolts have caused structural issues with the center section.  It appears that at least part of the issue of improperly torqued bolts has been due to the use of bolts that were too long.  As a result, builders were not aware that nuts had bottomed out and not been fully tightened or torqued to specification.

It will be several more days before I can finish assessing the practicality of installing nutplates on the cover plates as I had to pre-order those parts from the Finish kit, which I do not currently have.  I have also ordered a mixture of miniature nutplates and Riv-nuts to further explore this option.  Stay tuned!