Trent's RV14 Build: 29-14 -> 16: Riveting of Side Skins to Fuselage (Forward of Spars)

Continuing the riveting of the side skin forward of the spars - i started by backriveting the centre side plate using 25psi. The lower dome head rivets on this plate were fun - i used 70psi and due to the angle of the skins in this location, i was able to use the short cupped set.

I could not get the backrivet set onto the lower rivets on the centre side plate, as the gun was interfering with the seat ribs, so i used the mushroom set and bar.

The angle of the skins allowed me to use the short cupped set, which is a lot nicer than the long one.

This hole has been made in the parts, as a pilot hole for the fuel return line fitting.

It was a tight fit to get to some of the rivets between the side skins and the seat ribs.

I then completed the remaining rivets along the lower stiffner of the forward skin by backriveting these at 25 psi. I then moved onto the rivets along the lower longeron - these were all set using either the cupped set for the domed rivets, or the mushroom flush set on the remainder. I could not use the backrivet set on these, as i was not able to see what i was doing from above. To reach the aft most rivets, i removed the nut from the most outboard bolt in the lower drag fitting, and tapped the bolt out of the way. The only issue was the 2 rivets which connect the little tab from the seat cover rib to the lower longeron / skins - these were a real pain, as the rib had been bent out of shape a little. Took some convincing to get the tab to bend back in place and take the rivets.

The 2 rivets on this little tab were a real pain, as the tab had been slightly bent out of position and i did not notice this when clecoing on the lower longeron.

Some rivets around the seat ribs needed some creative bucking bars!

I tapped the bolt out of the way to set these rivets (and it needs to be out of the way to set the AD4 rivet in the lower skin anyway).

Next up was the rivets which join the forward engine mount to the lower longeron and side skins. These use a mix of AD4-7 flush rivets and flush CherryMax rivets - specifically -4 size in -5 length. Firstly, the -7 length AD4 flush rivets were too short, i elected to use -8 rivets in this location.

These are the -7 length rivet - much too short.

These are the -8 length - much better.

I had only 12 x -8 length flush AD4 rivets left in my box, for 12 holes in the longerons. Of course, i screwed one up and had to drill it out. For it's replacement, i just used a -7 length rivet, which set fine but does have a slightly smaller shop head.

This one got clipped by the bucking bar when setting another rivet - as they are all so close together.

This was the damage from the edge of the bucking bar.

But i managed to get it drilled out - right down the middle.

Cherry Trouble!

Van's specifies the use of some Cherry rivets in the lower engine mount bracket - apparently they have a much higher shear strength due to the steel mandrel remaining in the rivet. The rivets called out were CR3212-4-5 (the "3212" refers to the rivets being flush, the "-4" is the diameter in 16's (1/8") and the -5 is the length).

I used my pneumatic rivet puller to set these rivets, like i did with the rivets on the lower skins. All the rivets seemed to set ok, however after setting all of them i noticed that a number of them had much different shop head lengths:

I was not sure if these were acceptable, and whether they should be removed. So i I asked the question on the VAF Forums. What i learned from this discussion was that:

A Cherry Rivet has a very narrow 'grip length' window. The rivet needs to be selected to properly match the thickness of the parts being joined.
A properly set Cherry Rivet has a little steel collar - it looks like a tiny steel shaving (which looks like it shouldn't be there when you tap the mandrel out to lubricate it).
When the rivet is set to the proper length, the steel collar will nest into a groove in the steel mandrel locking the mandrel in place.
If tapped on with a pin pinch, the centre mandrel will not move. You are unable to punch out the mandrel on a properly set rivet. If you need to remove the rivet, the steel collar needs to be ground out (drilling is next to impossible, as the drill just skips off the hard steel collar and mandrel into the soft aluminium of the rivet body).
For an improperly set rivet, the mandrel has not been locked in place and can be easily tapped out.

So, after learning all these important facts about Cherry Rivets, i tapped on all the centre mandrels of each of the set rivets using a pin punch just slightly narrower than the mandrel of the rivet. Each of the mandrels in the poorly set rivets above, tapped out straight away. Proving that these rivets were not set properly.

But why? Cherry has published the Cherry Rivet Process Manual which clued me in that each rivet has the length of the rivet stamped onto the head. So i got out my magnifying glass to find that each of the poorly set rivets had a "6" stamped on it, while all of the properly set rivets had a "5".

A properly set rivet showing a "5" stamped on the head.

One of the improperly set rivets with a "6" on the head.

So, i thought i must have mucked up my inventory and mixed some -6 length rivets into the -5 box. However, checking the inventory list, the only CR3212 (Flush) cherry rivets we are sent are -5 length and -3 length. My assumption therefore is that some -6 length rivets made it into the bag of -5 length rivets sent by Van's.

I double checked the holes with the Cherry grip gauge, and the -5 rivets are the correct ones for these holes:

Past the 4 line, but less than the 5 line.

So how to fix it!

In the end, because the rivets were set improperly, i was able to tap out the mandrels with ease. The steel collar fell out by itself for all but one rivet, and i was able to pry it out for the remaining river. Drilling them out is now very simple as there is a ready made hole right down the centre.