[Design Deviation] Hstab Cherry MAX rivets

It's been a while, but I think you'll enjoy this one. It's a bit of a detective story, and if you've never touched an airplane kit before, stick around, I'll explain the bits as they come up.

Quick catch-up for anyone new here: I'm building a small two-seat airplane, an RV-14, from a kit, in my garage, one rivet at a time. It shows up as a giant stack of pre-cut aluminum pieces and a book of instructions, and my job is to fit and fasten it all together in the right order, which turns out to be the hard part. The tail is where I'm at, and the tail is where I goofed.

So here's the embarrassing part up front: I skipped a step. On the horizontal stabilizer (that's the fixed, wing-shaped surface sticking out either side of the tail, the one the moving flap hinges off of) I never actually riveted the elevator hinge brackets. And of course I didn't notice at the time. I only caught it much later, while I was test fitting the elevator and realized there was just no good way to bolt the two things together. Cue the sinking feeling in my stomach.

If you don't build, a rivet is just a little metal fastener you shove through a hole in two sheets of aluminum, and then you smash the back flat so it can't pull back out. The normal way to do it is with a solid 'driven' rivet: you hold a rivet gun on the front, and you press a heavy chunk of steel called a bucking bar against the back, and the gun hammers the tail into a nice little mushroom. Strong, cheap, been done a million times. The whole trick, though, is that you need to reach both sides of the joint, because that bucking bar has to go somewhere. Remember that, because it's the entire reason I'm in this mess.

And this isn't some throwaway rivet either. The elevator is the flappy bit at the very back of the tail that points the nose up and down, and it swings on these hinge brackets attached to the rear spar (the spar being the big structural beam that runs the length of the surface). Every ounce of load the elevator makes, all of your pitch control, gets funneled through that one joint. It's the lynchpin of the whole tail. Those rivets are not optional and they are not staying empty.

The problem is timing. The proper way to rivet these calls for a bucking bar in a spot that is now sealed up behind a bunch of structure I've already riveted shut. I can't get back there. So after swallowing my pride, I had to figure something out, because those brackets cannot stay unriveted.

The classic fix is genuinely miserable. You drill out 200-plus of the rivets you already put in, peel the structure back open, jam the bucking bar in, buck the hinge rivets, and then close the whole thing up again. Sounds fine on paper. In reality, drilling a rivet out almost never leaves a clean hole. The drill wanders a hair and the hole comes out a touch too big. Do that 200 times and realistically half of them end up oversized, which means reaching for special 'oops' rivets that are one size up to fill the sloppy holes. If I'm really careful maybe I get that down to 15%. Either way I'd be left with a bunch of enlarged holes in what's called primary structure, meaning the load-bearing bones of the airplane, the parts that absolutely cannot fail. Filling those bones with sloppy oversized holes is a great way to quietly weaken the exact thing I'm trying to save.

I wanted something smarter. This is where the blind rivet comes in. A blind rivet goes in from one side only, no bucking bar. It's got a little stem down the middle, and when you yank the stem with the right tool, the back end balloons out and clamps everything tight. If I could use blind rivets here, I wouldn't have to reopen anything, no drilling out 200 rivets, no wandering holes, no oversized mess. The one I had my eye on is an aerospace-grade blind rivet called the CherryMAX. So the whole question boils down to something simple to ask and annoyingly hard to answer: is a CherryMAX strong enough to stand in for the driven rivet the plans want, right here, in this high-load spot?

So first I turned to Van's builder support (Van's designed and sells the kit). They were absolutely unwilling to touch it, basically saying it was on me, the builder, to evaluate the decision. Thanks.

Then I went to the experts on the VAF forum, which is the big community hub for these planes. The advice there was to just read the shear ratings off the CherryMAX product page and be done with it. 'Shear' being the load that tries to slice the rivet sideways like a pair of scissors, which is the number most people grab first. And look, I knew better than to stop there. It's not that the ratings are wrong, I trust them completely. It's that a driven rivet and a CherryMAX are different alloys with different tensile strengths (tensile being how hard you can pull on something before it lets go), so the actual limit of the joint is different, not just the number stamped on one rivet's page. Grabbing one rivet's rating and multiplying by however many rivets there are doesn't really tell you what the assembled joint will hold. I needed to actually quantify the difference between the two, not squint at it and hope.

So I did what any reasonable person with too much CAD software would do and ran some FEA in Autodesk Inventor. FEA (finite element analysis) is basically a simulation of how a part bends and where it gets overstressed when you load it up. You hand it the 3D geometry and the forces, and it paints you a heatmap of where the metal is fine and where it's about to file a complaint. And I didn't model some lonely rivet by itself, I built the whole thing, bracket, spar, and skin all together as one system, with simulated rivets carrying load the way they actually would on the airplane. That's the part 'just read the product page' quietly skips over, because in real life the load shares itself across the joint in ways a single number can't see.

From the current worst-case load, the original driven rivets came out to roughly 150 kg of capacity per bracket pair. Swap in the CherryMAX properties and that same joint handled about 120 kg. Is 120 kg good? Honestly, I still don't know what the required number even is. I asked Van's straight up and got told the numbers are 'readily available online.' They are not. I looked. But at least now I had a real, apples-to-apples comparison instead of a hopeful guess.

Somewhere in here a few folks floated a different fastener called an M-lok, which would've been great, except it doesn't come small enough. Rivets are sized by their diameter in thirty-seconds of an inch, so a '-4' is 4/32 (an eighth of an inch) across and a '-5' is a hair fatter. This joint is drilled for -4 holes, and the smallest M-lok they make is a -5. You can't put a fatter rivet in a skinnier hole. Dead end.

And then, finally, one guy on the forum came through, and honestly I wasn't expecting it. He pointed me straight at Van's own Service Bulletin SB-00036. (A service bulletin is an official notice the manufacturer puts out after the kit ships, usually to fix or clarify something.) And this one calls for CherryMAX rivets in the exact location I'd been losing sleep over.

Just sit with that for a second. Van's builder support wouldn't say a single word about whether CherryMAX was okay here, meanwhile Van's had already published an official document saying yes, CherryMAX, right there, this spot. The answer existed the entire time. I could scream.

So the decision's made. I'm going CherryMAX, they're on order as I type this. And the FEA wasn't a waste either, having those numbers written down means I've got them ready for the next time a question like this pops up, and in this hobby there is always a next time.

All told this little detour cost me the better part of a week. It's not finished, the rivets still have to show up and go in, but I'll report back once they do.