What's new
Van's Air Force

Don't miss anything! Register now for full access to the definitive RV support community.

Double Flush Criteria

DanH

DanH

Legacy Member
Mentor
I've missed something somewhere, or maybe not.

We're all familiar with Section 5, but I'll repeat it here for clarity.

Sec 5.jpg


Question is, how big is an acorn? When I inspect, I see them in all sizes, quite a few being small enough to make me nervous. Here are some pretty ones, at least to my eye:

Excellent.jpg


A little less wonderful, good enough for most, but...

Less.jpg


And I've seen some small enough to pull through without much effort.

So, is there a published criteria specifically for the shop heads on trailing edge double flush rivets? I've been assuming it is the standard 1.5 x shank size minimum, so please correct me if wrong.

I doubt many folks are using vernier calipers with pointy little jaws to measure a shop head down in a recess. I assume most are doing it visually. Here's a visual representation, to scale, comparing manufactured head diameter (i.e dimple diameter) to minimum (I think) shop head diameter.

Comments?

Double Flush.jpg
 
Last edited:
MIL-R-47196A, the riveting spec from Van's, does not have information about double-flush rivets. I think I'd aim for a shop head diameter that meets the spec's Table III. For a -3 rivet, that's .122 inches minimum, to avoid pull-through and ensure that the rivet is fully set and its body adequately fills the hole.

While flushness is very desirable here, I think it would be good not to exceed that diameter by very much.

Just my opinion.....
Dave
 
The minimums standards for shop heads on the riveted training edges is the same as for any other shop head for the same sized rivet (as posted by David) as detailed in MIL-R-47196A .
So the accepted minimum diameter is actually less than 1.5 D. I agree that the only way to accurately check the diameter would be using a pointed tip caliper.
The shop head thickness standard for rivets in the trailing edge dimples has always been whatever thickness results from setting the shop head flush with the skin surface when a proper diameter is achieved (it can't be set any lower than what the skin surface is).

As for the "acorn sitting in a dimple" comment....:rolleyes: I have no idea where that came from and I have never noticed it or had anyone else point it out to me.
For certain I know it didn't come from engineering. I see no value in it being in the text and think it should be removed.

The comment about not attempting to fill the entire dimple on the shop head side with rivet material is from engineering.
Doing so causes a lot of plastic deformation of the skin material in the vicinity of the dimple because of the high setting force required, and the trailing edge will look awful.
 
Full disclosure: Never built a riveted trailing edge surface, buuuut.... I would think alternating direction on each rivet in the line would at least leave a full head in the dimple and then into the wedge to (maybe?) reduce a skin "un-zippering" threat. Seems like if all are the same direction, then the shop head side could have more of a tendency to pull-thru the skin. (as you mentioned)
 
Never really thought about this. Looks like the thickest part of the shop head (assuming it’s the same height as the skin) is a bit less than the nominal rivet radius, a bit smaller than the usual recommendation.
 
David Paule said:
MIL-R-47196A, the riveting spec from Van's, does not have information about double-flush rivets. I think I'd aim for a shop head diameter that meets the spec's Table III. For a -3 rivet, that's .122 inches minimum, to avoid pull-through and ensure that the rivet is fully set and its body adequately fills the hole.

While flushness is very desirable here, I think it would be good not to exceed that diameter by very much.

Just my opinion.....
Dave
Click to expand...

Why would you not exceed that diameter (0.122 shop head for a -3) by very much? Structural consideration, or not deforming the trailing edge as Scott mentions above?
 
They will be getting fairly thin around the edges as the diameter increases. That's all. If it doesn't crack and is still flush, it probably won't hurt anything.

Dave
 
Hard to believe there is no aerospace standard specifically addressing double flush rivets. The failure mode I'm seeing in my mind's eye has the small shop head pulling through the bottom of the dimple, similar to the way the manufactured head pulls through a nested dimple.

Shop Head Side.jpg
Manufactured Head Side.jpg
Photo credits: Bill Marvel

A #40 hole expands to about 0.104" D when dimpled, so a 0.122 diameter shop head only has 0.009" of overlap at the edge of the hole, in the bottom of the dimple. And that's measuring at the largest diameter. Depending on the bulge of the shop head, the actual contact diameter may be as little 0.118

To scale, -3 rivet with 0.016 skins:

3 rivet 016 skins.jpg
 
Last edited:
DanH said:
Funny you should mention that...
Click to expand...
Just looking at the drawing, I’d agree that the shop head looks a bit small, 0.141” is probably ‘better’. Nice drawing, but don’t forget that this substrate is a wedge- the factory and shop heads aren’t parallel. I have no idea how that factors into the strength. But also, when it comes to ‘pulling thru’, the substrate is countersunk-it’s a #40 hole, no larger.
 
BobTurner said:
...don’t forget that this substrate is a wedge- the factory and shop heads aren’t parallel. I have no idea how that factors into the strength.
Click to expand...

Doesn't matter structurally. Load is mostly 90 degrees to the wedge angle, along the length.

It does ensure a higher percentage are poorly set, as compared to conventional rivet installation.

BobTurner said:
But also, when it comes to ‘pulling thru’, the substrate is countersunk-it’s a #40 hole, no larger.
Click to expand...

As drawn.
 
Acorns.
I agree with Dan in being concerned with small or unfinished ‘flush’ shop heads. I have seen trailing edges that had rivets squeezed with flat sets but not finished off flush. I’ve seen trailing edges that had the wedge prosealed with a thick bead & only cleko’d till cured (resulting in an over thick sandwich) than was riveted together with the now too short rivet & resulting undersized shop heads…
I’ve also seen pictures of a failed rudder that had the trailing edge with small acorn shop heads come apart like a zipper (just one portion of a catastrophic failure).
 
Coupon pull tests are ongoing, but our intrepid tech doesn't have enough trailing edge wedge to build all the coupons required to do it right. Apparently Vans is backordered for a while. Anyone have spare wedge material?

Break...

Bored this evening? Participate in a simple field experiment

Drill three or four #40 holes in 0.016 sheet. Dimple the holes using your favorite standard die set. Use your number 39 through 30 drill bits to find the GO - NO GO diameters. Report back here.
 
Last edited:
3 holes. .016” sheet drilled #40
Cleveland AC Tool dies circa 2019.
Go #36
No Go #35
the holes measure 0.107” with pointed calipers.
 
Last edited:
You must log in or register to reply here.
Back
Top