Trailing edge construction

HomeBuiltAirplanes.com

Help Support HomeBuiltAirplanes.com:

addaon

Well-Known Member
Lifetime Supporter
Joined
Feb 24, 2008
Messages
1,696
Location
San Jose, CA
This is a follow-up to a thread I posted earlier today, but I'd rather broaden the discussion and have folks chip in ideas. I'm trying to figure out how to build my trailing edges. What would you do?

Given (tentatively):
Trailing edge angle 16.8°
Skin thickness (each) 0.020"
Target bend radius (6061-T6) 0.040"
Rivet diameter (-4) 0.125"
Minimum rivet edge spacing 0.219"
Target rivet edge spacing 0.250"
Maximum rivet edge spacing 0.500"
Rivet blind clearance (-4-01) 0.355"
Aluminum density 2.7 g/cm^3
Delrin density (glass filled) 1.6 g/cm^3
Minimum TE thickness 0.060"
Target TE thickness 0.125"
Maximum TE thickness ​ 0.250"


Skin thicknesses, rivet sizes, etc are all straw-men to allow comparison, and its possible that tweaking these numbers could lead to a much better approach. I'm interested in finding that tweak!


A note about area and weight numbers… these arbitrarily look at the weight of the portion of the wing that would be the last two inches, given an infinitely sharp trailing edge, ignoring rivets if applicable. Perhaps the best way to think about them is in comparison with Approach 1. The aircraft has about 20 ft of total trailing edge.


This last-two-inch portion of the trailing edge is what's illustrated… I should have shown more scale, sorry. Rivet thickness is a good marker for scaling.


Finally, all of this is conceptual. I'm deliberately ignoring some real-world factors like machining precision on the wedge, under the theory that it doesn't effect the comparison too much.
Approach 1: Bent trailing edge
Cross-sectional area: 66600 mil^2
Weight per linear foot: 0.0780 lbs/ft (0 lbs/ft vs. Approach 1)
TE thickness: ~0.119"
Pros: Simple, light
Cons: Sheet size limitations, rounded TE aerodynamics
1.jpg


Approach 2: Bent-and-magically-riveted trailing edge
Cross-sectional area: 73800 mil^2
Weight per linear foot: 0.0864 lbs/ft (0.0084 lbs/ft vs. Approach 1)
TE thickness: ~0.138"
Pros: Light
Cons: Bad aerodynamics, cannot use blind rivets
2.jpg


Approach 3: Bent-and-blind-riveted trailing edge
Note: By shifting the lap joint forward, we gain enough clearance to use blind fasteners
Cross-sectional area: 73800 mil^2
Weight per linear foot: 0.0864 lbs/ft (0.0084 lbs/ft vs. Approach 1)
TE thickness: ~0.119", with weirdness
Pros: Light, simple, constructible
Cons: Even worse aerodynamics?
3.jpg


Approach 4: Formed trailing edge insert
Note: The max rivet edge spacing means that we can't use blind rivets and take the sheet to the end… this might be fixable
Cross-sectional area: 97000 mil^2
Weight per linear foot: 0.1136 lbs/ft (0.0356 lbs/ft vs. Approach 1)
TE thickness: ~0.119", with weirdness
Pros: Easiest construction, might be able to get a clean trailing edge with different size or BK rivets
Cons: A bit weird ith the current trailing edge, hard to bend a TE sheet and get good straightness
4.jpg


Approach 5: Machined trailing edge wedge
Note: I've arbitrarily chosen a constructible, nice TE thickness here; very tweakable
Cross-sectional area: 111400 mil^2
Weight per linear foot (aluminum): 0.1304 lbs/ft (0.0524 lbs/ft vs. Approach 1)
Weight per linear foot (Delrin): 0.1025 lbs/ft (0.0245 lbs/ft vs. Approach 1)
TE thickness: ~0.070", selectedpr
Pros: Best trailing edge quality and thickness control, easily formed with good straightness
Cons: Strange angle for squeezing rivets, heavy
5.jpg


Other approaches…
A machined/extruded trailing edge in the style of Approach 4 could be used, giving potentially better straightness, TE quality, etc...


Thoughts…
The weight penalty for a machined wedge really isn't as bad as I thought, IF that trailing edge is built to near-minimum dimensions based on rivet spacing (and assuming rivet size). At only a half inch chord, there's just not a lot of material there… going to even a bit bigger (one inch, say) makes the situation MUCH worse. So Approach 4 is very sensitive to changes in rivet size. It also remains to be answered how feasible squeezing rivets on a 17° slope is… the RV guys do have difficulty at 9°.

 
Last edited:

PTAirco

Well-Known Member
Joined
Sep 20, 2003
Messages
3,610
Location
Corona CA
I seriously doubt getting this concerned about T.E.s is warranted. I would bet that none of the above would show any measurable advantages over one another. The air at the trailing edge is likely to be a little messed up anyway (that's an aeronautical engineering term...).

The only consideration I would give to trailing edges is to avoid really thick, rounded ones. Sharp, blunt, thick edges give far better control feel and flutter resistance from what I have read and I doubt they slow the aircraft down in any measurable way.

Just use whatever method is the least hassle.
 

addaon

Well-Known Member
Lifetime Supporter
Joined
Feb 24, 2008
Messages
1,696
Location
San Jose, CA
Not a big deal of the skin-to-skin style trailing edge... just don't love the look, and it tends to get wavy with wear (but then, 150s tend to be older).

Getting this concerned is definitely not warranted, but I'm blocked on a few other things and figured I'd make the decision. I mean, the fact that there's not much to choose between the different options is more or less the point... they're very close in weight, performance, and producibility. But I do need to choose one, so I figured (since I wasn't going to be pulling rivets yet anyway) I'd get some feedback on /which/ is likely to be the least hassle.
 

addaon

Well-Known Member
Lifetime Supporter
Joined
Feb 24, 2008
Messages
1,696
Location
San Jose, CA
I'm leaning towards #5 myself, despite the minor weight penalty. Reasons:

1) I'm pretty comfortable machining straight aluminum bar into straight wedge; keeping straightness for the same length bend seems harder.

2) My full trailing-edge length is 10', and I can easily get 10' aluminum bar, where as it's a pain to get sheet longer than 8'.

3) It looks good.

My concerns are:

1) Is it really sane to build the wedge to minimum rivet clearance on both sides?

2) How the heck am I going to rivet at a 16° angle?
 

Matt G.

Well-Known Member
Joined
Nov 16, 2011
Messages
1,253
Location
Kansas, USA
I think #2 is essentially impossible...there's not enough room for a bucked rivet or a blind one. I'd go with either #5 or D Hillberg's suggestion.
 

autoreply

Well-Known Member
Joined
Jul 7, 2009
Messages
10,763
Location
Rotterdam, Netherlands
5 has a very high weight at the rear of the aileron. Not a problem for a wing, but very problematic for ailerons and possibly flaps (you have to balance it out. A triangular (glued) foam core would be my personal preference, simple and light.
 

Himat

Well-Known Member
Joined
May 5, 2011
Messages
2,868
Location
Norway
#5, but with a difference.

If the rivets are not to close spaced and the alloy allow it, press a recess in the lower skin. The rivet would then be sitting just as when two sheets are riveted together. There will be a “sharp” trailing edge, less a recession on the underside at each rivet position. This recession could be levelled out with filler afterwards.
 

addaon

Well-Known Member
Lifetime Supporter
Joined
Feb 24, 2008
Messages
1,696
Location
San Jose, CA
I think #2 is essentially impossible...there's not enough room for a bucked rivet or a blind one. I'd go with either #5 or D Hillberg's suggestion.
#2 is a pain in the neck, but in my particular case it's constructible. Rivet, then put the whole piece in the brake and bend. Happens to work reasonably well since the top skin is flat in the final part. Still, it seems to be losing the popularity contest.
 

addaon

Well-Known Member
Lifetime Supporter
Joined
Feb 24, 2008
Messages
1,696
Location
San Jose, CA
5 has a very high weight at the rear of the aileron. Not a problem for a wing, but very problematic for ailerons and possibly flaps (you have to balance it out. A triangular (glued) foam core would be my personal preference, simple and light.
I'm using elevons, so this was indeed my concern, and is still on my mind... but if it's really possible to make a wedge that small, we're talking 1 lb per elevon. Even if they're fully mass balanced, it's not /that/ extreme. But there is indeed a weight penalty.
 

addaon

Well-Known Member
Lifetime Supporter
Joined
Feb 24, 2008
Messages
1,696
Location
San Jose, CA
I guess another approach is something like this (#6), currently draw to minimums but obviously tweakable to get a flusher TE:

Cross-sectional area: 53900 mil^2

Weight per linear foot: 0.0631 lbs/ft (0.0149 lbs/ft LIGHTER THAN Approach 1)
TE thickness: ~0.347"

The trailing edge thickness is quite high, but in my acceptable range (this is a pretty big-chord wing); blind rivets are usable, which I tend to prefer where viable; bending a C-channel isn't ridiculous (though keeping straightness for one this dainty might be tricky)... and the TE is truncated enough that weight is actually better than the nominal case. Concerns: Bending the thing accurately.

I need to cut some wedges and see what it feels like to rivet on that big a slope (for #5)... unfortunately I'm not near my shop.

6.jpg
 

autoreply

Well-Known Member
Joined
Jul 7, 2009
Messages
10,763
Location
Rotterdam, Netherlands
I'm using elevons, so this was indeed my concern, and is still on my mind... but if it's really possible to make a wedge that small, we're talking 1 lb per elevon. Even if they're fully mass balanced, it's not /that/ extreme. But there is indeed a weight penalty.
The problem is that the weight "avalanches" quickly. You usually can't balance much fwd of the aileron (elevon), so you need a multiple of that weight as balance weight, easily 5 times as much. Still by no means massive numbers though.
 

addaon

Well-Known Member
Lifetime Supporter
Joined
Feb 24, 2008
Messages
1,696
Location
San Jose, CA
The problem is that the weight "avalanches" quickly. You usually can't balance much fwd of the aileron (elevon), so you need a multiple of that weight as balance weight, easily 5 times as much. Still by no means massive numbers though.
I was trying to keep this out of the thread, but this is actually a rigid control system, so rotational moment of inertia is important, but mass balancing is not. (Linkage failure is already critical failure, and designing linkage to be rigid and high-tolerance enough to avoid control surface flutter shouldn't add much weight.)
 

OldT6Flyer

Member
Joined
Mar 3, 2013
Messages
6
Location
United States
You probably know that the Van's planes used the folded trailing edge approach on ailerons, flaps, and tail feathers in the earlier models and switched to the riveted wedge in the later models. Riveting isn't that big a deal as they use a "double flush" approach. Benefit is that you can buy the machined wedge from them that may be "close enough" to you exact airfoil (within manufacturing tolerances) if its profile approaches those used on the Van's designs...

The challenge with the riveted wedge approach is not getting a "wave" into the TE while riveting the assembly together. Tons of advice on technique in the usual Van's forums...

Richard
 

Hot Wings

Grumpy Cynic
HBA Supporter
Log Member
Joined
Nov 14, 2009
Messages
7,635
Location
Rocky Mountains
The aerodynamics of bluff trailing edges are pretty well understood, and pretty positive. Any information on undercut / gapped trailing edges?
If you wanted to reduce the gap - the reversed "C" section could be made much smaller than the drawing in post #14. Drill and counter sink the "C" before bending. Machine and polish one steel filler strip to use as a bucking bar. Insert the "C" section and drill the skin to match then countersink if desired. Put counter sunk rivets in from the inside of the "C", slide the bucking bar strip/filler under the rivets and buck both top and bottom rivet???

Fill the resulting gap with a bit of spray foam and coat with some polyester gelcoat for a flat and flush finish.

It's too bad we can't still get explosive rivets :depressed
 

TinBender

Well-Known Member
Joined
Mar 4, 2010
Messages
218
Location
Charlotte, NC, USA
Two ideas for brainstorming :

The extrusion in red: You can countersink the extrusion then dimple the skins. Produce a shop head on each side proud of the surface, then shave them flush. One could conceivably tailor the trailing edge thickness thicker after testing as required by sanding it down. Heavy
te.png



Joggle the TE's together and rivet, then fill the areas in blue with aerodynamic sealant. Heavy
te2.pngMethod used on Airbus elevator TE, albeit in composites. It may be important to point out that Airbus used a hard filler that would occasionally shed little pieces. I worked on some of the engineering authorizations in removing the filler during overhaul and re-installing it on the plane without any filler (after weighing.)
 
Top