Wing spar load testing question

Homebuilt Aircraft & Kit Plane Forum

Help Support Homebuilt Aircraft & Kit Plane Forum:

rtfm

Well-Known Member
Joined
Jan 3, 2008
Messages
3,754
Location
Brisbane, Australia
Hi,
I've decided to test my spar only, not my entire wing. If the spar fails, no big real. If the wing itself fails, I'll be thoroughly pissed off.
I recently watched a video by Mike Patey () where he tests the "Scrappy" spar. I found this a novel way to test the wing's ability to allow you to land alive.

However, I have a number of issues with his testing method:

(1) He applies force to the tip of the wing
(2) His method applies a point load at the centre of the wing.

Neither of these are applicable to real-world situations.
I would have applied the pulling force inward of the tip, but I'm having some difficulty determining exactly where. Looking at the lift distribution of a typical rectangular wing plan *seems* to suggest somewhere near the centre of the wing half. Would this be reasonable?

And if this is a reasonable assumption, then how much pulling force is required? The Fleabike fully loaded will weigh in the region of 256kg. The rear wing (for example) carries 40% of the load. Under 4G this equates to 409,6kg, and therefore each wing half needs to support 204.8kg

My problem is: do I apply 205kg at the half-span? Is my thinking correct? My wing is strutted, with the strut half-way between the wing centre and the half-span position. It seems to me that the strut is going to carry the lion's share of the load, with the 560mm outboard section (ie from the strut to the half-span position) bearing a significantly reduced load. This is important, because the wing folds at the strut, and all outboard loads will be referred back to the hinge.

I really like Mike's flat-on-the-floor testing method, by the way.

Regards,
Duncan
1662883487881.png
 
Last edited:

wsimpso1

Super Moderator
Staff member
Joined
Oct 18, 2003
Messages
9,498
Location
Saline Michigan
Duncan,

The issue is how closely do you want to mimic the in-flight loads.

First thing to know is your lift distribution is almost always pretty close to elliptical. This combined with how the lifting load is reacted into the rest of the airplane produces the shear and bending moment diagram for the wing.

So, if you want to test the whole spar for its capability relative to the whole wing loads in flight, you have to anchor the spar like it is attached to the rest of the airplane and load it like the lift is distributed in flight.

Now Scrappy's wing is basically a Super Cub wing scaled up. The Super Cub and many other strut braced high wing airplanes use main and drag spars of constant cross section. The main spar is bigger than the drag spar, but each has its own section from tip to root. More than that, the struts are attached at about 65% of the semi-span, which minimizes the bending moment seen in the spar set. See the worked example below. This type of arrangement allows the designer to use a constant section spar with only very modest weight penalties, which is done in many strut braced designs. In this example , gross weight was 500 lbs, span was 360 inches, and max g is 4.2.

The cantilever wing is also shown, note how the biggest shear in the cantilever spar is almost 4x that of strut braced, and bending moment is almost 10x in the cantilever spar vs strut braced.

1662908625663.png

So, if you know your max shear and bending moment, and you are using a constant section spar, you can just pick the anchors and loading point so you get your design shear and bending moment. That appears to be what Mike Patey did...

But if instead, you have a cantilever or near cantilever wing (one piece wing supported on a set of cabane struts above the fuselage), you have widely varying loads along the span. Use a constant section spar big enough for the root and it is way oversize and perhaps heavy in the outer 75% of the semispan. You get to judge if tailoring the spar is worth the effort and analysis. If you decide to use a constant section, you too can pick off the most severe spot and design your testing scheme to suit that. But if you tailor the spar to the loads, you do not know where your strength might be short of the loads - you pretty well have to anchor per your airplane arrangement and load per your airloads.

PM me...
 
Last edited:
Joined
Nov 14, 2009
Messages
8,764
Location
Rocky Mountains
I've posted this in other threads but I think it is worth repeating:

Jeff Hanson on You tube should be a must watch for everyone trying to design their own airplane, and haven't had a formal Statics class. He has 94 videos in this series. Some of the newer ones I haven't watched so I can't comment on their relevance. He starts out with simple vectors so it is an appropriate course for someone just starting down this path.
The whole series will require about 30 hours of face time on YouTube. This video is of particular relevance to the question asked:

 

Aesquire

Well-Known Member
Joined
Jul 28, 2014
Messages
3,358
Location
Rochester, NY, USA
His method seems sound. It's not the same as you load a wing for testing, but for a A to B test of a single, worst case it's good. A distributed load should get the same result, but at much higher loads. & danger.

The real world equiv. would be a plane falling, ( softly) off a jack after you removed the landing gear on one side, onto a wing tip stand.

Very revealing how important rib spacing is.

Keep in mind, this is a test against the calculations. The figures on paper are for the single parameter of that tip load. That's just one of many calculations that applied to the different parts. And frankly, not one I'd think I'd even bother with doing unless I'd thought of this test to confirm the spar strength.
 

Aesquire

Well-Known Member
Joined
Jul 28, 2014
Messages
3,358
Location
Rochester, NY, USA
My problem is: do I apply 205kg at the half-span?
That would test the strut and fuselage strut attachment point?

My only personal experience is with an entire hang glider hung upside down and loaded as calculated. Testing the entire craft, and the weakest part fails. Which on our test wasn't the thing we expected. Instead of the cross spar, the control bar uprights deflected first, which then caused the cross bar to get twisted out of line. That tracked with video of an in flight failure during aerobatics. But the failure was as expected, an S bend around a central bolt under compression of a 1 3/4" aluminum tube.
 

rtfm

Well-Known Member
Joined
Jan 3, 2008
Messages
3,754
Location
Brisbane, Australia
That would test the strut and fuselage strut attachment point?

My only personal experience is with an entire hang glider hung upside down and loaded as calculated. Testing the entire craft, and the weakest part fails. Which on our test wasn't the thing we expected. Instead of the cross spar, the control bar uprights deflected first, which then caused the cross bar to get twisted out of line. That tracked with video of an in flight failure during aerobatics. But the failure was as expected, an S bend around a central bolt under compression of a 1 3/4" aluminum tube.
Hi. Mmmm The Fleabike is small enough to physically flip upside down, attach the wings and test, as you describe. I'd be very comfortable with doing that AFTER I'd proof-loaded the spars flat on the ground, the way Mike Patey did. This would, essentially, be testing the rest of the airframe, the attach points and so on. Great idea.
 
Last edited:

rtfm

Well-Known Member
Joined
Jan 3, 2008
Messages
3,754
Location
Brisbane, Australia
Duncan,

Scrappy's wing is basically a Super Cub wing scaled up. The Super Cub and many other strut braced high wing airplanes use main and drag spars of constant cross section. The main spar is bigger than the drag spar, but each has its own section from tip to root. More than that, the struts are attached at about 65% of the semi-span, which minimizes the bending moment seen in the spar set. See the worked example below. This type of arrangement allows the designer to use a constant section spar with only very modest weight penalties, which is done in many strut braced designs. In this example , gross weight was 500 lbs, span was 360 inches, and max g is 4.2.

But if instead, you have a cantilever or near cantilever wing
Hi Bill,
First, thank you for taking the time to write your reply. believe I misled you (and everyone else) with my original sketch. I omitted to draw in the struts (see updated sketch below).
1662929503262.png

The wing is attached to the fuselage with two cabane struts, but also has struts which attach to the 50% chord point (where the wing hinges.) I've indicated both the hinges (on top) and the locking mechanism (bottom).

Also, my original scenario assumed all the load was to be borne by the single spar. But there are, in fact, two spars, so the original scenario was the worst case.

I will PM you also.

Duncan
 

rtfm

Well-Known Member
Joined
Jan 3, 2008
Messages
3,754
Location
Brisbane, Australia
Very revealing how important rib spacing is.
Yes, I found that interesting also. I haven't built the wing yet, so I still have time to revisit the rib spacing. However, bear in mind that the Fleabike weighs 256kg all-up. So we're not talking heavy. But the principle is the same.
 

rtfm

Well-Known Member
Joined
Jan 3, 2008
Messages
3,754
Location
Brisbane, Australia
I've posted this in other threads but I think it is worth repeating:

Jeff Hanson on You tube should be a must watch for everyone trying to design their own airplane, and haven't had a formal Statics class.
Gosh - what a cool teacher. Thank you.
 

rtfm

Well-Known Member
Joined
Jan 3, 2008
Messages
3,754
Location
Brisbane, Australia
Duncan,

The issue is how closely do you want to mimic the in-flight loads.

First thing to know is your lift distribution is almost always pretty close to elliptical. This combined with how the lifting load is reacted into the rest of the airplane produces the shear and bending moment diagram for the wing.

So, if you want to test the whole spar for its capability relative to the whole wing loads in flight, you have to anchor the spar like it is attached to the rest of the airplane and load it like the lift is distributed in flight.

Now Scrappy's wing is basically a Super Cub wing scaled up. The Super Cub and many other strut braced high wing airplanes use main and drag spars of constant cross section. The main spar is bigger than the drag spar, but each has its own section from tip to root. More than that, the struts are attached at about 65% of the semi-span, which minimizes the bending moment seen in the spar set. See the worked example below. This type of arrangement allows the designer to use a constant section spar with only very modest weight penalties, which is done in many strut braced designs. In this example , gross weight was 500 lbs, span was 360 inches, and max g is 4.2.

The cantilever wing is also shown, note how the biggest shear in the cantilever spar is almost 4x that of strut braced, and bending moment is almost 10x in the cantilever spar vs strut braced.

View attachment 129719

So, if you know your max shear and bending moment, and you are using a constant section spar, you can just pick the anchors and loading point so you get your design shear and bending moment. That appears to be what Mike Patey did...

But if instead, you have a cantilever or near cantilever wing (one piece wing supported on a set of cabane struts above the fuselage), you have widely varying loads along the span. Use a constant section spar big enough for the root and it is way oversize and perhaps heavy in the outer 75% of the semispan. You get to judge if tailoring the spar is worth the effort and analysis. If you decide to use a constant section, you too can pick off the most severe spot and design your testing scheme to suit that. But if you tailor the spar to the loads, you do not know where your strength might be short of the loads - you pretty well have to anchor per your airplane arrangement and load per your airloads.

PM me...
S Braced/M Braced?
 

daveklingler

Well-Known Member
Joined
Jan 22, 2013
Messages
180
Location
Albuquerque
I've posted this in other threads but I think it is worth repeating:

Jeff Hanson on You tube should be a must watch for everyone trying to design their own airplane, and haven't had a formal Statics class. He has 94 videos in this series. Some of the newer ones I haven't watched so I can't comment on their relevance. He starts out with simple vectors so it is an appropriate course for someone just starting down this path.
The whole series will require about 30 hours of face time on YouTube. This video is of particular relevance to the question asked:


I watched a couple of the series videos, but got distracted by the fact that a guy teaching statics doesn't know how to spell "scalar" (he spells it "scaler", like the dental tool, hundreds of times throughout all his videos) or "graphical" (he spells it "grafical").

I'll grant that an engineer who can't spell is more common than not, but those particular words...that's like an aeronautical engineer who can't spell the word "airplane". Or, well, "scalar", because few words in math or engineering or science are more basic than that one. I guess I've seen something new. :)

There are also other comparable series on statics and dynamics throughout Youtube and on Khan Academy. Keep in mind that engineering is not a spectator sport. You're better off if you go buy a mechanics of materials text and do some of the problems.
 

mcrae0104

Well-Known Member
Supporting Member
Joined
Oct 27, 2009
Messages
4,290
Location
CO
I watched a couple of the series videos, but got distracted by the fact that a guy teaching statics doesn't know how to spell "scalar" (he spells it "scaler", like the dental tool, hundreds of times throughout all his videos) or "graphical" (he spells it "grafical").
I’m confused; you said you watched a couple of videos but then said he misspells a particular word “hundreds of times throughout all his videos.” So did you watch them all?
 

SpruceForest

Well-Known Member
Joined
May 23, 2022
Messages
240
I watched a couple of the series videos, but got distracted by the fact that a guy teaching statics doesn't know how to spell "scalar" (he spells it "scaler", like the dental tool, hundreds of times throughout all his videos) or "graphical" (he spells it "grafical").

I'll grant that an engineer who can't spell is more common than not, but those particular words...that's like an aeronautical engineer who can't spell the word "airplane". Or, well, "scalar", because few words in math or engineering or science are more basic than that one. I guess I've seen something new. :)

There are also other comparable series on statics and dynamics throughout Youtube and on Khan Academy. Keep in mind that engineering is not a spectator sport. You're better off if you go buy a mechanics of materials text and do some of the problems.
At least he's not the guy with the finger on the 'nut-shaped' button... and before someone trots out that 'regional variation' stuff, same chuckleheads talk about 'the nuclear family' without the mispronunciation.

I watched a few of Jeff Hanson's vids and wish this guy would have been around when I was wading through my statics/dynamics/mechanics of materials lower level course sequence. I'd rather have some mildly annoying affectations (talking about you, Dr. Winkie, and you, Bruce D.) than crap instruction pretty much any time. If I decide to carry any more guitar-building students, there are a few Hanson vids that will be added to the long list of stuff I make them suffer through on the way to a guitar.

For those that stayed away from engineering in college, the statics/dynamic/mechanics of materials sequence is pretty fundamental stuff for engineers... wade through Hanson's online vids and refresh on some math, and you'll have much of what you need to work through undergrad and industry structures stuff as well as exercising the metal muscles needed for some of the design stuff discussed here. Sure - absolutely pick up a ninth or tenth edition copy of Hibbeler's combined statics/dynamics series (check Abe's Books...usually under $9) and either his Mechanics of Materials text (again, $6 or so) or Timonshenko, etc. - no substitute for working problems, right?
 
Last edited:

nestofdragons

Well-Known Member
Supporting Member
Joined
Jun 8, 2016
Messages
819
Location
Near Antwerp, Belgium
I've posted this in other threads but I think it is worth repeating:

Jeff Hanson on You tube should be a must watch for everyone trying to design their own airplane, and haven't had a formal Statics class. He has 94 videos in this series. Some of the newer ones I haven't watched so I can't comment on their relevance. He starts out with simple vectors so it is an appropriate course for someone just starting down this path.
The whole series will require about 30 hours of face time on YouTube. This video is of particular relevance to the question asked:


Just replying to have this post in my listing for further reading. This is a must safe post. Thanks!!
 

dog

Well-Known Member
Joined
Dec 29, 2019
Messages
816
the video in question has one simple good atribute up front,he defines his terms and as long
as he gets the overall "shape" of the word right the spelling might just be invisible to me
got "new pipe" installed just so I can watch it
without the adds
 

SpruceForest

Well-Known Member
Joined
May 23, 2022
Messages
240
Whiteboards and blackbirds can be pretty unforgiving of errors. No spell-check or autocorrect as you work through a problem or derivation.

Back in the closing days of the 1980's, I had to take an early final with John D. Anderson (if you've taken an aero course in the US in the last 40 years, you've likely used or at least referenced his texts), my Aero 445 prof (due to a deployment with my unit at the time, one of the Army's special forces groups). Dr. Anderson always had a couple derivations or proofs to work through, a definition or two, and the usual worked problems from the material covered in the syllabus. He also loved to throw in a bit of history as well, and even an ad hoc final thrown together for my very singular enjoyment had all that content. There was over 120 linear feet of blackboard in the Aero Staff Room, and I think I used every bit of it to work things as Dr. Anderson peppered me with questions. When I finished, he did a quick walk around the room, then with a piece of blue chalk circled one punctuation error and one spelling error in the response to the history of aero engineering question (IIRC, a misplaced semicolon and a butchered proper name).

"That's a 98... and those lost points were wholly avoidable with a bit more focus on the details."

I guess spelling and punctuation does count, or at least enough on that day to keep me out of the top spot in the class that semester.
 
Last edited:
Top