blended think tank

[steve]

i did lots of math and so guessing, the stab is 59.5 deg slope back. i changed it to 58.3 and lost a pinch of drag, to offset a small lose of lift in the wing end(7" of flat to 9" of curve). i cut 7" off so as to end up with the stab in the same spot as planes(I'm over by 7/8").its early and i mite cut and lose the 1"'

there is a lot more going on with this then what meets the eye. i started the curve in the LE the plans is 4.5 back(so every thing is out the window).

the rudder is a biggie to because you lose the 9" in the curve but need to get it back or do the math:irked: i have not got mine cut out yet (tonight with some luck) i don't know when the plans say to cut it:p

id give Marc the bucks but he still wants his Ginni's:brocolli:

this is paint by numbers and i just will stay inside the lines.

when done, it will be the same as plans just different:p

where are the pic's of jacks blend (i want to see his rudder height).

[steve]

I'll make an offer to anyone who is really set on blending their wings:

 

I have in my possession two (2) wings and winglets. One has have some minor defects in the wing root area -- probably not enough to scrap them, but since I didn't build them I've already decided not to build new ones from scratch.

 

For a reasonable price, I will sell you these mostly good wings & winglets, which you can then use for destructive testing of your particular blended wing design..

do you know what shipping would be ?

 

 

Unless the design is validated with at least static load testing, I wouldn't fly under blended winglets..

can you do this ? i looked at it and many say it needs not a spar and sheer?

i could use more input if you have it ?

 

 

-- One more thing. Dust's blog (not so much a forum any more) has interesting pictures on how to shape the foam for a blended winglet, but if he is using no spar or other reinforcements to hold the winglet on, I would steer far clear of that design.

it was not the look i wanted, i believe it was a compromise to what he wanted.

[Lynn Erickson]

I'd hoped to stay out of this mess, but in the interest of on-lookers' safety, I've got to say some stuff.

 

First, as the person who advised on the aerodynamics of Jack's blended winglet implementation and designed the structural modifications that he used successfully (and so far, the ONLY successful implementation of blended winglets on a Rutan Derivative canard aircraft), I think that I have a bit of authority on this matter.

 

Secondly, there are only three people that have asked for and received my assistance with blended winglets - one on an E-Racer, one on a COZY, and one on a Long-EZ. Only the E-Racer (Jack's) has flown. I don't know when the others will do so.

 

Thirdly, I have been approached for engineering assistance with this modification by a few other Long-EZ builders, none of whom have followed through with me on getting that assistance (yet). At least one of those folks is here on this forum stating that they're implementing this modification. They have stated that they will not be proceeding with actual fabrication until they've received my input.

 

Fourth, there is at least one person (not on this forum - the current admin of the defunct "Canard Community Forum") who, in my opinion, is building an extremely dangerous blended winglet design from a structural standpoint, going from what he writes on his blog. (And who knows what he's doing from an aerodynamic standpoint - I wasn't able to decipher the description of the winglet positioning or angles). He is essentially following Steve Innova's question #3 from the third post in this thread:

 

"#3. Why bother? The winglet doesn't see enough force on it to require a thick spar. The plans method of using BID layups seems to work fine. I'd rather not add weight."

 

First of all, the plans method uses a large assortment of mostly UNI and also BID layups, both internally and externally - it's not just BID.

 

Secondly, the answer here is that the blended winglet has completely different load paths and skin buckling issues than does the plans winglet, and leaving out either a shear web/spar cap arrangement, or another internal support structure (one of which I provided to Jack as an alternative when we were working on his E-Racer) is begging for trouble. If this path (no internal structure) is taken, I would suggest a FULL Limit-Load test plan for the new winglets (and this is something that I have implemented with Jack for his new aircraft design - just ask him what's involved in the test procedure, and how much weight is needed, if he'll tell you before the unveiling).

 

Unless a builder either follows a proven structural methodology or full Limit Load static proof testing, I would not touch their aircraft with a blended winglet implementation with a 10-foot pole, much less fly in it. My reading of Mr. Parkins' actions along this line put his airplane into this category (not withstanding any other issues I've got with it).

 

Steve Innova asks some good questions, and unless there's another qualified engineer out there advising canard folks on blended winglet designs, I'm not aware of these answers having been provided to those others building or contemplating this modification. It's a lot more complex than just taking some subset of the plans layups and calling it good, and there are more aerodynamic issues than just what the radius of the blend is.

 

This is NOT something where free advice will be dispensed, given the extreme safety issues involved.

Thanks, Mark for the comments on this subject because these guys a scaring the hell out of me. the design work you did on jacks plane looks good and I believe it will be successful. but jacks plane was the only one built with this mod and was it completely tested under all conditions? what happens if someone does not build it exactly to your plans and some one gets hurt? the guys that are building this mod without your help better have some engineering support from somewhere. this not a place to make a mod by holding up your thumb and thinking if it looks good, it will fly.

[Marc Zeitlin]

Thanks, Mark for the comments on this subject because these guys a scaring the hell out of me.

Some of them, yes. The problem is that many folks don't know what they don't know. The smart ones know what they don't know, and are willing to get real help - not hand waving on the internet from others who don't know what they don't know.

 

... but jacks plane was the only one built with this mod and was it completely tested under all conditions?

Well, Jack's a very thorough guy, and stated that he tested the plane incrementally to the limits that I recommended to him, which mimicked, as much as possible, the worst case conditions in flight. He did not do static load tests, but he did include structure that ensured a VERY high safety margin - much higher strength than the plans provide.

 

It would have been a good idea to static load test the winglets, as he's done with his new plane, but the design there is SO different from plans that there was no way I'd have let him fly it without the static load tests.

 

what happens if someone does not build it exactly to your plans and some one gets hurt?

The same can be asked about the stock LE, VE, and COZY plans. I don't have a good answer for that - when someone goes off into the weeds on their own, they're on their own...

 

the guys that are building this mod without your help better have some engineering support from somewhere. this not a place to make a mod by holding up your thumb and thinking if it looks good, it will fly.

I agree completely. I'm sure that there are numerous engineering consultants out there that can assist with work such as this - I'm not the only one - others can be found in the back of Sport Aviation or Kitplanes. But I'm the only one that's assisted with it so far, AFAIK.

 

I believe that Klaus is working on such a mod for his next plane, but I don't think he's flown it yet, or what the structural or aerodynamic design is. He probably wouldn't tell us anyway.

[steve]

it was a short night but the numbers are there and the look for the most part is there. i think all my deg's were right because things lined up so well.

i have lots of stuff to rethink(rudder cable, wire, antenna, nav, est so will hold off till i finger them out.

it has a nasty twist/load and i don't know if no sheer and spar is so hot a thing.. so to help get the twisting and side loading stronger i will cut the sheer web all the way up the winglet(same as wing) then do a small spar 1/2 to 3/4 up. i have lots of pics but cant load here because they are 500kb and up so you can see them on Dust blog, ill try one:rolleyes: 530kb 524kb all to big sorry

[mplafleur]

it has a nasty twist/load and i don't know if no sheer and spar is so hot a thing.. so to help get the twisting and side loading stronger i will cut the sheer web all the way up the winglet(same as wing) then do a small spar 1/2 to 3/4 up.

I advised dust to do this, but he insists it's unnecessary.

 

i have lots of pics but cant load here because they are 500kb and up so you can see them on Dust blog, ill try one:rolleyes: 530kb 524kb all to big sorry

Get yourself a photobucket account and upload your pictures there. Then just insert the image urls into your text here.

[Steve Innova]

Load testing wings & winglets:

 

This page http://www.cozybuilders.org/Canard_Pusher/RAF_LE-Structural-Test.pdf describes (with pictures) a static load test on a Vari-EZ wing, canard, and winglet.

 

While the Vari-EZ wing is somewhat different from the Cozy, and quite different from a blended Cozy winglet, this would be a good starting point for learning how to set up and test a modified winglet.

 

This particular canard failed at only 8.4 g's, which was much less than the often [mis?] quoted 12gs. Why did at fail at a lower limit? Probably builder technique, which goes to show that even if the modified design, layup schedule etc... is good, variances between builders can have a big impact on the strength of the part. I've heard some advice on adding a 2.0 safety factor to homebuilt designs to account for those variances.

 

But bear in mind -- this is a static, ultimate load test. We simply don't have the equipment neccessary to conduct dynamic load tests on wing structures. That's what the test period is for, I guess...

 

 

For those that care, in the next month I'll post pictures of the static load test of my modified composite "A" fram seat-back, which is intended to meet the FAR 23 seat / harness requirements. The test rig is proving to be quite the design challenge -- I've had to re-build a 18" fuselage cross section to attach the seatbacks into. Fun!

 

If anyone has a line to where I can get lead shot bags cheaply, please let me know!

[TMann]

Get yourself a copy of SolidWorks w/Cosmos. Give the landfill a break.

Most of what your talking about can be proven with software. You can weed out the failures without pumping any epoxy or cutting any foam.

 

If I was static failure testing all my parts that's all I would be doing is rebuilding parts (and I do enough of that now.) This is why we buy plans and incorporate proven designs from reliable/qualified sources.

 

I want to be flying this bird ......... soon. The $100/hr for the Archer is really cutting into my budget these days.

[Steve Innova]

Get yourself a copy of SolidWorks w/Cosmos. Give the landfill a break.

Most of what your talking about can be proven with software. You can weed out the failures without pumping any epoxy or cutting any foam.

"Proven with software" = garbage in, garbage out. FEA is a good starting point for composite design, but what you seem to be missing is that it's not just the theoretical performance of the materials in the design configuration, it's also the material variances and builder technique tolorances.

 

Remember your canard that came out 10 lbs heavier the first time? You followed the design both times, but somehow ended up with 30% more wieght? I guess that wouldn't have happened if you'd just "proven it" with SolidWorks?

 

Sure, you can account for the known and and unknown unknowns with a high safety factor, but you're missing the point. Aviation design is about compromises, how to acheive an optimal compromise between strength and weight. Sure, add 20lbs of S-glass to that winglet and it'll hold, no doubt. But that's a big waste of material that you'll pay for in AVGAS for the next 20 years.

 

If I was static failure testing all my parts that's all I would be doing is rebuilding parts (and I do enough of that now.) This is why we buy plans and incorporate proven designs from reliable/qualified sources.

What you are "incorporating" is not a proven design. If you're only the second person doing it, then that's pretty much the definition of experimental.

 

I have great respect for Marc's engineering work on Jack's blended winglet, but I'm willing to bet that he would agree that it's impossible to know that it's an optimal design without actually testing it.

 

I've seen Jack's workmanship and it's high -- I woudn't count on every other builder achieving the same level of worksmanship and safety factor. Maybe it matters, maybe not... You won't know.

 

Are you aware that Cirrus has tested their wings by deliberately leaving peel ply and other foreign objects in the layups? Just testing the "optimal" design doesn't answer questions about all the sub-optimal executions of that design.

[TMann]

"Proven with software" = garbage in, garbage out. FEA is a good starting point for composite design, but what you seem to be missing is that it's not just the theoretical performance of the materials in the design configuration, it's also the material variances and builder technique tolorances.

What you are missing here is the fact that much in the way of design CAN be proven as unacceptable before anything is built physically.

 

The "builder" variable will always exist no matter how many parts you stress to failure as has been proven in the past. Given your argument, all plans built canard aircraft should be grounded until they can be stress/load tested to failure.

 

As to the structural integrity of the blended winglet, I see the continuation of the shear web layup combined with the spar cap, being extended through the blend and up into the winglet as a real plus ....... and I'm not an engineer. That combined with input from an engineer as well as the designer/builder leads me to believe that this also functions as a structural enhancement as well.

 

Now, given that, I would suggest that you plan on putting up some of your own $$$ and hiring a professional if you plan on pursuing this on your aircraft.

 

I will share building techniques such as I did for construction of the nose shell, but I will not be posting structural design information that I am spending my cash for that will not even relate to the design others are building. That's just trouble for me and trouble for them.

 

In Marc's original write-up on the blend he states that the Long-EZ is a viable canidate for this design.

 

I've spoken with Jack as well, many times over this but the subject we have NEVER touched on is the structuaral makeup of the design. He spent the cash to get the qualified design as it applies to his application and I plan on doing the same strictly out of respect for the work both he and Marc put into this.

[Marc Zeitlin]

I have great respect for Marc's engineering work on Jack's blended winglet, but I'm willing to bet that he would agree that it's impossible to know that it's an optimal design without actually testing it.

You would win that bet. One of the reasons that I recommended the design that I did to Jack was because he WOULDN'T be static testing it (or varieties thereof), and therefore we wouldn't know for sure what the safety margins were. So we made them REALLY high, and then tested the winglets to the limits of use.

 

FEA (and ESPECIALLY SW/Cosmos) is, as you say, an interesting starting point, but hardly adequate. In particular, unless you're using some pretty sophisticated analysis to determine ply properties in a layup, and predict ply strains individually, using a bulk solver like Cosmos (or any number of other solvers) can give you extremely misleading information unless you're an expert in sandwich composite analysis (which I by no means am).

 

I've seen Jack's workmanship and it's high -- I woudn't count on every other builder achieving the same level of worksmanship and safety factor.

Exactly so. In fact, it's guaranteed that only a very few will - all you have to do is wander around and look at canards at fly-ins to see the large variability in builder capabilities.

 

Maybe it matters, maybe not... You won't know.

I guarantee you that it matters - your discussion of the VE canard that went to far less than the expected Ultimate Load shows that there are large variabilities in construction workmanship.

 

Are you aware that Cirrus has tested their wings by deliberately leaving peel ply and other foreign objects in the layups? Just testing the "optimal" design doesn't answer questions about all the sub-optimal executions of that design.

Bingo again. When Scaled does static tests of airframe structures, including deliberate flaws and MFG mistakes is part of the process.

 

For one off designs/builds, such as Jack's, it's unnecessary to do that, since you're only building one, but the testing to the expected Limit Load (with a conservative calculation of what that Limit Load should be) gives high confidence levels in the safety of the component.

[Marc Zeitlin]

What you are missing here is the fact that much in the way of design CAN be proven as unacceptable before anything is built physically.

That's certainly true. The analysis, if done reasonably, can indicate when there are problems and when the design needs to be modified. But if the safety margins are not extremely high, then testing is the only way to prove that what you've analyzed will actually be the case.

 

I don't really think that you and Steve are so far apart in your philosophies here - just looking at it from slightly different angles.

 

In Marc's original write-up on the blend he states that the Long-EZ is a viable canidate for this design.

Viable in the sense that you can do it and it can work without screwing the plane up, and with less disadvantageous issues than in a 4-seater like the COZY MKIV.

 

Personally, I don't think that a retrofit of an existing, stock Long-EZ with blended winglets is worth the effort for the very small speed gain that will be realized. But in a new build, where the extra work is minimized, and in O-360 or higher powered LE's where the speeds will be substantially higher than a stock Long-EZ, it may very well be worth the effort.

[TMann]

FEA (and ESPECIALLY SW/Cosmos) is, as you say, an interesting starting point, but hardly adequate.

Exactly the point I was trying to make.

If a part design fails in Cosmos, then try again. It would make no sense to build a part that Cosmos already has stated you're wasting your time on.

 

If it passes in Cosmos, then move on to the next step.

[mplafleur]

Fourth, there is at least one person (not on this forum - the current admin of the defunct "Canard Community Forum") who, in my opinion, is building an extremely dangerous blended winglet design from a structural standpoint, going from what he writes on his blog. (And who knows what he's doing from an aerodynamic standpoint - I wasn't able to decipher the description of the winglet positioning or angles). He is essentially following Steve Innova's question #3 from the third post in this thread:

 

"#3. Why bother? The winglet doesn't see enough force on it to require a thick spar. The plans method of using BID layups seems to work fine. I'd rather not add weight."

 

First of all, the plans method uses a large assortment of mostly UNI and also BID layups, both internally and externally - it's not just BID.

You gave Jack two methods, plans layups or web/spar cap? Jack choose the later which I believe was the better choice. Mike however has chosen the former and I guess from your response you think he has used the wrong layup schedule and I'm guessing was just BID and that the UNI is missing?

 

Woops! I'll have to change that signature. I'm not in China any more.

[Marc Zeitlin]

You gave Jack two methods, plans layups or web/spar cap?

Nope. That's a misunderstanding of the two structural scenarios that Jack and I discussed. Given Mike's known inability to understand either written or verbal communication (go back and read the "crosswind landing capability" thread from a year or two ago for confirmation of that point - in fact, in communication with Jack, no less), this is not surprising.

 

While the second method (no shear web/spar cap extension through the blend into the winglet) was an adaptation of the plans methodology, it did NOT leave out the internal structure that exist in the plans system - it adapted it for the curved geometry. In neither case were we relying solely on the external skin layups to take the loads and resist skin buckling.

 

Jack choose the later which I believe was the better choice.

Jack and I felt that the spar cap/shear web choice was simpler and easier to evaluate. The plans adaptation (including internal layups) would have been more complex.

 

Mike however has chosen the former...

No, as far as I can understand what he's written and from looking at his pictures, he's got no internal structure in the blended area whatsoever. ONLY external layups. This was NEVER one of the choices presented to Jack, and is not something that I would consider (or fly with). There's a reason the plans have internal structure to support the joint between the winglet and wing, and there are bending, shear and buckling loads that must be considered.

 

and I guess from your response you think he has used the wrong layup schedule and I'm guessing was just BID and that the UNI is missing?

Not really - it's that the whole concept of internal structure seems to be missing. Even with full, extended external layups top and bottom, there's nothing to prevent the top skin from buckling down/inward with a large horizontal load on the winglet inward. The internal structure (in the standard plans methodology) take care of that, and so does the spar cap/shear web system by ensuring that the skin layups don't need to take the bending loads - they're just redundant load paths at that point.

 

I think that you can see the issue.

[Steve Innova]

At minimum, Mike (aka Dust) should have used a higher density foam for the blend, if he's only relying on external skin.

[Steve Innova]

Here's another technique that another builder is using:

http://www.iconixresearch.com/aircraft/chapter20.html

[TMann]

Here's another technique that another builder is using:

http://www.iconixresearch.com/aircraft/chapter20.html

I've seen this before.

Not even close.

[Marc Zeitlin]

I've seen this before.

Not even close.

This was a little too cryptic for me to understand - I don't know what it's not even close to.

 

At any rate, it's fairly obvious that Rick Irwin is creating a substantial internal structure, as WELL as using something resembling a spar/shear web.

 

And if he's getting advice from Todd Parker, then I have complete confidence that he'll have a structurally sound system.

 

With respect to the aero aspects, personally I don't think that the 4.5" inside radius that Rick is using is large enough to develop the full advantage of drag reduction, and I'd have the radius increase as it travels from the leading edge to the trailing edge, but none of that has anything to do with the safety aspects of the installation.

 

I have no idea how Skorija's "encouragement" plays into the structural arrangement that Rick is using.

[TMann]

This was a little too cryptic for me to understand - I don't know what it's not even close to.

The items I think are shortcomings are:

 

• The radius
• The linkage
• The structure that facilitates the blend (and I may be wrong on that part).

I think the first item was dictated by design issues of the other two.

Perhaps I'm just Monday morning quarterbacking after seeing Jacks installation. When compared to the e-Racer solution ......... it's not even close.

[mplafleur]

I think that you can see the issue.

Yeah, I think so...

[dust]

Given Mike's known inability to understand either written or verbal communication

Quite the contrare, my rule is to take advice I ask for. In this case I attempted to pay for it. Marc, you are a piece of work and i don't miss reading your dribble one bit.

 

On the plans internal structure of the junction, mmmmm, i guess you don't remember it very well as you only attached 2 vert stabs to 2 wings, I on the other hand have done it 4 times(on two planes) and remember the internal structure, it was the second crappiest foam we use, the carve-able disintegrating one.

 

I wrote this one in your style, sorry it was not as good as you could have insulted and demeaned, I have to try to do it, for you it comes naturally.

[Steve Innova]

I have entirely abandoned the use of Urethane foam in my aircraft. There are superior substitutes available.

[mplafleur]

On the plans internal structure of the junction, mmmmm, i guess you don't remember it very well as you only attached 2 vert stabs to 2 wings, I on the other hand have done it 4 times(on two planes) and remember the internal structure, it was the second crappiest foam we use, the carve-able disintegrating one.

I think there are some layups in the internal structure as well, not just foam. Correct?

[Richard Schubert]

I think there are some layups in the internal structure as well, not just foam. Correct?

Yes there are, here is the drawing.