Twisted wing or something else?

Searched the forum but couldn't find anything will answer my question.

I have mated the wings to fuselage. All went well. The hardware store bolts slid right in, no issue. Wings are square with the same distance from wing tip to tail on both sides. No fwd or aft sweep, the four plumb bobs are all 1/8" or less from the straight line tip to tip. Until it is time to set the incident angle and drill the rear spar. I used the van's 3 inch block method and a bubble level to do the job. All is well for the left wing, the bubble is centered from root to tip. The surprise came on the right wing. If I centered the bubble at right wing root, I had to raise the 3" block about 0.06" to keep the bubble centered at the wing tip, indicating the wing is slightly twisted. I don't have a digital level that can measure the angle, but the distance between front and rear spars is about 26.5", with 0.06" difference, the angle is about 0.15-degree. So, the right wing tip will have about 0.15 degree more incident angle. My questions are:
1) Is this really due to a twist. If it is due to a twisted wing, the angle is more than what I expected. When I clamped down the wing skeleton, the difference on distance between the rear spar and the plumb bob line at root and tip is only about 1/64, instead of 1/16 showing right now. see my build log: http://rv7.yolasite.com/09-14-2011.php

2) Anything else could cause this?

3) if there is a twist of about 0.15-degree, how that will affect its flying? I would expect a heavy left wing, but how much? Should I set the incident angle of the right wing based on the angle measurement in mid span?

I will hold on drilling rear spar before this is cleared. Appreciate anyone's input on this.

I would take measurements at each rib on the right wing, +/- from root to rib and then come up with an average measurement for that wing.
Then try to get that average measurement as close as possible to the 3" as suggested making sure that you will have adequate edge distance in the rear spar stub for the bolt.
If that all checks out then set the left wing to match the average dimension of the right wing.
I suspect that you will not have any problems. As always check with Van's before drilling any major holes.

I found about the same on my quick built wings. A little less than 0.2 deg. Thought about it awhile then averaged out the twist drilled the rear spars and kept building. I didn't think there was a way to fix it so did worry about it too much. Almost all RV's have a heavy wing.

Brian

I have mated the wings to fuselage. All went well. The hardware store bolts slid right in, no issue. Wings are square with the same distance from wing tip to tail on both sides. No fwd or aft sweep, the four plumb bobs are all 1/8" or less from the straight line tip to tip. Until it is time to set the incident angle and drill the rear spar. I used the van's 3 inch block method and a bubble level to do the job. All is well for the left wing, the bubble is centered from root to tip. The surprise came on the right wing. If I centered the bubble at right wing root, I had to raise the 3" block about 0.06" to keep the bubble centered at the wing tip, indicating the wing is slightly twisted. I don't have a digital level that can measure the angle, but the distance between front and rear spars is about 26.5", with 0.06" difference, the angle is about 0.15-degree. So, the right wing tip will have about 0.15 degree more incident angle. My questions are:
1) Is this really due to a twist. If it is due to a twisted wing, the angle is more than what I expected. When I clamped down the wing skeleton, the difference on distance between the rear spar and the plumb bob line at root and tip is only about 1/64, instead of 1/16 showing right now. see my build log: http://rv7.yolasite.com/09-14-2011.php

2) Anything else could cause this?

3) if there is a twist of about 0.15-degree, how that will affect its flying? I would expect a heavy left wing, but how much? Should I set the incident angle of the right wing based on the angle measurement in mid span?

I will hold on drilling rear spar before this is cleared. Appreciate anyone's input on this.[/QUOTE]

The resolution of a digital level is +/- .1 deg.

Thanks for all the input. More data points make me feel less worried. From my RC model building and flying experience, that much twist is not a big concern unless you want to get the last bit of performance out of your airplane. You can always trim them out. Well, I guess the trimming on a RC is different from a RV. Never flown a RV so I don't know how much trimming is required due to this slight twist. I only plan to install manual aileron trim kit van's provides, not sure if it is enough to trim out the heavy wing.

As Brian mentioned, there is probably not much I can do at the moment to fix the twist. I will do what Tom suggested, setting the incident so the average reading on all ribs is zero.

I can live with a slightly draggy airplane due to this, but what got me is I have been always thought I had straight wings until yesterday.

Same here

Same here. QuickBuild wings. The left wing was dead nuts on. The right wing had about 0.15 deg of twist (negative washout). The QuickBuild factory must have a twisted right wing jig.

I'm not flying yet, but I suspect this slight amount of twist will not cause any perceivable difference in flying characteristics. It's also well within the 1/8" difference at the trailing edge that Van's specifies as tolerable. We'll see.

But to balance things out as best as possible when setting the wing incidence, I actually took a slightly different approach than just using the mean incidence or mid-span of the twisted wing. Rather, I used a point a little further outboard, approximately 2/3 span, as the incidence reference point. Doing the math (1st order linear approximation), I think this the point that should neutralize the roll moment.

Note that the total difference between setting the incidence at 2/3 span versus 1/2 span will amount to an almost immeasurable difference, about 0.025 deg of mean incidence. So that's also probably imperceptible. But I wanted to get the rigging as close as I could, so that's what I did.

Read about it in greater detail here:
http://www.kalinskyconsulting.com/rvproj/wingattach.htm#20100508_1030

But again, DISCLAIMER: This is just what I did, not flying yet, your mileage may vary, etc.

Getting interesting

Still no answer from Van's yet (on how the twist will affect the flying characteristics). To have a more complete picture of this issue, I compared the measurements on left and right wings using bottom skins. I used a 1" block to have the bubble centered. I can't see the similar twist on right wing by measuring bottom skin. The bubble was centered all the way from right wing root to tip. This makes compensating the "twist" a little more interesting. Maybe I should just set the incidence angle of right wing by equating the bottom skin measurements to the left wing, which is already set at 1-deg per Van's instruction, and deal with the possible heavy wing issue during flight test? Anyone see problem with that?

I also did a quick calculation on off balance moment on the right wing assuming linear twist (0 deg at root, 0.15-deg positive washout at wing tip). Assuming crusing speed of 180mph at 5000 ft MSL, right wing will create about 230 ft-lbs roll moment. This is equivalent to about 4 gal fuel at left wing tip. Can this amount of heavy wing be trimmed easily? Maybe someone has done wing tip fuel tank mod can provide some insight.

Back in the day, the consensus was that if a wing was twisted, set the incidence based on a reading at 2/3 span.

So, if you haven't drilled the aft attachment, adjust until incidence is right at 2/3 span on the twisted wing.

probably not a bad idea

Kyle Boatright said:

Back in the day, the consensus was that if a wing was twisted, set the incidence based on a reading at 2/3 span.

So, if you haven't drilled the aft attachment, adjust until incidence is right at 2/3 span on the twisted wing.

Click to expand...

The bulk of the lifting effort is from the inboard half of the wing, but that portion also has less ability to create roll moment. The outboard half is making less lift, but it creates more moment. So, use a measurement point somewhere between 1/2 and 2/3 span.

By the way, as an indication of how much this will affect speed, I did a test yesterday. Cruising along at 168 kt true at 10K ft, I stomped on the rudder enough to create a 1/2 ball width deflection on my Trutrak ADI pilot. It of course took some aileron pressure to fly straight. I held that until the speed stabilized after a while. 166 kt.

So thats how much a BADLY out of rig wing might cost you. With your amount of twist, I bet you can't even measure the speed difference.

.15? I dunno. This strikes me as the curse of the digital level. People used to make these with bubble levels. I doubt very much you'll find any negative flying characteristics.

I predict Van's response will be...

..."you're building an airplane - not a Swiss watch" ;-) Favorite response of an unnamed individual at Van'.

No offense to Pierre S., but that remark is usually followed with another about looking at the leading edges of an AG plane


Larry Tompkins
N544WB -6A
W52 Battle Ground, WA

LettersFromFlyoverCountry said:

.15? I dunno. This strikes me as the curse of the digital level. People used to make these with bubble levels.

Click to expand...

Bob, you can easily measure a 0.15 deg difference with a bubble level. In fact, even most digital levels don't give you readouts at that level of accuracy, nor do they need to. Whatever you're using, digital or bubble, you just need a level that is fairly sensitive at level. And then it's all about determining what shim thickness you need to put underneath one end to get it level. Followed by doing some basic trigonometry.

LettersFromFlyoverCountry said:

I doubt very much you'll find any negative flying characteristics.

Click to expand...

Maybe, maybe not. But there do seem to be a lot of RVs out there that don't want to fly straight, and end up being compensated with trim tabs, etc. Makes you wonder why. There could be various common reasons, but twisted wings and/or sloppy rigging would be an obvious one. Is that disastrous? No, within reason. But when it comes to rigging, may as well do the best you reasonably can to get it right in the first place. At least that's my take.

You will never know how it may or may not affect performance since you don't have a base line to test against. It doesn't really matter anyway unless you want to rebuild a wing. Do the best you can and there are several suggestions given that should let you do just that.

I had one of the very first QB 6 fuselages. The factory built the Center Section out of square, which was only discovered after the wings where installed. I have one wing with no sweep and one with 1/2" sweep. If it affected the performance, balance, or anything else I can't tell.
Doesn't matter.

I keep up with my buddies. When we are flying together and they ask me what power setting I am at I simply tell them "1 inch, 100 rpm, and 1/2 gallon less than you". They don't ask again, and sometimes it is even true.
I haven't told my airplane yet either as she is kind of sensitive and thinks she is perfect.

roee said:

But when it comes to rigging, may as well do the best you reasonably can to get it right in the first place. At least that's my take.

Click to expand...

Of course, you want to do the best you can in rigging and from what I can tell, that's been done here.

I'd be curious to see some engineering on a .15 "twist" and flying characteristics, all other things being equal (actually, assuming the incidence is based on an average, we're talking a .075 "twist.") . I'd be real surprised if it is close to noticeable as opposed to, say, a flap that isn't adjusted quite right or a trailing edge that wasn't bent exactly the way it should be.

I look forward to seattleworm's first flight. My money is on it being just fine.

LettersFromFlyoverCountry said:

Of course, you want to do the best you can in rigging and from what I can tell, that's been done here.

I'd be curious to see some engineering on a .15 "twist" and flying characteristics, all other things being equal (actually, assuming the incidence is based on an average, we're talking a .075 "twist.") . I'd be real surprised if it is close to noticeable as opposed to, say, a flap that isn't adjusted quite right or a trailing edge that wasn't bent exactly the way it should be.

I look forward to seattleworm's first flight. My money is on it being just fine.

Click to expand...

Yep, agreed.

I really doubt that the 0.15 deg twist is in itself a problem of any major significance. Partly because it falls well within Van's stated tolerance for incidence difference between the two wings. And partly because, from this accounts on this discussion, I suspect that many if not all QuickBuild 7's are flying with a 0.15 deg twisted right wing!

So, yes. With the incidence set accurately using a point 1/2 to 2/3 span, my money is on it being just fine too. And I sure hope so, because I'm in exactly the same boat. I'm not worried.

Rear spars drilled

Thanks again for everyone's input. I finally drilled the critical rear spar holes last night. I set right wing incidence angle using measurement at rib between 1/2 and 2/3 of the span, so there is a slight washout at inboard, and a slight washin (a new term I just learned, meaning positive angle) at outboard. Glad to put this task behind. The drilling process itself went well, all margins are greater than 5/8. The only thing I wish could be better is the bolt fit. I used 0.3115" reamer, but the AN5 bolts I have are all 0.309" dia. I will buy more AN5 bolts and hopefully I can find some bolts at high end of the tolerance (0.312") , so the fit could be tighter. But I don't think this is going to be an issue as is.

With all things set, I will deal what ever the consequence later during flight test phase. I guess how airplane flies is a big unknown, and anxiety to every builder until the day of first flight. I will keep my fingers crossed.

seattleworm said:

so there is a slight washout at inboard, and a slight washin (a new term I just learned, meaning positive angle) at outboard.

Click to expand...

Someone please help me understand this a little. First, I know that this is a very small amount of washout / washin, but isn't that bassackwards from what is desired? Or do I have a misunderstanding of the terms used?

BPoletti said:

Someone please help me understand this a little. First, I know that this is a very small amount of washout / washin, but isn't that bassackwards from what is desired? Or do I have a misunderstanding of the terms used?

Click to expand...

Right, more or less.

The RV is designed with zero washout wings, so zero washout is desired on an RV.

But for other aircraft designs that do make use of washout, the desired direction of twist is such that the outboard portion of the wing has a lower incidence angle than the inboard portion. That way the inboard section stalls first.

So I suppose if you were going to have twist in an RV wing, that would be the preferred direction of twist. Unfortunately, the twist we've been finding on the QB right wings is the opposite, i.e. "negative washout" or "wash-in". Yes, that's undesirable.

LettersFromFlyoverCountry said:

People used to make these with bubble levels. I doubt very much you'll find any negative flying characteristics.

Click to expand...

...took the words right outta my mouth. For YEARS, people built these with a 2' or 4' carpenters level. While digital-precision-inspired perfection is satisfying, it's not necessary for a good flying airplane. I had that very discussion with a couple of the guys at Vans three years ago.

Rupester said:

...took the words right outta my mouth. For YEARS, people built these with a 2' or 4' carpenters level. While digital-precision-inspired perfection is satisfying, it's not necessary for a good flying airplane. I had that very discussion with a couple of the guys at Vans three years ago.

Click to expand...

Yes, but a bubble level would give a shim different of about 1/10 inch on the shim dimensions for 0.15 degrees if my trigonometry is correct..

Even with old fashioned bubble technology I bet most builders aim for a tighter tolerance than 0.1 inch on the shims.

However, a friend bought a RV-3 that had a twisted aileron - the twist was almost 1/2 inch and looked bad even to the casual observer. His RV-3 was the fastest of the group at Compton though.

The aileron looked so bad he built another one, straight this time. He noticed no improvement in speed or handling.

I think the Vans designs can tolerate more error than we would want to build in...

The digital level vs. bubble level debate, round 2

Rupester said:

...took the words right outta my mouth. For YEARS, people built these with a 2' or 4' carpenters level. While digital-precision-inspired perfection is satisfying, it's not necessary for a good flying airplane. I had that very discussion with a couple of the guys at Vans three years ago.

Click to expand...

There seems to be a mistaken assumption going around that digital levels are necessarily more accurate or more precise than bubble levels. Generally, they are not. They are easier to use for determining non-level angles, and are easier to read off-axis than a bubble level. But precision is actually not their strong suit.

Most common digital levels give readouts in 1/10ths of a degree, which places an upper bound of 1/10th deg on their useful accuracy and precision. (Note that their actual accuracy and precision might be even worse than that)

Whereas with a common 4' bubble level, a set of shims, and a basic understanding of trigonometry, you can easily measure angular differences down to a few 1/100ths of a degree.

So, we can wave our hands and say that precision doesn't matter. Fine, that's one thing. But let's be clear about the fact that the ability to set the incidence angle very precisely did exist long before digital levels came along. There was no excuse for sloppy rigging then any more than there is today.

roee said:

So, we can wave our hands and say that precision doesn't matter.

Click to expand...

Nobody that I know of, or can tell from this thread, is saying precision doesn't matter. This is more a question of perfection. I think one of the most valuable skills in building an airplane is knowing when an *ongoing* quest for perfection at a particular point has a low reward.

If there's a desire to chase a .075 difference from root to tip (that's what you'd be chasing using the average) , I have no problem with that. A little OCD isn't a bad thing. But I don't think you'd be chasing performance here.

I don't also don't see accepting .075 as "sloppy" rigging.

Shawn is going to be just fine and I look forward to his "it flew perfectly!" first flight report.

Good to see you got the "big drill" done. That might be the scariest hole to drill in the entire project.
I have seen the repair required as a result of the "drill jig" slipping without notice on a buddy's RV8. It is not an easy repair.

Working on this now, and I observed the same problem as others in this thread. Left wing is exactly level across the entire length of the wing. Right wing (according to the all-knowing digital level) is 0.2 deg down at the root and 0.2 deg up at the tip, for a total twist of 0.4 deg. I'm planning to confirm with Vans, but from this thread it sounds like they're going to say build-on unless you want to re-build the wing.