E

elippse

Well Known Member
Jim has finally been able to make three sets of data runs at 4000', 6000', 8000' and 10,000' baro altitudes to compare with three sets of data he had taken previously with the standard tips. I took both data sets and obtained a second-order polynomial fit of each tip type vs density altitude. The results are presented below of the curve fit of TAS mph at 1000' intervals from a density altitude of 4000' to a density altitude of 12,000'.

dalt orig new
4000 189.9 197.0
5000 191.4 196.4
6000 192.0 195.6
7000 191.7 194.6
8000 190.4 193.4
9000 188.3 192.0
10,000 185.2 190.3
11,000 181.2 188.5
12,000 ----- 186.5

I didn't list the 12,000' value for the original tips as I felt that was at the extreme edge of the second-order curve and the closest three density altitudes with the original tips were 10,851, 10,885, and 10,998.

For those of a mathematical bent, the coefficients for the original tips are:
-4.669584E-7 * V^2 + 5.762168E-3 * V + 174.2846, and for the new tips:
-1.001908E-7 * V^2 + 2.854875E-4 * V + 197.5071
 
Which tips are these again? Do you have pictures? A 9mph increase at 4000' would be worth investigating for me... A lot of my flying around here is at or below 4000'.
 
worthless.gif


Well, maybe not "worthless" -----but photos sure would be nice
 
Hard to tell for sure from that one pic, but it looks like he should have a placard on the plane thanking Steve Whitman...

Charlie
 
Bob Kuykendall said:
The real question is, what do the short tips do to stall speeds and power-off glide ratios.
Click to expand...

These extended tips definitely lower takeoff, landing and stall speeds, while increasing rate-of-climb. Jim took a picture of a bunch of tufts near and at the tips just before a stall and they were straight back.
 
gasman said:
I like the looks of my stock flat top wing tips too much..... I don't know how much speed it would take for me to switch.
Let's just say........ ugly:eek:

How many G's would it take to tear one off?
Click to expand...

I told Jim that he would have to curtail his aerobatics because first off, his roll-rate would be decreased, but also with the longer span, the bending forces would be increased so that he should decrease his allowable G-limits.
What might be displeasing to one person's perception might be a delight to another who sees things more in their utilitarian aspect, aesthetcs, and elegant simplicity.
My plane, Peter Garrison's Melmoth, and the Katanas, all make use of similar designs, which give a dramatic performance increase. Two things you can't do without, HP and AR. No! Three things; HP, AR, and low drag. No! Four things; HP, AR, low drag, and an efficient prop. (with apologies to Monty P. and the Spanish Inquisition)
 
tips

These tips were designed by the late Steve Wittman, based in part on studies by the late August Raspet. The Wittman tips on the Wittman airplanes are a "clipped triangle". Wittman wanted to build the tips as on the RV6 but was unable to do so because of rear spar strength considerations on the much thinner airfoil on the Wittman airplanes. The tips were first used on one of Wittmans personal Tailwinds, the O&O, Buttercup and V Witt. Wittman did not use the small tip plate.
 
Elippse, How are these tips held on? Is there a spar extension? Are they just held on with existing wing tip screws?
 
jrs14855 said:
These tips were designed by the late Steve Wittman, based in part on studies by the late August Raspet. The Wittman tips on the Wittman airplanes are a "clipped triangle". Wittman wanted to build the tips as on the RV6 but was unable to do so because of rear spar strength considerations on the much thinner airfoil on the Wittman airplanes. The tips were first used on one of Wittmans personal Tailwinds, the O&O, Buttercup and V Witt. Wittman did not use the small tip plate.
Click to expand...

I wonder if the flow at the tip is the same on the double triangle of Wittman's as opposed to the swept tip on mine and the others. Using the Schrenk method I plotted the lift distribution on the RV-6 wing with the original tips and the new tips, and the new ones are much closer to elliptical. I designed my tips based on a NACA paper that said a 55? sweep was ideal.
 
Paul,

A couple thoughts and questions:

First, I echo the query about structure and attachment. Are there internal supports along the horizontal axis, or any other strengthening mechanics at play inside, or is all held by the screws and the skin overlap (tip joggle inserted into wing skin)? Do you have pictures of the guts of the tips?

A concern of mine is the g limiting required. My wings are clipped due to the Super 6's higher weight, to reduce the bending moment and allow standard g limits at the higher weight. Given that these tips are for speed and efficiency, a g sacrifice is tolerable when they are used...but what is the actual cost there (in your estimate). Would you limit it to 2g, 3g? Whatcha think?

We've talked about these tips in person, and I was under the impression that the advantage of increased span with less area (as provided by the triangular shape) is greater performance at higher altitudes. Your results in the opening post show a curved result, peaking at 4K' and 11K', and least impacting at 7-8K'. I don't have a clue what a second order polynomial fit is (I cut class that day in trig...or was it adv geo...or calc...OK, heck, I R a Pilot! :rolleyes:). However, that curve makes me think that lower might show more speed too...if the curve extrapolates in the same direction. That would get my attention for racing...but it doesn't make sense to my mind, that feels lower wing area is better down low and fast. Do you have comparitive data at sea level? I'm assuming that the before and after figures are from tests using identical MP and RPM settings for old and new tips, and that ambient conditions were close (and the new prop was run on both)...in other words, the tips were the only change. I know you test a lot, but I didn't see the data mentioned...so I'm just being a diligent watchdog! ;)

I never realized you had a fence out there either. Did you run tests with and without the fence? Any results to share on that? Any aileron effectiveness issues or changes? Interesting to note that others have tried fences on flat tips, and found them to load up the ailerons uncomfortably (ie scarily). Your tip puts some distance between the aileron and the fence, perhaps solving that issue. How's it feel to fly them? Then again, I saw these tips on Galloping Ghost, who ran very fast at Reno this year...but never had the chance to ask what impact the tips did and how they flew...looks cool though!

ggtip2.jpg


ggtip1.jpg


I'd love to run a side by side by side test of these tips and your tips versus my normal and flat tips! Anyone want to help a composites-challenged tinkerer fabricate some? Any Avgas sponsors out there? :D

Boy, this sure would mess with my FFI buddies' sight picture though, wouldn't it! :eek:

Cheers,
Bob
 
Last edited:
For questions related to the construction and attachment, you would have to contact Jim; I only supplied the planform and tip airfoil drawings. Jim hotwired these from foam using the wing outer shape to the 16" chord tip airfoil (23013.5).
As to G load, either someone on this forum with knowledge in this area could provide the insight, or I would have to hit the books and calculate a better estimate than I did initially.
I find myself in wonder at the speed increase below 6000' dalt which didn't show up in my estimates when running my program with and without the tips. I have spoken with a friend who markets software that shows the airflow over a plane, and he is working on it.
I estimated the increase in parasite drag based on the extra 6 sq. ft. vs the decrease in induced drag at higher speeds and higher air density and got the crossover at about 6000' dalt. One of the things that I puzzled over is the body attitude in flight. Jim says that the nose no longer points up at higher altitude the way it did previosly, with the fuselage at a much flatter attitude.
Perhaps, too, the flow pattern at the outer wing is giving it a much better L/D; I'll have to send you the pix he took in flight with tufts as he was almost at stall. They are really impressive with all of them being straight back with no curls evident.
Those "fences" you see are just to block the light from the tip strobes, but maybe they, too, are helping.
In all of his flights he is leaning to best power and doing two-way GPS-measured runs, so basically the plane is the same now as before, as far as I know. Obviously, another "puzzler"!
 
Jim's tips are made of fiberglass over foam, with a graduated six ply layup reinforcement at the end where it attaches to the wing through the same screws used to hold on the original 12" tips or the later 17" tips. The ply schedule starts 6" out from the inner end, and then has additional plys from approximately 5", 4", 3", 2", and 1". Because the tip has the inner filler of foam which provides the crushing strength, it has no need of ribs or spar, and it just becomes a stressed-skin addition to the main wing as with the two types of regular tips.
Since these tips provide lift all the way to the end, as opposed to the regular tips which only act as a buffer to the span-wise flow from bottom to top, they may be more efficient since the lift tapers off toward the end, giving a progressively reduced pressure differential. I'd be willing to bet that these tips have much less vortex as a result of this lift roll-off as compared to the regular tips. Any energy use in generating a vortex is lost energy or drag. This may be where the additional speed is coming from!
 
Paul in the photo Vlad took, it looks like the tips angle upwards from the wing plane slightly? what was the purpose of that?
 
What they do, Danny, is the top surface is a continuation of the top of the rest of the wing; only the lower surface tilts up. For comparison, I'm listing the results of Jim's testing woth his original 68-72 Aymar-DeMuth prop. I did the same second-order polynomial evaluation from his two sets of data, which gave me the following:

4000 193.8
5000 191,8
6000 189.8
7000 187.9
8000 185.9
 
Here are the pics Paul sent to me for posting. Paul, if you do a quote on this post, you can probably imbed explanations (better than me trying to 'splain it! :))

Cheers,
Bob

smithwingtip004.jpg


smithwingtip0052.jpg


smithtipsplane.jpg


smithsht1.jpg


smithsht22.jpg
 
What you are seeing, Danny, is that the top surface of the tips is a straight continuation of the top of the rest of the wing; only the lower surface tilts up.
For comparison, I'm listing the results of Jim's testing with his original Aymar-DeMuth prop and the original wingtips. I did the same second-order polynomial evaluation from his two sets of data, which gave me the following:

dalt tas,mph
4000 193.8
5000 191,8
6000 189.8
7000 187.9
8000 185.9
9000 183.9
10,000 181.8
11,000 179.8

The polynomial coefficients are:
-4.763081E-9*V^2 - 4.763081E-3*V + 201.5444

On these runs the 68-72 Aymar De-Muth averaged 100-120 rpm higher than did the 64-74 Elippse three-blade used in the wing-tip tests, so that would cause the 150 HP O-320 to put out about 4.1% more power or about 1.3% more speed. To better compare the A-D to the three-blade, subtract about 1.3% of the speed with the A-D prop.

To answer a previous question, a first order curve is one in which the resulting data is in a straight line; it would be evaluated as C1*V + C2. A second order curve is one in which the resulting data follows a slope that has some curvature to it. Typically our plane's speed vs altitude will follow a second-order curve. It is evaluated as C1*V^2 + C2*V + C3. You can subject data to even higher order curves if the nature of the process justifies it, but it can result in some really wild estimates, especially when trying to estimate values past the ends of the original data span. If the original process follows a lower order, evaluating it as a higher order will not usually corrupt the data. This process comes from what is known as a least-squares fit, which tries to give the minimum error between the data and the fit.
 
Paul,

Look back one post...I posted your pictures there.

On the first, second and higher order curves: why use curves rather than raw data...does it give more apples to apples?

I'd kinda like to see raw data results of before and after (new tips) speeds. with power settings used and speeds obtained at each altitude and power setting...if you or Jom have that data. Just interested in what one would see in day to day flight regimes. Thanks!!

Cheers,
Bob
 
Thanks for posting the pix, Bob! They show the approach-to and almost-at a stall. The only tuft that is showing some flip-up is the one at the gap at the end of the aileron. When I got these from Jim I was totally surprised that this section of the wing just kept right on flying. Since the tufts are straight back all the way to the tip, apparently there is no vortex curling around the tip causing them to show an inboard flow. I guess Wittman and Raspet and NASA got it right!
The drawings are what I supplied to Jim for making the tips. The very end of the tip is a clear plastic lens which houses the strobe and nav lights.
 
You must log in or register to reply here.