What's new
Van's Air Force

Don't miss anything! Register now for full access to the definitive RV support community.

Big Surprise

elippse

Well Known Member
Those of you who have followed my postings about Jim Smith's RV-6 and the big boost he has gotten in performance with his new wingtips know that I was somewhat taken by surprise at how much better they performed than what I had predicted. It finally came to me this morning what factor might be making the difference.

For those of you familiar with the Oswald efficiency factor of a wing and how it enters into the calculation of a wing's CL and induced drag, you know that a rectangular wing planform, with a really good tip shape, will have an OEF of about 0.8 to 0.82 for the best. Since that is in the denominator of the equations, that means that a wing's required CL and its CDI will be about 22% to 25% worse.
But were you also aware that an elliptical wing has an OEF of 1.0. That got me thinking about whether the excess improvement in Jims' performance may be do to an OEF that is greater than the 0.81 I had used. I had done a determination of Jim's before and after lift distribution using the Schrenk method, and the after version was much closer to the elliptical ideal than the before.

So I did some calculations based upon his multi flight averages at 10,000' dalt with the original tips and the new tips. When I ended up, after making power adjustment for his different speed and rpm, I came out with an Oswald efficiency factor of 0.91; that's 12.4% better than what I had assumed in my original performance estimates, and can be shown to be the difference factor for my estimates.

Coupled with this are the pix Jim took of tufts on his wing when approaching a stall that show the tufts straight back up to and just before the stall. This indicates that there is almost no spanwise flow and a very minimal tip vortex.

So it appears that by proper shaping and extension of the tip region, the wing can be made much more efficient both in terms of lift and drag than can be obtained by just one of the range of curved under, flat, or curved up shapes that are available.
 
Yes, but can you get the same efficiency with tip sails rather than the shape of Jim's tips?

As a friend said to me when discussing his tips, ?I don?t care how efficient they are, they are ugly and I?m not flying an ugly airplane.?
 
Coupled with this are the pix Jim took of tufts on his wing when approaching a stall that show the tufts straight back up to and just before the stall. This indicates that there is almost no spanwise flow and a very minimal tip vortex.

While tufts pick up the spanwise flow, they don't seem to pick up the tip vortex. The tip vortex is a larger-scale flow system and the radial components, at the wing surface, are difficult to see. Even a foot and a half away from the tip they are almost undetectable with tufts.

An easier way to see the tip vortex is a low pass over a dusty runway, the slower the better.

And just for clarification, when you calculated the OEF, did you use the airplane's exposed wing area with the new tips? And the weight at the time of taking the data?

Dave
 
So what do the tips look like?

img0978a.jpg
 
While tufts pick up the spanwise flow, they don't seem to pick up the tip vortex. The tip vortex is a larger-scale flow system and the radial components, at the wing surface, are difficult to see. Even a foot and a half away from the tip they are almost undetectable with tufts.

An easier way to see the tip vortex is a low pass over a dusty runway, the slower the better.

And just for clarification, when you calculated the OEF, did you use the airplane's exposed wing area with the new tips? And the weight at the time of taking the data?

Dave

Actually Jim did some tuft testing with long tufts out on the end of the tips, one outboard of the aluminum lightshield attached to the tip light lens, and one just inboard of the shield. The only activity is on the outpoard tuft, and at about 18" back it is whirling maybe in a 2"-3" diameter circle.

The increased wing area was used in calculating the increased parasite drag, The 230135 CD of 0.0065 X 6 sq. ft. was used to increase the equivalent parasite drag area from 2.2 to 2.239, and all testing was at the same 1440 lb.

Too bad Bill R's friend would trade his personal opinion of a plane's looks for a real boost in performance. Jim's speed at 10,000' dalt increased form 185.2 to 190.3, his stall speed lowered, his takeoff, ROC, and speed increased, and his landing speed dropped. With the reduced induced drag and better span efficiency, the plane's speed will be higher when at higher weight, pulling Gs, and at higher altitudes. This would show up as being able to do tighter turns around SARL turn points with less loss of speed. If I was running an RV in the SARL races I would say to H**l with waht anyone thinks about the plane's looks; I want to win!
 
Last edited:
Correct aspect ratio

It appears that new wing tips were added to the stock wing. The new wing has a higher aspect ratio. Did you account for the change in aspect ratio when you backed out the efficiency factor? It's not surprising that induced drag decreased since the span was increased. That the reduction in induced drag was apparently larger than the increase and parasite drag is interesting. One concern I would have is that he has changed his wing root bending moment which will reduce his plane's operational envelope. Also, it appears that the tips are swept back. This will move the aerodynamic center a little bit aft, so his cg limits will need to be changed as well.
 
Are you certain about the sweep? It looks to me like it's tapered instead?
If the tip generates no lift, just bleeds off tip vortex, then how would it increase the effective span?
So which is it going to be? More span or elimination of tip vortex? or the American Answer: 'yes' ??:confused:
 
Yes, but can you get the same efficiency with tip sails rather than the shape of Jim's tips?

As a friend said to me when discussing his tips, ?I don?t care how efficient they are, they are ugly and I?m not flying an ugly airplane.?


Ditto on the ugly
 
Where do I get the design?

Where do I get the design so I can build a set for my wings?

I don't care if I fly the fugliest plane in the airpark - a 5kt increase, lower gas consumption, and better stall speed - I'll take it..
 
Tips

Keep up the good work Paul! As Adm. Perry once said "**** the Torpedos, full speed ahead"!!!!!!!!!
 
Paul,

We've talked about this a few times, and the concept is intriguing. You've mentioned that this tapered tip design increases the span, while not increasing the wing area as much as a rectangular extension of the same length. I thought that was its "secret" to increased effeciency in high speed cruise.

Can you elaborate on how it increases Oswald Factor, as I thought that was the attachment of the vortex (or lack thereof) due to tip edge design (sharp versus rounded edge corners). I'm probably oversimplifying, but would like to know more, and Oswald Factor wasn't in my aero class (will break out my faded copy of Aerodynamics for Naval Aviators from college and look it up in the meantime).

This is the first time I've seen you mention SARL racers as a possible benefactor of the design. At AirVenture Cup, when its West to East, and flown up high for tailwinds, I can see the benefit. But in most other races, its down low (near sea level) and flat out. $64K question, will this tip be faster than my flat tip in the weeds at max speed, or is does my flat tip trump it in those conditions.

I see it as a high cruise booster at the moment, but am open to being convinced about low/fast. Only downside I can see is inreasing the bending moment on the wing for gust loading and acro in my heavier aircraft (why my wings are clipped). As for looks...heck, put checkerboards on them, and they may start to look cool...maybe! :D

By the way, I'd be a willing beta tester number 2 to see if I can match Jim's performance increase, and add in the low altitude racing tests to the mix! ;)

Cheers,
Bob
 
Where do I get the design so I can build a set for my wings?

I don't care if I fly the fugliest plane in the airpark - a 5kt increase, lower gas consumption, and better stall speed - I'll take it..

What you're largely doing is increasing the span (and bending moments, as someone posted). That is beneficial for climb, stall, and higher altitude cruise. It also reduces your structural margins and roll rate.

Unless someone posts verifiable data proving otherwise, this mod may turn your RV-8 into a tandem RV-9. Which isn't a bad thing, you just need to be comfortable with the trade-off's.
 
It appears that new wing tips were added to the stock wing. The new wing has a higher aspect ratio. Did you account for the change in aspect ratio when you backed out the efficiency factor? It's not surprising that induced drag decreased since the span was increased. That the reduction in induced drag was apparently larger than the increase and parasite drag is interesting. One concern I would have is that he has changed his wing root bending moment which will reduce his plane's operational envelope. Also, it appears that the tips are swept back. This will move the aerodynamic center a little bit aft, so his cg limits will need to be changed as well.

I designed these wing tips for Jim because when I was reducing his speed data with my propeller I noticed how much his speed bled off with increasing density altitude. That is when I realized that his low 4.8AR was the culprit. After sending him the design data and talking for several weeks about them I could tell that he was really reluctant to make and try them. Now he is so happy with his overall performance that you can't get the grin off his face! He's especially happy that the plane flies level at altitude and not nose up.

I told Jim that he shouldn't do any aerobatics with these tips on and to consider his plane to be 4G and not 6G with them. When you re-arrange the equation for induced drag AR does not appear, only pi, e and the squares of weight, span, and Q. The tip is swept at 55? as recommended in a NACA report. I also told him about the aero center moving back slightly. On my plane it amounted to about 1/4".
 
Ditto on the ugly

Well, if you're not racing, or concerned about more speed and better takeoff, landing, ROC, and overall performance along with better fuel mileage, I say stick to your esthetic preferences. However if you do race, and somebody passes you by with these unusual tips, just grin and bear it. They'll soon be out of sight. BTW, do you feel the same way about the Katana?
 
Hey, Barnstormer! I see you're in Gridley, Kansas, about 75 miles ENE of Jim in Rose Hill, Kansas, east of Wichita. You two ought to arrange to do a side-by-side comparison flight!
 
Where do I get the design so I can build a set for my wings?

I don't care if I fly the fugliest plane in the airpark - a 5kt increase, lower gas consumption, and better stall speed - I'll take it..

The speed increase is less at lower altittude, but it was still 2 mph at 4000' dalt. But it really goes up at 11,000' and 12,000' with an increas of 8.5 mph at 11,000' and 9 mph at 12,000'. Send an e-mail to Jim at: Jim Smith, [email protected] and I'm sure he will give you whatever info you need.
 
Paul,

We've talked about this a few times, and the concept is intriguing. You've mentioned that this tapered tip design increases the span, while not increasing the wing area as much as a rectangular extension of the same length. I thought that was its "secret" to increased effeciency in high speed cruise.

Can you elaborate on how it increases Oswald Factor, as I thought that was the attachment of the vortex (or lack thereof) due to tip edge design (sharp versus rounded edge corners). I'm probably oversimplifying, but would like to know more, and Oswald Factor wasn't in my aero class (will break out my faded copy of Aerodynamics for Naval Aviators from college and look it up in the meantime).

This is the first time I've seen you mention SARL racers as a possible benefactor of the design. At AirVenture Cup, when its West to East, and flown up high for tailwinds, I can see the benefit. But in most other races, its down low (near sea level) and flat out. $64K question, will this tip be faster than my flat tip in the weeds at max speed, or is does my flat tip trump it in those conditions.

I see it as a high cruise booster at the moment, but am open to being convinced about low/fast. Only downside I can see is inreasing the bending moment on the wing for gust loading and acro in my heavier aircraft (why my wings are clipped). As for looks...heck, put checkerboards on them, and they may start to look cool...maybe! :D

By the way, I'd be a willing beta tester number 2 to see if I can match Jim's performance increase, and add in the low altitude racing tests to the mix! ;)

Cheers,
Bob

I was particularly surprised that his speed continued to rise below 6000' dalt as I had estimated that 6000'-7000' would be a crossover between reduced induced drag and increased parasite drag from more wing area. He's now getting 197 mph average from four GPS-measured flights at 4000' dalt. That's really a lot from 150 HP!

I sent the data and tuft pix to a well known aircraft designer and monthly columnist asking why Jim was getting so much more performance than what I had estimated, and he replied that a lot of wings were sort of submarginal in performance.

But that answer didn't really satisfy me since I'm the kind of guy who can't stand having loose, unexplained stuff floating around in his head. I had never found out until yesterday after doing a search that the OEF on an elliptical wing is 1.0! This got me to thinking that since I had previously done a comparison of the lift distribution on his wings with the original tips and with the new tips by using the Schrenk method, that his new tips gave him a lift distribution a lot closer to the ideal elliptical.

That got me thinking about setting up an equation to solve for the OEF. I've since come to the realization that the usual quoted values of OEF in the range from 0.75 to 0.82 may only apply to rectangular planforms, and that when there is a planform near the tips that is tapered to near a point that it is probably possible to get into the region between 0.82 and 1.0! So far nothing I have found in the literature addresses this.

As far as performance at tree-top heights, I'd say stick with wings with some span clipped from them, but for cross-country to Oshkosh at the higher altitudes, I'd say that if any who have submitted to the "ugly" claim and cut off their nose to spite their face, and not equipped their plane to get more speed, then they're more concerned with other's opinions that winning races.

BTW, I sent Jim a list of the SARL races and he's definitely planning on entering at least one. If he does, when he taxies up, a bunch of you look and point at his plane and hold your noses and laugh! 'Careful, though! He's a little banty rooster and when that stub of a cigar in his mouth gets to twitching, better look out! You met him at Reno, Bob, so maybe you know what I mean!
 
Surprise vs. Van's

Van's says an RV-6 with 150 HP has a top speed at 8000' of 198 and an RV-6A has a top speed of 196. Both solo at 1400#.

The new-tip "6" gets 197 at 4000'. Notwithstanding that drag declines with greater altitude for the same speed, HP goes down, too. The fastest speed the airplane can fly is usually at sea level. So 197 is faster than 190.3. But the Van's specification for the same weight (-40#) at 8000' is 198 mph. The Van's airplane is a few mph faster than the new-tip version on the same nominal HP for an equal altitude.

You can now skip to the last paragraph. Optional stuff follows this:
--------------------------------------

Let's see what the 6 would do at 10,000'. I used the CAFE drag figures for the 6A and adjusted for the drag that slows the 6A by 2 mph. I did this iteratively on my triangle spreadsheet on my website. Try it yourself. You have to download it to be able to change the values.

You have to change the G/V ratio to 12.615 from 12.3, leaving the speed for best glide at 110. This is not completely correct, but we have no better data. The best answer would be a lower best glide speed and a lower glide, but still higher than 12.3.

The bottom line is that at 10,000' and 190.3 mph the Van's RV-6 using CAFE drag data needs 76.88 THP. That's an increase of 4.64 THP from 185.2

The new tips appear to give the RV6 4.64 more THP at 190 mph at 10,000'. That's net of increased parasite and decreased induced drag. The induced drag for the standard model 6A, adjusted to be a 6, at that speed and altitude, is 25.74# and parasite is 125.75#. Paul's friend's numbers will differ a little, of course, even before the new tips.

---------------------------------------------------
I won't get into the more technical stuff because I am not qualified to do so. I would suggest, though, that before we get exited about the new tips we look at what the unaltered airplane should be doing. It is not unusual for RV's to go faster than Van's numbers and most go at least as fast as they "should". This one is still not doing that.
 
Ditto on the ugly
Well, if you're not racing, or concerned about more speed and better takeoff, landing, ROC, and overall performance along with better fuel mileage, I say stick to your esthetic preferences. However if you do race, and somebody passes you by with these unusual tips, just grin and bear it. They'll soon be out of sight. BTW, do you feel the same way about the Katana?

While not astatically pleasing, your results speak for themself.

You missed part of my post where I asked if you could get the same results with tip sails rather than the shape of Jim's wing tips.

I realize tip sails used to be THE fashion statement 20 years ago but we haven't seen or heard much about them lately.
 
Van's says an RV-6 with 150 HP has a top speed at 8000' of 198 and an RV-6A has a top speed of 196. Both solo at 1400#.
.......
I won't get into the more technical stuff because I am not qualified to do so. I would suggest, though, that before we get exited about the new tips we look at what the unaltered airplane should be doing. It is not unusual for RV's to go faster than Van's numbers and most go at least as fast as they "should". This one is still not doing that.

Are you making assumptions about how much hp his engine is making? what else do you think could be different from the vans numbers if the plane is a bit different in that area?
 
Van's says an RV-6 with 150 HP has a top speed at 8000' of 198 and an RV-6A has a top speed of 196. Both solo at 1400#.

The new-tip "6" gets 197 at 4000'. Notwithstanding that drag declines with greater altitude for the same speed, HP goes down, too. The fastest speed the airplane can fly is usually at sea level. So 197 is faster than 190.3. But the Van's specification for the same weight (-40#) at 8000' is 198 mph. The Van's airplane is a few mph faster than the new-tip version on the same nominal HP for an equal altitude.

You can now skip to the last paragraph. Optional stuff follows this:
--------------------------------------

Let's see what the 6 would do at 10,000'. I used the CAFE drag figures for the 6A and adjusted for the drag that slows the 6A by 2 mph. I did this iteratively on my triangle spreadsheet on my website. Try it yourself. You have to download it to be able to change the values.

You have to change the G/V ratio to 12.615 from 12.3, leaving the speed for best glide at 110. This is not completely correct, but we have no better data. The best answer would be a lower best glide speed and a lower glide, but still higher than 12.3.

The bottom line is that at 10,000' and 190.3 mph the Van's RV-6 using CAFE drag data needs 76.88 THP. That's an increase of 4.64 THP from 185.2

The new tips appear to give the RV6 4.64 more THP at 190 mph at 10,000'. That's net of increased parasite and decreased induced drag. The induced drag for the standard model 6A, adjusted to be a 6, at that speed and altitude, is 25.74# and parasite is 125.75#. Paul's friend's numbers will differ a little, of course, even before the new tips.

---------------------------------------------------
I won't get into the more technical stuff because I am not qualified to do so. I would suggest, though, that before we get exited about the new tips we look at what the unaltered airplane should be doing. It is not unusual for RV's to go faster than Van's numbers and most go at least as fast as they "should". This one is still not doing that.

That's interesting, Howard, because the information I copied from Van's performance data says that a 150 HP RV-6, solo (1400lb) at 8000'/75% power is 187 mph and at gross is 186 mph and with 160 HP is 191 mph solo and 190 gross.

If that 187 is actually supposed to be knots, then that says a -6 should go 215 mph; not bad, and Jim's is definitely slower than that. 'Course maybe my data from 3 years ago is outdated and you have an updated version. But if my data is still correct, then Jim's GPS-measured speed of 193.5 mph at 1440 lb, from four separate tests, is 3.5% faster, or like having 10.8% more power.
 
While not astatically pleasing, your results speak for themself.

You missed part of my post where I asked if you could get the same results with tip sails rather than the shape of Jim's wing tips.

I realize tip sails used to be THE fashion statement 20 years ago but we haven't seen or heard much about them lately.

If you mean tip sails that extend at some vertical angle from a wing tip, the additional span they add they will reduce induced drag, but don't really add to the wing area, so that the performance increase isn't quite as great as with these tips.

Both will add to the parasite drag area which comes into play at lower altitudes. What I was trying to get across is that I think their tip planform brings the wing's lift distribution closer to the elliptical ideal, and as such, improves the Oswald efficiency factor, making the wing's span efficiency even better. I'm really waiting for some aero experts to weigh in on this with their opinions.
 
For you of a more sagacious or intellectual acumen, you might want to obtain this AIAA paper which deals with several types of slashed or swept tips, which an aero acquaintance has so kindly sent to me:

AIAA-45747-473.pdf
 
Oops again

That's interesting, Howard, because the information I copied from Van's performance data says that a 150 HP RV-6, solo (1400lb) at 8000'/75% power is 187 mph and at gross is 186 mph and with 160 HP is 191 mph solo and 190 gross.

If that 187 is actually supposed to be knots, then that says a -6 should go 215 mph; not bad, and Jim's is definitely slower than that. 'Course maybe my data from 3 years ago is outdated and you have an updated version. But if my data is still correct, then Jim's GPS-measured speed of 193.5 mph at 1440 lb, from four separate tests, is 3.5% faster, or like having 10.8% more power.

Oops. Vans says 198 top speed and 187 at 8000. Too many distractions down here in Manzanillo (vacation). No, it's not knots, it's mph. Jim's airplane did 197 at 4000' so it is faster than the standard. I did not find a speed for Jim's plane at 8000'. Did I miss it? Did Jim's plane do 193.5 at 8000? Your note wasn't specific on that. That would be faster, no question.

My 7A does 200.2 mph at 8000' (and higher, but that's not useful here). Van says 198 at 1400# (mine was at over 1500). I'm not bragging, just reminding everyone that besting Van's numbers is not unusual. There are many faster 7A's than mine. I remember that Roberta's 7A was noticeably quicker than mine. When you limit RPM on a FP airplane, you get distorted results. Mine can go 199 or more at 10,000. I'm not comparing to Jim's plane, just to Van's.

As Paul already knows, but to answer Danny7, it is not necessary to assume or know the engine power when you are looking for drag. I did the analysis based on drag only. THP is drag times velocity. BHP would have to include a factor for prop efficiency and I wanted to avoid that aspect. Whatever was done to Jim's airplane was drag modification. That said, Van's numbers are for the 150 HP version and I don't know what prop he used or assumed.
 
no surprise at all

...sorry, not to interrupt the engineers here, but should we not recognize that Steve Wittman was onto this 40+ years ago?
Doesn't make it any less impressive; (my -9a would certainly look ok with them if I could be confident of the engineered attachment.!)
Keep on experimenting...or we'll have to take the label off our planes! :)
 
Are you making assumptions about how much hp his engine is making? what else do you think could be different from the vans numbers if the plane is a bit different in that area?

Jim's O-320 is 150 HP operating on auto fuel.

Here's Jim's speed vs dalt based on three separate GPS-measured runs, and then run through a 2nd-order least-squares fit with density altitude to account for the variations in OAT at the different baro altitudes of 4000', 6000', 8000' and 10,000'. This is, to my way of thinking, the best way of combining data where the dalt is somewhat different.

4000' 197.0
5000' 196.4
6000' 195.6
7000' 194.6
8000' 193.4
9000' 192.0
10,000' 190.3
11,000' 188.5
12,000' 186.5

It's interesting to note that Jim's speed at 12,000' dalt is almost the same as Van's numbers for 8000' dalt.

Guys, understand that I put out stuff like this because I think that for the most part, you are experimenters as am I. If I find a something that looks like it fixes a problem others have or will give better performance, I like to share it with other experimenters. I stand to make nary a penny on the wingtips, so I present the info because I think that there are some of you who might want to do something like this. It's not to get into a p****g contest with someone who wants to prove how much smarter he is than me. I'm sure there are a lot of you on this forum whose intellect puts mine to shame, and that you hold back often so as not to embarrass me!
 
That's it!

I'm sure there are a lot of you on this forum whose intellect puts mine to shame, and that you hold back often so as not to embarrass me!

Ummm. Yeah, that's it! I hold back because I'm afraid of making YOU look stupid.

--Stephen

p.s. Reality: I've learned so much from your posts, Paul. Thank you!
 
Ummm. Yeah, that's it! I hold back because I'm afraid of making YOU look stupid.

--Stephen

p.s. Reality: I've learned so much from your posts, Paul. Thank you!

Yep Paul, I'd have to agree with Stephen. The only time I hold back is when I'm trying to ensure that I don't "remove all doubt" (if you follow the old expression!) and I'm sure I fall short of that goal often! I learn much from your posts too. You also mentioned our "sittin' under the cowl" meeting at Reno with Jim, and I found him to be quite the gentleman as well.

As for this wingtip design...it really looks like it has its merits in economical performance. It might be considered an arrow in the quiver of wingtips...for those who don't mind changing them when needed.

I say that, because something Kyle said back in post 14 keeps resonating with me (about turning an 8 into a tandem 9). Perhaps this one is the cruiser tip (I know it has other advantages, but that seems to be the highlight (IMHO). There are sacrifices to the "total performance" package that is a 6, 7, or 8 if one straps a set of these on. But for a trip to the Bahamas or to somewhere in a galaxy far, far away, these may be perfecto...but no acro or dogfightin'enroute...better stay out of Injun Country, lest you get jumped by another RV...you'll be a grape...a "target"! ;) But these tips may have their place.

On the other hand, the racer's (and perhaps the "total performance RV" pilot's) holy grail of wingtips is the one that adds speed with no g-limit degradation (good looks is a plus, but like I said before, put a checkerboard or flames on it, and almost anything looks cool! :p). Every racer I've met has said its a quest that is difficult, to say the least. Many have tried...differences/gains due to wingtips have been small, and hard to measure.

What sayeth you Paul...you up for the challenge? I'll test 'em! :D

Cheers,
Bob
 
BTW, if you'd like to get an estimate of what your speeds would be relative to Jim's if you equipped your plane with similar tips, but with different HP, here're some multipliers to get you a ball-park estimate:

160 - 1.0217
180 - 1.0627
200 - 1.1006

These factors are based on the cube-root of the power ratio. For instance, you could expect at 8000' dalt, based on Jim's 193.5 mph, about 203.5 mph with 180 HP or about 213 mph with 200 HP.
 
Looking at Serendipity

I am not afraid to risk looking ignorant or of making a mistake if it will help me to learn and possibly to help others to learn. This could be one of those times, but here goes.

Paul's friend's airplane gained some speed with new wingtips. Let's take that as a fact. Paul says the parasite drag increased slightly. If I understand his numbers, it is a 1.77% increase in parasite drag. So the speed increase had to come from reduced induced drag and that appears to be what Paul is saying he did. He also says he got a surprise when it worked better than he expected. He attributes the improvement to better Oswald efficiency of the tips. BTW, the CAFE 6A test said the Oswald factor was .851 but Paul said he used .81 for the original tips. At least I think that is what he said.

Unless I'm mistaken on the basics (always a possibility) the parasite drag goes up with the square of the speed and the induced drag goes down with the square of the speed. At the point of best L/D they are equal. For an airplane like the RV, there is not much induced drag at 190 mph TAS at 8000'. It is around 25 induced to 175 parasite for the CAFE test. I don't know any reason to expect that set of rules to change with different tips. (Laminar flow would violate it.)

Paul estimated a 10 or 11 HP improvement. If that is BHP, I agree. At 190 mph at 8000' that means around 8.1 THP which is about 16 pounds of drag net of the expected increase in parasite. Hmm. 175 pounds times 1.77% is about 3 pounds. So we need to reduce the induced drag by about 19 (16 + 3) from a starting point of around 25. Do you see where this is going? If so, you can skip to the last paragraph.
--------------------------------

I was curious, so I created a special spreadsheet set of columns for the drag curve of the CAFE RV6A. Then I created a new set of columns for the same airplane at 1440 pounds instead of 1650, the test weight. Then I modified the parasite drag enough to account for the 2 mph gain going from A model to t.d. Now I have a model of a 6t (not 6A) at 1440 pounds. Then I created another set of columns for the 6e (as in Ellipse). Each set of columns has one for speed, one for parasite, one for induced and one for total drag. The columns for the 6e are set up to allow easy iterative alteration to get numbers to where we want them.

I used the 6e's 6.4 mph gain over Van's numbers at 8000'. I ignored the CAFE performance numbers which are for 180 HP. The 6e has a nominal 150.

Since the CAFE drag numbers are for CAS, I had to figure the CAS numbers for the TAS speeds at 8000'. I then figured THP from the drag times the TAS. The idea is to alter the induced drag, see the reduced total drag such that the same THP gets the correct mph increase. For the 6e, I increased the parasite drag by 1.77%. Then I tried to find the numbers that will give the same THP as the Van's numbers, but at the higher speed (6t vs 6e). I knew that the 6e had a lower stall speed so I expected a lower best L/D, too. This is where I ran into trouble. In order to get the drag and THP numbers to come out right I had to reduce the L/D speed to 53 and the drag at that speed to 33 pounds and the resulting glide ratio was over 40. I don't consider these numbers realistic.

There are obvious problems with this analytical technique. I'm combining the CAFE drag numbers with Van's performance numbers. I'm using conventional techniques for changing from 1650 to 1440 pounds but I could have done that wrong. I could have made a mistake converting from 6A to 6t. None of these look wrong to me, though. Just to put possible error in perspective, the speed for the 6e before the new tips at 8407' DA was 187 mph. If you plug that into my model for the 6t you get less than 1 THP difference from Van's numbers. When I get home next week I can put the whole spreadsheet on the web so that anyone can check it. There are very likely small errors in all of this, but the result for the 6e with increased parasite drag is grossly unreasonable.
----------------------------------------

The 6e performed very closely to the CAFE 9A (can't be precise - too much missing data). That's very good for a shorter wing with longer chord and allegedly less efficient airfoil. I think that something in the design of the 6e improved parasite drag (too). If you use my triangle spreadsheet and use a L/D speed of about 95 mph (a guesstimate of five less than the 100 for the 6t at 1440#) you get a pretty reasonable result. The problem is that when you do that you get lower parasite drag along with lower induced drag and that is not what Paul says he expected. This is merely a guess on my part. I don't know what the experiment did to either kind of drag. When you encounter serendipity it is good to pursue it. It would be good to better understand how those tips actually do what they do.
 
The wing tips here increase span and AR. For a tighter analysis of the resulting planform, a vortex lattice model should be done to study the spanwise loading. The Oswald factor is a pretty dumb correction to estimate induced drag on a non elliptic planform, and isn't a good analytical tool.

D = Cd_0 + CL^2/Pi/A/e where e is the Oswald Factor. Pretty unsophisticated.

I must admit I'm surprised these helped at all, but it just goes to show how badly the short wing RV's need more AR to reduce induced drag.
 
Last edited:
For induced drag I prefer the simpler CDi=W^2/(Q^2*B^2*pi*e), which shows immediately the contribution to span in overcoming induced drag, and you don't have to do additional calculations involving span and area to arrive at AR. If you set this equal to CDo and solve for Q, you have the estimated speed at which L/D is maximum.
And I must apologize, Harry, but after wading through what you wrote, I'm not really sure what you were driving at. Could you just put into a simple declarative statement what it was I got right or wrong? :confused:
 
What I said..

For induced drag I prefer the simpler CDi=W^2/(Q^2*B^2*pi*e), which shows immediately the contribution to span in overcoming induced drag, and you don't have to do additional calculations involving span and area to arrive at AR. If you set this equal to CDo and solve for Q, you have the estimated speed at which L/D is maximum.
And I must apologize, Harry, but after wading through what you wrote, I'm not really sure what you were driving at. Could you just put into a simple declarative statement what it was I got right or wrong? :confused:

I answer to Harry, too. Just not late for dinner.

The short version: the performance gains observed cannot reasonably be accounted for by reducing induced drag alone.
The model is assumed inviolable and so the data must work within the model. The most reasonable guess is that, somehow, parasite drag was also reduced.
 
I answer to Harry, too. Just not late for dinner.

The short version: the performance gains observed cannot reasonably be accounted for by reducing induced drag alone.
The model is assumed inviolable and so the data must work within the model. The most reasonable guess is that, somehow, parasite drag was also reduced.

Sorry that I must disagree with your guess, but I think you're making an assumption that you can't support. First off, I must tell you that as far as C.A.F.E.'s numbers on OEF are concerned, I don't consider them to be the last and greatest word on anything to do with airplanes. They did some interesting testing years ago, but I don't credit them for anything beyond that.

You must consider that when testing any individual airplane, the performance numbers that you get apply only to that particular airplane and not apply to the fleet as a whole, as we're talking about homebuilts, not factory-built planes. They are engineering estimates!

Now I asked a well known acquaintance who actually designs real honest-to-God, high performance airplanes and he said that these short, wide wings are known for not being very efficient. In the book Design of the Aeroplane, by Darrell Stinton, 1983, he shows on P. 139 a maximum of 0.82 for a wing planform somewhat similar to an RV-6 with its raked tips if it has sharp rear corners. I'm afraid again that because Darrell actually designs real aeroplanes, that I would take his research over a single C.A.F.E. test.

Darrell gives the rake angle vs AR as 0? for infinite AR, 20? for AR = 6, and 25? for AR= 1 to 5. 'Don't know if that's a linear progression, but doubt it. The references for these rake angles are: Shaw, H. (1919), A text Book of Aeronautics London: Charles Griffin and Company Limited, and Barnwell, F.S. and Sayers, W.H. (1916) Aeroplane Design and a Simple Explanation of Inherent Stability London: McBride, Nast and Company. (Note, Barnwell was the designer of the Bristol F2B Fighter (1916-1917). The RV-6 tips with 1' span and 5' chord are about 11.3? for their 4.8:1 AR. Go figure if that's correct or not!
 
An error here:

The correct form for the dimensional induced drag in units of force is
Di = W^2/(pi * Q * B^2 *e).

So, if you divide by Q again (as shown in the quote) you have an equivalent "flat-plate" drag area. Not a drag coefficient yet. To get CDi, you must also divide by the reference area.

It is often very handy to work in terms of the dimensional forces for a given airplane.


For induced drag I prefer the simpler CDi=W^2/(Q^2*B^2*pi*e), which shows immediately the contribution to span in overcoming induced drag, and you don't have to do additional calculations involving span and area to arrive at AR. If you set this equal to CDo and solve for Q, you have the estimated speed at which L/D is maximum.
And I must apologize, Harry, but after wading through what you wrote, I'm not really sure what you were driving at. Could you just put into a simple declarative statement what it was I got right or wrong? :confused:
 
Off Point

Sorry that I must disagree with your guess, but I think you're making an assumption that you can't support. First off, I must tell you that as far as C.A.F.E.'s numbers on OEF are concerned, I don't consider them to be the last and greatest word on anything to do with airplanes. They did some interesting testing years ago, but I don't credit them for anything beyond that.

You must consider that when testing any individual airplane, the performance numbers that you get apply only to that particular airplane and not apply to the fleet as a whole, as we're talking about homebuilts, not factory-built planes. They are engineering estimates!

Now I asked a well known acquaintance who actually designs real honest-to-God, high performance airplanes and he said that these short, wide wings are known for not being very efficient. In the book Design of the Aeroplane, by Darrell Stinton, 1983, he shows on P. 139 a maximum of 0.82 for a wing planform somewhat similar to an RV-6 with its raked tips if it has sharp rear corners. I'm afraid again that because Darrell actually designs real aeroplanes, that I would take his research over a single C.A.F.E. test.

Darrell gives the rake angle vs AR as 0? for infinite AR, 20? for AR = 6, and 25? for AR= 1 to 5. 'Don't know if that's a linear progression, but doubt it. The references for these rake angles are: Shaw, H. (1919), A text Book of Aeronautics London: Charles Griffin and Company Limited, and Barnwell, F.S. and Sayers, W.H. (1916) Aeroplane Design and a Simple Explanation of Inherent Stability London: McBride, Nast and Company. (Note, Barnwell was the designer of the Bristol F2B Fighter (1916-1917). The RV-6 tips with 1' span and 5' chord are about 11.3? for their 4.8:1 AR. Go figure if that's correct or not!

I am not arguing about tip efficiency. All I'm saying is that you can't get that much of a performance increase from induced drag alone without changing the max L/D speed and ratio to a degree that is very unlikely in this case. Try your own drag curve estimates for that airplane (which just happens to perform, pre-tip mods, at the same speeds with the same engine as the Van's prototype).

Do you really think CAFE's drag curve can be off from Jim's airplane that much? Or do you think that the rules for how drag changes with speed are not "true"? Is the airplane now at L/D max at 53 mph when it was around 100? Does it now have a 40:1 glide ratio?

The model is a framework within which the observations must fit. That model does not care about tip efficiency as a separate factor. Parasite drag goes up, induced goes down, each as the square of the velocity. They cross at max L/D and drag min. If you don't accept this, say so. If you do, then do the curves and see what you get.

Here I am saying that you found a way to reduce parasite drag with a tip change and you don't like it? Better you should try to figure out how you did it.
 
They cross at max L/D and drag min. If you don't accept this, say so. If you do, then do the curves and see what you get.

Here I am saying that you found a way to reduce parasite drag with a tip change and you don't like it? Better you should try to figure out how you did it.

If you'd look at the AIAA paper that I reference you'd see that they found that the overall wing efficiency went up; this is more than just reduced induced drag due to increased span. The wing's induced drag decreased even more than would be obtained from the equation for CDi. Since the other terms in the equation are fixed - weight, Q, pi, and span, the only thing that must have changed is the OEF! They show that the wing's L/D increased with the swept-slashed tips.

Howard, Harry, the references show that in order to obtain an OEF of 0.82 for a rectangular wing that the tip would have to be at a 20? sweep angle; the RV-6 is 11?, so I stand with my use of 0.81, which is still very good, and I totally reject C.A.F.E.'s 0.85. You know, I really looked hard but couldn't find any plane that C.A.F.E. designed. Does that mean I dismiss them totally? No, they have done some interesting testing, but between the testing and analyzing the results there is a great gulf. I know a lot of people in aero who sort of look down their noses at some of C.A.F.E.'s conclusions.

Everything that is done in analyzing data relies upon making certain assumptions, as you did in accepting C.A.F.E.'s 0.85 OEF even though it flies in the face of other analysis. Now it is up to you and your assumption of reduced CDo to show exactly how the rest of the airplane's drag went down as a result of installing these wingtips. When you do, send the results to the SARL mailing list so that these guys who love to race can make their planes go faster.

I really am pleased with the results of the tips, but even more so is Jim who went through a lot of anguish in whether the effort was worth the time and cost of $600-$800 to make them. As the title of this posting said, the outstanding results he obtained, much better than I had estimated, really puzzled me and came as a surprise until I was able to find how the wing's lift distribution, tending more toward the ideal elliptical with its OEF of 1.0, could also have an OEF tending toward more toward 1.0 than the assumed 0.81.
 
The correct form for the dimensional induced drag in units of force is
Di = W^2/(pi * Q * B^2 *e).

So, if you divide by Q again (as shown in the quote) you have an equivalent "flat-plate" drag area. Not a drag coefficient yet. To get CDi, you must also divide by the reference area.

It is often very handy to work in terms of the dimensional forces for a given airplane.

So right, Steve! I used the wrong symbol because I usually add that to the equvalent parasite drag area to get the total drag area. Thanks for pointing that out!
 
Just to put possible error in perspective, the speed for the 6e before the new tips at 8407' DA was 187 mph.

Actually Jim's RV-6, the -6e you speak of, really went 190.4 mph avg. from three separate test flights at 8000' dalt, which is 3.4 mph faster than Van's 187 mph. That's 1.8% faster, or like having 5.6% more power. I know of a lot of SARL racers who would give a lot to see a 1.8% speed improvement.

If you'll look up the AIAA paper that I referenced, P.208, in it you will find that they clearly state that there was an reduction in Ki, which is the reciprocal of the OEF, meaning that the wing is giving less induced drag, not a decrease in the parasite drag. At high CL, as at low speed or in a climb, the L/D was also improved over a standard tip.
 
Backing up my assertion of your data

Actually Jim's RV-6, the -6e you speak of, really went 190.4 mph avg. from three separate test flights at 8000' dalt, which is 3.4 mph faster than Van's 187 mph. That's 1.8% faster, or like having 5.6% more power. I know of a lot of SARL racers who would give a lot to see a 1.8% speed improvement.

If you'll look up the AIAA paper that I referenced, P.208, in it you will find that they clearly state that there was an reduction in Ki, which is the reciprocal of the OEF, meaning that the wing is giving less induced drag, not a decrease in the parasite drag. At high CL, as at low speed or in a climb, the L/D was also improved over a standard tip.
--------------------------------
http://www.vansairforce.com/community/showpost.php?p=473136&postcount=26:
"8000', 8407' dalt, 2700 rpm, 187 mph"
That was before the new tips if I understand what you wrote. That is the basis of my assertion that the pre-new-tip performance approximates Van's. You can look it up. You said it. I have no data of my own, only what you posted. If in actuality it went faster before, then the improvement from the tips was less. What is the actual, correct set of numbers?
------------------------------------------------------------------------
Please understand what I am saying. I think I am making it clear. Yes, the new tips likely reduced induced drag. But, no, the performance gains cannot reasonably be explained by induced drag alone. It's really very simple and the curves are what they are. All the discussion of the details loses sight of the big picture. I don't know what changed parasite drag but there are only the two basic kinds of drag and they move in opposite directions as the square of the velocity. If induced alone cannot fully explain what happened, then there is no place else to look except parasite. You say parasite increased. That seems even less likely.

You have produced, according to your data, a significant improvement. You are overlooking or even denying something important about that. It would be good to further investigate it. It is not appropriate to issue challenges to someone like myself who is only reading what you wrote and applying known relationships. All I'm doing is asking what I think are good and reasonable questions. I think many would benefit from the answer(s).
 
The Spitfire has one of the most pleasing wing shapes ever designed. Based on what I have read about elliptical wing shapes, it is also the most efficient wing for subsonic flight. I can't help but think someone is reinventing the wheel here while losing the aesthetics and beauty of the design.

Kudos for gaining speed without adding horsepower to our hershey bar wing.
 
It's cool to see one of Paul's posts reanimated. He added so much to this forum. I miss his contributions.

--
Stephen
 
Photo and mention Of Paul Lipps in FLYING Magazine

Yes, I miss his posts, too. Peter Garrison wrote a nice mention of Paul Lipps in the March issue of FLYING on page 70.
 
Back
Top