Airfoil Testing With Denny Dart II

by Bill Kuhl

Having built several of the standard Denny Dart II rubber-powered model planes, I thought it would be interesting to test two additional airfoil types using the same the airplane with the exception of the airfoil. One plane was built with a Two Surface Flat Bottom airfoil with covering on both sides of the wing and another plane was built with a single surface airfoil with a 6% arc with the high point at the midpoint of the chord.

Stock Flat Plate Wing	6% Curved Single Surface Wing
	Two Surface Flat Bottom Wing Covered Top and Bottom

To make the comparison as fair as possible, I wanted to get the weights of the two planes identical. The extra weight of the additional covering material on the Two Surface Flat Bottom wing made this plane considerably heavier. Shaving all the balsa on the motor stick and tail surfaces possible, was still was not enough to make the weights equal, so I added a small piece of clay to the single surface plane to make the planes equal in weight at 8.1 grams less rubber.

The same loop of 3/32" width rubber was used in all flights. It was only wound to 40 winder turns to keep from over-stretching the motor and to keep the plane from bashing the steel girders.

To compare with the original flat plate wing, a flat wing was constructed that was switched on the plane with the 6% curved wing. It was interesting to note that the flat wing had to be moved forward ½ inch from the position that the curved wing flew with.

Five flights were flown for each plane and times were recorded. Times were pretty equal except the times of the flat plate wing were normally slightly lower. Observation of the flights did reveal more differences.

The plane with the Two Surface Flat Bottom-type wing flew noticeably faster and climbed higher. All of the winds of the rubber motor were used up with the plane still fairly high.

Watching the 6% curved wing plane, it flew slower and did not climb quite as high as Two Surface Flat Bottom but almost. There were a few winds left over on landing.

Putting the flat plate wing on the curved wing plane required a big CG change before it would climb at all. The plane flew fairly slow with the tail dragging slightly, the highest altitude reached was consistently lower than with the other two planes.

Note how much farther forward flat plate
wing was than the curved wing.

Flight Time Comparison

Center of Gravity Positions for Airfoils

Two Surface Flat Bottom Center of Gravity was 1.75" from leading edge

Curved 6% Center of Gravity was 1.25" from leading edge

Flat Plate Center of Gravity was 1 7/8" from leading edge

Additional Test

Additional pitch was twisted into the 6" Midwest plastic prop on the plane with the Two Surface Flat Bottom airfoil and five flights were flown. Times were consistently higher than the with the unmodified prop.

1. 30 seconds

2. 28 seconds

3. 28 seconds

4. 27 seconds

5. 29 seconds

Feedback on My Results

The link to this website were passed to a couple of free flight listserves for comments on my results and methodology.  Received a variety of comments, only a small portion I have quoted here.  There is widespread belief that in small, slow-flying models that a flat plate airfoil is equal to a curved airfoil. Yet, from more scientific research than I have done, there are indications this is not true.

 "I should just like to comment on the oft quoted statement that flat plates are as good as anything at low Reynolds numbers. With the common flat plate airfoil, just covered on the top surface, there is usually a trailing edge spar acting as quite an effective 'Gurney' flap. So there are 'flat plates' and 'flat plates' and some of them can work quite well. However I would draw attention to low-ceiling indoor HLG and CLG. These aeroplanes work at very low Reynolds numbers. I do not think that a glider with flat plate balsa wings would stand a chance against the highly cambered beauties that are usually flown."

"Flat plate --poor lift -- high drag -- result poor L/D - Top forms laminar separation bubble --some lift, bottom goes turbulent and creates drag.

Curved plate - Very good lift -- higher drag - better L/D - Top has curved shape and if 5 to 6% may not have separation bubble good lift and lower drag than flat plate. Bottom goes severely turbulent at leading edge causing more drag than flat plate but back has curve down which acts as a flap giving good lift.

Covered top and bottom --Clark -Y -- intermediate lift --Lower range --Better still L/D, Top same as curved plate. Bottom may manage to be fully laminar when flown gently at a positive angle of attack as we do in Free Flight rubber. This will give less lift than the curved bottom but also much less drag.

Summation--

The efficiency is L/D and when D is reduced the plane can pick up a bit of speed so we have a new balance of forces.

Thrust = speed squared times coefficient of drag

Lift = speed squared times coefficient of lift

Bills plane with Clark Y was staying aloft and using all the turns in the motor where the curved plate landed with some turns. This is and indication of better efficiency since it stayed aloft with the reduced power at the end of the motor turns."

"I found one set of experiments by S. Suzuki with a rotating rig(Zaic 1955-56 YB) which did cover single surface flat and arcs as well as double surface flat bottom airfoils at a Re=45,000 which, unfortunately, is much larger than your Denny Dart's flight Re, but "probably" the results can be of service.

 In these experiments the thin flat plate had a best Lift/Drag ratio of 8.2 when the AOA was 2 degrees. A circular arc with medium camber of 5% had a best L/D of 14.5 at 2.5 degrees AOA. Increasing the camber of the arc to 10% reduced the best L/D to the flat plat value. In fact the best L/D for the arcs is about 5% at this Re. The Clark Y had a best L/D of 8.3 at 3 degrees, again similar to the flat plate, but this can be explained, as JB noted, by the small camber (about 3%) of the Clark Y.

Roughening the top improved the performance of the double surface airfoils. Thus an 8% thick airfoil with flat bottom and sharp nose and spars on top had a 10% advantage in L/D with respect to the same airfoil with a smooth top.

Conclusions? Camber matters more than thickness, too much camber seems to increase the drag out of proportion with the increase in lift. The closed airfoils tended to have the best L/D at higher angles of attack (near stall) in comparison with the single surface ones. The minimum drag is very high, a fact that surprised Suzuki.

 Quoting Suzuki's conclusions: "It is advisable to use airfoils for duration models that have "high" mean camber, about 4 to 5% and make the thickness of the wing as thin as paper".

Suzuki's little article is very interesting for what he tells about the conduct of experiments, which took about 4 years and required many set of wings of the same profile to get consistent results. If you have Frank's 1955-56 YB, do give this article a look. There are also some German wind-tunnel tests which are worth quoting later."

Return to AMA Cub / Delta Dart Webpage