Ridding Our Aircraft of P r o t e c t y o u r a i r p l a n e ’s p o w e r TO D D S. PA RK ER

any sea creatures have their sleek shapes disrupted by the presence of other creatures seeking free rides. Barnacles and remoras are among these freeloaders. While not true parasites, biologically speaking, barnacles and remoras suck energy from the host vessel. How many barnacles or remoras do you have stuck to your vessel—your aircraft? Drag is the non-useful energy expended while we fly. It can be thought of literally as the amount of energy we expend dragging air along with us on our journey. You have probably stood next to a busy road and felt the disturbed air following behind cars and semitrailers as they passed. What you felt was the air that had picked up momentum as the cars and trucks passed through it. You can sense how much drag those vehicles have by how forcefully the air blows when they pass. Engineers discovered early on that they could measure drag from wings and other bod46

FEBRUARY 2008

ies by placing pressure probes behind the object in a wind tunnel or in flight and measuring how much momentum was lost compared to the air with no obstructions. This method is called a wake survey and is an accurate means of determining the drag of any object. There are two types of drag affecting the flight characteristics of an airplane: induced drag and parasitic drag. Induced drag is a byproduct of lift, and it’s at maximum at the point the aircraft stalls. As speed increases, induced drag decreases with the inverse square of velocity. In con-

EAA Sport Aviation

47

induced drag. It will likely add cooling drag as well. Engine cooling is often the largest single drag component. Reducing this component of drag was addressed in the March 2006 issue of EAA Sport Aviation (“Control the Flow”) and will not be addressed here. The purpose of this article is to help identify those little parasites sucking power, and with it speed, from our aircraft. It may surprise you how much power is being sapped away by these seemingly small parasites.

THE RIVET

A metal aircraft is covered with thousands of tiny parasites. The rivet is probably the most common element of all aircraft. Individually they are insignificant, but collectively they can represent huge power losses. It should come as no surprise to anyone that a flush-type rivet has the lowest drag, but it can have a range of drag coefficients depending on how it is installed. If the rivet is filled and matched to the surface, it will offer no more resistance than the skin would without the rivet. This is the best condition. The Germans found that a domehead rivet created 20 times as much drag as a flush rivet in laminar regions of flow. Recessing the flush TOP: Wheel pants may cut drag in half, even though they are larger. rivet about 0.001 to 0.002 inches reBOTTOM: One of the subtle parasites on your aircraft include antennas BOT duced the drag by half. If you have protruding into the smooth airflow across your aircraft. A 1/4-inch diameter, pr a laminar airfoil, a non-flush rivet 2-foot long antenna will dissipate approximately 1 hp at 125 knots. 2 is not for you. Once the boundary layer is about 30 to 40 times thicker than the height of the rivet, the differences in drag are trast, parasitic drag is directly proportional to the square tra minimal, but not zero. of the velocity, increasing dramatically with speed. Paro asitic drag is caused by many factors: form (pressure), cooling, intersection, prop wash, excrescence, and wet- JOINTS, BUGS, AND THINGS THAT STICK OUT ted area. The lowest drag condition occurs where the in- Another subtle parasite is the type of skin joints used duced and parasitic drag are equal. This point is usually and the quality with which they are constructed. A lap much closer to stall speed than cruise speed but moves joint with the top surface pointing into the flow will farther from stall when the parasitic drag is reduced. have twice the drag of the lap facing the other direcThe more drag produced, the more power must be tion and 40 times as much drag as a flush- or butt-type expended to counter it. The power needed for this ef- joint. On a forward lap joint, just rounding the protrudfort directly subtracts from power available for useful ing edge can reduce the drag by a factor of 10. A slightly flight. (For sailplane folks this means a higher descent lifted forward facing thin edge can have 60 times greater rate.) Therefore, we either fly slower or add more power drag than the flush joint. to counter the effects. Adding power will offset the efThe condition of the leading edge of an airfoil can fects, but it’s not the most efficient means because it make a huge difference in the drag and performance of will usually require a larger engine and more fuel for the it. When airfoils are rough due to bugs, ice, paint chips, same range. Both of these add weight and increase the dents, or any other disruption of the leading edge, the 48

FEBRUARY 2008

minimum drag coefficient of the airfoil is typically doubled. On some highly laminar airfoils, the lift may also be severely reduced, perhaps as much as half, because of the rough edge forcing turbulent transition. Your propeller is also an airfoil, and all of those little chips and dents are robbing you of power right at the source. Wire, tubing, and pipes are convenient shapes for many things, but aerodynamics is not one of them. A cylindrical shape, compared to a streamlined shape, creates anywhere from four to eight times as much drag. Cylindrical shapes should be avoided or faired whenever possible. Exhaust stacks are especially important to fair or turn into the flow. Aerodynamically speaking, exhaust exiting from the pipes looks like long poles sticking out into the flow. NACA researchers showed they could place pipes to represent flows from exhausts when doing wind tunnel studies. Turning those poles as close to parallel to the flow as possible sounds like a good idea, and it is. Wire antennas protruding into the flow also create huge drag considering their small shape. A 1/4-inch diameter, 2-foot long antenna protruding into the flow will dissipate approximately 1 hp at 125 knots. Blade type or streamlined antennas can The rivet is one of the most common elements on an aircraft and can collectively vely significantly reduce the drag, or betcreate a large amount of drag. The flush-type rivet has the lowest drag if installed led ter yet, bury the antenna in the strucproperly, while the dome-head rivet can create 20 times as much drag. ture. On composite/fiberglass aircraft, embedding the antennas is common. Carbon composite and metal aircraft must have non-con- the retract mechanism is heavy (more induced drag). Control cables, pushrods, torque tubes, control horns, ducting covers or fairings to achieve the same gains. Fixed landing gear can be another major source of drag. and bell cranks often look like barnacles stuck to the sides Frequently, landing gear have round struts. As discussed of our aircraft with the same drag-producing effects. When earlier, these should be faired whenever possible. More im- possible these items should be contained within airframe portantly, wheel fairings should be added. Unfaired wheels structures, but the mechanical gearing and control tolerare messy aerodynamically, with tubes and sharp corners, ances may not permit this luxury. These protruding items small air passages, and interferences everywhere. Unfaired should be faired with gaps kept to a minimum. Some of wheels are similar to an engine without a cowling, a lot the new Reno racers like the AR-6 or the new Air Force of messy, useless turbulence. Wheel fairings typically cut fighters, F-22 and JSF, exhibit superb examples of what can the drag from landing gear in half even though their di- be done to control horn fairings. Many modern aircraft mensions are larger. Retractable gear may have more net have slotted or Fowler flaps with the axis of rotation bedrag than fixed gear unless the holes the gear fits in are low the wing. The mechanism typically involves a series closed. Some aircraft whose gear doors are attached only of hinges. Careful fairing of these can pay large benefits in to the struts do not cover the wheels completely. Such ar- drag reduction. Commercial jets are also good examples of rangements may not reduce the drag beyond what can be how to do this. For non-pressurized aircraft, inlets for cabin air should achieved with a good fairing on a fixed gear, especially if EAA Sport Aviation

49

Control surface gaps contribute Cont to excrescence drag (yes, that’s explained in the article) and should ex be kept as small as possible. Gap seals—as simple as strips of tape—help se tto reduce the effect.

be NACA flush scoops whenever possible. The drag of these scoops is very low regardless of whether the vent is open or closed.

WHAT IS EXCRESCENCE DRAG?

This brings us to one of the least obvious parasites of all. It is broadly known as “excrescence drag,” or drag from air leaking into or from places you don’t have a need for it to go. Engine compartments are usually the biggest source of this type of drag. A loose fitting cowling, or one without a seal strip, can leak copious amounts of relatively high-pressure air, disrupting the flow around the cowling

50

FEBRUARY 2008

and reducing the air intended for engine cooling. Sometimes a cowling is not stiff enough, deforming to create leaks when in flight. Loose fitting or leaky baffles inside the engine compartment can rob you of efficiency in the same way. Seal it up as tightly as you can. Only let air in and out of places you want air to flow. Cabins can also contribute to excrescence drag. Improperly located or designed vents may bring air into or let air out of places where little benefit is obtained. I remember hot days leaning over in the little Cessna 150 trying to catch the breeze coming from the little pullout vents bleeding air from the leading edge of the wings. Ahhh…. Non-pressurized cabins can either be pressurized close to pitot pressure or they can be less than static pressure, it all depends on where the inlet and outlet air comes from. This is the reason it is usually not a good

idea to have your static pressure source taken from inside the cabin. Again, drawing from experiences with cars, I am sure most of us have felt the pressure change inside a car when a window is opened. It may either increase or decrease the pressure in the car depending upon which window is opened and where the air source and exits are located. Sealing the cabin windows, hatches, and doors so that all of the entrances and exits are controlled is important in reducing this source of drag. In non-pressurized aircraft, this problem is made more difficult by the fact that the controls and other systems are sharing the same volume with the cabin. This means that anywhere a wire, cable, pushrod, or other system exits or enters the airframe, there is the potential for excrescence drag to occur. Where seals are not practical or desirable, a close fit is encouraged. Another source of excrescence drag is control surface gaps. This was a major research area in the ’30s and ’40s, with many simple and complex solutions developed. This type of drag is simply the air that leaks from the highpressure side of the wing to the low-pressure side of the wing. The effects are to increase drag, reduce lift, and reduce control effectiveness. There is nothing good about control gap leakage. Gaps should be kept as small as possible. Some choose to close the gaps with simple strips of plastic taped to the wings or fins with vinyl tape. Appar-

ently, the drag from the gaps is larger than the drag from the tape and plastic. This system is simple and effective. Some car and airplane designers make the top and side surfaces with nice sleek lines and flowing streamlined shapes, and then stick all the messy drag-producing stuff underneath where it may go unnoticed by the consumer. Often an assortment of struts, wires, hinges, drains, vents, antennas, and other protrusions can be found here. Air is not impressed by aesthetics and will find all of those drag-producing add-ons no matter where you stash them. Depending upon how they are arranged, collecting all of these items into one location or line may either increase or decrease the net drag effect. A thick, turbulent boundary layer will likely be found downstream from these “porcupine farms.” Streamlining and reducing the individual effects as much as possible is the best chance of sorting this mess out. Next time you walk up to your airplane, think about the parasites stuck to its surface looking for a free ride. How many are necessary and how many are just stealing power from you? I hope that some of the identifiable parasites can be removed or at least changed from barnacles to remoras. Todd S. Parker is president of EAA Chapter 23 in Ogden, Utah, and works as an engineer for the U.S. Air Force.

EAA Sport Aviation

51