October 2012

Experimenter is a magazine created by EAA for people who build airplanes. We will report on amateur-built aircraft as well as ultralights and other light aircraft.

Issue link: http://experimenter.epubxp.com/i/84816

Contents of this Issue


Page 28 of 47

when that turbo does not raise top output but only allows maintenance of sea-level power to some alti- tude, raises engine output at any given altitude, and thus raises cool- ing requirements. Drag increases faster than the square of the speed (many claim the cube; some, more than that—it depends somewhat on the speed range one is discussing), but the amount of heat the air carries away increases only linearly with speed. For example, doubling the speed in- creases drag by at least four times, but the amount of air going past the fins merely doubles. So, the faster we fl y, the higher we fl y, the more critical becomes the cooling design. Some Important Considerations Air separates from the surfaces it covers when the angle increases beyond a certain point. In general terms, that's about 15 degrees; in a tube, the walls should converge or diverge no faster than about 7 degrees each. We know that when a wing's angle of attack exceeds about 15 degrees, we quickly trade lift for turbulence; the same thing happens to any air interface. So, when we're ducting air in our cowlings, gentle "ramps" are re- quired. However, once we know we need to break that flow (22 degrees, say, is as bad as 90), we can then abandon the pretense of keeping airflow attached, and just deal with the drag and build the "easiest" solution. Air cools as it expands and heats up as it is compressed. An extreme example is a diesel, i.e., compression engine, where combustible gases are compressed in the cylinders until they ignite simply from the heat generated by compression; no spark plug required! Photography by Tim Kern Turbulence can be used to our advantage, if we are aware of it and exploit it. Pressure, both high and low, can and should be exploited in the cool- ing design. With propeller planes, the amount of air the blades move is asymmetrical in climb. We all recognize P-factor when climbing; the same phenom- enon is in play when we consider how much cooling air the prop is forcing into the cowl. (Did you ever notice the little blanking plate in front of the "down blade" cylinder on many certificated airplanes? That's why it's there.) Your cooling requirements extend to more than just your cylinders. Consider cooling your alternators, magnetos, batteries, etc., and don't forget the oil cooler! Although crude "eyebrows" work on many slow airplanes, an integrated enclosed cowl/internal baffl ing system is more elegant and ever more effi cient as aircraft speeds rise. Keeping the cooling air focused on its job is the duty of the inner cowl design, specifi cally the baffl ing. In "fl at" (opposed cylinder) engine airplanes, the simplest enclosed designs consist of an opening in the front of the cowl for inlet air, a hori- zontal "tray" separating the top and bottom of the cowl, and an opening in the bottom of the cowl, to let the air out. The purpose of the tray is to di- rect the fl ow of cool inlet air through the engine's fi ns, rather than allowing it to fi nd the easiest way out, bypass- ing the engine. The next step is to build a vertical wall atop the tray, behind the cylin- ders, to concentrate the airfl ow. The gap between the cowl and the top of the metal "box" and the sides of the tray is typically closed with a sheet Jeff LaVelle's Glasair topped 400 mph at Reno, partly due to effi cient use of cooling air in a proper plenum. of fl exible reddish silicone cloth. For slow airplanes, this can be suffi cient; but it's ineffi cient. Other approaches are to attach the baffl ing directly to each cylinder bank. Jabiru, for instance, furnishes formed boxes that can be ducted forward to the cowl inlets, which fi t well to the cylinders and heads. Put cold air in the front, and the factory- supplied ductwork carries it right to the hot spots. It's important to put the cool air where it does the most good, and Tony Higa's Pitts, set up for Reno racing, features an internal pressure plenum inside the original cowl. W.G. Coppen's Corvair-powered SA-7 gathers all its air in the cowl and forces it through the six cylinders' fins. EAA EXPERIMENTER 29

Articles in this issue

Archives of this issue

view archives of Experimenter - October 2012