Fairchild PT-19:
On Sept 22nd, 1939, the PT-19 had just won a fly-off against 17 other aircraft and was awarded a contract for 270 copies from the Army at $5800 apiece. How they beat out the competition is the story. Arman Thiebolt, from France, and Alexander Kartueli, a Russian engineer who had worked with Alexander De Seversky, came to the U.S. in 1928. After designing a few aircaft together, such as the "Uncle Sam" and the XFA-1 Navy Fighter, Armand ended up at Fairchild in 1933. In 1937 he became Chief Engineer and was assigned to design a new trainer for the US Army. His impeccable logic derived a clean and advanced design for it's time. It had a low wing to be a lead in for the new fighters, used a Fairchild designed Ranger engine and used non-strategic materials as much as possible. The fuselage is a welded truss using square tubing and the wing is primarily molded plywood over wood ribs and spar. The plywood gave the wing a smooth surface with no rivets. The wing design interests me most because here is where the famous handling qualities were created by trial and error. The PT-19 is probably one of the nicest flying aircraft I have had the privilege of flying and I have flown some, From De Havilland Chipmunks to Boeing 747s. Some of the homebuilts you fly have acceptable handling characteristics, but some are a little ill-mannered. The Long-eze I fly, while generally having good manners, at low speeds in the flare-landing phase becomes more and more pitch sensitive to a point where overcontrolling is "very easy" just before touch down. The PT by contrast maintains a constant response all the way to the stall. So how did they do that? Windtunnels for a full scale model were not available so all the testing was done in flight with tufts on the wings. They knew that for lateral control during the stall, and this is really were it counts most. They needed the stall to start at the wing root trailing edge, and then move outboard. So they selected a reliable turbulent flew airfoil (NACA 2416) for the root transitioning to NACA 4408 at the tip. The lift curve of the 4408 is more gentle at the stall than the 2416. 2 other methods for controlling the tip stall were tried: Leading edge slots of the outboard wing and sharp leading edges or spin strip on the inboard wing. The problem with the slots is that they became effective just when the rest of the wing is completely stalled. Snap rolls became impossible and looked more like a giant barrel roll. Stall strips were tried at mid span with only a slight loss of maximum lift. Unfortunately there aggravated turbulence over the horizontal stabilizer. Thiebolt wanted to stall to start at 20 to 30% of span outboard of the wingroot to avoid buffeting the horizontal stabilizer. He achieved the stall pattern but the design was still plagued with pitch buffeting. The solution turned out to be raising the horizontal stabilizer 10". Analysis of all the flight test data of the different configurations tried resulted in the original PT-19, but increasing the washout to 5% and using a slightly thicker airfoil at the tip and no slots or stall strips. This work was done from May 15, 1939 until they won the fly-off in September. Thiebolt later designed the AT-21, C-82 "flying boxcar" and the C119.