Test Methods and Tools
The majority of tests were conducted using the World War II Online airplane wind tunnel system. This is a system that I wrote for WW2OL when I worked at CRS in order to help verify the accuracy of the flight models with respect to historical data. The folks at CRS were kind enough to let me use this tool for my testing. In a nutshell, it loads the aircraft into a version of the game without graphics or other interface, runs the plane through a series of tests, and prints out the performance results. Much time and effort was spent on this tool to ensure that the data it produced matched the airplane in the game as close as possible. I would have loved to have had a tool like this when I did my original web site for Warbirds.
All tests were conducted with a full load of fuel, full load of ammunition, but no bombs.
The discussion below will use m/s for wind tunnel speed descriptions, as the basic unit for speed in the game engine is meters per second. Conversion to kph is 3.6kph per m/s, and 2.24mph per m/s.
Some tests could not be done with the tool, due to lack of certain features. These tests were conducted the old fashioned way, by hand, and with the aircraft's instrumentation.
The specific series of tests done for the pages are:
Mass determination
This was straightforward for the wind tunnel tool to determine. Add up the masses of all the components, and you have the total mass of the airplane.
Weapons configuration, rate of fire, fire duration, and ammunition load
Weapons configuration was determined by visual inspection, WW2OL documentation, and by seeing which guns fire when each of the triggers were pulled. Fire duration was done via stopwatch, ammunition load determined by various means (emails to CRS, counting ammunition fired via log file, etc), and Rate of Fire was determined by dividing the ammunition fired by the fire time. Ammo load was determined by firing into a brick wall with each weapon type, and noting what rounds hit and in which order.
Stall Speeds (1g, sea level)
The wind tunnel tool took the aircraft and turned off the engine. It then placed the aircraft at sea level (terrain collision was turned off). The aircraft was subjected to wind velocity of 20m/s to 200m/s in 1m/s increments, and the angle of attack determined that resulted in exactly 1 g of lift while the aircraft was held in place. The lowest speed where 1g was possible was recorded. This was repeated flaps up and flaps down.
Best Climb Speed
As part of the climb rate test, the wind tunnel places the aircraft at the specified altitude, and goes through the speeds from 20m/s to 200m/s in 1m/s increments, determining the angle of attack with 1g of lift, this time with max power, no WEP. The plane was still held in place, but the propwash was needed to ensure accurate representation of propwash lift and drag. The excess propeller thrust beyond that needed to overcome drag was determined, and a climb rate calculated from that and the airspeed. The best climb rate through the speed range was used for the climb rate for a particular altitude, and the speed at which this occurred was recorded. This puts the climb rate within 10cm/sec of the actual climb rate, and the best climb speed within 1-2kph compared to real climbing performance, the error due to the fact that the climb rate was calculated and not measured directly. In-game testing verified the accuracy of this approach.
The sea-level best climb speed was recorded for the performance pages. This speed is roughly similar to the IAS (Indicated Air Speed) needed for best climb at higher altitudes.
Maximum Reasonable Angle of Attack (AoA)
While the wind tunnel can determine the maximum AoA for maximum lift, in the game (like reality) the maximum reasonable angle of attack is usually lower. This is because most planes become unstable beyond a certain AoA, even when flying coordinated, and tend to enter stall early (one wing or both). This test was conducted by hand, and the highest angle of attack in a flat sustained turn that was possible without excessive effort or stalling out was recorded.
Top Speed, max power (no War Emergency Power/WEP) vs altitude
The wind tunnel did the same performance run as the climb test, except this time it looks for the speed at which the thrust exactly matches the drag at 1g lift and a particular altitude. This is the maximum possible speed in level flight for that aircraft and power setting. This speed was recorded for each altitude checked.
Top Speed, 200m altitude, WEP
The wind tunnel did not use WEP for its test runs for a number of reasons. As a result, this test had to be done via hand. A special front end with some extra instrumentation was used, and the plane taken off and flown at WEP at 200 meters altitude. The top speed attained without climbing/diving was recorded.
Climb Rate, max power (no War Emergency Power/WEP) vs altitude
The Best Climb Speed test data was used, this time recording the top climb rate at 1g, repeated at each altitude.
Roll Rate vs Speed, IAS, sea level
The airplane was held at sea level, and at different speeds, from 20m/s to 220m/s, zero degrees airframe pitch (this is usually the angle the plane attains at around 200-300kph). The ailerons were fully deflected (as far as the game lets the pilot deflect the stick for a given airspeed). The plane was spun at different rates until the rate that resulted in no rolling torque was found. This rotation rate indicated the maximum sustained turn rate for that speed.
Turn Rate, Flaps Up and Flaps Down
Turn performance was not part of the wind tunnel, for a variety of reasons I cannot go into here (NDA limitations and other reasons). Thus turn performance data had to be determined by hand. This was done by taking off, dropping any bombs, and turning the plane at maximum reasonable angle of attack and Max power (not WEP) (right hand turn for all planes). Altitude was maintained at 200m. Maximum AoA was determined during this test as well. When the speed settled, the speed and G load was recorded. Later, from geometry and dynamics equations, the turn rate and turn radius was computed. The test was done with WEP power.
WEP boost percent
The extra instrumentation on the Front End I used for the tests showed the output power of the aircraft (actually the manifold pressure into the cylinders, while running at maximum RPM, which makes manifold pressure proportional to the power output). The difference in output was recorded between Max power and WEP, and that percentage was used for this statistic.
Safe WEP time
The plane was taken off in WEP, and flown level at roughly 200m, until either 15 minutes passed, or ticking could be heard from the engine. Several minutes were removed for a margin of error, and the approximate time recorded.