Airspeed tube nose

Designed, not yet tested

This is the design of the airspeed tube. No tube has been made yet, so nothing on this page is measured on the aircraft. Build the airspeed tube has the procedure.

Total pressure only

TerraHawk’s tube measures total pressure only. The static pressure comes from the two flush ports in the Fuselage Front, because static holes in a tube beside the camera would read about ±26 % of airspeed wrong across the flight range (±28 % to an angle of attack of 16°), while the tube’s tip at the same station read a total pressure coefficient of 0.985 to 1.006 at every angle of attack from −5° to 16° (th-aeronautics, ed70acfb). The static figure ranks the station rather than calibrating it, but its conclusion is firm: static holes there are unusable. With static gone from the tube, nothing constrains the nose shape except one thing: how well it reads total pressure when the air does not arrive straight on.

The angles at the mouth

That angle is not small. At the mouth (Y −25.3, on the 7° tip-down axis) th-aeronautics’ CFD found incidence from −16.4° to +16.3° across an angle of attack of −5° to 14°, +19.2° at 16°, and a worst combined angle of 19.5°, also at 16° (th-aeronautics, e68dc651). In cruise it is −2° to +4°.

What NACA measured

The classic Prandtl nose, a hemisphere with a small hole, is a poor choice for that. William Gracey’s 1956 wind-tunnel survey of total-pressure tubes for NACA left hemispherical noses out of its tests on purpose: they “were not included in the present investigation because they were known to be very sensitive to angle of attack”.

The survey counts a tube as usable while it reads within 1 % of the impact pressure. Square-ended, thin-walled cylinders with an opening of 0.96 to 0.98 of the diameter stayed usable out to ±23° at Mach 0.26. The best tube without a shield was a cylinder whose opening equals the tube diameter, with a conical chamber of 30° or less behind it: about ±28° at Mach 0.26, the highest usable range of any unshielded tube in the survey. The report gives that chamber as 15° in its discussion (±27.5°) and as 30° in its summary (about ±28°); it also says the sensitivity angle stays essentially constant for cone angles below 30°, so either reading puts a 30° chamber in the best group. Cylindrical outer shapes beat every conical and ogival nose tested.

TerraHawk’s tube

The tube follows that result: an open mouth, a 30° internal cone, and a land of only 0.1 mm, so the opening is 0.95 of the diameter (the pitot-tube project). At an angle of attack of 16°, where the worst combined angle of 19.5° occurs, that leaves about 8.5° of margin to Gracey’s ±28°, against 3.5° for a plain square end. That is why the internal cone stays, and why the outer edge is only deburred: rounding it moves the nose toward the hemisphere. th-aeronautics found no loss at the station: the total pressure coefficient reads 0.988 to 0.990, the mesh’s own free-stream level.

Two smaller decisions

The tip sits at Y −25.3, 5 mm aft of the original tube’s station, for the camera’s picture. A move of 2 to 3 mm changes the flow angle by less than the CFD’s ±1°.

The collar sits behind the Cone Plate, so the tube loads and unloads from the rear with the cone glued on. It sets the tip’s station and takes the sleeve’s push, and the M2 clamp screw holds the tube against the sleeve’s pull when the cone comes off (th-blender W102).

Sources

• Gracey, W. Wind-Tunnel Investigation of a Number of Total-Pressure Tubes at High Angles of Attack: Subsonic, Transonic, and Supersonic Speeds. NACA TN 3641, Langley Aeronautical Laboratory, May 1956. NTRS 19930084398. The quotation, the 1 % criterion, the ±23°, the ±27.5° with the 15° chamber (discussion), the ±28° with the 30° chamber (summary) and the cone-angle result are checked against the report’s text.
• th-aeronautics’ probes of the tube beside the camera (ed70acfb): the static holes’ error and the tip’s total pressure coefficient of 0.985 to 1.006.
• th-aeronautics’ mouth-flow CFD (e68dc651): the incidence at the mouth, the total pressure coefficient of 0.988 to 0.990 and the margins.
• The pitot-tube project (commit 171d465): the tube’s dimensions.
• th-blender W102 (fit pass) and W104 (888383ba): the collar, the clamp screw and the installed position.