In the past, the problems of obtaining accurate measurements from air data sensors or probes during very high angles of attack have been recognized, and different approaches have been used for obtaining correction factors. For example, U. S. Patent No.

5,205,169 provides for inverting the ratio of pressure differentials from sensing ports at high angles of attack, generally above 45°. In Patent 5,205,169 the normal angle of attack ratio (Pl-P2)/qcm) is inverted to be qcm/ (Pl-P2) when the denominator qc approaches zero.

Other methods have been used, such as an iterative calculation of measured pressures which requires substantial computer capacity for carrying out accurate analysis of the measured pressures.

SUMMARY OF THE INVENTION The present invention relates to a method for correcting measured pressures at high angles of attack, by determining Mach number using an alternate pressure signal and providing in the computation, derived correction factors from a calibrated"look-up"table in an air data computer. The calibration data or factors are either directly obtained from wind tunnel tests of

the sensor probe configuration in actual operational use, or from computer simulations of such sensors at various Mach numbers and angles of attack.

The calibration values are used in a conventional manner for determining accurate angle of attack, static pressure, and pitot pressure from the measured pressures. The pressures that are measured at the ports are correctable up to angles of attack of in the range of 65°.

The alternate pressure signal is obtained from an angle of attack measuring port located on the sensing probe at a location where pressure increases as angle of attack increases, and the measured alternate pressure is used when the pressure from one angle sensing port of an angle of attack sensor is larger than the measured pitot pressure. The conventional local flow calibration approach is used when the pressure of the angle measuring ports is smaller than the measured pitot pressure.

The method of the present invention is applicable at angles of attack of between about 30° and 65°. Different shaping of the probe could extend the use of the present invention at angles up to about 90°.

The calibration values are provided for measured pressures indicating Mach numbers at angles of attack above about 35°. The calibration values can be easily placed in memory of an onboard computer to derive accurate readings of angle of attack, static pressure, and pitot pressure from the measure pressures.

The method is used in general with a cylindrical probe having angle sensing ports on the top and bottom of the probe, but it also is useful for flush mounted probes having angle of attack sensing ports that are sensitive to a change in angle of attack, in

conjunction with a forward facing (pitot) pressure sensing port.

BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 is a schematic representation of a typical sensor in the form of a probe used on an aircraft, having sensing ports and instrumentation that can be used with the methods of the present invention; Figure 2 is a fragmentary enlarged longitudinal sectional view of a forward end of the probe of Figure 1; Figure 3 is a graphic representation of information stored in a look-up table from actual tests including factor P1/Pm for different Mach numbers plotted as a pressure quantity versus probe angle of attack above 30° where Pi is the higher pressure at the angle sensing ports (these calibration values are used to determine angle of attack); Figure 4 is a graph of corrected plots of a pressure quantity Pe/Pm versus angle of attack of a probe and stored in a look-up table (these calibration values are used to determine static pressure); Figure 5 is a graph of corrected plots of a pressure quantity Pte/P1 for use in obtaining pitot pressure plotted versus angle of attack and stored in a look-up table (these calibration values are used to determine total pressure).

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A strut mounted air data sensing probe illustrated generally at 10 is used to obtain pressure measurements for determining angle of attack pitot pressure and from the sensed pressure. A support strut 11 is attached to a mounting base 12, that is generally fastened onto the fuselage 13 of an aircraft or air vehicle. Probe 10 includes a probe barrel 14 that has

a longitudinal axis 16 that is oriented in a selected relationship with respect to a reference axis of the fuselage 13 of the aircraft. A forward tapered end 15 is provided with top and bottom angle of attack pressure ports shown at 18 and 20 and a pitot port 22 which is shown at the leading end of the probe barrel and has a center axis coincidental with the barrel axis 16. The barrel axis is a center line between the ports 18 and 20.

The pressure ports 18 and 20 are used for determining the angle of attack or angle of local flow relative to the reference line which coincides with axis 16 which is a reference axis or line of the air vehicle or aircraft. When flow direction is represented at line <BR> <BR> <BR> <BR> 24, angle of attack is indicated as a in Figures 1 and 2.

At low angles of attack or at low angles a, the standard equation for calculating the angle of attack works well. These compensation techniques are disclosed, for example, in U. S. Patent No. 4,096,744 as well as U. S. Patent No. 5,205,169. In determining angles of attack for positive angles, the bottom pressure port 18 is considered to have a pressure P, provided along a pressure carrying line 23 and the upper or top pressure port 20 that is sensitive to angle of attack is considered to have a pressure P2 provided along pressure carrying line 28.

The pitot pressure is P, and provided along a pressure carrying line 25. Additional designations are used with P to indicate whether it is measured pressure (Ptm) or local (Pte) pressure. The pressures are provided to suitable pressure sensors 45,46 and 47 that provide electrical signals to an air data computer 33.

The air data computer then provides corrected outputs

along lines 34 to instrument displays and controls shown generally at 35, that display measured parameters such as angle of attack, Mach numbers, altitude and the like.

The computer outputs can be used in fly by wire aircraft controls.

In a flush mounted sensor the angle sensing ports would have an axis in a plane in which an angle of attack is to be measured. As angle of attack in the plane changes, the ports would sense different pressures, to determine angle.

The basic equation for determining angle of attack is (P1-P2)/qcm. Pl is pressure measured by the lower angle of attack pressure port 18, and P2 is the pressure measured by the upper angle of attack pressure port 20 when positive angles of attack are measured. q,-is the impact pressure, which is the measured pitot pressure Pt'm minus the static pressure (Pt, m-Pm) Pm is (P1 + P2)/2. The equation (Pm--Pe)/qce, provides local static pressure Pe. The static pressure ratio for PI is equal to (P1-Pe)/qce-Finally, the static pressure ratio for P2, the upper port, is (P2-Pe)/qCe In these ratios, the quantities are defined as follows: Where P1 = Static pressure measured by the lower static pressure port P2 Static pressure measured by the upper static pressure port Pt'm = Probe measured pitot pressure Pm = (P1 + P2)/2 derived measured static pressure <BR> <BR> <BR> <BR> <BR> <BR> qcm = (Pt'm - Pm)<BR> <BR> <BR> <BR> <BR> <BR> <BR> <BR> Pe = Local static pressure Pt'l = Local pitot pressure qc< = (Pf ?- P ?)

Me = Local Mach number aprobe = Probe or sensor angle of attack Studies show that at high positive angles of <BR> <BR> <BR> attack using measured pitot pressure (Pt, m) in the pitot error relationship (Px-Pt, e)/qcl, results in greater error in the measured angle of attack than when P1 is used as Px.

In U. S. Patent No. 5,205,169, where the ratio of pressure differentials was inverted when angle of attack was greater than approximately a 45° angle of attack with the type of probe shown in Patent No.

5,205,169, there still was a need for providing substantial mathematical correction. The prior art system shown in Patent No. 5,205,169 used the measured pitot pressure divided by the average of the pressures <BR> <BR> <BR> at the angle measuring ports, or Put, mu [ (P1 + P2)/2] for measured Mach number.

It has been discovered that using the pressures measured from the ports arranged as shown in Figure 1, it is possible to obtain accurate angle of attack, local static pressure and corrected local pitot pressure simply. The mathematics that are necessary for correction are substantially reduced, and high accuracy is achieved by substituting the sensed pressure PI at the lower port (for positive angles of attack) for P,, (the pitot pressure) and dividing P, by the measured <BR> <BR> <BR> static pressure (P, = (P + P2)/2) to determine measured Mach number for calibration information stored in an onboard computer after probe calibration in the wind tunnel.

Summarized below is the method of the present invention for correcting measured pressures at high angles of attack: When Pi < Pt, m use the prior art local flow calibration approach, where the ratio Ptm/Pl was used for obtaining measured Mach number.

When Pi > Pt, m use the following calibration approach: <BR> <BR> <BR> Define aprobe by<BR> <BR> <BR> <BR> <BR> aprobe = f (P1/Pm), (Pl Pt, m)/(P1 Pm)} Define Pl by (P,/PJ=f[a,(P,/PJ} Define Pte/Pm by (P,,,/P,)=f{c<pe.(Pi/PJ} The present method uses measured pressures to determine Mach number and convert directly to corrected local pitot pressure, local static pressure and probe angle of attack. A key feature of the method is the interchanging of P, and Plm (measured pitot pressure) at high angles of attack to determine the local pitot pressure, Pt, e and static pressure. All the calibration values for the present invention are determined using available probe calibration data from wind tunnel tests, but using P1 instead of Pt, for the Mach number for correlating the calibration data. A parallel path using this method can be developed to convert from specific measured probe pressures to true values of P (static pressure) and Pt (pitot pressure) and aircraft angle of attack. This case requires knowing the flow field on the aircraft at the probe mounting location. Extension of the method to solve for true conditions is straight forward.

Calibration data for the present invention are: Figure 3 (P1-Pt, m)/ (P1-Pm) for lines (Angle of Attack Signal) of constant (Pl/Pm) as a function of sensor angle of attack.

Figure 4 (Pe/Pm) for lines of (Static Pressure constant sensor angle of Calibration) attack as a function of (Pl/Pm) Figure 5 (Pt/P1) for lines of (Pitot Pressure constant sensor angle of Calibration) attack as a function of (Pl/Pm).

Example calibration values for obtaining angle of attack when the pressure ratio on the vertical scale is sensed and calculated from measured pressures are shown in Figure 3. The signal shows uniform variations with increasing probe angles of attack to about 70° at lower Mach numbers as determined by P1/Pm. A range up to a probe angle of about 64° shows the best accuracy at higher Mach numbers. Reduced sensitivity of the angle of attack signal occurs at the highest Mach numbers.

Static pressure and pitot pressure calibration curves in Figure 4 and 5 show smooth uniform variations with angle of attack and P1/Pm. It is recognized the use of different probe shapes or additional multiple ports could benefit this signal.

Again, the data in Figures 3,4 and 5 are from ports on a strut mounted probe, but the ports could be flush ports on the nose or side of an air vehicle, rocket or the like.

In Figure 3, the results of wind tunnel tests that place the probe at different angles of attack, which can be measured, and at different known Mach numbers are plotted for constant measured values of

Pi/Pni on the test probe or sensor. Pi/Pn, as measured and determined in flight provides measured Mach number.

Figure 3 is for angles of 30° through 90°, which are considered high angles of attack and using P1/Pm for Mach number rather than Pt, m/Pm. The measured values of P1/Pm for the listed Mach numbers are also shown in Figure 3. By measuring the pressures (P1, P2 and Pt, m) and calculating the ratio shown on the Y axis in Figure 3, namely (P1-Pt, m)/(P1-Pm) l the angle of attack value can be directly obtained using an air data computer to interpolate the signal. This gives a value for the ratio using measured pressures, while measured P1/Pm give the measured Mach number. The look-up table 36 (Figure 1) having the correct angle of attack correlated to the Y axis ratio of measure pressures at measured <BR> <BR> <BR> <BR> Mach number (P :/Pm) is stored in onboard computer 33, so the correct angle of attack is obtained. For example, <BR> <BR> <BR> <BR> at measured Mach. 45, (as determined by using Pl/Pm) if the value of the ratio (P1-Ptm)/(P1-Pm) using measured pressures is 2, the angle of attack would be slightly over 60°. The accuracy is achieved by providing the <BR> <BR> <BR> <BR> Pl/Pm ratio, which at Mach. 45 is 1.15, since Pi becomes a stronger signal as the Pt, signal decreases when angle of attack increases. In the prior art, and at lower angles of attack, the value used for measured Mach number is Ptm/Pm. Additional iterative calculations are necessary and a calculation based on those measured pressures is not as accurate or as direct as the disclosed method when P > P m.

Once the angle of attack of the probe has been determined, the corrected static pressure, which can be called P, (local static pressure), or true static pressure, is obtained from the results of wind tunnel tests illustrated in Figure 4.

The predetermined relationships are in the look-up table 36 in the memory of the onboard computer 33. The values in the table 36 are shown in the graph <BR> <BR> <BR> <BR> of Figure 4. The probe angle of attack, aprobe t iS plotted against the values of the ratio P/Pm obtained by calibration in the wind tunnel or by computer simulation and stored in the look-up table.

When the angle of attack has been determined, Pm can be obtained by the quantity (P1 + P2)/2 and the local static pressure Pe can be easily solved in the onboard air data computer 33 using look-up table values.

To determine the true pitot pressure, as shown <BR> <BR> <BR> <BR> in Figure 5, a plot of probe angle of attack, arobes versus the quantity Ptse/P1 as corrected is used. This is done for several different Mach numbers in a wind tunnel or by computer simulation, as indicated. The Mach number graph plots are represented using the same angle of attack determined from Figure 3 data. The angles of attack shown in Figure 5 are between 30° and 90°, and the plotted values obtained from wind tunnel tests or simulation are stored in a look-up 36 table in the air data computer so that the particular pressure on the vertical axis can be determined by knowing the angle of attack, which in turn is determined as previously disclosed.

With a measured P1 pressure, and a known angle of attack, the local pitot pressure Pt, can be easily derived or solved. It will equal the value on the vertical axis for a particular angle of attack, times PI. This also can be considered to be the true pitot pressure. The look-up tables provide the information that is necessary for obtaining true pitot pressure, true or local static pressure, and angle of attack. The

pressures used can be directly measured by the three ports on the probe shown in Figures 1 and 2.

Accuracy, ease of operation, and a reduction in the amount of calculations that are necessary to obtain correct parameters is achieved using the method of the present invention.

The measured Mach number is determined by directly measuring P1 and P and arriving at the ratio P1/Pm, when P1 > Pt, m which in turn is correlated to angle of attack as shown in Figure 3. Determining measured Mach number by providing directly measured pressures P1/P as opposed to the previously used Pt,/Pm, enhances operations, and does not require the inverted ratio to be used. It also can be seen that the measurements are more direct, and more easily obtained.

The present invention reduces the instability of prior art algorithms used at high angles of attack, generally at probe angles above about 35°. It can be seen that the angle of attack curves that are developed and correlated to the quantity P1/Pm do not have instability across the range of operation between 30° and 90° probe angle of attack. Further, direct pressure measurements are used with the present invention, as opposed to the iterative correction process necessary for mathematical calculations of the local pressures or angle of attack using conventional methods.

It should be noted that the term port used for the angle sensing ports means not only a single opening, but also two or more openings on the bottom or top of the probe or sensor that are plumbed together and provide a signal which indicates a pressure changing with angle of attack.

Although the present invention has been described with reference to preferred embodiments, workers skilled in the art will recognize that changes may be made in form and detail without departing from the spirit and scope of the invention.