Professional Documents
Culture Documents
December 1996
The use of trademarks or names of manufacturers in this report is for
accurate reporting and does not constitute an official endorsement,
either expressed or implied, of such products or manufacturers by the
National Aeronautics and Space Administration.
Acknowledgments
This effort would not have been possible without the support and as-
sistance of the staff of the Wallops Flight Facility and the flight crew
of the Electra airplane. Their cooperation and effort in executing pre-
cise flight maneuvers and ground calibration procedures ensured the
quality and overall integrity of the TAMMS calibration. We would
also like to thank Riley S. Bull, Prentiss J. Moore, and Charles S.
West of Wallops for their cooperation and efforts in the instrumenta-
tion of the tower used for correlative ground measurements.
NASA Center for AeroSpace Information National Technical Information Service (NTIS)
800 Elkridge Landing Road 5285 Port Royal Road
Linthicum Heights, MD 21090-2934 Springfield, VA 22161-2171
(301) 621-0390 (703) 487-4650
Contents
Symbols .......................................................................... iv
Abstract ........................................................................... 1
1. Introduction ..................................................................... 1
Figures .......................................................................... 14
iii
Symbols
CL lift coefficient
P s,ref static pressure interpolated to airplane sensor height by using tower measurements, hPa
Ts,ref static air temperature interpolated to airplane sensor height by using tower
measurements, °C
vi
Abstract
V = V a + Vp (2)
_" = w'c" (4)
The transformations and actual expressions used to deter-
mine the ambient horizontal (u and v) and vertical (w) where
wind components from airplanes into the standard mete-
orologieal frame of reference have been derived and dis- t
w = w - _, (5)
cussed in detail by Axford (1968) and Lenschow (1972).
These complex equations, however, can be simplified
r
to the following approximate calculations (Lensehow c = c - _ (6)
1972):
and the bar over the symbol and prime denote the mean
u = -U a sin(_+ 13) + Up (3a)
value and the fluctuation from the mean value, respec-
tively. Sample periods are typically 10 to 20 min in dura-
v = -U,, cos(w + 13)+ vp (3b)
tion to obtain a statistically significant sampling.
2
3.1. Nose Boom and Meteorological Temperature measurements needed to determine Ua
Instrumentation are made with a Rosemount model 102 non-deiced total
air temperature Tt sensor housing with a fast-response
The nose boom is a conical composite structure fab- platinum sensing element (E102E4AL) (DeLeo and
ricated from high-modulus graphite preimpregnated with Werner 1960; Werner, DeLeo, and Rogal 1961;
an epoxy resin. It is 3.7 m in length, 25.4 cm in diameter Stickney, Shedlov, and Thompson 1990). The
at the base, and 7.6 cm in diameter at the tip. The wall E102E4AL element has a nominal 50-_ resistance and a
thickness of the cone decreases from 0.61 cm at the base 20-ms response time. The Rosemount signal conditioner
to 0.46 cm at the tip. An NACA research-type combina- was replaced with an inhouse signal conditioner designed
tion pitot-static tube is attached to the tip of the graphite to have a 0- to 10-V output. Rosemount total temperature
boom. This position places the meteorological instru- sensors normally exhibit a measured temperature Tm of
ments and airflow sensors as listed in table I at a distance -0.995T t. The sensor is mounted on the starboard side of
forward of the fuselage which significantly reduces the the pitot-static tube-boom adapter. A high correlation
influence from the airflow around the airplane. Con- between the yaw angle and total air temperature Tt mea-
structing a boom system of sufficient length that will surements observed during in-flight calibration maneu-
totally remove the influence of airflow about an airplane vers indicates a shadowing effect from the pitot-static
is not physically possible. Differential and absolute tube and/or adapter, which is discussed later. Also, due to
pressure transducers for impact and static pressure mea- the support structure for the temperature sensing ele-
surements, respectively, are housed in the pitot-static ment, the T t system actually has a two-term exponential
tube-boom mounting adapter. Fast-response temperature response (Lenschow 1972; McCarthy 1973). A numeri-
and humidity sensors are mounted to the adapter. Flow- cal filter was employed to recover the high-frequency
direction vanes are located on an NACA combination signal that is lost because of the presence of the support-
pitot-static and flow-direction probe. The overall instru- ing structure for the element (Ritter, Smith, and Cahoon
ment configuration is shown in figure 3 and weighs 1987).
approximately 31 kg. The boom is mounted at the base to
Airflow angles are determined from a pair of free-
a support frame housed within the radome which accom-
rotating balsa vanes mounted orthogonally to the instru-
modates the INS. This support frame is rigidly attached
ment head as shown in figure 3. The vanes are oriented to
through truss structures mounted to the forward fuselage
measure angle of attack and flank angle, that is, the inci-
bulkhead. A Honeywell Primus 800 weather radar
dence angles of airflow relative to the airplane in the ver-
mounts above the support frame between the truss struc-
tical and horizontal planes, respectively. Sideslip is
tures without field-of-view restrictions in the forward
assumed to be equal to flank angle for small angles of
or zenith directions. The lowest natural frequency of
attack. The vane assembly is attached to a shaft which
the overall INS, support structure, and boom system is
rotates in a hollow strut and flange machined from a
15.3 Hz, which is sufficiently beyond the desired
0-10 Hz bandwidth. solid piece of stainless steel for mounting. A bronze
sleeve beating supports the outboard end of the shaft
with a synchro repeater attached to the inboard flange
The static and differential pressures, which are criti-
end. A 16-bit synchro-to-digital converter was used to
cal parameters in determining true airspeed U a, are mea- obtain the data. As a result, airflow angles of 0.005 ° were
sured by Paroscientific Digiquartz pressure transducers,
capable of being resolved. With equation (3c), this flow
models 215-AW-002 and 5006-D-002, respectively.
angle translates into a sensitivity of 1 cm-s -1 for the ver-
These transducers employ a crystalline-quartz resonator
tical velocity component.
which measures change in pressure via their pressure-
dependent oscillation frequency. Busse (1987) provides a Richardson (1971) discusses the design of the vanes
detailed explanation of transducer mechanics and opera- and provides data on the dynamic properties as employed
tion. Application accuracy of the transducers is cited by the TAMMS on the Electra airplane at Wallops.
by the manufacturer to be 0.01 percent of full scale with Advantages of free-rotating vanes include the measure-
1 x 10 -8 resolution. Both the differential and absolute ment of the airstream angles directly; therefore, the need
pressure transducers are housed in a thermally insulated is alleviated for accurate knowledge of aerodynamic
and thermostatically controlled "cocoon." Thermal sta- coefficients for data interpretation, and the elimination of
bility of the transducers is obtained with Du Pont Kapton acceleration sensitivity by mass-balancing each vane
laminated etched-foil heating elements controlled with (Lenschow 1971; Richardson 1971). The natural fre-
an inhouse feedback circuit. The thermal environment of quency of free-rotating balsa vanes was determined by
the pressure transducers is set at a control temperature of Richardson to be approximately one-half times the
44°C + 0.5°C to maintain temperature stability during square root of the dynamic pressure measured in pascals,
normal operating conditions. which translates to 25 to 50 Hz throughout the range of
3
theElectraflightenvelope.
Thelower detectable limit of electronics, (2) a pulse code modulation (PCM) system,
the length scale for turbulent eddies would then be and (3) a data recording and display system. A block dia-
approximately 4 to 8 m with the assumption of an aver- gram of the overall system is shown in figure 4.
age airspeed of 100 m-s -1. Spectral analyses of flight
data from these vanes support these findings and indicate Analog, synchro, frequency, and ARINC 561 format
good response over the 10-Hz bandwidth of interest. data signals received from the TAMMS components are
sent, via signal conditioning and interface electronics,
A Lyman-Alpha hygrometer (Buck 1976) is to the PCM system which uses a remote multiplexer/
mounted on the boom tip adapter aft of the pitot-static demultiplexer unit (RMDU) for processing data. Synchro
tube and flow angle vanes. A slower response EG&G and analog data are recorded at 16- and 12-bit accuracies,
137 hygrometer is mounted in a window port on the fuse- respectively. Selected analog signals are filtered by using
lage and used to normalize the Lyman-Alpha signal. 8 pole, 6 zero elliptic low-pass filters. Signal cutoff fre-
quencies varied accordingly with the response character-
3.2. Inertial Navigation System istics of the respective instrumentation (e.g., EG&G 137
hygrometer, 2 Hz; total air temperature sensor, 5 Hz; ver-
The Litton Model LTN-72RH gyro-stabilized iner-
tical velocity and fast-response measurements of chemi-
tial navigation system is employed on the Electra aircraft
cal species, 10 Hz). Data are recorded at 66 Hz by a
for TAMMS. This system is the high-precision version
dedicated 386 personal computer onto a dual disk drive.
of the standard LTN-72 model. The RH model has been
The dual drive allows a full data disk to be swapped
primarily developed for scientific applications with strin-
without loss of data. A flight data recorder is also used as
gent requirements such as those needed for scientific air
a storage backup. A second 386 personal computer per-
motion measurements. It provides an increased update
forms system control, data analysis, and display tasks in
rate of approximately 24 data frames/s on both a binary real time.
and binary-coded decimal (BCD) bus in ARINC 561 for-
mat. The position error drift is approximately 0.5 km/hr
4. Analysis
(personal communication from Charles Robinson of
Litton Industries, Inc.). The INS is mounted to the base
of the boom within the Electra radome by way of an 4.1. Static Pressure Error
environmentally controlled housing.
The flow field in front of an airplane is influenced
The airplane attitude angles obtained from the INS for a relatively large distance at subsonic speeds. Far
were electrically fed through a 16-bit synchro-to-digital upstream from the airplane, where fuselage effects on the
converter to yield an angular resolution of 0.005 ° . streamlines are minimal, the total pressure is the sum of
Equation (3c) indicates that for an airplane speed of the free-stream static and dynamic pressures,
I00 m-s -l, the required angular resolution for 0 is 0.06 °
in order to meet a short-term resolution requirement of Pt = P + qc (7)
1 cm-s -1 for the vertical velocity fluctuation. The INS
The shape of the fuselage nose, primarily, determines the
used by the TAMMS platform, therefore, easily meets
distance ahead of the airplane that the flow field is
this requirement. The vertical velocity of the airplane Wp
altered. Because of the blunt nose of the Electra airplane,
is derived by integrating the vertical acceleration output
of the INS and bounding it by the third-order barometric- the streamlines of the oncoming air must diverge rapidly;
thus, differences are created between the measured and
inertial loop algorithm as suggested by Lenschow
(1986). free-stream pressures. Total pressure Pt is conserved for
any point along the aircraft body for subsonic flight as
The long-term accuracy of the horizontal velocities given by
Up and Vp are dictated by INS drift rate. A thorough dis-
cussion of inertial systems and the errors present in pt = p'+q_ (8)
the resultant velocity measurements are presented by
Broxmeyer (1964) and Kayton and Fried (1969). where the prime denotes local values. The static pressure
Lenschow (1972) gives a general discussion on the types error Ap defined as
and orders of magnitude of errors associated with inertial
systems.
Ap = p'- p (9)
3.3. Aircraft Data Acquisition System is substituted into equation (8) to get
4
Subtracting
equation
(7)fromequation
(10)gives during flight, angle of attack is usually correlated with
nondimensional position error:
qc = qc-AP (11)
A__p_p
= f(ot) (15)
The magnitude of Ap can be substantial for very qc
blunt fuselages. Wind tunnel tests have previously dem-
onstrated that the magnitude of static pressure error, as a This correlation is possible because C L and angle of
ratio of Ap/qc, is influenced by nose shape and distance attack are linearly related for small angles and a given
upstream of the nose (Letko 1947). The presence of the airplane configuration (Gracey 1980); this allows angle
airplane fuselage, therefore, causes an increase in the of attack to be substituted for CL in these conditions.
local static pressure as well as a decrease in the local Thus, from equation (14) the nondimensional static pres-
dynamic pressure. Changes in the local pressure field sure error at the static port varies with angle of attack.
may also result from airflow distortion induced by the Altering the airplane configuration also affects the flow
presence of the pitot-static tube. The determination of the field and position error. Pressure errors due to accelera-
overall static pressure error is a requisite and can be tion are relatively small and become negligible with
determined through flight calibration tests (Gracey 1980; flight leg scenarios which are primarily kept at steady
Lenschow 1986; Dommasch, Sherby, and Connolly conditions and below M = 0.5. Consideration of a much
1961). wider dynamic range of flight parameters is recognized
for the calibration of high performance aircraft and
An additional source of error in pitot-static systems
expanded range of flight conditions (Haering 1990;
arises from resonance in the lines between the pressure
Bjarke and Ehernberger 1989).
port and the pressure transducer (Iberall 1950) and,
therefore, is a function of the length and diameter of tub-
ing and air density (Nicholls 1982). In the NASA Electra 4.2. Static Air Temperature
configuration, the length and diameter of tubing are
approximately 0.75 m and 0.25 cm, respectively; thus, The static air temperature (i.e., the temperature of
the resonant frequency is ensured to be above 10 Hz. the static, undisturbed air through which the airplane is
traveling) is actually the last of four temperature values
Gracey (1958 and 1980), Etkin (1959), and that must be determined for any given point. The four
Dommasch, Sherby, and Connolly (1961) primarily temperatures in order of derivation are (!) measured
associate changes in airflow, which influences Ap/qc, temperature Tm, (2) recovery temperature T,, (3) total
with Mach number, angle of attack, and/or lift coeffi- temperature Tt, and (4) static temperature Ts. These
cient. The Mach number is given as parameters are discussed in detail by Stickney, Shedlov,
and Thompson (1990). For our purposes a short sum-
M 2 2 _(qc + 1 - mary will suffice.
= _---L-_L\p )v l"t- 1] (12)
(13)
_'= _'o(1-0.07129p) where M 1 denotes the local Mach number as determined
by local values of static and dynamic pressure. Estima-
where e is partial vapor pressure and % = 1.4028 for dry tions of the difference in local static and dynamic pres-
air. sures were made for the Rosemount total temperature
probe located 0.6 m downstream from the pressure ports.
As mentioned, Ap/qc is influenced by the lift coeffi-
The increase of the local static pressure at the Rosemount
cient CL, which is defined as
sensor was thus estimated to be approximately 10 percent
L of the free-stream dynamic pressure.
CL - -- (14)
- qc S
The recovery temperature T r is related to T t by
6
30sbeforeandafterpassage of thetower.Towerflyby the reference values, which are near measurement accu-
calibrationdataweretakenduring21passes at5 differ- racies of the instrumentation.
entspeeds. A clean,aerodynamic profilewasmaintained
(i.e.,no flaps)at all times.Passes weremadewith the 5.2. Speed Variation Maneuver
flight path offset approximately 2 to 3 wingspans
(100m)fromthetower.Datatakenontheairplanewere The speed variation maneuver was the first of sev-
averaged at 10Hztomatchtheradarsampling rate.Ref- eral in-flight maneuvers required to correct measure-
ments of flow angles due to the distortion airflow around
erence valuesforairtemperature Ts,re f and pressure Ps,ref
the airplane. The maneuver consists of flying at a con-
were determined for comparison with values obtained
stant heading and altitude while varying airspeed through
from the TAMMS platform within 3 s of passage of the
the speed range of the airplane. This maneuver modu-
tower. These reference values were based on the height
data obtained from the radar and the temperature and lates angle of attack cx and pitch angle 0 due to the rela-
tion of lift to ot and airspeed and allows the alpha vane to
pressure reference measurements made on the tower.
be directly calibrated to 0, which is obtained from the
Because pitch attitude changed with airspeed of the air-
plane during the tower flybys, the reference height INS. Because of upwash effects in front of the airplane, a
(determined from radar tracking) was geometrically cor- correction or sensitivity factor is found to be 0.75 for ¢x
rected for the difference between the vertical position of as indicated by the slope of the plot of ¢x versus 0 shown
the radar transponder and static pressure sensor. in figure 11. The sensitivity of the alpha vane is not only
affected by upwash, but a zero offset is also determined
Letko (1947) published wind tunnel data indicating to be 1.35 ° from the same data set. This offset includes
the variation of Ap/qc with x/D for a body with a blunt- any rotational misalignment of the alpha vane synchro
shaped nose (x is distance of static port in front of nose repeater caused by the nose boom configuration and
and D is fuselage diameter). Wind tunnel data shown in installation. This procedure was repeated during mis-
figure 6 indicate Ap/qc is constant during level flight for sions to compensate for shifts in offset and sensitivity of
Mach numbers within the Electra flight envelope and x/D the flow angle vanes.
ratio (O'Bryan, Danforth, and Johnston 1955). Roe
These maneuvers are also used to check the depen-
(1951 ), however, also presents data (fig. 7) that illustrate
dency of T s on airspeed. Adiabatic effects on the temper-
a slight Ap/qc relationship to angle of attack; thus, tower
ature sensor (i.e., measured temperature) vary with
flyby data for the TAMMS configuration (fig. 8) support
airspeed. The removal of these effects is done through
a relationship between Ap/qc and angle of attack.
computations which require a known recovery factor for
Static pressure error (difference between TAMMS the sensor. Errors in determining this value or the mea-
static pressure and free-stream static pressure interpo- surement of airspeed show a periodic relationship
lated from tower to airplane height), obtained from tower between Ts and airspeed; this indicates a dependence of
flybys, ranged from approximately 3 hPa at Mach num- Ts on airspeed. The time series plot of T s and airspeed
ber 0.2 to 9 hPa at Mach number 0.45. An approximate shown in figure 12, however, indicates no such relation-
value of 0.075 was determined for Ap/qc with a slight ship and reinforces the validity of airspeed and tempera-
correlation to angle of attack as indicated by the tower ture sensor calibrations.
flyby data shown in figure 8. These results are highly
correlative to the wind tunnel data in figures 6 and 7. 5.3. Skid Maneuver (Yaw or Sideslip)
Figure 9 shows the experimentally obtained values Airplane heading V, horizontal velocity components,
of the recovery factor for the temperature probe as a and sideslip [3 are modulated during the skid maneuver
function of M l. These data indicate that although a while altitude and roll are kept constant. Skid maneuvers
second-order curve is shown to describe the variation of r are necessary for in-flight calibration of the sensitivity
with airspeed, a value of 0.98 could be used over the factor and estimation of errors associated with the beta
speed range of the airplane with minimal error. Calcu- vane. Sensitivity and offset values are unique to each
lated total and static air temperatures as a function of M l flow vane installation and must be determined accord-
are shown in figure 10. The obvious correlation of air- ingly. Sideslip flight calibrations, however, cannot
speed with Tt is not present, as should be expected, for assume that the horizontal winds are negligible in con-
the calculated T s data curve. The difference between the trast to the calibration of the angle of attack when vertical
reference tower temperature and the airborne static tem- winds are assumed to be zero. Tjernstrom and Friehe
perature measurements for each pass of the tower was (1991) describe a flight calibration technique employed
determined and indicated no dependence on airspeed. for sideslip, which assumes for homogeneous low-
Although no speed dependence is noticeable, the air- turbulence conditions that all variations in the computed
borne measurements were found to be =0.3°C lower than horizontal wind are induced by the skid maneuver only.
The sensitivity factor determined under homogeneous pitch maneuver is to indicate any errors in the determina-
and low-turbulence conditions during skid maneuvers is tion of vertical winds w or, conversely, demonstrate the
assumed to be highly representative of the beta vane. The overall ability of the air motion system to remove vertical
offset value, which may include a bias error related to airplane motion and make ambient vertical wind mea-
from the INS alignment, however, is determined by surements. Lenschow (1986) suggests a period of 10 to
adjusting the offset value until the cross-track wind com- 20 s for the pitching motion with an airplane vertical
ponent is equivalent in magnitude for both directions of a velocity of 2.5 to 4 m-s -1 for this maneuver. The airplane
reversed heading maneuver. Calibration errors associated vertical velocity and ambient vertical wind determined
with the beta vane are indicated by a periodic variation in from a pitch maneuver are plotted for comparison in fig-
computed horizontal velocity during skid maneuvers. ure 15. The amplitude of airplane vertical velocity oscil-
The beta vane sensitivity factor was found to be 0.81 lation is varied between 6 and 20 m-s -1 with a 20- to 30-s
with an offset of 1.10 °. These values rendered horizontal period. The general performance criteria for an air
wind field measurements of less than 10 percent of motion system are that errors in calculated vertical winds
Ua(sin 13), which meet the criteria for satisfactory system be less than 10 percent of the airplane vertical velocity.
performance. These criteria, as illustrated, are easily met during the test
conditions.
Sideslip, similar to angle of attack, influences static
pressure measurements. A time series plot of the aircraft
6.2. Reversed Heading Maneuver
sideslip motion, measured static pressure, and corrected
static pressure for a skid maneuver is shown in figure 13. The reversed heading maneuver demonstrates the
The large sinusoidal curve indicates the sideslip angle as ability of the air motion system to measure the ambient
measured by the rotating vane. A split scale is used, for horizontal wind components u and v by modulating air-
data presentation purposes, to offset the measured static speed and sideslip angle errors. From equations (3), _,
pressure (p', upper curve) from the final static pressure Up, and Vp are obtained through the INS leaving U a and [3
values (p, lower curve) which are determined through as determined by the pitot-static tube, T1 sensor, and
calibration as a function of 13and q_. beta vane measurement; thus, the calibrations of these
sensors are critical and directly influence horizontal wind
Data from sideslip maneuvers indicate an asymmet-
measurements.
ric relationship of sideslip with calculated static air tem-
perature T_. A time history plot of sideslip, T_, and Reversed headings are performed by flying at a con-
corrected Ts is presented (in similar fashion as the previ- stant altitude and heading for approximately 3 min and
ous static pressure data set) in figure 14. The variation in then reversing the course in a manner in which the return
T_ was removed for positive sideslip angles by using the course passes through the same air mass. These maneu-
corrected static pressure to determine M I values. This vers are performed above the boundary layer in a homo-
approach, however, was not sufficient to remove the geneous air mass to ensure the sampling of the same
larger variations in T_ during negative sideslip angles. meteorological conditions as realistically possible. A
The nonsymmetrical relation of T s with sideslip is time history plot of along-track winds during a reversed
attributed to the nose boom configuration (fig. 3), which heading maneuver performed along the prevailing wind
shows the total temperature probe mounted horizontally direction is shown in figure 16(a) and demonstrates the
on the starboard side of the boom. This position produces ability of the system to make accurate, true airspeed mea-
a shielding effect during a negative sideslip angle. An surements. A reversed heading is immediately performed
empirical multiplication factor was determined and adjacent to the first and perpendicular to the prevailing
applied to previously determined static pressure correc- wind. Cross-track winds obtained in these conditions are
tions to yield local static pressure conditions and subse- shown in figure 16(b) and are similar, as expected, to the
quently M t encountered by the total temperature probe along-track winds. This similarity indicates that the sum
during negative sideslip. The resultant Ts with the major- of sideslip and heading has been adequately determined.
ity of the asymmetric effects from shielding removed is Horizontal wind, obtained by vector addition of along-
indicated by the lower curve in figure 14. and cross-track winds from a previous flight track is
shown in figure 16(c).
6. Flight Validation Maneuvers
7. Concluding Remarks
6.1. Pitch Maneuver
The calibration of a turbulent air motion system from
Vertical airplane velocity, airspeed, and angle of in-flight and tower flyby maneuvers using a tracking
attack are modulated during a pitch maneuver, whereas C-band radar has been discussed. A description of the
roll and sideslip are kept constant. The purpose of the gust probe measurement technique and instrument
8
package, flight calibration procedures, and results were and flight data showed a high correlation of position
included. The primary goals that have been addressed error to impact pressure with a slight dependence on
consist of: (1) defining any dependence of static air tem- angle of attack and/or lift coefficient C L. Because of the
perature to airspeed, (2) determining systematic static difficulty in obtaining a reasonable approximation of air-
pressure position error as related to platform configura- plane weight during flight and the linear relationship of
tion, and (3) demonstrating the overall system consis- angle of attack with C L for small angles, angle of attack
tency to make accurate three-dimensional ambient wind was substituted for CL. As a result, static pressure error
field measurements. was determined to be defined adequately as a function of
impact pressure and angle of attack. Pitch, skid, and
The total air temperature probe recovery factor of reversed heading maneuvers demonstrated the overall
0.98 was determined experimentally through analysis of ability of the air motion system to remove aircraft motion
the tower flyby calibration data. Comparison of airplane- and make ambient vertical and horizontal wind measure-
derived static air temperature Ts with tower air tempera- ments. The general rule applied to airborne platforms
ture data and speed variation maneuvers performed used to make air motion measurements states that the cal-
remotely from the tower illustrated the lack of depen- culated vertical winds should be <10 percent of the plat-
dence between Ts and airspeed; consequently, the form vertical velocity and calculated horizontal winds
calibration of the air temperature probe was verified. should be <10 percent of Ua(sin _) (where U a is true air-
Further flight tests (skid maneuvers), however, indicated speed and 13 is sideslip angle). These criteria have been
a shadowing effect on the total air temperature probe met and provide ample confidence in the system to
induced by the pitot-static tube mounting configuration provide reliable measurements of the ambient three-
on the nose boom. This shadow effect was successfully component wind field during straight and level flight
removed through the application of a numerical filter to conditions.
the air temperature data.
9
Appendix
The Lockheed 188 Electra airplane based at Wallops Flight Facility and designated NASA 429 is a low-wing,
medium altitude, moderate range transport airplane powered by four Allison turboprop engines.The specifications of the
airplane are given as follows:
Wing:
Area, m 2 ................................................................................. 120.8
Span, m .................................................................................. 30.2
Fuselage:
Diameter, m ............................................................................... 3.45
Overall length (excluding boom and pitot-static tube), m ........................................... 32.25
Weight:
Takeoff gross weight, kg ................................................................... 52 730
Payload capacity, kg ...................................................................... 10 450
Maximum fuel loading, kg ................................................................. 15 450
Performance:
True airspeed, m-s -1 ..................................................................... 90-160
Flight endurance at 155 m-s -l, hr ............................................................... 5.5
Maximum operating altitude, km ................................................................ 7
Cruising altitude, km ....................................................................... 4.5-7
10
References cal Observations and Instrumentation of the American Meteo-
rological Society, pp. 426---432.
Anon. 1971" Flight Test Procedures for Static Pressure Systems
Installed in Subsonic Transport Aircraft. SAE ARP 921. Gracey, William 1958: Measurement of Static Pressure on Air-
craft. NACA Rep. 1364.
Anon. 1991a: Rosemount ® Model 858 Flow Angle Sensors.
Bul. 1014, Rosemount Inc. Gracey, William 1980: Measurement of Aircraft Speed and Alti-
tude. NASA RP-1046.
Anon. 1991b: Wind Measurement. MRF Instrument Note 10,
British R.A.E. Greenhut, G. G.; and Gilmer, R. O. 1985: Calibration and Accu-
racy of the NOAA/ERL Gust Probe System and lntercompari-
Armistead, Katherine H.; and Webb, Lonnie Dean 1973: Flight son With Other Systems. NOAA-TM-ERL-ESG-22.
Calibration Tests of a Nose-Boom Mounted Fixed Hemispheri-
cal Flow Direction Sensor. NASA TN D-7461. Haering, Edward A., Jr. 1990: Airdata Calibration of a High-
Performance Aircraft for Measuring Atmospheric Wind Pro-
Axford, D. N. 1968: On the Accuracy of Wind Measurements files. NASA TM-101714.
Using an Inertial Platform in an Aircraft, and an Example of a
Measurement of the Vertical Mesostructure of the Atmosphere. Iberall, Arthur S. 1950: Attenuation of Oscillatory Pressures in
J. Appl. Meteorol., vol. 7, pp. 645-666. Instrument Lines. J. Res. Natl. Bur. Stand., vol. 45, no. 1,
pp. 85-106.
Bjarke, Lisa J.; and Ehernberger, L. J. 1989: An In-Flight Tech-
nique for Wind Measurement in Support of the Space Shuttle Kayton, Myron; and Fried, Walter R., eds. 1969: Avionics Naviga-
Program. NASA TM-4154. tion Systems. John Wiley & Sons.
Brown, E. N.; Friehe, C. A.; and Lenschow, D. H. 1983: The Larson, Terry J.; Flechner, Stuart G.; and Siemers, Paul M., I11
Use of Pressure Fluctuations on the Nose of an Aircraft for 1980: Wind Tunnel Investigation of an All Flash Orifice Air
Measuring Air Motion. 3'. Climat. & Appl. Meteorol., vol. 22, Data System for a Large Subsonic Aircraft. NASA TP- 1642.
pp. 171-180.
Lemone, Margaret A.; and Pennell, William T. 1980: A Compari-
Brown, E. N. 1985: Calibration of the NCAR (National Center for son of Turbulence Measurements From Aircraft. J. Appl.
Atmospheric Research) Sabreliner Research Static Pressure Meteorol., vol. 19, pp. 1420-1437.
DeLeo, Richard V.; and Werner, Frank D. 1960: Temperature Lenschow, Donald H., ed. 1986: Probing the Atmospheric Bound-
Sensing From Aircraft With Immersion Sensors. Preprint No. ary Layer. American Meteorol. Soc.
NY60-91, Fall Instrument-Automation Conference and
Letko, William 1947: Investigation of the Fuselage Interference
Exhibit, Instr. Soc. America, pp. 91-NY60-1-91-NY60-10.
on a Pitot-Static Tube Extending Forward From the Nose of
DeLeo, Richard V.; and Hagen, Floyd W. 1966: Flight Calibration the Fuselage. NACA TN 1496.
of Aircraft Static Pressure Systems--Final Report. SRDS Rep.
List, Robert J. 1971: Smithsonian Meteorological Tables,
No. RD-66-3, REC Rep. 76431, FAA.
Smithsonian Inst.
Dommasch, Daniel O.; Sherby, Sydney S.; and Connolly, Thomas
McCarthy, J. 1973: A Method for Correcting Airborne Tempera-
F. 1961: Airplane Aerodynamics, Third ed., Pitman Publ. Corp.
ture Data for Sensor Response Time. J. Appl. Meteorol.,
Etkin, Bernard 1959: Dynamics of Flight. John Wiley & Sons. vol. 12, pp. 21 !-214.
Frost, Walter; Paige, Terry S.; and Nelius, Andrew E. 1991: Guide Meissner, Charles W., Jr. 1976: A Flight Instrumentation System
to Measurement of Winds With Instrumented Aircraft. NASA for Acquisition of Atmospheric Turbulence Data. NASA TN
CR-188601. D-8314.
Gilmer, R. O.; McGavin, R. E.; and Reinking, R. F. 1978: A Small Nicholls, Stephen 1982: An Observational Study of the Mid-
Aircraft Gust-Probe System for Studies of Boundary Layer Latitude, Marine, Atmospheric Boundary Layer. Ph.D. Diss.,
Convection and Transport. Fourth Symposium on Meteorologi- Southampton Univ.
11
O'Bryan,
Thomas
C.;Danforth,
Edward
C. B.; and Johnston, Boundary Layer Flux Measurements of Trace Species Over
J. Ford 1955: Error in Airspeed Measurement Due to the Canadian Boreal Forest and Northern Wetland Regions. J.
Static-Pressure Field Ahead of an Airplane at Transonic Geophys. Res., vol. 99, no. DI, pp. 1671-1685.
Speeds. NACA Rep. 1239.
Roe, M. 1951: Position Error Calibration of Three Airspeed Sys-
Richardson, Norman R. 1971: Dynamic and Static Wind-Tunnel tems on the F-86A Airplane Through the Transonic Speed
Tests of a Flow-Direction Vane. NASA TN D-6193. Range and in Maneuvering Flight. Rep. No. NA-51-864, North
American Aviation, Inc.
Ritter, J.; Beck, S.; Hedgepeth, R.; Barrick, J.; and Butler, C. 1987:
The Development of an Air Motion Measurement System for Scott, Stan G.; Bui, T. Paul; Chan, K. Roland; and Bowen, Stuart
NASA's Electra Aircraft. Sixth Symposium on Meteorological W. 1990: The Meteorological Measurement System on the
Observations and Instrumentation of the American Meteoro- NASA ER-2 Aircraft. J. Atmos. & Ocean. Tech., vol. 7,
logical Society, pp. 140-143. pp. 525--540.
Ritter, John A.; Smith, G. Louis, and Cahoon, Donald R. 1987: Shrager, J. J. 1964: Calibrating Static Pressure Systems at Low
The Use of a Numerical Filter to Correct Airborne Temperature Altitudes--Final Report. RD-64-37, FAA.
Measurements for the Effects of Sensor Lag. Sixth Symposium
Stickney, Truman M.; Shedlov, Marvin W.; and Thompson,
on Meteorological Observations and Instrumentation of the
Donald I. 1990: Rosemoant Total Temperature Sensors. Tech.
American Meteorological Society, pp. 261-264.
Rep. 5755, Rev. B, Rosemount, Inc.
Ritter, John A.; Barrick, John D. W.; Gregory, Gerald L.; Sachse,
Telford, J. W.; and Wagner, P. B. 1974: The Measurement of Hor-
Glen W.; Hill, Gerald F.; and Lenschow, Donald H. 1990: Air-
izontal Air Motion Near Clouds From Aircraft. J. Atmos. Sci.,
borne Flux Measurements and Budget Estimates of Trace
vol. 31, pp. 2066-2080.
Species Over the Amazon Basin During the GTE/ABLE 2B
Expedition. J. Geophys. Res., vol. 95, pp. 16875-16886. Telford, J. W.; Wagner, P. B.; and Vaziri, A. 1977: The Measure-
ment of Air Motion From Aircraft. J. Appl. Meteorol., vol. 16,
Ritter, John A.; Barrick, John D. W.; Sachse, Glen W.; Gregory,
pp. 156-166.
Gerald L.; Woerner, Mary A.; Watson, Catherine E.; Hill,
Gerald F.; and Collins, James E., Jr. 1992: Airborne Flux Tjernstrom, Michael; and Friehe, Carl A. 1991: Analysis of
Measurements of Trace Species in an Arctic Boundary Layer. a Radome Air-Motion System on a Twin-Jet Aircraft for
J. Geophys. Res., vol. 97, no. D15, pp. 16601-16625. Boundary-Layer Research. J. Atmos. & Ocean. Tech., vol. 8,
pp. 19--40.
Ritter, John A.; Barrick, John D. W.; Watson, Catherine E.;
Sachse, Glen W.; Gregory, Gerald L.; Anderson, Bruce E.; Werner, F. D.; DeLeo, R. V.; and Rogal, B. 1961: Total Tempera-
Woerner, Mary A.; and Collins, James E., Jr. 1994: Airborne ture Probes. Flight.
12
Table I. TAMMS Sensor Characteristics
13
/
I
North /
z/ Xaxe
,Y'
J East
A ,/
, /
North r--.-.._ _¢
i / _stream
r \. .J
I X-Y plane / 0
Figure 1. Coordinate system, airplane attitude angles, and reference axes used to calculate air velocity components.
14
WI-84-368-6
Figure 2. Lockheed 188 Electra airplane (NASA 429) at Wallops Flight Facility.
15
Instruments & I Signal conditioning Pulse code modulation system Data recording
data input I & interface & display system
devices I electronics
1
I
Inertial I
navigation I ARINC 561
unit : -- Interface
I
t
-_[ Signal
box
I
I : _ demultiplexer unit
I Remote (RMDU)
multiplexer/[
Nose boom }-- I
I Data formatting & ]
I storage computer I
1
I
' I
I
I
I
I
_ Frequency-
to-digital
converter
r
I
I
16
91.5 -- T, ref' Ps, ref P"qc' T's
Az
76.2 T1
C-band
radar
Tower at
Wallops
'r
3483 JJ .-I
Figure 5. Illustration of tower flyby test setup. Dimensions are in meters.
17
.20 -
.16
.12
x/D
_Lplqc 0.75
.08
1.00
.04
1.50
2.00
ii
*l
, 1 , I , I , 1 , I',,' , I
0 .2 .4 .6 .8 1.0 1.2
Mach number
Figure 6. Wind tunnel data illustrating variation of static pressure error as influenced by Mach number at different
distances ahead of aircraft fuselage (O'Bryan, Danforth, and Johnston 1955).
t D = 16.5 cm
__!__
_, deg
.10 0
\ \ ',\ - .....
', " ',\ ----- 20
.08 - \,, "\ ". _ .... 30
.06 -
Aplq c \ "\'-.
.04 \
\,
.02 -
Figure 7. Wind tunnel data showing effect of angle of attack on static pressure error forward of airplane with blunt-
shaped nose similar to Electra airplane. (Data from Roe 1951 .)
18
.080
.078
.076
_/q_ .074
.072
.070 D ÷
.068
I I I I I I I
-3.0 -2.0 -1.0 0 i .0 2.0 3.0 4.0
Figure 8. Static pressure error determined from tower flyby test data as function of angle of attack.
.99 -
.98
Recovery
factor
.97
Ml
Figure 9. Recovery factor for TAMMS total air temperature probe as function of Mach number.
19
314
F
310 --
306
r,
Air
temperature,
oC
302 '_ Ts
298 - .
294 I I I I [ I I I I ] I I I I [
.2 .3 .4 .5
MI
Figure 10. Tower flyby data indicating dependent nature of total air temperature and required isolation of static air
temperature on Mach number.
21)
2
Angle of attack,
or, deg
-4 I
-2 0 2 4 6
Pitch angle, 0, deg
Figure 11. Calibration results for angle of attack vane from speed variation maneuver.
_150
u 140
a
130
120
True airspeed,
m/s
110
100
90
80
I I I I
120 240 360
Time, s
Figure 12. Time series plot of static air temperature during speed variation maneuver.
21
8
/ 812
810
Static pressure,
p',mb
Sideslip
angle,
_,deg
8O6
810
808
Static pressure,
8O8 p, mb
I , J , J I , J J , I
0 10 20
Time, s
Figure 13. Time series plot of sideslip, measured static pressure p', and corrected static pressure p obtained during skid
maneuver.
8F
Static air
temperature,
Tj, mb
Sideslip angle,
13, deg 0
-2
Static air
-4
temperature,
T., mb
-6
-8 I
0 10
Time, s
Figure 14. Time series plot of sideslip, calculated static air temperature T_, and corrected static air temperature T s
obtained during skid maneuver.
22
20
10 _ 10
Md
Vertical
airplane w /_ Vertical wind
velocity
component,
0 0 velocity component,
Wp, m-s -1
-10
' V
/
V
-20
0 50 100 150 200
Time, s
Figure 15. Time series plot of aircraft vertical velocity and vertical wind component as determined by TAMMS during
pitch maneuver.
23
Along-track Cross-track
wind, 0 wind, 0
m_s-I m_s-I
i I i I J I I i I I I I
100 200 300 400 500 100 200 300 400 500
Time, s Time, s
10
Horizontal
wind, 5
m_s-I
J
i t I i I J I i I t I
0 100 200 300 400 500
Time, s
24
Form Approved
REPORT DOCUMENTATION PAGE OM8 No. 0704-0188
Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources,
gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden eslimate or any other aspect of this
collection of information, including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson
Davis Highway, Suite 1204, Arlington, VA 22202-4302, and to the Office of Management and Budget, Paper_vofk Reduction Project (0704-0188), Washington, DC 20503.
1. AGENCY USE ONLY (Leave blank) ,2. REPORT DATE 3. REPORT TYPE AND DATES COVERED
6. AUTHOR(S)
Unclassified-Unlimited
Subject Category 01
Availability: NASA CASI (301) 621-0390
A03
17. SECURITY CLASSIFICATION 18. SECURITY CLASSIFICATION 19. SECURITY CLASSIFICATION 20. LIMITATION