Performance Data for Beechcraft T-34 Mentor

of these studies used ground-based, fixed (or in a relatively few cases, moving-base) trainers or simulators, in which little or no acceleration forces were present. The few studies which did use actual aircraft involved high-performance military planes in which "smooth arc" tracking tasks were performed by highly experienced military fighter pilots and test pilots. Unfortu- nately, these provided little indication of what the results might be during routine IFR condi- tions with ordinary civilian pilots. This is prob- ably one of the major reasons why little serious thought has been given by operational personnel to switching over to the use of the moving- airplane instrument. Despite the weight of re- search evidence favoring its adoption, it seems that many operational people still feel the va- lidity of results from ground-based simulators and training experiments has not been sufficiently established for situations in which physical ac- celeration cues are bound to be important. Per- haps they also feel the results of the few flight experiments conducted in the past have been too unrealistic in task requirements to be useful in making such a far-reaching decision. An inter- esting exception, however, is that the USAF Development Engineering Inspection Board for the North American F-108 long-range inter- ceptor unanimously decided to adopt the mov- ing-airplane steering display for the F-108 air- plane. Unfortunately, the F-108 program was cancelled too soon for operational experience to be gained on the use of this "new" display concept. In general, previous research, involving little or no acceleration forces, showed that: 1. Low-time pilots and non-pilots responded more rapidly and more often correctly to the moving-aircraft instrument. 2. Positive transfer for all pilots and non- pilots was greater when switching from the moving-horizon to the moving aircraft instru- ment. 3. All pilots and non-pilots demonstrated fewer reversals with the moving-aircraft instru- ment. 2 4. Low-time pilots and non-pilots subjectively "felt" that the moving-airplane instrument was more "natural" and "easier to interpret." 5. Initially, experienced high-time pilots sub- jectively were more "at ease" with the moving- horizon instrument. Concerning these findings, however, Johnson and Roscoe (1970) pointed out "it is essential that certain critical experiments be conducted in- flight to eliminate the possibility of drawing spurious conclusions from a simulated flight en- vironment. Both the speed of learning by rela- tively inexperienced pilots and the ease of transition of highly experienced and currently proficient pilots must be measured"; they also emphasized that "flight tasks must be opera- tionally realistic and representatively difficult and stressful." In order to examine the problem in an environ- ment representative of several general aviation instrument flying situations-in which accelera- tion loads and some form of psychological flight stress would be present-an FAA-CAMI in-flight study was initiated. Designed to measure pilot performance while using the moving-airplane- symbol instrument and the moving-horizon-bar instrument, the in-flight study-utilizing a gen- eral aviation type aircraft-included the follow- ing maneuvers: (1) recoveries to level flight from shallow and steep turns; (2) performing a series of left and right turns while maintaining a level pitch attitude; and (3) maintaining a given airspeed while performing a series of spiraling descents. Subjectively, these tasks appeared to be highly stressful to many of the subjects and, in fact, a few subjects lost control of the aircraft and gave up after unintentionally getting into "graveyard" spirals some involving more than 5g normal acceleration. Because of this, as well as on the basis of the other results of this in-flight study, it is suggested that the data in this report may be viewed as being reasonably representative of pilot performance in certain "real life" instru- ment flying environments. However, these facts should be noted: (1) the attitude indicators used in the study were not exact duplicates in "face- format" (the aircraft symbol in the moving- aircraft instrument [A, Fig. 2] was smaller than the one in the moving-horizon instrument [B, Fig. 2]; (2) roll and pitch indices were not www 200 B www PULL CAGE FIRURE 2. A CAUTION CAGE DIRECTIONAL GYRO FOR AEROBATICS Outside-in (moving-aircraft) attitude indicator (left) utilized fixed-horizon and aircraft symbol that moved both vertically and rotationally for pitch and bank information. To provide space for these move- ments required use of a smaller aircraft symbol than that used in the contemporary inside-out (moving-hori- zon) attitude indicator (right). Both indicators in this photo depict an attitude of approximately 48° right bank and 10° nose down. However, the moving-horizon bar (B) can be mistaken for an aircraft symbol, giv- ing initial impression of climbing turn to left. Pointer (C) moving toward left may also tend to psychologically say, "We're turning left." identical in each indicator; (3) vertical displace- ment of the moving aircraft symbol for a given pitch change was twice that in the moving- horizon instrument; and (4) sky-ground colors on the instrument faces differed. It would, of course, have been ideal to have had identical appearance and pitch displacement in both in- struments, but this could not be done within time and financial restraints. II. Equipment and Methodology A Beech T-34 two-place military trainer (Fig. 3) with tandem seating and a blind flying hood over the rear cockpit was used in the study. (This aircraft was particularly appropriate for the study because its design strength-more than 8g positive and 4g negative-eliminated much of the hazard involved in permitting the sub- jects to lose control of the aircraft, as some did.) A Lockheed model 417 recorder was installed in 3 the rear baggage compartment for the acquisi- tion of pitch and roll data. The test attitude indicators were installed in the instrument panel of the rear cockpit. Each subject was exposed to four different instrument displays (Figs. 4a, 4b, 4c, 4d), in a statistically designed sequence, during the recoveries from banks to level flight. These four consisted of two full panel displays (all flight instruments available for use) and two part panel (attitude indicator only) displays. In the alternating turns and the descending turns, only the part panel was utilized. However, the airspeed indi- cator was added to the part panel in the descend- ing turn maneuver. Cardboard disks with adhesive tapes were used to cover instruments not used during the flights. A bank and pitch calibration unit was installed ahead of the wind- shield (Fig. 5); this was used by the safety pilot to calibrate the recorder and the visual indica- tions of the rear cockpit attitude indicator in- FAA FIGURE 3. Aircraft used in study was a two-place Beechcraft Mentor (T-34) whose performance and handling characteristics are similar to contemporary, light 4-5 place general aviation aircraft; however, pitch and roll is by use of a stick rather than by a wheel. KNOTS FIGURE 4A NDING 0 CAUTION CAGE DICT PS-JOK CAUTION CAOS Con GYRO FOR AMOMICS TAKE OFF FIGURE 4C CAUTION CAGE DICTIONAL CAUTION FIGURE 4B FIGURE 4D FIGURE 4. Each subject was exposed to four different instrument displays in a statistically designed sequence- two utilizing the moving-horizon attitude instrument and two with the moving-aircraft instrument. Each of the two displays was divided into full (4a and 4b) and part (4c and 4d) panels. As shown in 4c and 4d, the only instrument available for aircraft attitude information was the attitude indicator. In descending turns, the airspeed indicator was also available in the part panel display. 4 B FIGURE 5. In-flight view showing sighting device (A) mounted ahead of windshield and windshield lines (B) used by safety pilot to place aircraft in proper attitude while he remained visually alert for other aircraft in area. strument, and to accurately place the aircraft in proper attitude prior to letting the subject take control during each test segment of the flight. Subjects. Thirty-two male, FAA-certificated pilots from 24 to 60 years of age (mean age, 43.5 years), were used as subjects. Nine held ATR ratings; another nine were commercial pilots; the remaining 14 held private pilot cer- tificates. Only one (a commercial pilot) was not instrument rated; he was “in-training" for an instrument rating. Total flying time of the subjects ranged from 80 to 18,000 hours, with a mean of 4031 hours. They were divided into two groups of low time and high time. The low-time subjects had a total flight experience of less than 400 hours with a mean of 178 hours; the high-time subjects had more than 1000 hours, with a mean of 7885 hours (see appendix). Statistical Methods. The flight protocol was designed to provide statistically valid data. The data were tested using two methods: Anal- ysis of Variance (type SPF-P.qr Design, Roger 5 Kirk, pp. 299-307) and Simple Effects Test (Roger Kirk, pp. 303–306). Briefing. Each subject received a standard pre-flight briefing, and was asked not to discuss the flight with other subjects-to-be. The pre- flight briefing consisted of informing the subject that he would be recovering to straight and level flight from medium and steep turns, flying a sequence of alternating left and right turns while attempting to maintain a level pitch attitude, and maintaining a given airspeed during de- scending left-hand and right-hand spirals. He was also informed that the aircraft would never be in an inverted attitude when control was given to him. The aircraft was started, taxied and taken off by the safety pilot, with the subject in the rear cockpit. After departing the traffic pattern, and with the aircraft trimmed for cruise climb, con- trol was turned over to the subject with instruc- tions to climb to a given altitude. During this time, as well as during the subsequent familiari- zation turns and calibration maneuvers, the sub- ject remained "in-the-open" with the blind flight hood in a stowed position. The flight task which each subject performed consisted of: 24 recoveries to level flight from 30° to 45° bank turns; eight 45° banked turns, rolling consecutively from bank to bank; and four 45° banked descending turns-or a total of 36 turns. A third of the first 24 turns were in a coordinated condition when control was given over to the subject; another third involved "slipping" entries; and the remaining third were in a "skidding" condition. Half of the 24 turns involved use of full (all flight instruments) panel conditions and half involved part (atti- tude indicator only) panel conditions. Slipping and skidding conditions were superimposed on the turns by the safety pilot for a time sufficient (more than 30 seconds) to confuse the subjects as to the direction of turn before the subjects took over control. During the time the safety pilot was placing the aircraft in the appropriate turn condition, the subject kept his eyes closed and covered by his left hand to preclude any inadvertent cueing from shadows passing across his head. Also, he kept his right hand away from the control stick. Upon a spoken interphone command from the safety pilot, the subject opened his eyes and took control of the aircraft as he scanned the instru- ment panel. The sequence and conditions of the turns (panel configuration, bank angle, direction of turn and "coordination condition") were sys- tematically counterbalanced. A typical flight protocol for one subject is shown below: S# Flight Protocol Name Recovery Sequence #1. Moving-Aircraft Indicator (Full Panel) 45 R SK 30 L SL 30 L SK 30 R CO 45 L CO 45 R SL Date #2. Moving-Horizon Indicator (Full Panel) 30 L SL 30 L SK 45 R SL 30 L C 45 R SK 45 R C 6 #3. Moving-Aircraft Indicator (Part Panel) 45 L SL 45 L SK #4. Moving-Horizon Indicator (Part Panel) 45 R CO 30 L CO 30 R CO 30 R SL 30 R SL 45 L SL 30 R SK 45 L SK 45 L CO 30 R SK Alternating Turns (45° Bank) Moving-Aircraft Indicator Moving-Horizon Indicator Descending Turns (45° Bank) Moving-Horizon Indicator Moving-Aircraft Indicator Moving-Horizon Indicator Moving-Aircraft Indicator L-R-L-R R-L-R-L Left Right Right Left Rate of entry into the turns was controlled by the safety pilot to prevent the subject from ac- curately assessing the direction of turn and the pitch attitude. III. Results Recovery from 30° and 45° Bank Angles. The mean values appearing in the various tables and figures of the text should be used only as com- parisons rather than as absolutes. Not only were different magnitudes of initial bank angles em- ployed, but switching from one instrument and panel combination to another may have intro- duced transfer or sequence effects that could have influenced the gross numbers used to express performance. However, the values are compar- able because all such effects have been system- atically counterbalanced. Two magnitudes of initial bank angles and two directions of turn were employed to intro- duce a means of minimizing anticipatory esti- mations of required corrections by the subjects. All initial conditions were counterbalanced and the data combined for the purpose of statistical analyses. Initial Control Reversals. A subject's control response was scored as a reversal if the indicated bank angle increased by two degrees or more above the value recorded at the time he took control of the aircraft. To avoid interpreting minor control irregularities during transfer of control as potential control reversals, the bank angle trace had to demonstrate a specific depar- ture from the established value (or trend of values where absolute consistency of bank angle could not be attained) in order to be scored as a reversal. The mean number of bank angle reversals re- corded for each sequence of six recoveries to level flight are presented in Table 1. ATTITUDE INDICATOR (in seconds, measured to the nearest quarter second) between the moment the experimenter relinquished control of the aircraft to the moment the subject had reduced the bank angle by 10°. This elapsed time included the period involving transfer of control, the subject's reading and interpreting the applicable instrument(s), and the time used to reduce the bank angle by 10°. Where initial control reversals were experienced, the time expended by these control actions were also included in the elapsed time. The mean times to recover the first 10° of bank angle are presented in Table 2. Pilot Experience Moving Aircraft Moving Horizon Full Panel Part Panel Full Panel Part Panel High 0.81 1.00 0.56 0.31 ATTITUDE INDICATOR Low 1.56 0.69 1.75 1.75 Pilot Experience Moving Aircraft Moving Horizon Full Panel Part Panel Full Panel Part Panel High 3.06 3.28 2.61 2.55 Low 4.88 3.91 4.58 4.13 TABLE 1. Mean number of bank angle control reversals during each sequence of six recoveries to level flight. Analysis of variance showed a significant in- teraction between experience level and type of attitude indicator, F (1, 30)=6.11, p<.05. A simple effects test of the interaction indicated that there was a significant difference between the two experience levels when the moving- horizon indicator was used, F (1, 60)=27.56, p<.001, but not when the moving-aircraft indi- cator was used. The components of the inter- action are graphically depicted in Figure 6. MEAN NUMBER OF REVERSALS 2.0 1.0 0 100° HIGH EXPERIENCE GROUP LOW EXPERIENCE GROUP MOVING AIRCRAFT INDICATOR MOVING HORIZON INDICATOR FIGURE 6. Mean number of reversals for each sequence of six recoveries to level flight (confidence level of significant difference is indicated where applicable). Time to Recover First 10° of Bank Angle. This performance measure was based on the time TABLE 2. Mean time (in seconds) to recover first ten degrees of bank angle, Analysis of variance showed a significant dif- ference between recovery times for the moving horizon indicator (3.46 sec.) and the moving- aircraft indicator (3.78 sec.), F (1, 30)=4.32, p<.05. There was also a significant interaction between pilot experience level and instrument panel configuration (full or part), F (1, 30) = 5.63, p<.05. A simple effects test showed that the high experience group performed significantly faster than the low experience group on both the full and partial panels. However, the low experience group recovered faster with the par- tial panel while the high experience group did equally well with either the full or partial panel. Rate of Recovery from Established Bank Angles. Rate of recovery to wings-level flight is expressed in degrees-per-second rather than as elapsed time so as to minimize the effects of ATTITUDE INDICATOR Moving Aircraft Pilot Experience Moving Horizon Full Panel Part Panel Full Panel Part Panel High 5.63 4.90 6.16 6.79 Low 3.71 3.89 3.76 3.83 TABLE 3. Mean rates of recovery (deg/sec) to level flight from established bank angles. 7 High Low High Low - - Groups - - High Low High Low - - - Mean Rate F Ratio (Deg/sec) & df Significance Level Aircraft - Full 5.63 10.35 (1, 60) .005 Aircraft - Full 3.71 Aircraft Part - 4.90 2.65 (1, 60) N.S. Aircraft - Part 3.89 Horizon Full - 6.16 14.72 (1, 60) .001 Horizon Full - 3.76 Horizon Part 6.79 22.44 (1, 60) .001 Horizon Part 3.83 Aircraft Horizon - - High High Aircraft - High Horizon High - - Full 5.63 2.35 (1, 60) N.S. Full 6.16 - Part 4.90 29.08 (1, 60) .001 Part 6.79 Aircraft Horizon Low - Low - Full 3.71 <1 - Full 3.76 Aircraft Horizon Low Low - Part 3.89 <1 Part 3.83 Full - High - Aircraft 5.63 7.53 (1, 60) .01 - Part High - Aircraft 4.90 Full Part High - High Horizon 6.06 5.74 (1, 60) .05 - Horizon 6.79 Full Part - Low - Aircraft 3.71 < 1 - Low - Aircraft 3.89 Full Part - Low Horizon 3.76 <1 - Low Horizon 3.83 Legend: High, Low - (pilot experience) Aircraft, Horizon - (type of attitude indicator) Part, Full - (panel configuration) TABLE 4. Component comparisons of three-way interaction for rate of recovery from banks. 8 variations in initial bank angles. Individual recovery rates were calculated by dividing the initial bank angle by the time (to the nearest quarter second) required to bring the aircraft to an effective and constant wings-level attitude. For those subjects who failed to establish a rela- tively precise wings-level attitude, the total period was based on a time point after which no further bank corrections were made by the subject. The mean rates of bank angle recovery are presented in Table 3. Analysis of variance showed a significant in- teraction between pilot experience and type of attitude indicator, F (1, 30)=10.00, p<.005. There was also a significant three-way interac- tion, F (1,30)=5.43, p<.05. A simple effects test of pilot experience by attitude indicator in- teraction showed that three of the four compo- nent comparisons were significant. Only the performance by the low experience group on the two types of attitude indicator failed to be sig- nificantly different. The interaction is shown in Fig. 7. 7.0 6.0 .001 The results of the simple effects test of the three-way interaction have been tabulated by component comparisons in Table 4. The panel components of the interaction can probably be ignored for practical purposes since they may represent an overlap from the other combinations especially since there was no sig- nificant panel effect. Furthermore, the panel variable did not show a primary interaction with either of the other two variables. Pitch Control During Recoveries. Initial pitch attitude (at the time control was given to the subject) was not systematically, nor pre- cisely, controlled by the safety pilot; but there is no indication of systematic bias relative to magnitude or direction. The mean for all initial pitch attitudes was 3.53° from level flight; the algebraic mean was 0.66° nose-down. Pitch control was measured by two criteria. One quantified the excess pitch movement of the aircraft. Excess pitch was defined as the total range of pitch change made during each recovery minus the amount which would have established a stabilized, level pitch attitude. The second performance criterion related to the rate at which pitch attitude was changed to obtain a stabilized pitch condition. The results for excess pitch movement are presented in Table 5. RATE OF RECOVERY (DEG/SEC) 5.0 4.0 3.0 T .005 ATTITUDE INDICATOR Pilot Experience Moving Aircraft Moving Horizon .001 Full Panel Part Panel Full Panel Part Panel High 3.19 3.21 2.59 2.64 Low 5.85 4.35 8.61 9.73 HIGH EXPERIENCE GROUP LOW EXPERIENCE GROUP MOVING AIRCRAFT INDICATOR MOVING HORIZON INDICATOR FIGURE 7. Mean rate of recovery from established bank angles (confidence levels of significant differences are indicated where applicable). 9 TABLE 5. Mean rates of pitch movement (in degrees) in excess of required pitch correction to establish effective control of aircraft. An analysis of variance indicated a significant interaction between pilot experience and type of attitude indicator used, F (1, 30)=14.53, p<.001. A simple effects test indicated that the low ex- perience group made significantly larger excess pitch corrections than did the high experience group when the moving-horizon indicator was used, F (1,60)=45.71, p<.001. However, the low experience group made significantly smaller ex- cess pitch corrections with the moving-aircraft indicator than with the moving horizon indi- cator, F (1, 30)=22.23, p<.001. On the other hand, there were no significant differences be- tween the two groups when the moving-aircraft indicator was used, or between the two types of attitude indicators when used by the high ex- perience group. The interaction is shown in Fig. 8. 60 EXCESS PITCH CHANGE (DEGREES) N + 0 .001 - - 100° were relatively small, averaging between 3° and 4°. An analysis of variance indicated that only the difference between pilot experience groups was significant, F (1, 30)=12.59, p<.001. Bank Angle Control During Alternating Turns. Attempts to quantify bank angle control during alternating turns on the basis of accuracy and consistency of performance were unsuccess ful because performance by both pilot experience groups using either attitude indicator was toc inconsistent to provide meaningful data. Pitch Control During Alternating Turns. The criterion used to evaluate performance was the ability of the subjects to maintain zero pitch attitude. The means of the total ranges of pitch during the maneuver are presented in Table 7. ATTITUDE INDICATOR Pilot Experience Moving Aircraft Moving Horizon High 12.13 11.00 Low 20.13 34.69 0 HIGH EXPERIENCE GROUP LOW EXPERIENCE GROUP MOVING AIRCRAFT INDICATOR MOVING HORIZON INDICATOR FIGURE 8. Mean range of excess pitch change during recovery to level flight (confidence levels of significant differences are indicated where applicable). The results for rate of pitch correction are summarized in Table 6. ATTITUDE INDICATOR Moving Horizon Full Panel Part Panel Pilot Experience Moving Aircraft Full Panel Part Panel High .80 .78 .85 Low .49 .70 .44 .43 TABLE 6. Mean rates of pitch correction during bank angle recoveries. .76 (deg/sec) The relatively small values in Table 6 are attributed to the fact that the initial pitch angles 10 TABLE 7. Mean rates of ranges of pitch changes (de grees during alternating turns). An analysis of variance indicated that there was a significant interaction between pilot ex- perience and type of attitude indicator used F (1, 30)=6.63, p<.05. A simple effects test showed that the low experience group made sig- nificantly larger changes than the high experi- ence group when the moving-horizon indicator was used, F (1, 60)=32.25, p<.001. The low experience group made significantly smaller pitch changes when using the moving-aircraft indicator than when using the moving-horizon indicator, F (1, 30)=11.43, p<.005. The inter- action is shown in Fig. 9. Bank Control in Descending Turns. The ob- jective of this maneuver was to maintain a 45° angle of bank in a series of descending turns. Each turn was terminated at the command of the safety pilot after the aircraft had descended at least a thousand feet. The percent of time the aircraft was held in a 45° bank (±5°) is presented in Table 8. An analysis of variance indicated there were no significant differences between any of the values in Table 8. MEAN RANGE OF PITCH CHANGES 40 30 20 10 .005 100° HIGH EXPERIENCE GROUP LOW EXPERIENCE GROUP MOVING AIRCRAFT INDICATOR MOVING HORIZON INDICATOR FIGURE 9. Mean range of pitch change during alter- nating turns (confidence levels of significant differ- ences are indicated where applicable). ATTITUDE INDICATOR Pilot Experience Moving Aircraft Moving Horizon High 49.03 Low 47.07 47.43 43.63 instrument type, F (1, 30)=11.50, p<.005. A simple effects test showed that the high experi- ence group demonstrated significantly smaller ranges of pitch change with the moving horizon indicator than did the low experience group, F (1, 60)=40.23, p<.001. The high experience group was also superior with the moving-aircraft indicator, F (1, 60) = 13.03, p<.001. There were no significant differences between the two indi- cators for the high experience group, but the low experience group did significantly better with the moving-aircraft indicator, F (1, 30)=21.31, p<.001. The interaction is graphically illus- trated in Fig. 10. 30 20 MEAN RANGE OF PITCH ADJUSTMENTS 10 (DEGREES) 30 1 .001 .001 100° HIGH EXPERIENCE GROUP LOW EXPERIENCE GROUP TABLE 8. Percent time pilot maintained 45° (±5°) of bank during descending turn. Pitch Control in Descending Turns. Pitch control was examined relative to pitch changes nade by the subjects in their attempts to main- cain the desired airspeed. Relatively large pitch changes were judged to indicate less satisfactory bitch control than smaller pitch changes. The pitch range means are presented in Table 9. Moving Horizon ATTITUDE INDICATOR Pilot Experience Moving Aircraft High Low 15.50 23.22 15.28 28.84 TABLE 9. Mean rates of ranges of pitch attitude (de- grees) during descending turns. An analysis of variance indicated a significant nteraction between pilot experience level and 11 MOVING AIRCRAFT INDICATOR MOVING HORIZON INDICATOR FIGURE 10. Mean range of pitch changes during de- scending turns (confidence levels of significant dif- ferences are indicated where applicable). The results may be summarized as follows* : Roll Recoveries. (1) Reversals (initially increasing bank angle while attempting to return to level flight): (a) there was no difference between attitude indi- *NOTE: The summations given in this section refer to findings whose validity is based on the use of accepted statistical testing methods. Differences in performance are expressed only where they were statistically significant at the .05 level of confidence or higher and not on the basis of the size of the numerical difference between the mean values. Performances are expressed as being equal, compar- able or identical when there was no statistically sig- nificant difference between the means regardless of the size of the numerical difference and in which direction it was weighed.