ControlsDisplays

ControlsDisplays - ENsc-1o4 Controls and Displays 91-2...

Info iconThis preview shows pages 1–14. Sign up to view the full content.

View Full Document Right Arrow Icon
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 2
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 4
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 6
Background image of page 7

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 8
Background image of page 9

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 10
Background image of page 11

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 12
Background image of page 13

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 14
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: ENsc-1o4 Controls and Displays 91-2 Displays Displays convey the current state of a device to the user. Depending on the context, the user will seek different information. Instructional information tells the user how to accomplish their goals (eg. a help facility). Command information comprises direct orders to the user to perform specific actions, while the related advisory information are recommendations or cautions (eg. a dial indicator with the unsafe ranges highlighted in red — a check gauge). Conversely, the user may request answers to inquiries (eg. a controller bringing up the flight identifier and altitude information on a his/her radar screen). Visual Displays Being visual creatures, visual displays give the greatest bandwidth for communicating with the user. However, vision is directional, so anything outside of the approximately i30° vertically and 180° horizontally about the line of sight will likely be missed by an operator. One consequence of this is that control panel real estate quickly becomes a scarce commodity in complex systems. In spite of their directionality, visual displays are the preferred choice under the following conditions (Bailey, 1989): ~ The user’s auditory system is already at or beyond capacity (eg. fighter cockpits). - The receiving location is so noisy that auditory messages would be missed or misheard. o The message is complex and/or long. 0 The message deals with a particular area of the control panel (eg. alerting the operator to a cluster of controls/displays related to a subsystem). - The message must be referred to later. 0 The user works (and looks) primarily in one location. Visual displays can be broken into static and dynamic displays. Static displays, or labels, provide instructional information, and are of greatest use to the novice user. Convention recommends that labels be placed above the display or control they identify, so that they will not be obscured by the operator’s hand(s). Dynamic displays, which convey information that changes over time, can be further subdivided by whether or not they convey discrete or continuous changes. An example of a discrete display is an indicator light on an automobile dashboard (ie. an annunciator). Independent of display technology (eg. mechanical, LED, LCD), continuous displays can be characterized as digital or analog. Digital displays (eg. a calculator display) directly show a numerical quantity and are ideal for conveying precise numerical data to arbitrary precision in relatively little panel space. They are difficult to read, however, if the numeric value changes rapidly, and are a challenge to determine rate-of-change. Also, because the display is incapable of showing the range limits on the numeric variable, it is difficult to gauge how close the value is to one of its limits and so operates poorly as a check gauge. Analog displays, like the hands on a clock, typically show a single variable, and can incorporate a variety of styles (see figure 1): - Shape can be linear, arranged horizontally or vertically, circular, semicircular, or less commonly some other general shape. ‘ - Show a full (ie. limit-to-limit) scale, or a partial (ie. subrange) scale. Mm 1 5mm. Controls and Displays 91-2 - Be scaled incrementally in units increasing linearly, logarithmically, or some other fashion. 0 The scale may show advisory ranges or zones (ie. check-gauge). - Have a fixed scale with a moving pointer, or a fixed pointer (or fiducial mark) with a moving scale. - Present a wide variety of symbol and pointer designs. This includes bar-graph styles of pointers. 0 Incorporate 1 or more pointers (cg. clock, altimeter). o I 1 : LA_!_J_L_L.J I l l | I) I 2 J Pulmud Poo: alum-mm Nut-Ill]! mun-Id Palm"! Imufllnml nun-nah Figure 1. Examples of ‘ analogue displays. (a) pointers in dial displays. (b) scales in dial . displays. elm-M Nmmul mains" sun I . 0| Independent of the style of the indicator, common sense dictates that elements of the display must not obscure other elements (eg. the pointer covering up the number), the scale numbers should not crowd each other and they should proceed in an orderly fashion, and the pointer should align with the scale to eliminate the need for interpolation. The smallest scale division should be the minimum that the task requires. Finally, for speed and accuracy in reading the indicator, the smallest feature should be no less than 15 minutes of arc, otherwise the operator will spend a lot of time peering at the display. Scales should increase up, to the right, or clockwise. ’ Generally, the fixed scale with moving pointer is the preferred style because it shows not only the current value, but the range of values which is ideal for qualitative or check readings. Scanning and comparing several such indicators can be done quickly and reliably if all the indicators are arranged so that the normal operau'ng value of each is arranged at the 9 o’clock or 12 o’clock position. The disadvantage of fixed scale with moving pointer displays is that they tend to occupy more panel space than other types, especially if the range is large and the scale divisions small. One way around this is to use multiple pointers, where one division of one pointer represents one complete range/revolution of the next lower one. For example, a clock face is divided into 12 major divisions, and the small (hour) hand moves through 1 such division while the long (minute) hand makes one complete revolution. On aircraft altimeters, which have a scale with 10 major divisions, the long slender hand indicates 1,000 foot/division (10,000 feet/revolution) while the short wide hand indicates 10,000 feet/division. The problem is that multi-pointer displays are easy to misread by swapping which hand means whichl, and that it is also easy to misinterpret the direction of dial movement (Fitts, 1951). Inf rm i n in In addition to the above guidelines and display styles, the format of the information can be optimized to assist the user. If the information needn’t be an exact replica or analog of the 1 Misreading the altitude as 42,500 feet instead of 24,500 feet can prove embarrasing to the pilot’s widow. ENsc-104 Controls and Displays 91-2 source, it may be coded variously. Coding schemes include: Colour Foley and Wallace suggest that, for reliable operation, no more than 6 distinct colours be used for coding. Bailey also points out that colours be used consistently (eg. red for alarm/critical/emergency conditions, amber/yellow for warning, and green for normal/OK conditions), and that any words associated with the colour— coded display jive with the colour (eg. a red indicator labelled “grass” would be inappropriate). Also, be aware that many men are colour blind or colour impaired. B rightness Foley and Wallace suggest limiting brightness coding to two levels (normal and bright). For many indicators, this could be simplified to off/on. “On” brightness can be 50 to 75 times the background brightness, but the resulting glare at higher ratios can impair the operator’s vision. B linking Blinking should be off/on with about a 25% duty cycle and should be used only to draw the operator’s attention. A blink rate of 3 Hz has been found most effective, with a 6-9 Hz rate sometimes being used for emergency conditions. Note that rates above 9 Hz approach the critical flicker frequency and may be interpreted as a steady light. Size Size is a limited coding scheme because the visual system tends to normalize stimuli unless there is a contrasting size nearby. Extrapolating from Foley and Wallace, size coding is probably limited to 2-5 sizes. S hape Foley and Wallace suggest a maximum of 10 shapes for reliable communication, though this limit could likely be raised when dealing with visually complex shapes such as icons. Text Similar to shape, but requiring different translation centres, text can readily identify a condition. However, use brevity to maximize the accuracy and reduce panel area. The designer can also use redundant coding (ié‘f‘c‘ombinin g two or more coding schemes) to improve reliability. Examples are indicator lights with explanatory text labels, or stop lights which combine geometric and colour coding (ie. Red = stop = top light, yellow = caution = middle light, green = go = bottom light). ' You should limit the amount of coded information to that which is sufficient to avoid overloading the operator. This is especially true in crisis situations where the operator is already taxed. A case in point is the 1979 Three Mile Island nuclear accident, during which over 100 alarms were simultaneously active. Most of these were warning lights, and with no way of suppressing the unimportant ones, the entire control panel became one confusing “Christmas tree”. Warns Since most complex equipment incorporates several displays, their arrangement is important to ensure reliable operation. Displays should be grouped by sequence of use. Figure 2 shows part of an aircraft instrument panel, where the links represent the frequency of pairwise glances at instruments. Analogous to Fitts’ law (covered later), displays should be grouped so that those with the strongest links are closest together. For example, interchanging the airspeed and cross pointer would significantly improve the efficiency. Note however, that the link weights may change over time (eg. takeoff, cruising flight, landing). When there is no dominant scanning sequence, it is ENSC-‘l 04 91 -2 Controls and Displays useful to group displays by common function (eg. placing all engine instruments in a cluster) to simplify f‘within-function” tasks. GO Tum mu hank All-mun Ya"! all [unit a b Figure 2. Link chart of eye motions between aircraft instruments during instrument landing approach (links with less than 3% have been omitted). The original configuration (a) could be improved for this task by interchanging the airspeed indicator with the cross pointer, as in (b). Roll M climb “an 0' elm-o Since operators most easily see those displays nearest their normal centre of View, we find instruments clustered below and above vehicle Windshields. In situations where the operator cannot afford to divert their gaze to the instruments, it may be necessary to overlay their normal field of view with the displays. An example of this is the fighter pilot’s head-up display, or HUD, where the critical displays are projected onto a half-silvered mirror directly inthe pilot’s field of view. Even the HUD is proving inadequate as military pilots are often looking elsewhere for enemy threats, so the current trend is towards head-mounted displays. Many visual displays are reinforced by an auditory display, usually alarms/annunciators to alert the user, however they can be used for continuous monitoring/tracking such as sonar, biofeedback devices or a fighter pilot’s missile lock-on tone. Much less frequently, it is used as an alternative communication channel for visually impaired users. Auditory displays are preferable under the following conditions (Bailey, 1989): o The message is short (eg. yes/no). 0 Response time to the message is critical (eg. “Pull upl”). - The information is transitory (ie. no need to refer to it later). 0 The visual channel is already loaded to capacity or beyond. - The receiving location is unsuitable for visual communication (eg. to bright or dark). 0 The user must be away from the workstation but must still be alerted, or where their gaze may be away from an important visual display). In the alarm/annunciator role, the designer should consider the following criteria (Bailey, 1989): o The signal should be loud if the user will be a considerable distance away from the sound source. - The signal should be low frequency if the user must go into other rooms, behind partitions, or some distance away from the source. ENscm Controls and Displays 91-2 - The signal should be readily distinguishable from the background. 0 The signal may need to be modulated in addition to the above techniques if they prove inadequate. 0 The alarm should cease only after the user responds appropriately to the cause of the alarm. Controls Controls are the means by which the operator communicates with the machine. These can be grouped by three criteria of 1) discrete vs. continuous operation, 2) linear vs. rotary motion, and 3) sin gle-dimensional control vs. multi-dimensional control (examples of combinations are shown in figure 3). There are no widely accepted criteria for choosing between linear and rotary controls—the choice often becomes one of applicability, population stereotype and aesthetics. A single multi-dimensional control is generally preferable to multiple single- dimensional controls in those cases where the dimensions are related (eg. an aircraft joystick). ‘ An exception to this is where the display information is in the form of two scalar values. One—dimensional control TWO ~ dimemional comm. One — dimensional control Two — drmemronal control .,., _._,.._. . . ; .:.,. it )ucrete : n r l _ o to DISCFCIE central f“.-—r—_—’-T.~-_.W,...1.-;fi.w—I—.q ' J - rntinuous '. - ' I ‘ - - ' I. - " lh" -. ' ;' ; ' ‘ Continuous control r w - I" - . . ' m5 I X— Y Polentrorneur Aummobile bake Dad-l Jovmck ‘" “j” Spring relmn on! Yuu ball Flgure 3. Control examples (left) linear controls, (right) rotary controls. Discrete controls are preferred when there are a limited (ie. < 25) number of control states, or alphanumeric information must be entered (eg. a keyboard), it is a binary setting (ie. on/off), or if only small mechanical forces are required or available. Continuous controls (eg. car steering wheel) are preferred when there are a large number of control states, the operator must exert considerable force, or the speed of operation is more important than the absolute accuracy. You should consider the following design principles when designing or specifyingcontrols (Bailey, 1989): 0 Critical or frequently used controls should be within easy reach of the operator from their normal position. 0 The characteristics of the control (eg. force, speed, accuracy, range of body movement required) should not exceed the limits of the least capable user. 0 Minimize the number of controls, possibly by automating some of the control relationships. ENsc-104 Controls and Displays 91-2 0 Make the control movements as easy and natural as possible. 0 When controls must be powered, provide artificial “feel” to the user (eg. airliner controls). 0 Controls should positively indicate their actuation (eg. keyboard click). - Control surfaces should be designed to prevent the operator’s fingers, hands, etc. from slippin g off them. 0 Controls should be located and designed to prevent accidental operation, especially when such accidental operation would result in a critical situation. Possible methods include recessing the control, locating the control where it is unlikely to be hit, orienting the control axis normal to the most likely operating force, cover the control with a guard, locking the control, require a sequence of operations to activate the control, or adding resistance as a stiffer control is less likely to be accidentally operated. MmLQQang Making controls easy to identify with function improves the usability of the system. In particular, for more than casual use, controls should be coded so that they needn’t be looked at to determine their function. 492%" 9e landing "up Laud-n9 pm Fur nunguunu ii; <2} <25 3 ' V Q Pot-II hit-om!) “Lu. lill II) "but: power . Suntan!»va Miuuu Cub-nun air I Figure 4. Standard Aircraft controls using Figure 5. Alphabet of shapes discriminable by shape coding. touch alone. Shape Shape is the most powerful non-visual means of identifying controls. Figure 4 shows the standard shapes for levers in aircraft, while figure 5 shows a standard set of non—aircraft shapes. All of these shapes can be distinguished by touch alone while wearing gloves. Size Size is not as effective as shape at non-visual identification. Control knobs need to ' vary in size by at least 20% to be discriminable. Colour Colour in and of itself cannot be used to touch-distinguish controls, but is excellent for redundantly~coding other methods, and for controls which the operator must look at. Colour works best when the meanings associated with the colour match the use of the control. Red should be used for emergency controls because of its stop connotation, while green should be used for go or safety connotations. Labeling Labels, like colour, are good for redundantly coding controls, and require little operator training, which makes them good for novice and occasional users. Labels should be located consistently (usually above the control), and should indicate what 5mm Controls and Displays 91-2 is being controlled. They should be brief, avoiding technical terms unfamiliar to the operator, and should avoid abstract symbols with the exception of meaningful icons. Lastly, the designer must ensure the labels are adequately illuminated Control-display relationships Displays and controls do no operate in a vacuum—control input will affect display output, which causes the user to modify the control input... It is important that there be a rational relationship between the actions of the controls and the displayed results—if not, data will be misread and controls will be operated incorrectly. The biggest constraint on control-display relationships is in meeting the users’ expectations based on past experience. Plinr Within any population (ie. an identifiable group of people) there are certain generally accepted expectations about how things will work, especially if the machine provides no contrary information. Some population stereotypes for North America include the assumptions that a light switch is flipped up to turn the light on, or rotating a volume control clockwise will increase the radio’s volumez. Population stereotypes evolve as a consensus of the general population to address those situations where controls have no intrinsically “correct” orientation (eg. the light switch operates equally well in any orientation), and become internalized to become “common sense”. These stereotypes are carried over by analogy to new situations. For instance, if the light operates with the switch in the “up” position, then all electrical devices are expected to operate with the switch in the “up” position too. Take this opportunity to fill out the population stereotype quiz in appendix C...NOW (answers are in appendix E). Population stereotypes are something of a double-edged sword; while they greatly simplify the designer’s choices, problems will arise if the designer and the user population use different stereotypes. Because population stereotypes are so integrated in our lives, the designer is often unawares that there may be a potential mismatch between their assumptions and the users’3— another case for Hansen’s “know the user”. Where there are no established stereotypes, you must perform studies with typical user populations to determine what the populations to determine what the populations to determine what the Equally, many of the stereotypes we use without thinking have come about, not because of good ergonomic planning, but simply through historical accident —— for example, aircraft controls (see Appendix D). The direction of a control’s movement should be considered in relation to a) the location and orientation of the user relative to the control, b) the position of the display relative to the control and the nature and direction of the display’s response (figure 6), and c) the change resulting from the control movement, either in terms of motion of moving components (eg. landing gear is lowered by pushing a lever down) or in terms of some dimensional quantity (eg. radio volume is increased by rotating the knob clockwise). 2 Conversely, German expectations are exactly opposite (ie. switch up = off). > 3 A case in point is the old battle over labelling fuel tankers as flammable or inflammable. To the chemical engineers who labelled the trucks, flammable meant that the material could be safely cleaned with a flame—hence the trucks contained an inflammable liquid. The general populace regarded flammable as meaning it could burst into flames (ie. flame-able). Eventually the general populace won. M 7 “$0-104 Controls and Displays 91-2 Control-movement relationships are particularly important when they result in vehicle movement. A movement of a control to the right should result in a movement to the right, a right turn, or right bank of the vehicle. The relationship between the actions of visual : 30 displays and controls and the relationship w 2° \ _ f ‘0 between these and orientation of the operator is especially important in vehicles. Consider the case of an aircraft’s artificial horizon, an M an instrument that shows the pitch and roll (ie. bank) orientation of the aircraft. Early versions (figure 7a) showed a fixed aircraft in the dial centre with a moving horizon that mimicked the real horizon visible through the cockpit windows—an “inside-out” relationship. Critics . stated that this relationship was backwards and counterintuitive, and that an “outside-in” relationship with a fixed horizon and a moving ‘ V f»\\ aircraft (figure 7b) was more natural because 0 the other instruments had a similar “outside-in” m relationship. To break the ensuing “religious war” Fogel (1959) proposed a “kinalog” artificial horizon in which both the aircraft symbol and the horizon moved. When the control stick was deflected, the aircraft symbol would quickly bank against the horizon (figure 70), then would return to a level position as the horizon rotated to catch up with the bank angle (figure 7d). This system is outside-in to display control inputs, and inside-out to display system responses, and mimics the view from the cockpit if one were pursuing an aircraft. - ”' Figure 6. Display-control compatibilities. (a) compatible location and movement. (b) Compatible location but incompatible movement. (c) Incompatible locations and movement. (d) Compatible orientation of controls to displays but direct association is slightly unclear. (e) Incompatible orientation and obscure association. ' a b Figure 7. Alternative versions of artificial horizons. (a) Moving horizon, (b) Moving aircraft, and Kinalog (c) initial response, (d) steady-state response. ENsc-104 I Controls and Displays 91-2 Other general design principles are that hand controls should move horizontally rather than vertically, and fore and aft rather than side-to—side. Also, all equipment that the same person uses should have the same control-display motion relationship. In keeping with population stereotypes, display scales should increase up, right or clockwise, and that movement of the display pointer up, right or clockwise indicates that the value is increasing.Control and display actions should also agree with the user population stereotypes (figures 8, 9). @‘fl - It? © |@ Figure 8. The solid arrow shows the clockwise Figure 9. Effects of scale side. The arrow shows population stereotype for increasing the display the direction of control rotation to increase indicator indicator. They conflict in (a) and agree in (b). values. References Bailey, R.W. Human Performance Engineering. Prentice Hall, 1989. Ferrari, J .P. Ten Principles of a Good User Interface. START The ST Monthly, August 1989. pp. 25-26. Folley, J.D. & Wallace, V.L. The Art of Natural Graphic Man—Machine Conversation. Proceedings of the IEEE, April 1974. Appendix A: Fitts’ Law Since many tasks involve hitting a target (cg. click on a light button with a mouse, put a cotter pin in a hole), it is important to minimize the time required while maximizing the accuracy. Fitts’ law predicts that as the distance to a target (D) increases, so will the time to reach it, and as the size of the target (W) decreases, it will take longer to accurately land on it once you get there (see figure 10). Fitts defined an index of difiiculty (ID), relating both distance and target size to a single number, measured in bits. It is defined as ID = 10g2 (ZD/w) where the 2 in the numerator avoids having to take the log of 1 for the case where D = W. Movement time, to accurately smite the target is then related to the ID by T = a + b ID = a + b logg (2D/w) (Fitts’ law) where the parameter “a“ IS related to reaction time, and “b” is a proportionality constant — about 100 milliseconds/bit for humans. The implications for dialog design, and particularly for GUIs, is that items that are often used together should be placed close together. Similarly, targets (cg. check boxes) should be as large as possible, and have enough separation between them to allow for positioning error. Conversely, W 9 Emma Controls and Displays 91-2 dangerous options should be separated from the rest by quite a margin. Target i ' ~ Figure 10. Index of difficulty is directly proportional to distance (D) from starting point to target, and inversely proportional to the target width (w). Fitts’ law applies to conventional controls as well. The standard QWERTY keyboard was designed in th 1860’s to limit the typists’ speed to prevent the typewriter keys from jammin g (done by placing frequently used letters as far apart as possible). The DVORAK keyboard groups frequently used letters close together and is 2.6% faster, while the Montgomery wipe-activated keyboard goes further by grouping the most frequent4 triads of characters. Figure 12. Keyboard layouts (a) QWERTY, (b) Dvorak, (c) Montogomery Appendix B: Practice Effect The Power Law of Practice, or “muscle memory” states that the time to perform a task decreases as the task becomes more familiar. VTn = T1 n-a (Power law of practice) where Tn is the time taken on the nth trial, and a is approximately 0.4. Again for GUIs, muscle memory can only be a benefit if the target is always in the same absolute location in the case of pull—down menus, check boxes and dialog boxes, or the same relative location from the cursor. Appendix C: Population Stereotype Quiz. Knob Ill". Numb-u? “V‘ a... haul! <2 /’g\\ /’§‘E\ 1 2 3 1. To move the arrow indicator to the centre of the display would you turn the knob a) clockwise or b) counterclockwise? 4 For English text. 10 2 I ENsc-104 Controls and Displays 91 2. A worker is required to type numbers as they appear on a screen by pressing 10 keys, one for each finger. Label the keys for the 10 digits 0-9. 3. Which way would you turn the knobs to turn the water on? Put arrows on each dotted line to indicate the direction. 4. Working with a fire crew, the hoseman yells down to you “pressure high”. Should you a) increase the pressure or b) decrease the pressure? Level comm! Nor-w" Ilnu Rim lunl 5 . Is the church on the east bank or the west bank? 6. Which is the outside lane (A or B)? 7 . To move the arrow indicator to the right would you a) push or b) pull the lever? 11 l ENSC-1 04 I. How skillful are you at the con- trols of various airplanes? If you were trained in a wheel-control type, do you have trouble switching to a slick? if' you learned in the left seat. how do you do in the right? Can you step out of a tricycle-gear model and handle a tail- dragger? For most present-day pilots, these transitions are relatively .minor—the differences aren't really as great as .some people imagine. Actually, getting Controls and Displays used to the different switch and instru- ment locations in different wheel-con- trol models Seems to bc a bigger prob- lem than the change from wheel to stick control. It was not always so. Back in the pi- oneer days before World War I, the controls of airplanes were a long way from being standardized. Various manu- facturers had their own'systems, and the differences were so great that some pilots could not make the transition. The famous Lincoln Beachey at the controls at his special 1912 Curtiss stunter. The wheel works the rudder, lore-and-alt movement at the yoke works the elevators. and the brackets around Beachey's shoulders work the ailerons. The brake bar pressing against the nosewheel can be operated by the pilot's loot or by pushing lull forward on the control column. A pedal under Beachey'slelt loot operates the spark or_ throttle tor engine speed. 56 THE AOPA PILOT l FEBRUARY 1973 12 91-2 ' Appendix D: Historical Accidents (reprinted with the kind permission of the author)‘ The controls at a Wright Model "B". circa 1911. The pilot's lett hand is on the elevator control. His right is on the wing-warping control, which had a' supplementary rudder control on the top and was shared with the right-seat pilot. The lever at the right side of the right seat is that pilot’s own elevator control. Note the single loot pedal used by both pilots lor spark control. :YESTEHDAY'S WINES, Confusing Control Systems by PETER M. BOWERS /AOPA 54408 There were actual cases in which pilots who were trained on one system had to have that same system installed in a plane they bought from another manu- facturer that used a different one. A modern example would be a pilot who learned on a Champion insisting on stick control, heel brakes. and a left- hand, quadrant~type throttle when he ordered a Cessna 150. Let's try a checkout on some of these early-day systems, and see what they were like; First, of course, there was the Wright system. This was weird to start with. The pilot lay prone on the bottom wing and controlled pitch by rotating .1 hori- zontal rocking shaft that he held onto with both hands. Roll, which was tied into the rudder in a coordinated action. was accomplished by swinging the hips from side to side; the pilot's pelvis rested in a sideways-sliding cradle. Later, when the Wrights sat upright, control was by means of upright hand levers. As the first to teach others to fly. the Wrights had the first dual-control airplanes. These controls, however, were not fully duplicated; nor were they symmetrical. The Wright biplanes were usually solocd from the left seat for balance purposes, since the engine was off—center to the right (see photo). This put the instructor in the right seat. The lever at his right worked the pitch controls, forward for “down” and backward for “up"—a natural action that we still use today. This control was duplicated at. the left side of the left seat. Between the seats was a third lever, used by both pilots. This lever controlled roll, which still had the rud- der tied in. Forward stick gave left roll; back stick, right roll—hardly a "natu- ral" movement. If necessary, more rud- der could be fed in by an auxiliary right-left lever on top of the center ENSC—104 Controls and Dis » .. stick. This system. which went through resulted in recognized "right seat” and "left seat" Wright pilots—the transition from one seat to the other was not easy in those days. . ' ‘ The Curtiss system was considerably different and started with the pilot sit- ting up. By the time Curtiss was in pro- ized with a wheel to control the rudder. Push-pull on the wheel controlled pitch a just as it does today. but aileron control was by “body english." A yoke, pivoted behind the pilot's seat, bracketed his shoulders. To bank into a left turn, the ', pilot leaned to the left; to raise a low . wing. he leaned to the high side. This ‘ is probably the direct result .of Glenn "cycle racer beiore becoming 'a builder ~of airplanes. - Curtiss retained this unique system on some models until 1915. Curtiss, incidentally. developed the "throwover" control column for side-by-side pilots in 1911. A sort of mix betwoen Wright and - Curtiss control was introduced by Al- - berto Santos-Dumont in his 1908 "De- - moiselle.". The Paris-based Brazilian used two vertical levers for pitch and that he had run out of hands. He re- medied this situation with a special flying jacket that had a long tube sewed up the back. This tube fitted over a third stick behind the seat that con- trolled roll when the pilot leaned in the Curtiss manner. » ‘ ' The French "Antoinette" of 1909 sub- stituted wheels at each side of the cock- pit for the Wrights' levers. The left wheel controlled roll and the right con- of a foot bar to work the rudder. In ad- dition. there were separate engine con- trols, which meant that the pilot had to 4 several minor variations over the years. - ductlon. his controls had been standard-. . Curtiss’s having been a famous motor- ' rudder control. and found. like Curtiss. trolled pitch. An innovation was the use . ’ Chaos ' reigned—and crashes sometimes followed—when those - Imagnificent men went from one. - early-day flying machine to another What in probably the world's most unique alrcralt control system out the ends of the control Iurlacal right in the pilot's handal This is the Rhalnhold Plat: canard glider of 1923. The canard star-laces. worked independently. served both as elevators and as ailerons. With no vertical surfaces incorporated in this design, there was no need tor rudder control. let go of one of the other controls in order to manage the powerplant. The Wrights and Curtiss didn't have that problem at the time; both used a foot lever to control spark—the ‘oniy speed control on the engine. " The Wrights and Santos-Dumont used wing-warping for lateral control. while Curtiss used ailerons. The de- signer of the "Antoinette" tried every- thing at various times. first warping the wing. then pivoting each entire panel at the fuselage before standardizing on ailerons. ’ ‘ The first of the standardized control 'systems we use today was the stick- and-ruddcr-bar arrangement, developed by a Frenchman, Robert Esnault—l’eltrie. who gave us a single stick to control pitch by fore-and-aft movement and roll by sideways movement. Rudder control‘ was by the foot bar. as on the "An- toinette." (The bar soon had an alternate installation in the form of pedals. which are universal today.) This con- trol system came to be called the R.E.P. system. While the stick control seemed a fairly "natural" procedure for move- ment about the pitch and roll axes, the rudder control was definitely not a natu- ral. Anyone who ever used a home-built coaster wagon or sled with the steering controlled by the feet finds it natural to push on the outside of the turn. it is the same way with the bicycle or mo- torcycle—the push is on the outside. The foot-operated rudder control calls for the push to be on the inside of the turn. This gave many early-day pilots trouble. and some actually reversed the rudder cables on their personal planes. Some fatal accidents occurred when these pilots stepped into other planes and forgot the standard procedure, or when other pilots used the modified air- craft. ' 13 ' (as on the bow of a kite) 91-2 plays A relatively minor variation of the R.E.P. system was developed by another French manufacturer. Deperdussln. This used the R.E.P. rudder control, but had a wheel on a yoke instead of a stick. Fore-and-aft movement of the whole assembly still worked the pitch control. but rotation of the wheel moved the ailerons (or warped the wings; wing-warping _was fairly com- mon as late as 1916 and Guiseppe Bel- lanca used it on a' production lightplane in 1919). After the R.E.P. stick control became the accepted standard. wheel control was distinguished'by being re- ferred to as the "Dep" system. Standardization was still a long way off. In one of his attempts to beat the ~Wright patents, Curtiss tried one sys- tem, in 1914. in which only one aileron was operated at a time, and that by a- foot pedal. William E. Boeing came ’up with something else on his first air- plane, built in 1916. His system had a patented "three-way" control with a _ wheel on top of the stick to work the rudder. sideways motion of the stick for the ailerons. and fore-and-aft motion for the elevators. The throttle was a foot pedal, similar to that in a car. This system was soon changed to the "Dep." (The replica that Boeing built for its 50th anniversary in 1966 uses. stick- and-rudder.) - - - , ‘ Perhaps the most unconventional, yet direct, control system of all time was that used on the Rheinhold Platz can- ard glider of 1923. To appreciate the control system, one has to understand the design of this unique machine. This was an early ancestorof the present-day "llogallo Wing." in which 'the wing surface is a single thickness of cloth. In planform, the Platz glider looked like a small sailboat mated to its waterline mirror image. Like a sailboat pointed upwind, the airfoil was main- tained by the flow of air over the sail- like cloth wings. The -"mast" was a 24~ foot spar. held in a curve by a cable ' that formed the.leading edge of the main lifting surface. ' The canard surfaces were essentially like sailboat jibs, with their "booms'.' pi- voted on the fuselage near their for- ward ends. in addition to contributing to the total lift, the jihs also served as the control surfaces. There was no'rud— der. For control, the pilot held the free ends of the booms in his bands. which meant that he was also holding some 10% to 15% of the total lift in position with his own muscles.- (These surfaces do not "trail" with zero stick force as do conventional elevators and ailerons.) To lower the glider's nose. the pilot raised both hands evenly to decrease the jibs’ angle of attack; to raise the nose, he pulled down equally with both hands. For roll to the left. he raised his ,left hand and pulled down with his right. To raise a low wing, he pulled down on the low side and pushed up on the high side. - Now. did you say you had a problem in changing from a wheel. to a stick? El FEBRUARY .1973 I THE AOPA PILOT 57 W 5mm Controls and Displays 91-2 Appendix E: Answers to the Population Stereotype Quiz This questionnaire was given (Smith 1981) to 92 male engineers, 80 women who were friends or relatives of the engineers, and 55 (mostly male) human factors specialists. Note that even the experts, who should know, are sometimes at odds with the populace. Engineers Women Human Factors Specialists (%) (%) (%) 1. Counterclockwise 97 94 91 2. Ascending to right 70 70 84 Ascending outwards from thumbs 18 16 5 Other 12 14 1 1 3. Left Faucet Right Faucet CW CW 17 34 22 CW CCW 23 20 13 CCW CW 13 26 16 CCW CCW 47 20 49 4. Lower the pressure (ie. error) 66 48 78 Raise the pressure (ie. command) 34 53 22 5. Left bank 18 16 13 Right bank 82 84 80 (No answer) 7 6. A outside 50 51 20 B outside 50 49 8O 7 . Push the lever , 76 59 71 Pull the lever 42 41 25 (No answer) 4 14 ...
View Full Document

Page1 / 14

ControlsDisplays - ENsc-1o4 Controls and Displays 91-2...

This preview shows document pages 1 - 14. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online