KC-135 Airframe and Cockpit Upgrade

							KC-135 Airframe and Cockpit Upgrade
- Why the KC-135 is worth upgrading - Problems with PACER CRAG and Proposed Rockwell Collins GATM upgrade - What should be done in a KC-135 Cockpit Upgrade

David C. Fedors 452 FLTS 31 Jan 01

Why The KC-135 Is Worth Upgrading
A survey of commercially available aircraft reveals that in order to duplicate the aggregate capability of the KC-135 fleet, the KC-135 would have to be replaced either with a much larger aircraft or with a greater number of smaller aircraft of less capability. The high cost of a suitable replacement should make upgrading the existing KC-135 fleet very attractive. As the fleet of KC-135 ages, the Air Force is faced with the decision whether to modernize or replace these aircraft. Air Force planning for air operations assumes that tankers will be available in large numbers. The aggregate capability of the KC-135 fleet consists of two items of interest to air campaign planners: 1. The number of booms available 2. The amount of fuel that is available for offload. In this analysis, these two criteria will be used to measure the mission suitability of potential replacement aircraft. Since the fleet of potential receivers (bombers, transports, fighters and special mission aircraft) is fixed, a potential tanker replacement aircraft fleet will have to duplicate both of these capabilities in order to keep the combat capability of the Air Force the same. Assuming that the KC-135 replacement must be a commercial transport built by a US company, all of the potential replacement aircraft have two engines. The list of potential aircraft includes the following Boeing commercial aircraft currently in production: B-737-900, B-757-200F, and B-767-200ER. The “F” suffix designates a freighter version, an aircraft without passenger accommodations. All of these aircraft were originally designed for the extremely competitive airline market to carry passengers. The primary need of this market is for a certain volume of space in which to carry passengers at a minimum cost per seat mile. The economic imperatives of the airline industry along with the high thrust and reliability of modern engines has lead to the domination of the twin engine configuration for current production transport aircraft. Even with the high reliability of modern engines, both civil and military takeoffs are planned assuming the lost of an engine during the takeoff ground roll. After the engine loss, the aircraft is required to continue the takeoff with a positive climb gradient. Climb gradient is fundamentally a function of thrust to weight ratio. This engine out takeoff climb requirement is more severe for a twin engine aircraft, since they must continue the takeoff with only 50 % of the thrust remaining. This limits the amount of payload weight for a twin engine aircraft more severely than it does for a 3 or 4 engine aircraft. Twin engine aircraft have a payload weight fraction of maximum takeoff gross weight of about 0.50. The comparable number for the KC-135R is 0.62. This ratio represents the ability to lift fuel into the air and deliver it to receiver aircraft, and is the measure of efficiency for a tanker aircraft. There is no current aircraft of comparable size on the market that can do the KC-135 mission as efficiently as the KC-135. The chart on the next page summarizes savings realized in upgrading, rather than replacing, the C-135 fleet. The following assumptions were made in the calculations. 1.Crew costs are comparable between aircraft types. 2. All of the aircraft have comparable engine technology (High bypass turbofan). All else being equal, fuel consumption is a function of aircraft weight. Therefore, in a first order approximation, hourly operating cost is a function of maximum takeoff gross weight. Operating cost in dollars per hour was estimated at 10 times the gross weight in thousands of pounds. 3. Unlike four and three engine aircraft, twin engine aircraft will not benefit from militarization of their takeoff data. Twin engine aircraft are almost always limited in takeoff performance by second segment climb gradient requirement. FAR Part 25 requires a second segment climb gradient of 2.1 % for a twin engine aircraft, while the military usually specifies a 2.5% climb gradient regardless of the number of engines installed. Rather than accept a reduction in capability, it is assumed the military would adopt

1

the civil climb gradients. The military could gain some increase in takeoff weight by the elimination of the FAR mandated 35-foot screen height. The 35-foot screen height requirements means that following the loss of an engine when the takeoff is continued, the aircraft must cross the departure end of the runway at 35 feet. The military requires only that the aircraft be airborne at the departure end of the runway in the same scenario. Eliminating the 35-foot screen height requirement would benefit military operations were the takeoff weight is limited by runway length. As a practical matter this would be a small benefit as twin engine aircraft are infrequently in runway limited situations. 4. The replacement aircraft must provide at least the same capability that the current KC-135 fleet provides, both in terms of the number of booms available and total fuel lift capability. Adding more than one boom per aircraft is not practical. Some proposed tanker designs have booms located on the wingtips. This is impractical for refueling large aircraft: For a large aircraft, half of the receiver’s wing is in the downwash field of the tanker wing, while the other wing outside the downwash field in relatively undisturbed flow. The result is a very strong rolling tendency on the receiver towards the tanker. This makes position keeping extremely difficult. Stabilizing behind another aircraft’s wing tip is the most difficult position to hold in the entire tanker flow field, and usually requires full lateral control deflection and use of sideslip Turns by the tanker aircraft require the receiver not only to roll, as in the case of a centerline boom, but to translate up and down as the wing rises or falls in the turn as well. This seriously complicates station keeping for the receiver pilot. 5. Comparing maintenance costs between an upgraded C-135 fleet and a modern commercial derivative aircraft are difficult. The increase in reliability of a modern commercial aircraft is some what offset by the generally high price of components bought on the commercial market and increasing complexity of these aircraft. . The lower price of military parts does not reflect the cost of maintaining the logistical infrastructure. It is difficult to estimate the cost of maintaining the logistical infrastructure since it is spread out over many types of airframes. For the purpose of this argument, it is assumed that the maintenance costs between an upgraded C-135 fleet and a modern commercial derivative aircraft fleet are compatible The author acknowledge this requires more study before a rational decision can be made. In order for upgrades to the KC-135 to make sense economically, the Air Force must get an accurate assessment of the remaining structural life in the KC-135 fleet. Operation of a fleet of aircraft whose origins date from the mid 1950’s is unprecedented in aviation. Corrosion control measures during manufacture were much less sophisticated and comprehensive in the mid 1950’s than they are today. Many of the areas were subject to potential corrosion are inaccessible without destructive inspection. It does no good to upgrade the fleet just as it is reaching the end of its structural life. The Air Force must aggressively work this problem and become the experts in aging aircraft issues. . The B-767 is a one for one replacement, while the others require additional aircraft since the aircraft individually do not have the payload of the KC-135. 6. Upgrade cost for the KC-135 is estimated at $ 15 million/aircraft. This includes any structural life extension/corrosion control measures necessary to extend the life of the fleet in addition to avionics and system upgrades. By comparison, FedEx is spending approximately $10 million per aircraft to upgrade its DC-10s to MD-10s.

2

7. Weights and acquisition cost for the new Boeing aircraft come from the Boeing Corporate Web site, www.boeing.com. When a price range was given, the median price was used. Not that the “F” suffix designates a freighter version of the aircraft without passenger seats, galleys, etc. KC-13R 322.5 122.5 200 0.62 1 $3225 $16.12 $3225 15 550 8.25 B-767-200ER 395 185.7 210.2 0.53 1.051 $ 3950 $18.80 $ 3950 160 550 88 B-757-200F 255 112 143 0.56 0.71 $ 2550 $17.83 $ 2550 96 775 74.3 B-737-900 174 93 81 0.46 0.4 $ 1740 $21.48 $ 1740 55 1375 75.6

Max Takeoff Gross Weight, 1000 lbs Operating Empty Weight, 1000 lbs Payload, 1000 lbs Payload weight fraction of MTOW Payload, fraction of KC-135R Operating cost/hour Operating cost hour per 1,000 pounds of fuel uplifted Operating cost per hour per boom Acquisition cost/aircraft (millions) Number of aircraft needed Cost of Fleet to upgrade/replace 550 aircraft C-135 fleet (billions)

Table 1. Comparison of current Boeing Production Aircraft with KC-135R This chart shows the advantage of a careful upgrade to the KC-135 fleet. The operating cost per pound of fuel uplifted for a KC-135 is about 10% less than the next cheapest option. The cost of a comprehensive upgrade is a fraction of the acquisition cost of a new airframe.

3

Problems with PACER CRAG and Proposed Rockwell Collins GATM upgrade
The basic architecture of PACER CRAG seriously limits its capability, especially with respect to future upgrades. Despite the civil origins of many of its components, the PACER GRAG system is a military unique system. There are serious human factors and cockpit integration issues that will cause operational problems. The system is difficult to program, and this problem is compounded by the fact that the displays make it difficult monitor the system’s performance. Most of the safeguards that allow error checking and many of the capabilities that are standard on modern, automated, transport aircraft are absent.

PACER CRAG Architecture
The fundamental, fatal flaw of PACER CRAG is that the EHSI and CDU display screens are too small. The FMS-800 CDU has only 8 display lines of 22 characters, for a total display capability of 176 characters. In contrast, the transport category standard (Boeing, Airbus, and McDonnell Douglas) is 14 lines of 24 characters, for a total capability of 336 characters. The FM-800 has only 52% of the display of the industry standard CDU. The EHSI is a 4 x 4 instrument, having only 25% of the display area of the industry standard 8 x 8 display. These display limitations: 1. The small FMS CDU screen means that multiple pages are required to enter and display data. Proper programming relies extensively on the rote memorization or excessive menu/page searching. This increases possibility for programming error due to omission, especially during times of stress. This also increases the amount of training required to become and maintain proficiency in the system. 2. The display limitations of both the CDU and EHSI limit the ability of the crew to verify and monitor that the system is programmed properly and will perform as intended. These two shortcomings compound each other, and are a flight safety issue that will cause automation surprises, navigation errors, and the potential for accidents. The following are shortcomings that directly result from the FMS-800’s small display screen: 1. The FMS-800 has the title of the page on the second line. It should be on the top of the page for easy identification. 2. The FMS CDU pages are not top down oriented. Many features require moving sideways to access pages. This nonstandard and not intuitive. 3. The main function keys (FPLAN, IDX, etc.) on the CDU do not always return to the top of the menu, but to the last page viewed. This is nonstandard and confusing. 4. More abbreviations are used. Rather than use clear text, waypoints are given single letter attributes that are not always intuitive, and require memorization. 5. More screens are required to display the information, which results in “deeper” menu. The FMS800 as installed in the KC-135 PACER CRAG Block 30 has over 200 separate screens. By contrast, the vastly more capable B747-400 FMS has only 20 basic screens. All of the individual functional areas are at most 3 pages deep. 6. There is no designation between user entered and machine computed data. There is also no variation in font size or use of color to make important information stick out. This makes difficult to find and already cluttered displays even more difficult to interpret. These display limitations that seriously limit the ability of the FMS-800 to support future enhancements, such as controller pilot data link (CPDLC). The Rockwell Collins proposed GATM upgrade to the KC-135

4

require flight plan logs and CPDLC info to be displayed on the center MFD. Since the center “MFD” is not an MFD in the sense that it is equipped with push buttons around its periphery, an “interactive hand controller” or IHC is installed on the arm rests of each of the pilot’s seats. These IHCs allow interaction with the buttonless center MFD. Billed as allowing “eyes up control of displays”, the IHCs and the center MFD are necessary only because of the inadequate display size of the FMS-800 CDU. In the commercial word, all the functions performed via the IHCs/Center MFD are simply accomplished on the multi-function CDU, where the buttons (line select keys) are immediately adjacent to the functions they control. The FMS-800 is heavily dependent on GPS for a satisfactory navigation solution. GPS is a radio navaid and is subject to interference/jamming/MEACONing, etc. Military aircraft, especially ones with Global Mobility commitments, should have a navigation solution suitable for long duration flights that is independent of GPS. A triple mixed INU solution would be sufficiently accurate over long duration (ocean crossings) and be independent of external radio navigation aids. Consider a selectable means to display aircraft position as measured by different sensors; i.e., IRU, GPS, VOR/DME on the EHSI. This will allow rapid and intuitive evaluation of the individual navigation sensors. The PACER CRAG displays uses nonstandard display symbology. On most automated aircraft, the active FMS route is shown on the navigation display as a magenta line. PACER CRAG displays the active route as a white line. When operating in arc mode, and with the heading bug not in the visible portion of the compass arc, a magenta line is shown off to the side where the heading bug is located. This is confusing for operators of any other automated aircraft where the active FMS route is shown with a solid magenta line. The PACER CRAG modification has compounded already poor integration of autopilot and flight director. The autopilot, FMS, and flight director all have separate lateral and vertical modes. The pilot can easily be presented with conflicting and confusing navigation information. The various autopilot, flight director and FMS modes are not clearly or consistently displayed. While the flight director modes are shown on the EADI, the only indication of the autopilot steering mode is switch position on the overhead panel. Consider this example. Assume the autopilot is engaged in NAV/LOC, the flight director in VOR/NAV and both are tracking the FMS route. If the “HDG SYNC” knob on the display panel is pressed (single button press, no confirmation required), the FMS will go into FMS Heading mode (FHDG). The change is indicated on the EHSI, but is easy to miss. There is no change in flight director mode display on the EADI, and the flight director commands continue to be satisfied. With the FMS in FHDG, the autopilot is actually in a heading hold mode – it will continue on present heading until intercepting the FMS route. If the aircraft is not on an intercept heading to the FMS route, the aircraft will continue on present heading indefinitely. No indication is given to the crew if the aircraft is not on an intercept heading. The autopilot remains in NAV/LOC mode the entire time. Some call this a feature! The EHSI steering guidance is selected via a “COURSE” rotary selector. This mode control has no labeled switch positions; rather, the various steering modes are sequenced in a preset order as the knob is rotated. This requires the pilot to crosscheck the display as he is rotating the knob until he gets the desired one. It would be much simpler for the pilot to have a selector with distinct, labeled switch position for each navigation mode, allowing the pilot to just reach up and chose the desired navigation solution. The flight director may be coupled to the displayed navigation steering solution with a single selection. To couple the autopilot to the displayed steering solution, it must be first engaged, and then coupled to the steering solution. The autopilot coupling is totally independent of the flight director selection. No where is the fact the autopilot is coupled to the displayed course shown in the pilot’s field of view. This is contrary to FAR Part 25.1329 which states in part that selector switch position is not an acceptable means of indicating the current mode of automatic pilot systems coupled to airborne navigation equipment. The various modes of the flight director, autopilot and FMS should be combined and simplified, and the operating modes displayed consistently and clearly in one place. Every modern automated transport has flight mode annunicators (FMA), which are fundamental to the operation of the autopilot/flight director. The WXR-800 weather radar installed on PACER CRAG, while impressive in capability, is being asked to do too much. Its function should be reduced to weather avoidance, rather than trying to use it as an air to air radar. The controls take up too much valuable space on the glare shield. One set of radar controls should be provided where they are accessible to both pilots, preferably on the center pedestal.

5

While aircraft will alert you if it determines there is a significant differences in the pilots and copilot’s attitude sources, there is no alerting provided for errant navigation solutions. This becomes more important when operating out of the range of ground based navaids, such as during ocean crossings (FAA calls this Class II navigation). This shortcoming is noted in AFI 11-2 KC-135 Volume 3, paragraph 6.33.3.5, “For Pacer CRAG, periodically check all position solution differentials on the INAV pages” Every half and hour or so should be adequate.” There should be automatic monitoring and annunciation of the navigation sensors.

FMS-800 Architecture

The purpose of a flight management system (FMS) is to simplify the pilot’s task of flying by: 1. Simplifying navigation, by allowing a programmed route to be entered, and providing an easy to use and intuitive navigation solution. Typically the aircraft’s progress along the programmed route can be monitored via a navigation display (ND). 2. Providing the information normally obtained from enroute charts, instrument departure and arrival procedures and approach plates 3. Providing information normally obtained from aircraft performance charts and tables. The FMS-800 has serious shortcomings in all three areas. These shortcomings are compounded by poor flight displays and autoflight integration

Shortcomings in navigation route programming: Waypoints on the FMS-800 “fpln” page have both a name and a number. The name remains constant, but anytime waypoints are added or deleted, they are renumbered. The names are meaningful to pilots, while the number must be crosschecked against the name on the “fpln” page. The waypoints are referred to by their number on the EHSI, holding, and the air refueling pattern pages, requiring careful cross checking on multiple FMS-800 pages to avoid errors.

ACT RTE 1 LEGS
085 5 NM

1/2
.785/FL330

HEC
078° 98 NM

EED
080 99 NM

.785/FL330 .785/FL330 98 NM .785/FL330 67 NM

DRK
072

PYRIT
073

ZUN .785/FL330 ------------------------------< RTE 2 LEGS RTE DATA >

Figure 1. Generic Honeywell/Boeing FMS “Legs” Page. The MD-11 “F-PLN” and A-320 “F-PLN” are comparable. Compare with FMS-800 Equivalent “fpln” page, next

6

CRS[085] ° [OFFSET]   fpln 02 HEC 078º/98nm 03 EED 04 DRK
↕[

AUTO: V  
]

Figure 2. FMS-800 “fpln” page. To access all the heading, altitude, and distance information shown on the Honeywell/Boeing FMS “Legs” Page for the same flight plan, 15 different FMS-800 pages would have to be accessed. Route “Discontinuities” are not inserted when adding, deleting waypoints or procedure stringing (adding an ATC coded instrument departure or arrival procedure or instrument approach). These route discontinuities should be inserted into the flight plan to give the operator a change to confirm the routing after changes are made, and prevent the FMS from assuming what the pilot wants. By not inserting route discontinuities, the FMS-800 does not give the pilot a chance to confirm the changes to the route in a positive manner. This, coupled with the difficulty in visualizing the intended route on the EHSI, will result in automation surprises and the potential for accidents. The operator can delete waypoints without confirmation. On the Block 30 FMS-800 confirmation is required only when editing the active waypoint. Downstream waypoints can be modified with a single keystroke. The display limitations make it very difficult to confirm that the modification is correct. The AA accident in Cali, Columbia was a result of a DIR-TO that didn’t work the way the pilots intended. On the B-737NG/747-400/757/767, the proposed routing is shown on the Navigation Display as a dashed line. This allows both pilots to cross check that the proposed amendment to the flight plan is correct PRIOR to making it take effect. To make the proposed routing active, a confirmation is required in the form of hitting the EXECute button. If the proposed route modification is incorrect, it can be rejected with a single keystroke (ERASE) At least the AA crew had the opportunity to avoid their navigation error, had either pilot crosschecked their navigation display prior to execution. Without the ability to easily visualize route changes prior to making them active, the KC-135 fleet should expect these types of navigation errors frequently. It is not possible to enter a flight plan via ATC route clearance. For example, consider a flight from Edwards AFB, CA to Tinker AFB, OK. The filed route on the flight plan is “EDW HEC J6 IRW TIK” On the Honeywell/Boeing FMS, this route would be entered as filed or as received from ATC: “EDW HEC J6 IRW TIK.” The intervening points on the jet routes are automatically filled into the flight plan from the database. The system also error checks and will not allow starting or ending points on a jetway that are not part of the jet way. The FMS-800 requires all the intervening points on the jet route read of the chart and manually entered: “EDW HEC EED DRK PYRIT ZUN ABQ TCC PNH CRUSR IRW TIK.” This vastly increases the possibility of error, and there is no indication in the FMS that a jet route is being flown. The difficulty in verifying and monitoring the flight plan on the EHSI increases further the possibility that a keyboard entry error will go undetected. This will also make it more difficult to uplink ATC clearances to the system.

7

ACT RTE 1
VIA

2/2
TO

DIRECT DIRECT J6 DIRECT

KEDW HEC IRW KTIK

.785/FL330 .785/FL330

- - - - - - - - - - - - - - - - - - - - - - - - - OFFSET < RTE 2 ---

Figure 3. ATC Route entry on the Honeywell/Boeing FMS “RTE” Page The FMS-800 can be set to AUTO (flyby) or FLYOVER mode on the “fpln” page. This should be a waypoint attribute, not a FMS steering function. (Usually only one or two waypoints in a procedure are flyover waypoints) Some instrument departures/arrivals (common in Europe) and almost all RNAV approach procedures (missed approach waypoints are almost always a “flyover” waypoint) require certain waypoints to be flown over, rather than leading the turn as is standard practice. The following symbol on charts designates with these FLYOVER waypoints:

The FLYOVER waypoint attribute should be designated by the same symbol on the EHSI. Actual winds at the current waypoint should be propagated forward, blended progressively with pilot entered winds at future waypoints. Waypoints entered using latitude-longitude must be entered in a rigid format. For example, the oceanic waypoint N55 W 10 must be entered as N5500.0 W01000.0. The FMS should support entries such as “N55W010”. The FMS-800 does not support Place-bearing/place-bearing waypoint or along track waypoint definition. Shortcomings in providing navigation information: Fundamentally, the DAFIF unique database used by the FMS-800 doesn’t comply with the commercial ARINC 424 FMS Database standard, either in completeness or naming convention. As an article in Professional Pilot magazine stated, “Virtually all avionics systems use databases produced according to the ARINC 424 standard” (Professional Pilot Magazine, April 2001, “Terminal Checklist”). The FMS-800 is a simple point to point navigator, that is the route of flight is defined by waypoint latitude and longitudes, and the courses and distances are calculated between the waypoints. This type of leg is called a Track-to-fix Leg, and is one of nineteen different procedural leg types defined in ARINC 424. A procedural leg has two parts: a Leg Path and a Leg terminator. The leg path defines the path over the earth that the aircraft will fly, and the leg terminator defines the end of that leg and transition to the next procedural leg. These type of procedural leg types are essential to completely define instrument departure procedures, arrival s and approaches. In addition to the Track-to-fix Leg , the FMS-800 supports Direct to Fix, Track from Fix to Fix legs, and Heading to intercept (FMS heading does this for GPS/Tactical/Visual approaches only). The FMS-800 does not support the following procedural leg types:

8

Leg Path Heading to Heading to Heading to Heading to Course to Course to Course to Course to Course from fix to Course from fix to Course from fix to Initial fix Procedure Turn DME Arc

Leg Terminator Altitude DME distance Vector Radial Altitude DME Distance Intercept Radial Altitude Along track distance (Along track waypoint) DME Distance (none) Turn fix Radial from fix

As a result of this shortcoming , the FMS-800 is essentially limited to use as an enroute navigator. Shortcomings in providing aircraft performance and fuel calculations: While much of the aircraft performance data required for operation of the KC-135 can be obtained from the FMS/FSAS, it is difficult to use. While the FMS provides the navigation information, it is separate component, the FSAS (Fuel savings advisory system), that accomplishes aircraft climb and descent performance available and fuel calculations. The FSAS and FMS are very poorly integrated. Waypoints are entered in the FMS flight plan, either manually on the “fpln” page or via transfer from PCMCIA data cartridge. 1. FMS air refueling waypoints designated by the “P” attribute must have a fuel transfer amount, altitude and airspeed entered on two different flight plan pages. If a “P” attribute is present in the FMS flight plan, the FSAS will not calculate a profile unless this information is entered for each air refueling waypoint. 2. A waypoint must be designated as a bottom of descent point for the FSAS to calculate a profile. The bottom of descent is the first waypoint where an altitude and airspeed is entered with no fuel transfer amount, even if the airspeed and altitude doesn’t indicate a descent. Again, altitude and airspeed must be entered of separate flight plan pages. 3. Subsequent FMS flight plan additions/changes may (no indication is given to the operator) update the FSAS automatically once all CDUs are no longer displaying a FSAS page. All of this is required in order for the FSAS to calculate climb, cruise, and descent airspeeds and thrust settings and fuel consumption for the active flight plan. The system does not provide any indication of what is missing in when the profile is not calculated. The result of the design of this system is that it is very difficult to have any confidence in the FSAS fuel calculations or take advantage of any of the fuel savings features. There is no integration between the FMS, which calculates required descent angles/rates to meet pilot entered altitude constraints (VNAV function) and the FSAS, which calculates actual aircraft climb and descent performance and the location of the top of climb and top of descent. These functions should be integrated to allow the aircraft’s available performance to be compared to what is required to meet altitude constraints. This has the potential to reduce fuel consumption by having the FMS calculate a min cost climb/descent solution. It should be possible to associate altitude/speed constraints with a waypoint

9

(VNAV or “Prof” style navigation). There is no way to designate an “at or above” or “at or below” altitude constraints such as might be encountered on an instrument departure or arrival. What would be easy to display on the EHSI is a “distance to altitude arc.” This arc would be a distinctive color and display , based on the current ground speed and vertical speed, where the altitude set in the altitude alerter would be met. This is an intuitive means of displaying actual aircraft climb/descent performance relative to the FMS route. The inputs required for fuel, time and distance calculations, are done completely differently depending weather the operator is working with the ACTIVE or ALTERNATE flight plan. To minimize workload, both should have the same interface and it should be possible to transfer the performance calculations from the ALTERNATE flight plan to the ACTIVE. With the BLOCK 30 configuration, all the ALTERNATE flight plan performance calculations are lost once the flight plan is transferred to the ACTIVE flight plan, and all the entries must be duplicated, which includes altitude/airspeed “tagging” of the waypoints for use in FSAS computations, and using the distinct FSAS air refueling and bottom of descent marking procedure. The FSAS should provide endurance and optimum/maximum range altitudes/airspeeds. It should display optimum/maximum altitudes so the pilots can respond quickly and intuitively to climb requests by ATC. It should also provide range/endurance until reaching fuel reserves and until the tanks are empty. It should be possible to get thrust settings, distance and times for intermediate climbs and descents without going through the FMS-CDU and FSAS. Currently, having a complete FSAS calculated profile is the only way to get MCT/MCL thrust settings. These thrust settings are required multiple times on every sortie. This is a prime example of the system not being pilot oriented. For better situational awareness, top of climb, top of descent and bottom of descent could be displayed as pseudo-waypoints on the “fpln” page, and integrated with the FMS VNAV functions. The FMS should have and updateable database of runway available, grade, and obstacle distance and height for each runway. This will reduce workload and decrease the possibility for crew keyboard errors when calculating takeoff data. The aircraft should be equipped with a reliable temperature sensor on the aircraft for use in takeoff calculations. Since they can be sensed, neither temperature nor pressure altitude should have to be manually entered for the takeoff calculation. The pilot should only have to select desired takeoff thrust setting (TRT or reduced), flap setting and runway. The calculated takeoff/GA/MCL/MCT N 1 settings should automatically be displayed on the N1 indicators. Provide a takeoff configuration warning if takeoff is attempted and flaps do not match the setting for which the takeoff data was calculated. Flap position is already available to the FMS as evidenced by its display on the landing data page. The FMS-800 uses the line six as an annunciation line. During ordinary operation messages are frequently encountered, and as a result, the crews get in the habit of clearing them without reading them closely. They are also tersely worded, making their meaning very unclear in many situations. Some of the annunciations are not related to the FMS, such as load fuel manifold pressure. These sort of system annunciations should be on a separate display.

PACER CRAG Displays
The EADI/EHSI displays are too small. Serious consideration should be given to upgrading to 8-inch displays. These are the de facto standard for modern glass aircraft. Attitude indicator/autopilot attitude switching should be independent of FMS-CDU. This is necessary to keep attitude information from being lost due to an FMS hardware or software malfunction. EADI PACER CRAG uses a single cue flight director, probably as a legacy of the previous FD-109 flight director The EADI should use dual cue flight director. When implemented electronically, a dual cue is much more

10

precise than single cue. Many pilots (including me) coming from single cue flight directors have a strong initial preference for them because they indicate intuitively indicated the desired bank angle during turns. A comparable dual cue flight director initially indicates only the direction of turn, and the cue must be tracked until the desired bank angle is reached. This extra interpretation is not present with a single cue. The advantage goes to the dual cue, however, when both are implemented electronically. With a dual cue, it is very easy to center the commands and to see when small changes are required. With an electronic implementation of a single cue flight director, it is difficult to tell whether the command is being satisfied or exceeded. This is due to the difficulty in telling whether the symbolic aircraft “just meets” or overlaps the flight director command. This is easily demonstrated in a simulator that has both implementations. Also, the aircraft symbol on the dual cue flight directors is typically one degree by one degree, which allows for very precise pitch control against the pitch ladder. While I am a big fan of TCAS, The current PACER CRAG Block 30 has TCAS displays in three places: on the VSI, on the center MFD, and as an overlay on the EHSI (which even if not selected, appears automatically in the event of an TA or RA). I think a single display is enough, and I think the EHSI is the most appropriate place. The TCAS display could be eliminated from the VSI, and RA commands indicated via no fly (red boxes) on the EADI. EHSI The EHSI display modes (Map, conventional HSI, HSI arc, etc) are selected via a “Mode Control” rotary selector. This mode control has no labeled switch positions; rather, the various EHSI modes are sequenced in a preset order as the knob is rotated. This requires the pilot to crosscheck the display as he is rotating the knob until he gets the desired one. It would be much simpler for the pilot to have a selector with distinct, labeled switch position for each EHSI mode, allowing the pilot to just reach up and chose the desired mode. The stick airplane symbol on the Map and HSI arc displays is easy to loose. It should be made more prominent. The Boeing Navigation Display uses a triangle for the aircraft symbol and it is much more prominent than a stick airplane that is drawn in the same weight as the other symbology. The scales and display of TERRAIN, TCAS, and weather overlays should be simplified. Eliminate ranges where all are not supported. The frequently used EHSI map mode is a “Heading up” display. Track is displayed with a small green “T” on the compass rose. The aircraft’s track is the aircraft’s heading, corrected for wind drift, and represents the travel of the aircraft over the ground. Anytime you are attempting to fly a course based on a ground based NAVAID or reference (Localizer, VOR, TACAN or FMS), what you are really trying to do is make the aircraft’s track coincide with the desired course. The MD-11, all glass Boeing transports and Airbus 320/330/340 have “Track Up” as the standard for the Map display. This makes it much more intuitive to see that the aircraft is properly following the FMS route. Each individual bearing pointer (TACAN, VOR1, VOR2) should have a distinct on/off switch. The current system requires selection of via momentary contact switch of one of two bearing pointers, and then multiple pushes of a second momentary contact switch that cycles the bearing pointers in a preset sequence. This requires the pilot to crosscheck the display as he is pushing the button until he gets the desired one. The pilot should be able to look at the control panel and select a desired bearing pointers for display with a single switch action. During aircraft turns, the weather overlay remains fixed on the EHSI display until updated via a radar sweep. This makes it difficult to navigate around weather when turning. The weather overlay should rotate to keep the relative position of the weather off the nose correct when in turns. When both pilots are in MAP mode weather overlay, the system will not support two different display ranges. The displays should be able to display different ranges. The “TO” waypoint is not distinguished in any fashion on the EHSI. On Boeing, Airbus and McDonnell Douglas automated cockpits the ACTive waypoint is designated in magenta on the Navigation display. Distinguish the ACTive waypoint in some fashion (different color)

11

A means should be provided to display via independent, separate overlays on the EHSI map NAVAIDS, airports, and NAS waypoints independent of flight plan waypoints. This will provide increase situation awareness and improved confidence when making flight plan changes.. Each of these overlay displays should be individually selectable/deselectable There should be a means of clearly evaluating proposed flight plan changes visually on the EHSI before making them take effect. The direction and route that the aircraft would take once the change is made should be clearly and unambiguously displayed. Ideally, it should be possible to do this by comparing with the unmodified active flight plan, NAS waypoints, NAVAIDS, and airports. If the proposed modification is incorrect, there should be a one key method to reject the change. There should be a means to preview the complete flight plan, at varying scales (ranges), centered each waypoint in turn, in a North up map presentation. The flight crew should be able to step each waypoint in this fashion, varying the displayed range as required to view each waypoint and its relationship with other waypoints. Airspeed Indicator The Block 30 PACER CRAG airspeed indicator is too small and is difficult to read, especially airspeeds above 200 KIAS. Move the Mach indication from the lower left corner of the EHSI to the airspeed indicator. The Mach display should be digital in thousandths of Mach a number. FMS commanded speed for required time of arrival (RTA) should be displayed via speed bug on the airspeed indicator. The Technical Order should standardize the use of the airspeed bus for takeoff and approach Altimeter The Block 30 altimeter is small and difficult to read. It has tick marks at 50-foot intervals, however MDAs on instrument approaches are in increments of 20 feet. It is possible to set an altimeter setting in millibars, however it requires changing the setting mode with using a pin or paperclip. The altimeters should indicate in 20-foot increments and allow the setting of either inches of mercury and Millibars simultaneously (separate windows for millibars and inches of mercury) There is no altitude bug available to set MDA/DA’s. This is standard equipment on most altimeters. Altitude Alerter The altitude alerter as installed on the PACER CRAG Block 25/30 is difficult to use. The adjustment knob should be below the display so the display is not obscured by the operator’s hand when making adjustments. The adjustment knob should have detents at even 1,000-foot increments to allow quick inputs. Consider a speed sensitive control. There is no integration between the VNAV FMS function and the altitude alerter. The FSAS is interfaced to the altitude alerter on Block 30 but requires manual toggling via the “MSN ALTR” between CLB/DES and CRZ to have the altitude alerter input altitude into the FSAS. To avoid confusion, it should be demonstratively impossible to have more than one adjustable altitude in the cockpit. In other words, as the crew is climbing and is cleared to a higher altitude, they should only have to set the altitude one place and that should be in the altitude alerter. This should automatically update the altitude everywhere required in the FMS/FSAS/Remote display line on the EADI. Clock It would almost impossible to design a clock more difficult to read than the clock installed on PACER CRAG. It is small and the hands and numbers are LCD on a very low contrast background. Furthermore, it is very poorly lit. It should be replaced by a large, easy to read clock. Ideally, the clock chronometer feature should be operable via a button on the yoke or side panel. It should display UTC time obtained fro the GPS receiver. This would eliminate the requirement to display the time on the already task saturated FMS annunciation line

PACER CRAG Operations
12

Preflight In general, the PACER CRAG preflight is complicated and relies excessively on rote memorization and a lengthy, detailed checklist list. The preflight of a PACER CRAG aircraft takes more time than the unmodified aircraft due to all the additional avionics checks. This reduces the crew duty day and mission capability. 1. The aircraft model, engine type, brake and antiskid configuration, and navigation database currency dates should be all on one page, and this should be the first page accessed. Changing the navigation database should delete all route data for the ACTive flight plan. 2. Tests of the IFF, TCAS, FSAS, navigation sensors should be performed automatically at power up, with a reply only if there is a problem. 3. There should be prompts to lead the pilot through the preflight, some sort of indication of items remaining to be accomplished, and some means to indicate that the preflight is complete. 4. There should be a means to differentiate mandatory entries from optional entries. On the Honeywell FMS, mandatory entries are shown with box prompts. 5. This is not normally an issue, because use of the GPS position to initialize the INUs works well on Block 25/30 PACER CRAG. To avoid gross navigation errors were GPS is not available, if the operator attempts to input a position different from the last FMC position (say by 1 NM or so), the insertion of this new position is most likely in error and should require a confirmation. To minimize the possibility of keyboard errors when entering a lat-long, perhaps it should be possible to initiate the navigation system via an ICAO airport identifier. As an example of the complexity involved in the PACER CRAG preflight, consider the steps required to transfer a flight plan from the PCMCIA data card to the active flight plan. This is something required on nearly every sortie. This is a two step process: First step; transfer flight plan from the PCMCIA card to an alternate flight plan in the FMS: 1. IDX START  “Start 3 fpln/load” page 2. Select “altn catlg” via LSK 2R 3. Scroll through alternate catalog and select desired fpln via corresponding LSK. 4. Select “fpln replace”. Note: This transfers the flight plan from the data cartridge to one of the alternate flight plans. Despite the words “fpln replace” it does not replace the active flight plan. Second step; replace the active flight plan with the alternate flight plan: 1. EDIT ALTN FLPN (LSK 1L) 2. Replace FPLN via LSK 3L Use of the PCMCIA data card does simplify the preflight. Its use should be expanded to allow the CLB, CRZ, DES and weight and Balance data to be retrieved in addition to the route data. Inflight There is no means of displaying MCT/NRT thrust inflight. For the R-model, it is possible to display MCT by selecting MCT CLB (usual mode is MCL) and being in an FSAS “CLB” mode (aircraft must be in a climb). Otherwise, obtaining an MCT or NRT thrust requires the use of the Checklist or Performance manual Racetrack patterns require the input of true heading for the inbound course. Since there is a magnetic variation database, the crew should be provided the option of selecting either true or magnetic. The default should be magnetic, since normally every other aspect of the FMS operation is referenced to magnetic heading.

13

Holding. Provide an option for using timing to determine leg length in addition to distance. The FMS should be aware of the aircraft altitude, the default leg timing and holding airspeed should be appropriate for aircraft’s altitude. The FMS should also calculate holding endurance until reaching reserve fuel, and until empty. These should be available without pilot action. Hold speed should automatically be selected on the CRZ page when a hold is active. Of course, this default holding speed should be able to be manually overridden, and the title of the CRZ page should change to reflect this. If the manually entered speed is deleted, the default speed should return. The hold page should calculate the holding time available. The hold page should refer to the waypoint by name, not by number. Currently, the hold page displays “Hold waypoint 23,” which requires crosschecking with the fpln page Position reports – Should be able to read the position report directly from the screen. Oceanic position reports are given in the following format: Last Waypoint, Time, Altitude, Next waypoint, ETA, and Following waypoint. The current page does not include an altitude display. This requires the pilot to remember the sequence of the report add in the aircraft’s altitude. An oceanic position report in the proper format on the FMS page will facilitate datalink operations since the information could be encoded for transmission directly.

Approach The “LOC” annunciation changes to “B/C” (back course) when the aircraft heading is more than 110  from the selected course. This is confusing, for example, when briefing the ILS on downwind and “B/C” is displayed. Suggest that all ILS/LOC approaches be assumed to be front course, and require manual selection by some means to select a planned back course approach. On INSTRUMENT DEPARTURE PROCEEDURE/STAR pages, the runways are not listed in numerical order. Runways should be listed in ascending order (01L, 01R, 19L, 19R, etc.) To insert an ATC coded instrument departure or arrival procedure into the flight plan, the pilot must go through the TRANSistion LSK, even if none is required or available in order to get to the FLPN INSR prompt. This is a direct result of the FMS CDU display limitation.

14

What should be done in a KC-135 Cockpit Upgrade
The KC-135 flight deck should upgraded with an integrated, coherent design. This must be pursued in order to keep the workload suitable for a two-person flight deck. This cockpit should be designed for the minimally qualified crew to effectively perform the mission in the weather, in formation, and allow enough of the pilots’ capability to be spared so they can deal effectively with malfunctions and emergencies. The cockpit upgrade should automate the many of the tanker pilot’s tasks where it makes sense to do so with a consistent, integrated, automation philosophy. In the past, C-135 system upgrades such as FSA/CAS and PACER CRAG were bought for reliability/maintainability improvements, rather than to improve mission utility or to improve safety. The past KC-135 upgrades were purchased as individual components without regard for an integrated, coherent flight deck design. The need to replace aging and expensive to replace components, rapid changes in the airspace system, and the Air Force institutional need to eliminate the navigator position on the aircraft are driving the current round of KC-135 cockpit upgrades. This sort of upgrade is not unprecedented in aviation; many examples exist in the commercial sector. These commercial upgrades also give an idea of the scope of work involved in upgrading an older aircraft: DC-10 to MD-11 or MD-10; both eliminated a crew position resulting in a two-pilot flight deck B-747 Classic to B-747-400 eliminated a crew position resulting in a two-pilot flight deck B-737-200 to B-737-400/500/600/NG; technology update/automation increase DC-9 to MD-80/90; technology update/automation increase MD-80/90 series to B-717; technology update/automation increase There are several advantages to be gained from using commercial systems that have a large, installed base. Over half of the KC-135 tanker crew force is in the Reserve or Guard, and many of these folks operate modern glass cockpit aircraft for the airlines. There are safety, training, and operational advantages to having systems that are similar to what the Guardsmen/Reservists operate in their civilian jobs. A large, diverse, installed base means that these systems: 1. Will not become orphans, unique to military, to be supported at great cost. It is interesting to note that the KC-135 will soon be the only major USAF platform left with the FMS-800 as it is being abandoned by the C-5, RC-135, and KC-10. 2. Allows the military to take advantage of civilian innovations, such as FANS, ACARS, and CPDLC, which are implemented on the large base of existing equipment. With a military unique FMS, these types of “applications” have to be ported at great expense and delay to the sometime less capable hardware. These adaptations are usually clumsy in their human factors. 3. Benefit from the combined operating experience in ferreting out problems, errors, bugs, etc, which plague any complex system 4. Worldwide parts logistics network The Upgrade requirements: Rather than recommending specific hardware, this section outlines the capabilities that the upgraded cockpit should have. Each feature is justified based on increasing operational capability, improving safety, or reducing workload to a level appropriate to a two-person flight deck.

15

Physical changes to the flight deck: Physical changes to the KC-135 cockpit will be required for any redesign/upgrade. The following general principals should be employed: 1.Avoid duplicating controls. An example is the PACER CRAG radar controls. Duplicating controls wastes valuable space. 2. Controls and displays for each system should be grouped together. The current hydraulic controls and displays are located all over the KC-135 cockpit. This makes it much more difficult to monitor these systems. 3. Controls should be located immediately adjacent to the display or system they control. As an example, the PACER CRAG controls for the center TCAS display are located in the FMS, some are controlled via the radar/display control panel. The TCAS range is controlled via the EHSI “Data” knob, which has an entirely different function when used with the EHSI. 4. Important controls and displays should be accessible to both pilots. An example is the engine antiice controls that are not visible to the left seat pilot. By making these controls visible, the situational awareness of both pilots is increased. 5. There should be a direct indication of system operation. For example, if fuel panel valve position does not agree with commanded position, there should be a “DISAGREE” indication. There should be lights or direct indications of electrical buses that are unpowered. This is necessary to improve the “Observability” of the aircraft systems and reduce the amount of mental energy required to determine operating status, especially in the presence of malfunctions. This especially true with respect to autopilot/flight director operation. The combined autopilot/flight director/autothrottle control should be moved to the glareshield between the pilots (where the fire switches and radar controls are currently located). This will allow heads up control of the aircraft’s flight path by either pilot. The fire switches and extinguisher panel should be moved to a less obtrusive location, and fire warning given via a bell or aural alert. Standard transport category crew seats are narrower than the unique seats fitted on the KC-135. New seats would allow additional space between the seats to allow the center pedestal to be expanded as well as improving crew comfort. Seat belts/shoulder harness that allow only the shoulder harnesses to be released without releasing the lap belt should be installed for additional crew comfort. Expand the center pedestal behind the throttle quadrant, This would provide space for a set of weather radar controls, communication and navigation radio control heads and interphone panels. Placing the interphone control panels between the pilots will allow the pilot doing the radio communications at a glance to determine what the other pilot is listening to and greatly simplify crew coordination. Freed of the communication and navigation radio control heads and autopilot, the overhead panel should be dedicated to system controls/displays. All of the controls and displays for the hydraulics, electrics, anti-ice, pressurization, lighting should be placed on the overhead panel where it would accessible by both pilots. The Environment These are the individual piloting tasks facing the tanker pilot. Each should be automated and simplified: 1. Fly 2. Navigate 3. Communicate between crewmembers 4. Communicate outside the aircraft 5. Station keeping in formation 6. Monitor weather 7. Monitor aircraft systems

16

Specific recommendations: Fly: An improved, integrated autoflight (autopilot/flight director/autothrottle) system is required to reduce the workload of the crew during high workload situations, such as formation departures in the weather or when dealing with aircraft emergencies/non-normal situations. In general, there should be distinct levels of automation available to the crew: 1. Hand flown 2. Low level automation, flown via autopilot (heading select, altitude hold, IAS hold, etc. 3. High level automation, programmed via the FMS (LNAV, VNAV, RTA) The autopilot and flight director modes and controls should be combined so there is only one set of controls for both the flight director and autopilot. Flying the flight director should result in the same flight path that would result if the autopilot were engaged. The Speckled Trout aircraft has installed the B-737 Mode control panel (MCP) and implements most of the functionality described below with the existing C-135 autopilot servos. Therefore, the autoflight systems could be implemented as a modification to the existing autopilot system. This autoflight system should have the following capabilities: 1. Autothrottles with the following modes: a. N1 mode (drive to a limit N1 determined by a dedicated thrust rating computer or the FMC). This should allow the crew to select a power setting (Takeoff/GO-around, Climb, or MCT) and have the autothrottle set and maintain the power setting. This will reduce the crew workload, especially during departures by eliminating the need to look up power settings in a checklist and then manually set the throttles. This results in heads down time during a critical phase of flight. The current PACER CRAG FMC/FSAS will provide MCT/MCL power settings only after complicated programming. The limit thrust setting should be displayed directly on the N 1 indicator. b. A speed mode. Used for cruise at constant altitude or vertical speed, with the autothrottle controlling speed. There should be an indication of commanded speed in the airspeed indicator. Reduces workload, will save fuel due to mode precise control of airspeed. Autothrottles could make large thrust changes by back driving the throttles, and make small changes without back driving the throttles using the authority of the PMCs. This would minimize thrust overshoots cause by friction in the autothrottle system characteristic of fully mechanical autothrottle systems and provide better fuel economy by reducing throttle excursions. Speed control will be more precise and consistently improving the ability to maintain formation position. 2. Autopilot pitch modes: a. IAS/Mach hold. For use during climbs and descents, when speed is controlled via the elevator. Again, more precise formation position and reduced workload for the pilots. b. Vertical speed/altitude hold mode. The current autopilot’s basic pitch mode of pitch attitude command should be replaced with a vertical speed command mode. This is really what the pilot is interested in controlling. With the current system the pilot performs the role of a feedback controller, adjusting the pitch attitude in an attempt to maintain either an airspeed or descent rate. Note that a vertical speed command of zero is the same as altitude hold. c. It should be possible to set altitudes in an altitude window on the autopilot control head. The combined autothrottle/autopilot system should be able to transition from climbs or descents and capture an altitude. This transition should be automatic and not require manual intervention. The autothrottle/autopilot would transition from an autothrottle N1 /autopilot IAS hold to an autothrottle speed/autopilot altitude hold at an appropriate altitude to capture the selected altitude. d. Glideslope tracking mode: Should be able to armed from an altitude hold mode and automatically capture and track the glideslope. The current autopilot glideslope mode is adequate.

17

e. Takeoff/go-around mode: The takeoff mode should be flight director only. The current flight director MAX takeoff mode is adequate. The go around mode should be an autopilot mode in addition to a flight director mode. f. All pitch modes should provide protection from stall. The installed AOA system could be easily adapted for this role. 3. Autopilot roll modes a. Heading select: Move the two heading select knobs from the individual HSI’s and combine them in a single heading select knob on the autopilot control panel. The currently implemented heading select feature with multiple heading control methods should be eliminated. b. Heading hold: Provide a means for heading hold, having the autopilot roll out on heading when heading hold is selected c. FMS Navigation: Should be able to be armed from heading select mode and automatically capture the selected course. Follow the FMS programmed route. d. VOR/TACAN/LOC navigation: Should be able to be armed from heading select mode and automatically capture the selected course. Present autopilot mode is adequate, except for the inability to arm them from heading select. Flight instruments should be provided by two 8-inch by 8-inch displays at each pilot position. This is the de facto standard for automated transport aircraft. The top display should be configured as a primary flight display (PFD), which would display aircraft attitude, airspeed, altitude, heading and vertical speed. The bottom display should be configured as a navigation display (ND). All flight director/autopilot/autothrottle modes should be displayed on flight mode annunicators (FMAs). As a minimum, there should be one area where autothrottle modes are displayed, one area for pitch modes, one area for roll modes, and an area for autopilot/flight director status. Three sets of independent attitude, airspeed and altitude should be plainly visible to both pilots. This will allow rapid and unequivocal determination of a bad instrument. The current dual system does not allow rapid determination of a bad instrument since there is no way to break the tie. The current system of attitude comparison (COMPAR WARN) provides adequate redundancy to identify a bad attitude source. There is no redundancy for airspeed or altitude. It must be possible for the crew to quickly and intuitively determine which is the bad instrument and determine which are the good ones, and then transition to the good ones. Three sets of instruments provide visual redundancy and quick determination, which allows rapid recovery. Navigate: A flight management computer (FMC) compatible with a large installed civil base should be included in the cockpit upgrade. Practically speaking, this means one capable with the Boeing glass aircraft should be installed. An example of an after market FMS is the Honeywell Trimble HT-9100. This will ensure compatibility with future airspace upgrades. Move the course selectors and VHF navigation radio heads to the glareshield on either side of the autopilot/flight director mode control panel. This will allow heads up tuning on Navaids. Since anytime you are flying in relation to the ground, such as tracking a VOR or LOC course, you are really flying a track (heading corrected for wind drift), the pilot’s workload would be reduced by providing a means of flying a “track” One way to do this is to provide a means of displaying a “Track up” navigation display. Then the heading select can be used to align the aircraft’s desired and actual track. Another , more advanced means would be to provide a mode that allows the aircraft’s track to be directly commanded.

18

Install dual TACANs, one with beacon inverse mode. This allows complex arrival and departure procedures to be flown, reducing the number of navaid changes Install an ILS receiver that can receive ILS signals that end in .05 MHz, i.e., 110.05, 111.05, etc. Currently, the KC-135 cannot fly ILS’s with these frequencies. This fact is not documented in the flight manual, despite the fact that the ILS head will physically display 110.05, 111.05. Communications between crewmembers: Currently, the combination of high noise levels and a poor intercom (inability to individually adjust radio volumes, blank out of incoming radio calls when using the flight interphone) make intercockpit communications very difficult. In a two person cockpit these barriers to communication must be reduced to allow more efficient and less error prone communication. Suggestions: 1. Soundproof the cockpit to a level that allows cross-cockpit voice communications without using the interphone. This should be given a high priority, and has several benefits: a. Headsets no longer required to provide noise attenuation. This would allow lightweight headsets and use of cross-cockpit voice communication. Also, the lightweight headsets are much less fatiguing to wear. b. Opens up the use of aural alerts. The fire switches should be moved from their current prime location on the glareshield to an unobtrusive position on the overhead panel or aisle stand. This would allow the frequently used autoflight/flight director controls to be positioned on the glareshield where they can be used in a heads up fashion. The use of a fire bell or other aural alert is much less likely to be missed than the fire light. The crews get conditioned to not looking at the fire lights since they are so infrequently activated. Aural alerts, such as windshear warning, fire, excessive cabin altitude, etc., should be independent of the interphone system. c. Move the interphone controls from their current location outboard of the pilot’s and co-pilot’s seating position to the aisle stand between the pilots. This will allow the pilot doing the radio communications at a glance to determine what the other pilot is listening to. Currently, the interphone panel is not visible to the other pilot, so the pilot had no idea what the other pilot has selected for receiving or transmitting. The current system requires coordination over the flight interphone anytime a radio is changed. This discipline frequently breaks down even with experienced crewmembers, resulting in missed radio calls and less situation awareness. d. Provide individual and independent volume controls for each source selectable as a receiver. The new interphone allows each source to be adjusted independently, but through a common volume knob. This makes it impossible to determine at a glance which radios are set at which volume. Communications with outside: Provide all communication radios with an active and inactive manual frequency window. This allows a new frequency to be selected without obliterating the frequency previously in use. The ability to rapidly return to the frequency previously in use drastically reduces the confusion and heads down time in the event the new frequency is incorrect. The active/inactive frequency windows also allow anticipated frequencies to be preset, reducing workload during departures and descents. Also a means should be retained to store preset frequencies. Station Keeping: Fit dual TACANs, one with beacon inverse mode (for air-to-air bearing). This configuration is fitted on the KC-10 and Speckled Trout. Allows daisy chaining in formation, provides bearing and distance in formation, resulting in much more precise station keeping. Provides position without requiring the undivided attention of a crewmember, as does the use of radar. Makes it easier for the receiver aircraft to find the tanker aircraft. An alternative would be to create a new station keeping system, perhaps based on the station keeping equipment (SKE) equipment used on C-17s, C-141s, and C-5s.

19

Autoflight systems that can fly precise airspeeds and vertical speeds will result in much more precise flight path control and better formation flying. Monitor the aircraft systems: The current aircraft has systems indications for quantities, temperatures that require constant monitoring and are difficult to read (try reading the TR loadmeters or AC load meters quickly). A comprehensive and somewhat automated means needs to be fitted to monitor the aircraft systems. Care needs to be given to avoid an excessive level of false or inappropriate alerts. Most tanker pilots have been conditioned to cancel the “COMPAR” warning and landing gear aural alert without first considering the reason for the alert. This makes these systems much less useful as warning devices. Each malfunction requiring action by the crew should be clearly and unambiguously communicated to the crew without the need to monitor quantities, voltages, etc. System status must also be clearly and unambiguously communicated to the crew (for example, valves whose positions disagree with selected position, electrical busses that are unpowered). The current system will result in the problems being overlooked by a busy crew. There should be centralized warning and caution alerting system. This system could take several forms. There must be a single master warning and master caution annunicator in the pilot’s direct line of sight (glareshield) and preferably an accompanying aural alert. The master warning /master caution annunciation should able to be reset quickly in order for it to capture additional events and quickly and efficiently guide the pilot’s attention to the appropriate indication. This could take several forms: 1. An EICAS (Engine indicating and crew alerting system), EAD (Engine and alert display) or ECAM (Electronic Condition and aircraft monitor). Provides computer-generated messages of problem. Usually several levels of alerts (Warnings, cautions, alerts) a with priority scheme. This is the current practice of automated two cockpit aircraft in production. 2. A centralized warning light panel. As is done on the DC-9, with all the system annunicator lights on a single panel. 3. Cue lights. Associated with the master warning/master caution light is a cue light, which directs the pilot’s attention to the appropriate system. This is the scheme used on the B-737-100/200 and for the flight engineer on the DC-10. This allows the individual system lights to be located next to the system controls. Engine Instruments: All of the engine instruments should be in visible from the design eye point. This is currently not the case with the PACER CRAG cockpit. The glareshield blocks the view of the N 1 gages, the primary engine instrument. The aircraft systems controls and displays should be physically grouped together in a logical fashion where they are visible/accessible by both pilots. The overhead panel is the best place for this. Hydraulics: All hydraulic controls and indications should be grouped together on the overhead panel so they are accessible by both pilots. Automated warning should be provided for low and rapid loss of quantity, as well as low system pressure Electrics. Individual warning lights should be provided for when busses are unpowered. For example, there should be light/EICAS Message that illuminates “TR Buss 2 OFF” along with triggering the master caution when TR bus 2 has lost power. This eliminates the need to monitor the TR voltage and loadmeters. Fuel. Disagree lights/EICAS Messages should be provided for all valves when the position of the fuel valve disagrees with the commanded switch position. The controls on the ground refueling panel currently located in the wheel well should be duplicated in the cockpit and integrated with the current fuel panel. This would allow more flexibility in transferring fuel in-flight.

20

Pressurization: A cabin rate of change indication should be provided. An aural warning independent of the interphone system should be provided when the cabin altitude approaches/exceeds 10,000 feet. Lighting: All lighting should be controlled from the pilot’s station (move lighting controls for the dome light from navigator’s station.) The lighting controls for common areas (overhead panel, engine instruments, center pedestal) should be accessible by both pilots. Every annunicator light should be on a single test/bright switch. Anti-ice: Engine anti-ice, pitot heat, q-inlet heat controls should be moved to the overhead panel where they would be accessible by both pilots. A means should be provided to verify pitot heat operation. Landing gear warning systems. The current landing gear warning conditions the pilots to ignore it because it goes off every time the power is reduced to idle. A landing gear warning horn should be installed that only goes off when appropriate. Some suggestions: Below a certain radio altitude, power near idle, and landing gear not down and locked Flaps 40 or 50, landing gear not down and locked Miscellaneous The APU or operating engines to provide sufficient bleed air to run an air cycle machine on the ground. Ground air conditioning is currently not practical because of the use of ground ejector air to cool the heat exchanger in the air cycle machine is very inefficient. The ineffective ground air ejector system should be replaced with electrical fans of sufficient strength to provide adequate airflow over the heat exchanger. To provide adequate air for crew comfort and to eliminate fatigue, two air cycle machines should be provided. Some of the special purpose C-135 variants are currently equipped in this fashion. Change “Do-list” philosophy to increased use of flows and checklists. Print normal checklist on a single card to allow easy updating and use in the cockpit. The first section of the interior inspection checklist should be separated into a Pre-electrical power application safety check. This checklist would ensure that it was safe to apply electrical power prior to the aircraft. Once electrical power was on, then the “Interior inspection” checklist would be run. After takeoff, the checklist format should be PNF/PF rather pilot/copilot since this more accurately reflects the way the aircraft is operated. Integrate the required approach and before landing checklists into the non-normal checklists when appropriate and provide required performance data. As an example, consider the loss of right hydraulic system. This non-normal requires running separate checklists to identify and attempt to mitigate the leak, then separate checklists for each of the inoperative components. Then performance data must be computed and normal checklists run. This very time consuming and error prone even for experienced crews, and requires the sustained, undivided attention of one crewmember. It also requires use of the full flight manual and performance manual. Consider eliminating boldface memory items for engine fire/shutdown and replace with a critical action card. United Airlines investigated the error rates for performing critical action items both from memory and by reference to a card. Use of a card resulted in 80% fewer errors and a negligible increase in the amount of time required to execute the procedures. The advantages of this approach could be easily verified in a flight simulator. Boldface should be reserved for items such as aborts, GPWS/windshear warnings. Get rid of checklists items that instruct the pilot to do instinctive things such as rotate and raise the landing gear such is the case in the “Engine Failure, Takeoff Continued” checklist. For abnormal/emergencies where aircraft performance is compromised, the appropriate checklist should direct that the TCAS be put in “TA” mode, since the TCAS computer assumes full aircraft performance when issuing resolution advisories.

21

Should have Technical Order procedures for GPWS alerts and PWS microburst/windshear alerts: GPWS Warning suggestion: AUTOPILOT/AUTOTHROTTLES – DISCONNECT THROTTLES – FULL FORWARD PITCH – ROTATE TO 20 OR GREATER SPEEDBRAKES – RETRACTED Microburst/windshear Warning suggestion: AUTOPILOT/AUTOTHROTTLES – DISCONNECT THROTTLES – FULL FORWARD PITCH – ROTATE TO 15 SPEEDBRAKES – RETRACTED Improve standby power system so left seat pilot has instruments, communication and navigation radios and the ability to shoot and instrument approach with no AC power. Currently, with loss of all generators, the right seat pilot has instruments, and neither pilot has navigation radio capability. A single UHF transceiver is available on DC power. Hydraulics upgrade: Move to inboard (formerly the right)/outboard (formerly the left) systems. Replace crossover valve with power transfer unit/reversible motor pump that would allow the inboard system to retract the landing gear and the outboard system to power the rudder. This would vastly decrease the probability of an outboard engine failure without the powered rudder operating. The crossover systems is nearly useless as currently implemented. To be used, it requires depressurization of both hydraulic systems, resulting in an unacceptable degradation in the flight controls (loss of both powered rudder and spoilers). Flap system. Replace the manual crank system with an electrically powered alternate flap operation system. Such a system is already installed in some C-135 models. The manual system takes to long to use and is not consistent with a two-person cockpit. Operations/Takeoff data For standardization with civil aircraft, change the “S 1” nomenclature to “V1 ”. Do systems safety analysis to determine if biasing the takeoff decision speeds towards going rather than stopping would improve safety. Currently, the KC-135 takeoff data is biased towards stopping (high decision speed). This is probably a legacy from the J-57 days when engine out takeoff performance was more critical. With the R-model, much more climb capability exists. The risks from a high speed, heavy weight abort are probably greater than a continued takeoff in many cases with the current decision speeds. Make rotate speed VR instead of VROT . Currently, refusal speed is abbreviated VR. This is confusing to pilots who operate civil certified aircraft. More than once someone has mistaken refusal speed for rotate speed. On a heavy weight takeoff, this could be a problem since refusal speed can be much lower than rotate speed. An early rotation could result in a tail strike and marginal acceleration. For standardization with civil aircraft and international operation, incorporate the standard ICAO distant and close in noise abatement profiles into the KC-135 flight manual. There is no physical or operational reason not to do so. Eliminate the FD-109 ACCL mode takeoff. With the increase in takeoff performance, with the R and E model tanker variants, it is not needed. Training An automation training program should be created which is based on the automation philosophy of the cockpit upgrade. This program should emphasize the use of the appropriate level of automation based on phase of flight and automation pitfalls and problems.

22