Category Archives: USAF

Cutaway Thursday: Convair B-36J Peacemaker

20140521-155441-57281211.jpg

Advertisements

3 Comments

Filed under Aeronautical Engineering, Air Force, Airplanes, Flying, History, Plane Pr0n, USAF

Iron Birds

Static test airframes, or more commonly called, “iron birds” are partially built, non-flying airframes or old formerly flying airframes that are used by agencies and manufacterers to test either the strength of than airframe, various design components or aircraft subsystems (avionics, flight control, engines, etc).

The iron birds used for strength testing are typically full scale representations of the aircraft that are rigged to gaint gantry cranes with weights and strain gauges attached. See the pic:

Lockheed's F-35 test airframe installed on gantry cranes with strain gauges.

Lockheed’s F-35 test airframe installed on gantry cranes with strain gauges.

Once installed on the cranes the airframe is literally pulled and pushed to properly simulate all the aerodynamic forces that the aircraft will encounter throughout it’s flying career.  Often the iron birds are tested till destruction.

This is a VC-10 undergoing wing fatigue testing. Note the bending wing.

This is a VC-10 undergoing wing fatigue testing. Note the bending wing.

Some iron birds are formerly flying airframes that have accumulated too many flying hours and are no longer consider safe to fly. These aircraft are typically stripped of most equipment (engines mostly) and used to test various aircraft subsystems in support of other programs.

This NASA's F-8 Crusader iron bird that was used to test software for NASA's Digital Fly-By-Wire program in the 1960s,

This NASA’s F-8 Crusader iron bird that was used to test software for NASA’s Digital Fly-By-Wire program in the 1960s,

 

As the latest example of NASA's iron bird, this is an F/A-18 Hornet used by NASA to support many of the F/A-18 test programs.

As the latest example of NASA’s iron bird, this is an F/A-18 Hornet used by NASA to support many of the F/A-18 test programs.

Iron birds aren’t limited to NASA. The US military also used them for the same purposes.

This B-2 at the National Museum of the USAF was never an actual flying airframe. This "aircraft" appropriately named "Fire and Ice"was used for fatgiue and climatic testing.

This B-2 at the National Museum of the USAF was never an actual flying airframe. This “aircraft” appropriately named “Fire and Ice”was used for fatgiue and climatic testing.

A close up of "Fire and Ice's" nose gear door.

A close up of “Fire and Ice’s” nose gear door.

You can learn more about that particular aircraft here.

As an aside, old airframes are also typically used as maintaince trainers in the military. These are called ground instructional airframes:

images 080613-F-1322C-001

 

 

4 Comments

Filed under Aeronautical Engineering, Air Force, Airplanes, Flight simulation, Flying, Naval Aviation, Other Stuff, Outside the Box, USAF

Cutaway Thursday: Lockheed F-104 Starfighter

My all time favorite of the “Century Series” fighters, the first flight of the F-104 was this week, on 4 March 1954. Strangely, it had a relatively short career with the USAF but enjoyed far more success with NATO countries.

f104g_cutaway

You can learn more here at the International F-104 Society’s webpage.

1 Comment

Filed under Aeronautical Engineering, Air Force, Airplanes, USAF

Cutaway Thursday: Douglas X-3 Stiletto

x-3

More on the X-3 Stiletto here.

2 Comments

Filed under Aeronautical Engineering, Air Force, Airplanes, Flying, USAF

USAF MiG-21F-13s at Tonopah.

Before you read on if you haven’t been in-briefed on Have Doughnut go read.

MiG-21f-13 "Fishbed-C". Image credit Wikipedia.

MiG-21f-13 “Fishbed-C”. Image credit Wikipedia.

Have Doughnut was the combat evaluation of the Soviet-built MIG-21F-13 (NATO codenamed) “Fishbed-C.” The F-13 version of the -21 is an early version of the ubiquitous MiG-21 family. It was the first short range day fighter version of the -21 to be placed in mass production and was the first variant to use the K-13 air-to-air-missile. The -21F13 was also in service with the North Vietnamese Air Force and saw regular combat against USAF and USN combat aircraft in-theater.

Constant Peg was a training program that took Have Doughnut a step further an under the auspicies of the 4477th Test and Evaluation Force. The “Red Eagles” as they became known, equipped with the MiG-21F-13,  provided a somewhat formalized training environment, for both USAF and USN fighter pilots that saw the MiG as the primary threat aircraft.

From the Wikipedia page:

By the late 1970s, United States MiG operations were undergoing another change. In the late 1960s, the MiG-17 and MiG-21F were still frontline aircraft. A decade later, they had been superseded by later-model MiG-21s and new aircraft, such as the MiG-23. Fortunately, a new source of supply of Soviet aircraft became available, Egypt. In the mid-1970s, relations between Egypt and the Soviet Union had become strained, and Soviet advisers were ordered out. The Soviets had provided the Egyptian air force with MiGs since the mid-1950s. Now, with their traditional source out of the picture, the Egyptians began looking west. They turned to United States companies for parts to support their late-model MiG-21s and MiG-23s. Very soon, a deal was made. According to one account, two MiG-23 fighter bombers were given to the United States by Egyptian president Anwar Sadat. The planes were disassembled and shipped from Egypt to Edwards Air Force Base. They were then transferred initially to Groom Lake for reassembly and study.[2]

In 1987, the U.S. Air Force bought 12 new Shenyang F-7Bs from China for use in the Constant Peg program. At the same time, it retired the remaining MiG-21F-13 Fishbeds acquired from Indonesia.[citation needed][3]

The United States operated MiGs received special designations. There was the practical problem of what to call the aircraft. This was solved by giving them numbers in the Century Series. The MiG-21s and Shenyang F-7Bs were called the “YF-110” (the original designation for the USAF F-4C), while the MiG-23s were called the “YF-113”.[2]

The focus of Air Force Systems Command (AFSC) limited the use of the fighter as a tool with which to train the front line tactical fighter pilots.[1] Air Force Systems Command recruited its pilots from the Air Force Flight Test Center at Edwards Air Force Base, California, who were usually graduates from either the Air Force Test Pilot School at Edwards or the Naval Test Pilot School at NAS Patuxent River, Maryland. Tactical Air Command selected its pilots primarily from the ranks of the Weapons School graduates at Nellis AFB.[1]

The 4477th began as the 4477th Test and Evaluation Flight (4477 TEF), which began 17 July 1979. The name was later changed to the 4477th Test and Evaluation Squadron (4477 TES) in 1980. The 4477th began with three MiGs: two MiG-17Fs and a MiG-21 loaned by Israel, who had captured them from the Syrian Air Force and Iraqi Air Force. Later, it added MiG-21s from the Indonesian Air Force.

Here are some newer photos of those MiG-21F-13s at the Tonopah Test Range sporting Soviet Air Force markings and cavorting about the Tonopah Test Range, probably in the 1970s.

20140205-085016.jpg

20140205-085053.jpg

20140205-085123.jpg

20140205-085157.jpg

20140205-085409.jpg

20140205-085440.jpg

20140205-085509.jpg

Several of these photos are probably in Steve Davies “Red Eagles” and Gail Peck’s “America’s Secret MiG Squadrons.” According to both of these books, after the aircraft reached the end of their useful lives, they mysteriously appeared at several museums throughout the United States. Reportedly, as the aircraft were dropped off, curators were told not to ask any questions about the aircraft.

If you haven’t read either “Red Eagles” or “America’s Secret MiG Squadrons” you should. You’ll get more information than what you read on the Wikipedia page.

Apparently, I have a cutaway drawing of the -21F-13 so if you’re wondering what it looks like under the skin, here ya go:

mig21f13

2 Comments

Filed under Air Force, Airplanes, Naval Aviation, Navy, Red Flag, Tomcats Forever, USAF, Vietnam

Rockwell’s HiMAT RPRV-870 (part 1)

The Rockwell International RPRV-8870 HiMAT.

The Rockwell International RPRV-8870 HiMAT.

Rockwell International’s RPRV-870 (Remote Piloted Research Vehicle) HiMAT (Highly Maneuverable Aircraft Technology) was a joint USAF/NASA program to develop advanced technology for future fighter aircraft. There were 2 aircraft flown for a total of 26 flights. The 0.44 scale RPVs tested advanced technology concepts like composite and metallic structures, digital integrated propulsion control, and ground and airborne relaxed stability digital fly-by-wire.

Many of the aircraft technologies that we use today came from HiMAT developments of the 1970s and 80s. HiMAT’s main advantage was that it was able to test several of these technologies in a synergistic manner, integrating as many of these technologies as possible. The program was operated and funded jointly by the USAF and NASA.

In August 1975, Rockwell International was awarded the contract to build 2 subscale unmanned vehicles. These were delivered to NASA’s Dryden Flight Research Center in March and June 1978.

HiMAT technologies and 3-view.

HiMAT technologies and 3-view.

One of the main research goals of the HiMAT program was to provide a mathematical foundation for the use of computational tools in comparative analysis of future aerodynamic and structural design and to provide high quality flight test data for correlation and comparison with flight test data. Goals for the vehicle itself included testing of advanced aerodynamic configurations and advanced design concepts like composites, metallic structures, a digital integrated propulsion system (IPCS) and ground to airborne digital fly-by-wire concepts.

Structurally, the HiMAT, design goals were to provide “compromises when necessary to integrate several new technologies into a system with real constraints that must allow a significant level of benefit of the technology to be realized.” As a result of this integration, designers realized that “late discoveries of these unpredictable problems may significantly impact cost and/or schedule performance.”

In turn,  the HiMAT vehicle was driven by several design goals and constraints:

1. Most importantly the design must integrate, to the highest fidelity possible the technologies mentioned above.

2. The configuration had to be representative of a realistic (future) air-to-air fighter.

3. The additional constraints of scaling effects and RPRV requirement.

4. The transonic sustained maneuver requirement of 8g’s coupled with efficient performance.

The overall structural approach was design a full-scale fighter that had the desired full technology integration and then develop the RPRV with minimum compromises. Keeping this in mind it was necessary to match thrust-to-weight (T/W) and wing loading (W/S) of a full-scale fighter so that equivalent maneuvering performance could be achieved with the RPRV. Another constraint imposed was the off the shelf availability of the J85-21 engine, so the best match in terms of size and performance was to be the 0.44 scale factor.

HiMAT fighter to RPRV comparison.

HiMAT fighter to RPRV comparison.

The design of HiMAT was also driven by 3 major factors:

1. Aerolastic tailoring to provide the aerodynamic required twist and camber distributions.

2. Maximum use of advanced materials, both metallic and nonmetallic.

3. Modular design so that components of the airplane could be replaced with alternate designs for research purposes.

HiMAT's modularity

HiMAT’s modularity

HiMAT's modularity

HiMAT’s modularity

The HiMAT design used several advanced materials and composite structure applications. HiMAT trade studies and fabrication experience dictated that graphite/epoxy be the material for primary and secondary lifting surface structures. Control and lifting surfaces were designed and fabricated to meet the program’s torsional stiffness requirements. This had the added benefit of saving 25 to 40% of what otherwise would have been the vehicle’s weight. Development of  HiMAT’s structure included the first utilization of the TSO composite design optimization computer program.*

HiMAT construction by material.

HiMAT construction by material.

A primary feature of the HiMAT, was its construction from advanced materials. The wing and canard were constructed from graphite-epoxy using a non-standard ply technique. This provided an aerodynamically beneficial spanwise aeroelasticity to reduce vehicle drag. The vehicles were also designed to fly in a relaxed stability configuration to further reduce drag. NASA established the goal that the HiMAT vehicle have the ability “to perform a sustained 8-g turn at Mach 0.9, at an altitude of 30,000 feet, with a mission radius of 300 miles.”

The RPRV (Remotely Piloted Research Vehicle), which the HiMAT was, had systems requirements that no single systems failure could result in the loss of the vehicle and the vehicle must fly statically unstable. This meant that the approach to flight control systems, vehicle systems, fault detection and failure management systems, and systems flight qualification all required an unheard of level of complexity. All of HiMAT’s systems could be divided into primary and backup. There were dual onboard microprocessor computers as well as dual electrical and hydraulic systems as well as duel flight control systems and redundant flight sensors.

The overall vehicle configuration was that of a “close coupled canard with a swept wing.” The control surfaces included twin all-movable twin rudders mounted on a boom. The rudders deflected symmetrically for yaw control and asymmetrically for speed brake control. The elevator was used for pitch control, elevons for pitch and roll control. The canard flaps were capable of pitch and sideforce control but were not used as such.

HiMAT gross weight was 3501lbs with 659lbs of fuel. Each vehicle was powered by a J85-21 turbojet engine rated at 5004lb of sea level static thrust. Engine control was provided through an IPCS in the backup computer. Vehicle subsystems were also powered by an engine-driven electrical generator with backup power supplied by a 35V silver-zinc battery. Primary hydraulic power was supplied by an engine-driven pump with a backup electrical pump. There were no single points of contact between the primary and backup hydraulic systems other than the dual tandem actuators for the rudders.

HiMAT's operational concept.

HiMAT’s operational concept.

HiMAT mounted on the pylon of NASA's B-52 mothership.

HiMAT mounted on the pylon of NASA’s B-52 mothership.

Flight tests were conducted with the HiMAT dropped from at B-52 mothership at about 45,000ft at Mach 0.68. The vehicle was placed on the right wing of the B-52 on a specially modified pylon. Typical missions lasted about 30 minutes. The pilot (or rather operator) sat in a fully instrumented, fixed base ground cockpit. The pilot would fly HiMAT with typical fighter-type controls, a center stick and left side throttle.

HiMAT cockpit.

HiMAT cockpit.

Control of HiMAT was maintained through ground control and uplink computers and aircraft response was downlinked to a ground station and indicated to the pilot via conventional aircraft instrumentation. A flight engineer sat next to the pilot and assisted the pilot with the conduct of maneuvers and navigation. NASA ground radar provided tracking, backup airspeed and altitude information. In the event of systems failure, the backup systems could be engaged by the chase aircraft (a NASA TF-104) or at the ground station.

HiMAT was equipped with skids for landing on the Rogers Dry Lakebed. Typical landing runs were 4500ft along the 15000ft runway.

Simplified diagram of HiMAT's different signal paths.

Under typical flight conditions HiMAT was flown via the Primary Control System (PCS). The PCS control laws ran from a ground-based Varian V-73 computer. In the event, the PCS failed, there was a Back-Up Control System (BCS) whose program was run either based on a failure of the PCS or could be engaged manually from the chase aircraft or the ground station. The BCS control laws were resident in the backup computer aboard the vehicle.

The downlink receiving station processed data from the primary on board computer to the flight test instrumentation system (FTIS) for transmission in a modulated data stream at 220HZ. The ground station then processed this data into usable words using a Varian V-77 computer.

Cockpit displays on the pilot’s fixed station included a forward-looking video monitor, attitude direction indicator (ADI), a radar altimeter, barometic altimeter, airspeed and Mach indicators, altitude rate, engine rpm, fuel flow, fuel quantity, and exhaust gas temperature (EGT), a computer select mode control (CSMC) box and pulse panel. As previously mentioned, the pilot’s interface to the PCS was through the standard fighter aircraft 3-axis controls made up of a stick, throttle lever, and rudder pedals. A speed brake switch was provided on top of the throttle lever. Interface to the BCS was via command switches located to the left and right of the main console, also in the pilot’s cockpit.

No vehicle data was provided to the chase TF-104 so the aircraft provided control and guidance visually.

The flight test engineer (FTE) communicated with the pilot for assistance with navigation, checklist and emergency procedures, pulse panel tests, system gain changes and landing emergency management. Navigation was done via a radar driven plot board showing the position in a track above the ground. Energy management during landing was also display on a radar driven slope plot board. Pulse panel inputs were initiated by the FTE and consisted of pre-programmed commands given to the pilot.

Three computers were used to perform the PCS control law, maneuver autopilot and navigational computation functions. These were the Varian V-73A, V-73B and V-72 computers, respectively. The vehicle downlinked sensor signals and were processed through the V-77 then combined with control inputs from the pilot via the V-73A computer (using the PCS control law commands). The V-73 also performed air-data calculations for display in the ground cockpit. Additional functions carried out by the V-73 was the automated prefilght program that tested all interface input/output.

The V-73 also had a mechanized flight test maneuver autopilot (FTMAP). The FTMAP “was designed to provide precise, repeatable control of HiMAT during certain prescribed maneuvers.” FTMAP acted as a “non-flight-critical” outer loop controller with the PCS. During FTMAP operation, the FTMAP computer replaced all pilot imputs. The FTMAP also passed landing guidance information to the cockpit’s ADI.

The V-72 received and decoded tracking radar data and calculated vehicle ground track information for display to the pilot. The V-72 also provided a limited ILS (Instrument Landing System) i.e. glideslope and localizer information for display in the cockpit.

The uplink encoder, combined signals from the pilot’s input and output from the V-73 computer before sending the commands to the vehicle. The encoder formatted to send data in four 16-bit words per frame at a rate of 106.6 frames per second. Two different frames were alternately sent for a total of 8 16-bit updated 53.3 times/second. The data itself was formatted as:

The first four words addressed vehicle decoder number one, and the last four words addressed vehicle decoder number two. The first 10 bits of each words were available only to the primary onboard computer and were designated proportional data. These proportional channels represented the PCS interface to the vehicle aerodynamic surfaces and throttle. The last 6 bits of each word were designated as manual command discretes and were hardwired directly to the encoder from cockpit switches. These discretes represented the pilot’s discrete interface to the onboard BCS and other vehicle systems.

In terms on airborne systems, the PCS required that both receivers and decoders be operational. Dual receivers-decoders received the uplink signal and provided the command input interface to dual onboard computers. Early in the program the vehicle experienced frequent automatic transfers from PCS to BCS because the either one of both of the receivers-decoders received inadequate uplink signals as a function of vehicle attitude. To alleviate this these nuisance transfers to the BCS, a diversity-combining concept was used in the hardware to prevent an interruption of the groundlink signal. The diversity combiner combined output signals of the dual receivers so regardless of vehicle orientation, the best continuous signal was available for uplink commands. As the diversity combining hardware was installed in vehicle number one, it eliminated nuisance transfers to the BCS.

Dual onboard custom built computers provided critical systems control for the HiMAT. These 2 computers were based on Intel 8080 microprocessors. and operated asynchronously but had identical computational and memory capacities:

Each computer contained 22,528 bytes (8 buts) of erasable, programmable, read only memory (ROM) and 1024 bytes of random access memory (RAM). Both computers were programmed entirely in 8080 assembly language and packaged in a common chassis with separate circuit cards and connectors. The dual computer chassis weighed 40lb and had a volume of 1198in3

The principal functions of the primary computer were:

1. uplink data processing.

2. downlink data processing.

3. failure detection for the computers, flight sensors, servoactuators, and power system (for both the PCS and BCS).

4. failure detection for the IPCS

The functions for the backup computer were:

1. uplink data processing.

2. primary IPCS.

3. BCS control laws.

The computers communicated with other vehicle systems via digital, analog and discrete channels. Each computer also had its uplink and downlink telemetry systems.

HiMAT had 7 redundant flight critical system sensor sets that provided the vehicle with a “fail-safe” capability. 5 of the sensor sets were triplex and the 2 other were duplex. The 7 sensor sets:

included the triplex three axis angular rate gyros, normal and lateral accelerometers, and the duplex air data system.

Air data rates were determined by an analog differentiation of the air data signals. A single sensor of each set was designated the backup to the BCS. A simplex all-attitude gyro and radar altimeter were provided but not deemed “flight critical.” The gyro provided data to the vehicle’s ADI and to a “direction cosine algorithm” in the BCS code. The radar altimeter provided data to the pilot at all altitudes below 5000ft during approach and provided critical input to the BCS in automatic landing mode. The BCS could even be used to land the vehicle even if the radar altimeter failed.

The servoactuator electronics (SAE) box provided an interface between the onboard computers and the control surface actuators as well as the engine nozzle. The SAE box functions included:

electrically closing all servo loops to the actuators, receiving all actuator command from the computers and feeding back all actuator positions to the computers.

Another primary function was failure detection of the elevon servoactuator system which was sensed faster in hardware than software. If a failure was detected the SAE would notify the primary computer.

Rounding out the airborne systems of the 2 HiMAT vehicles, is a flight test instrumentation system (FTIS). The FTIS processed all data to be downlinked into a PCM data stream. Inputs to the FTIS were both analog and digital. Some signals were downlinked directly through the FTIS and through the onboard computer. Some downlinked data was processed through the downlink processing routine in the FTIS. The downlink processing routine included midvalue selection of triplex sensors, synchronization logic and the packing of vehicle status and failure indication discretes. The routine formatted and processed data eighteen 10-bit proportional parameters and seven 10-but discrete words.

In part 2 we’ll discuss the HiMAT flight test program and results.

Another view of HiMAT at landing.

Another view of HiMAT at landing.

*The TSO (aerolastic tailoring and structural design optimization) computer program provides aerolastic tailoring information for lifting surfaces.

2 Comments

Filed under Aeronautical Engineering, Air Force, Airplanes, History, USAF

Cutaway Thursday: Convair F-106 Delta Dart

Today in 1956 was the first flight of Convair’s F-106 Delta Dart.

20131225-232314.jpg

3 Comments

Filed under Aeronautical Engineering, Air Force, Airplanes, Flying, Plane Pr0n, USAF