While the F-22 is unambiguously the most lethal air-to-air platform in existence, the F-22 was designed during the 1980s and 1990s under a different threat and technological environment. Namely the F-22’s antiquated internal computing capabilities, software, limited combat radius, and high maintenance requirements degrade the utility of the F-22 within the context of operating in the Asia-Pacific against increasingly capable great power threats. Part II will examine these deficiencies further in preparation for an analysis of what features an F-22C could include which would both correct these shortcomings and add new capabilities to the F-22 airframe in Part III.
1980s
Hardware & Software
Image 1: F-15C cockpit vs. F-22A. F-15C image courtesy of Eagle.RU forums.
The
avionics suite of the F-22 is among the most capable of any fighter in service
in terms of raw performance, the AN/APG-77 active electronically scanned array
(AESA) and ALR-94 radar warning receiver (RWR) provide unmatched active and
passive detection capabilities. Data collected from the F-22’s avionics suite
are fused and presented on six liquid crystal displays in the cockpit providing
unmatched situational awareness when compared to primarily analogue switches
and cathode ray tube based displays within 4th generation cockpits.
However, the original internal computing hardware and software that manages the F-22’s
avionics are obsolete.
Image 2: F-22 internal computing systems. Image Credit: F-22 avionics handbook, Ronald Brower, 2001.
Two Hughes
Electronics designed common integrated processors (CIP) provide the computing
backbone of the F-22 avionics and flight systems which enable dissemination of
radar, communication, electronic warfare, and systems data.[1] The CIP is a modular
design composed of 66 Standard Electronic Module Size – E (SEM-E) units each which
are in turn connected to Dual Data Processing Elements (DDPE) on each side of
the SEM-E units; the DDPEs feature two 32-bit, 25-MHz, Intel 80960 (i960)
processors which collectively provide the bulk of the F-22’s processing
capability to support its integrated avionics suite.[2] Polyalphaolefin liquid
coolant provides thermal management for both the CIP racks and AN/APG-77 radar.
Each CIP is capable of computing 10.5 billion calculations per second and have
a maximum memory capacity of just 300 megabytes each. The software which runs
the F-22’s hardware is equally dated.
Of the 1.7 million lines
of code responsible for running the F-22s various systems, 90% is written in
Ada - a prehistoric programming language developed in 1980. In a Wall Street
Journal editorial against the F-22 program, former Secretary of the Navy
John Lehman sarcastically remarked, “At least they [the F-22] are safe from cyberattack
since no one in China knows how to program the '83 vintage IBM software that
runs them”.[3]
Despite the limitations of the F-22’s current hardware and software, Lockheed
Martin engineers ensured the aircraft had significant growth margins to
accommodate future computing advances.
A
total of 19 SEM-E slots in CIP 1 and 22 SEM-E slots in CIP 2 are vacant to
facilitate future growth.[4] Furthermore, provisions
were made within the F-22 airframe to facilitate future incorporation of a
third CIP.[5] Production of the i960MX
ceased in 1997 and it’s likely that the CIP’s original hardware was upgraded,
but these upgrades are not well documented. Under the common configuration
program (CCP), Defense Industry Daily reports, “F-22A Block 10s were
retrofitted to Block 20/ Increment 2 status, but retain the original core processor [implying a new processor has
been fielded]”.[6]
In 2001, Military and Aerospace
Electronics, reported that PowerPC processors would be integrated into lot
5 production aircraft:
…an upgrade to a new PowerPC processor already is on
the drawing board, beginning with Lot 5 production of the aircraft around 2004…When
the time comes, designers say they expect to replace the signal processor with
a PowerPC using AltiVec technology, Motorola's high-performance vector parallel
processing expansion to the PowerPC RISC processor architecture. AltiVec adds a
128-bit vector execution unit operating in concert with the PowerPC's existing
integer and floating point units to provide highly parallel operations, as many
as 16 simultaneously in one clock cycle.
The full extent of the CIP’s upgrades are not apparent
from public sources, but it’s likely the original obsolescent parts were at
least partially replaced for sustainment purposes since Lot 5. Given the
additional sensors and networking capabilities envisioned in an F-22C, which
will be detailed in Part III, it’s likely the current baseline computing
hardware will require additional upgrades. Furthermore, the USAF ought to
examine the feasibility and relative utility of upgrading to a C++ or non-Ada
based operating system while also keeping cybersecurity in mind. The Integrated
Maintenance Information System (IMIS), the rough equivalent of AILIS for the
F-22, is currently being upgraded to the C++ standard.[7] The Air Force’s budget
materials for FY 2017 under “F-22 Small Projects” lists “Windows XP migration”
as a planned upgrade.[8]
[UPDATE 9/25/16]: Forecast International reports a third CIP was added on lot 5 production aircraft and beyond.
Range
Image 4: F-22A range comparison, the chart is somewhat biased against the F-15E given the HLLH configuration. Image Credit: Lockheed Martin.
Arguably the most
substantial limitation of the F-22 is its limited range. On internal stores
only, the F-22 has a subsonic combat radius of 590 nautical miles (nm). With
the addition of a pair of 500 gallon drop tanks, which are mounted from
detachable pylons on the wing to enable reestablishment of the F-22’s stealth
outline, is 850 nm. However, even a range of 850 nm is fairly limited when
compared to the vast geographic expanse of the Asia-Pacific. The original
advanced tactical fighter requirements were tailored to the strategic situation
of the Cold War in which the U.S. air campaign would be fought from a network of
bases in the U.K. and Western Europe which were comparatively close proximity
to Warsaw Pact forces.
In
order to both improve the relevance of the Raptor to the Asia-Pacific and
reduce the strain on aerial refueling assets during a high-end conflict, the
F-22C would add both variable cycle engines and conformal fuel tanks. Variable
cycle engines are likely to be among the defining traits of six generation
aircraft, provided such a platform centric approach is pursued, and provide
numerous performance benefits when compared to current turbofan engines:
To alter bypass ratio, variable-cycle engines add a
third airflow stream outside of both the standard bypass duct and core. The
third stream provides an extra source of airflow that, depending on the phase of
the mission, can be adapted to provide either additional mass flow for
increased propulsive efficiency and lower fuel burn, or to provide additional
core flow for higher thrust and cooling air for the hot section of the engine,
as well as to cool fuel, which provides a heat sink for aircraft systems.
During cruise, the third stream can also swallow excess air damming up around
the inlet, improving flow holding and reducing spillage drag.[9]
Variable cycle engines
have the potential to provide between 25% and 35% greater range and 10% greater
thrust when compared to traditional turbofan engines.[10] Furthermore, the third
stream of air provides additional heat sink capacity which would facilitate
both the incorporation of additional avionics – which often generate excessive
heat, and directed energy weapons.[11]
Image 5: F-22A drop tank test. Image Credit: Lockheed Martin.
The addition of conformal fuel tanks would greatly expand
the Raptor’s range at minimal cost to maneuverability, for example, the F-15’s Fuel
And Sensor Tactical (FAST) CFTs provide an additional 1,698 gallons of fuel
while the F-16C Block 52’s CFTs provide 900 additional gallons, and the
Advanced Super Hornet’s CTFs provide 3,000 pounds of additional fuel combined.
The addition of variable cycle engines and CTFs could expand the F-22B’s combat
radius to approximately 825 nm or greater than 1,180 nm with two drop tanks.[i] However, the addition of
CTFs would degrade the F-22’s stealth performance by virtue of disrupting the
careful balance of planform alignment, the process in which multiple flight
surfaces of an airframe share the same angle such that they reflect radar waves
way from the source; shaping techniques provide between 80-90% of radar cross
section (RCS) reductions while radar absorbent material (RAM) coatings provide
the remainder.
The relative utility of mounting CFTs for the F-22 would
depend upon the extent of RCS degradation and the expected threat environment.
For example, even if the CTFs would entirely negate the F-22’s stealth
characteristics, CTS would still be useful for ferry flights between distant
Pacific bases such as Joint Base Pearl-Hickam in Hawaii and Kadena Air Base in
Japan. However, it is unlikely the addition of CTFs would entirely negate the
F-22’s stealth when shaped appropriately and treated with RAM. Israeli
Aerospace Industries has explored adding CTFs to the F-35.[12] Similarly, both Boeing’s
Advanced Super Hornet and Silent Eagle proposals incorporate CFTs and are able
to maintain a relatively low RCS. If the addition of CTFs does not degrade the
frontal RCS of the F-22C significantly beyond that of the F-35, it might be
appropriate to use in moderately contested threat environments; it would not be
used in highly contested SEAD/DEAD missions against near-peer competitors.
Availability
Rates & Sustainment
Banal
details related to maintenance, repair, and overhaul (MRO) programs and their
impact fleet readiness is a dimension of defense analysis that is often lost
upon most armchair generals. Despite the unmatched air-to-air capabilities of
each individual F-22 and the extensive training of each Raptor pilot, the small
fleet of primary aircraft inventory airframes (PMAI) translates into an even
smaller number of planes which are ready for combat at any one time. For
example, the USAF has two broad terms to describe an aircraft fleet’s
readiness: mission capable rates and availability rates. Mission capable rates
(MCR) are equal to the mission capable hours divided by the unit possessed
hours while the mission availability rate is equal to the mission capable hours
divided by the total aircraft inventory (TAI) hours; MCR is generally a
satisfactory level of determining readiness at the unit level while
availability rates are indicative of broader fleet level readiness.[13] For example, of the 183
F-22s in the USAF inventory, on average roughly 115 are airworthy and able to
execute assigned missions at any one time with an availability rate of 62.8%,
the corresponding mission capable rate for the PMAI F-22 component fleet is
72.7% or roughly 89 of 123 PMAI aircraft would be ready to execute missions at
any one time.[14][15]
Image 6: Image Credit GAO, 2014.
The
U.S. can effectively increase its fleet of deployed F-22s by improving readiness
rates such that the existing limit fleet size translates into the most combat
capability possible. For example, a 10% improvement in MCR among PMAI aircraft
would effectively boost the available PMAI fleet size by 13 aircraft – more
than half a squadron worth, to a total of 102 up from 89. With such a small
fleet size and the prospect of restarting production low, ensuring maximum fleet
readiness is vital given the F-22’s unique role as the only high-end survivable
air superiority asset in the USAF inventory for the foreseeable future. The
USAF has a goal of achieving a fleet availability rate of 70% by 2018 up from
the current 62.8% through the reliability and maintainability maturation
program (RAMMP). RAAMP modifications include:
Mighty Tough Boot Development [toughens the seams between aircraft panels to facilitate easier maintenance and mitigate damage to RAM coatings], Aircraft Mounted Nozzle Shield (AMNS) Liner Redesign, Integrated Forebody (IFB) Rain Erosion Nose Cap, Canopy Topcoat Redesign, Stored Energy System (SES) Air Filter, Auxiliary Power Unit (APU) Plenum Sealing, Gland Redesign, Automated Backup Oxygen System, Secondary Multi-Function Display (SMFD) Backlight to Lower Power LED, Gland Redesign, and Driver B RF Circuit Redesign[16][17]
Image 7: RAMMP. Image Credit: Flight Global.
According to Lockheed
Martin, 50% of all maintenance activities for the F-22 relate to maintaining
its RAM coatings. The limited resilience of the F-22’s RAM coatings contributes
towards its astronomically high cost per flight hour to operate at $59,166
compared to $20,318 for the F-16 and $32,000 (projected) for the F-35 as of
2015 data.[18]
A total of $1.7 billion will be spent on RAAMP associated upgrades through 2020,
but additional modifications – particularly to the F-22’s RAM, are likely
required and ought to be incorporated to any F-22C.
Part III will detail capability improvements such as enclosed weapon pods, HMD, IRST, etc.
Part III will detail capability improvements such as enclosed weapon pods, HMD, IRST, etc.
[10] Ibid.
[11] Ibid.
[i] Assumes 3,000 pounds additional
fuel from CTFs and 25% greater fuel efficiency from variable cycle engines.
Does not factor drag or other important factors i.e. this is a “napkin math”
type calculation that provides a rough estimate of expected performance.
Hi Matt,great aeticle.I like to see Raptors with new engines and more range.
ReplyDeleteBut since 2011 the Raptor uses the F-35's coatings
http://www.dailytech.com/F35+Stealth+Coatings+Applied+to+F22/article21321.htm
Thank you! I had heard some of the more resilient F-35 developed RAM coatings were applied to the F-22 fleet but the high maintenance costs have not come down substantially. Also the USAF is still funding RAM related maintenance and sustainment upgrades. My theory is, of the two RAM layers for the F-35: (1) fiber mat which is baked into the skin of the airframe and (2) external top coating, the fiber mat is the most important for reducing maintenance needs. The F-22 only received the top coat and it can't receive fiber mat without totally redesigning the aircraft.
DeleteHi Matt! Long time reader, first time poster. Love your work, do keep at it.
ReplyDeleteI wanted to point that that Ada's age isn't really a fair complaint against it. If you go by first-appeared date, the C++ used on the F-35 isn't actually that much newer. If you go by latest update to the standard, Ada is relatively young - there was a preview of an upcoming release less than 18 months ago (Ada 2012 TC 1 Draft 3: http://www.ada-auth.org/corrigendum1-12.html).
I would recommend against "upgrading" the base operating system by porting it to another programming language. Not only does that consume time and money, but also introduces the possibility of new bugs. Writing new code that talks to the existing code base in something non-Ada is fine (which is what the F-35 program did), as is rewriting parts that are being actively worked on anyways (such as, presumably, IMIS). However, if the existing code works, works well, and was well designed and reviewed originally and you weren't going to touch it anyways, better to leave it be. Don't fix what isn't broken. The fact it wasn't written in the language de jour isn't a bug.
Ada is generally considered a good choice for projects that need performance, control and correctness. The language does a lot of things right to minimize certain classes of common programming problems which would be horrendous to have on an aircraft. The latest all new Boeing aircraft, the 787, uses Ada, as do many other aircraft manufacturers such as Airbus: https://www.seas.gwu.edu/~mfeldman/ada-project-summary.html. I fully expect new aircraft to continue the trend of using Ada, as despite its age it's one of the best choices here. Newer attempts at a language which encourages correctness, such as Rust, still miss a lot of things Ada did right back in the prehistoric days. AFAIK there's yet to be a language that is wholly better than Ada at what Ada tries to do, just better in some respects while behind in others.
Ada's main issue is that it's a bit tedious to work with, which makes it unpopular, which makes it hard to find good software engineers who know it, as well as associated tooling. I suspect the F-35 program and IMIS changes went with C++ to try and expedite development and save costs, not due to a design fault in the Ada language itself.
Hope that didn't come off too harshly. The Ada bit was a small part of an otherwise excellent article.