Editing Advanced Tactical Fighter (section)

===Demonstration and validation===
{{main|Lockheed YF-22|Northrop YF-23}}
The Dem/Val phase was intended to develop and mature ATF technologies that would facilitate the fighter's eventual full-scale development and production, and focused on three main activities: requirements and [[requirements engineering|system specification development]], [[avionics]] ground prototypes and flying laboratories, and prototype air vehicles.<ref name="A&Hp104">Aronstein and Hirschberg 1998, p. 104.</ref> During Dem/Val, the ATF SPO program manager was Colonel James A. Fain, while the technical director (or chief engineer) was Eric "Rick" Abell. The director of ATF requirements was Colonel David J. McCloud of TAC, and the draft System Operational Requirements Document (SORD), derived from the 1984 SON, was released in December 1987.<ref>Aronstein and Hirschberg 1998, p. 106.</ref> In addition to the government contract awards, company investments during Dem/Val would amount to $675 million and $650 million (~${{Format price|{{Inflation|index=US-GDP|value=675000000|start_year=1988}}}} and ~${{Format price|{{Inflation|index=US-GDP|value=650000000|start_year=1988}}}} in {{Inflation/year|US-GDP}}) for the Lockheed and Northrop teams respectively, not counting additional investments during prior phases or by subcontractors. Pratt & Whitney and General Electric would each invest $100 million as well (~${{Format price|{{Inflation|index=US-GDP|value=100000000|start_year=1988}}}} in {{Inflation/year|US-GDP}}).<ref name="A&H1998p164">Aronstein and Hirschberg 1998, p. 164.</ref>

With the ATF system specification, the SPO had set the technical requirements without specifying the "how"; this was meant to give the contractor teams flexibility in developing the requisite technologies and offer competing methods.<ref name="A&Hp104"/> Furthermore, the SPO was also open to adjusting requirements if necessary. Both the Lockheed and Northrop teams conducted performance and cost trade studies and presented them in system requirement reviews (SRRs) with the SPO periodically during Dem/Val. This enabled the SPO to adjust ATF requirements and delete those that were significant weight and cost drivers while having marginal operational value. For instance, the number of internal missiles (represented by the [[AIM-120 AMRAAM|AIM-120A]]) was reduced from eight to six to reduce weight and cost.{{refn|A clipped-fin variant of the AMRAAM, the AIM-120C, was eventually developed to increase the F-22's internal missile load back to eight.<ref>Aronstein and Hirschberg 1998, p. 60.</ref>|group=N}}<ref name="Hehs1998P2"/> Because of the added weight for thrust vectoring/reversing nozzles and related systems on the [[McDonnell Douglas F-15 STOL/MTD|F-15 STOL/MTD]] research aircraft, the SPO changed the runway length requirement to {{convert|3000|ft|m|sigfig=1}} and removed the thrust reverser requirement in late 1987.<ref name=Sweetman_p23>Sweetman 1991, p. 23.</ref><ref name=Miller_p23>Miller 2005, p. 23.</ref> The [[ejection seat]] requirement was downgraded from a fresh design to the existing McDonnell Douglas [[ACES II]]. However, both contractor teams still found the {{cvt|50000|lb|kg|-2}} takeoff gross weight goal unachievable, so this was increased to {{cvt|60000|lb|kg|-2}}, resulting in engine thrust requirement increasing from {{cvt|30000|lbf|kN|0}} class to {{cvt|35000|lbf|kN|0}} class. Furthermore, Dem/Val would be extended several times to better mature technologies and reduce near-term budgets.<ref name="A&Hp105-108">Aronstein and Hirschberg 1998, pp. 105–108.</ref>

[[File:Boeing 757 Prototype N757A F-22 Raptor Systems Testbed.jpg|thumb|left|The [[Boeing 757]] used for testing the Lockheed team's avionics and later modified into the Flying Test Bed during full-scale development.]]
Aside from advances in air vehicle and propulsion technology, the ATF would make a leap in terms of avionics performance with a fully integrated avionics suite that [[sensor fusion|fuses sensor information]] together into a common tactical picture, thus improving the pilot's situational awareness and reducing workload; the avionics were expected to make up about 40% of the ATF's flyaway cost. The avionics system was to employ the ''[[PAVE|PAVE PILLAR]]'' system architecture and leverage technology from the [[Very High Speed Integrated Circuit Program|Very High Speed Integrated Circuit]] program; software would primarily be written in [[Ada (programming language)|Ada]].<ref name="ReferenceB"/>{{refn|The ATF/PAVE PILLAR architecture was the basis for the Joint Integrated Avionics Working Group (JIAWG) formed in 1986, which was to develop a common avionics architecture for the ATF, Advanced Light Helicopter (LHX), and Advanced Tactical Aircraft (ATA) programs; the latter two programs, resulting in the [[Boeing-Sikorsky RAH-66 Comanche|RAH-66 Comanche]] and the [[McDonnell Douglas A-12 Avenger II|A-12 Avenger II]] respectively, would eventually be canceled.<ref>Aronstein and Hirschberg 1998, pp. 173-175</ref>|group=N}} The Dem/Val phase for avionics development was marked by demonstrations of the hardware and software with Avionics Ground Prototypes (AGP) to evaluate performance and reliability. The SPO gave the teams flexibility to pick their own vendors for some subsystems; for instance, the Lockheed team's [[infrared search and track]] (IRST) sensor was supplied by General Electric, while Northrop team's was from [[Martin Marietta]]; both teams chose the [[Westinghouse Electronic Systems|Westinghouse]]/[[Texas Instruments]] [[active electronically scanned array]] (AESA) radar.{{refn|group=N|The Westinghouse/Texas Instruments radar design would beat the Hughes/General Electric design and became the [[AN/APG-77]].<ref name="A&H1998p181">Aronstein and Hirschberg 1998, p. 181.</ref>}} The integrated electronics warfare and integrated communication, navigation, and identification avionics were selected by the SPO.<ref name="A&H1998p181"/> Although not required, both teams would employ flying avionics laboratories as well, with the Lockheed team using a modified [[Boeing 757]] and the Northrop team using a modified [[BAC One-Eleven]].<ref>Aronstein and Hirschberg 1998, pp. 113-115.</ref> The avionics requirements were also the subject of SRRs and adjustments; as avionics was a significant cost driver, [[side looking airborne radar|side-looking radar]]s were deleted, and the dedicated IRST system was downgraded from multicolor to single color before changing from requirement to goal and provision for future addition.<ref name="A&Hp105-108"/> In 1989, a $9 million per aircraft cost cap on avionics in FY 1985 dollars (~${{Format price|{{Inflation|index=US-GDP|value=9000000|start_year=1985}}}} in {{Inflation/year|US-GDP}}) was imposed by the SPO to contain requirements creep.<ref name="Hehs1998P2"/><ref>Mullin 2012, p. 36.</ref>

Finally, two examples of each prototype air vehicles were built and flown for Dem/Val: one with [[General Electric YF120]] engines, the other with [[Pratt & Whitney F119|Pratt & Whitney YF119]] engines.<ref name="USAF museum" /><ref>{{cite web |url=http://www.nationalmuseum.af.mil/factsheets/factsheet.asp?id=2383 |title=YF-23 fact sheet |publisher=National Museum of the U.S. Air Force |archive-url=https://web.archive.org/web/20110716073256/http://www.nationalmuseum.af.mil/factsheets/factsheet.asp?id=2383 |archive-date=July 16, 2011}}</ref> Contractor teams made extensive use of analytical and empirical methods for their air vehicle designs, including [[wind tunnel]] testing, RCS pole testing, and software for [[computational fluid dynamics]], RCS calculations, and [[computer-aided design]].{{refn|group=N|For example, the Lockheed team conducted 18,000 hours of wind tunnel testing during Dem/Val.<ref>Aronstein and Hirschberg 1998, pp. 121-125.</ref>}} Consistent with the SPO's willingness to give contractor teams the flexibility in determining how to achieve the ATF requirements, the flight test plans were created and executed by the teams themselves and the prototype air vehicles were not flown against each other for direct comparisons; neither the YF-22 nor YF-23 would share the same test points, which were set by their own teams to [[technology demonstration|demonstrate concept viability]] and validate engineering predictions.{{refn|group=N|The contractor teams were to give the SPO "sealed envelope" flight performance predictions against which their aircraft would be evaluated, rather than against each other.<ref name="A&Hp82-89"/>}}<ref>Aronstein and Hirschberg 1998, p. 137.</ref> Noteworthy is the Lockheed team's complete redesign of the YF-22's entire shape and configuration in summer 1987 due to weight concerns, with prototype design freeze relatively soon afterwards resulting in its shape being rather unrefined and immature.<ref name="Hehs1998P2"/><ref>Mullin 2012, pp. 29-30.</ref> In contrast, the YF-23 was a continual refinement of Northrop's design concept even prior to Dem/Val proposal submission, with the configuration remaining largely the same throughout. Accurate artwork of the prototypes, which had been highly classified due to the stealth shaping, was first officially released in 1990 ahead of their public unveiling; the aforementioned Dem/Val extensions also pushed flight testing from 1989 to 1990. While the prototype air vehicle designs were frozen in 1988 in order to build the aircraft and begin flight tests by 1990, both teams continued to refine their F-22 and F-23 designs, or Preferred System Concepts, for full-scale development.<ref>Aronstein and Hirshberg 1998, p. 119.</ref><ref name="Metz2017p25-27"/><ref>{{Cite journal|title=Pentagon relaxes ATF design secrecy|journal=[[Flight International]]|location=London|publisher=Reed Business Information|issue=4217|volume=137|page=4|date=23–29 May 1990|url=http://www.flightglobal.com/pdfarchive/view/1990/1990%20-%201446.html|access-date=24 June 2011|archive-url=https://web.archive.org/web/20121105070635/http://www.flightglobal.com/pdfarchive/view/1990/1990%20-%201446.html|archive-date=5 November 2012|issn=0015-3710}}</ref>

[[File:YF-22 and YF-23 formation.png|thumb|Northrop team's YF-23 (above) and Lockheed team's YF-22 (below) flying in formation.]]
The first YF-23 made its maiden flight on 27 August 1990 and the first YF-22 first flew on 29 September 1990.<ref name=Goodall_p99>Goodall 1992, p. 99.</ref>  Flight testing began afterwards at [[Edwards Air Force Base]] and added the second aircraft for each competitor in late October 1990.<ref name=Miller_p38-9/>  The first YF-23 with P&W engines supercruised at Mach 1.43 on 18 September 1990 and the second YF-23 with GE engines officially reached over Mach 1.6 on 29 November 1990, topping out at Mach 1.72.{{refn|The YF-23 with the General Electric engines was officially stated to have been able to supercruise at over Mach 1.6, and estimates from General Electric engineers suggest that the top supercruise speed was as high as Mach 1.8.<ref>Sweetman 1991, p. 55.</ref><ref name="Chong2016p237-238"/>|group=N}}<ref name=Miller_p38-9/><ref name="test_pilot_interview">{{cite AV media |people=Metz, Alfred "Paul"; Sandberg, Jim |title=YF-23 DEM/VAL Presentation by Test Pilots Paul Metz and Jim Sandberg |date=27 August 2015 |publisher=Peninsula Seniors Production |location=Western Museum of Flight, Torrance, California |url=https://www.youtube.com/watch?v=Vpkv1ErWIf8 |access-date=15 September 2015}}</ref>  The first YF-22 with GE engines achieved Mach 1.58 in supercruise on 3 November 1990 and the second YF-22 with P&W engines also achieved Mach 1.43 on 27 December 1990. Maximum speed of both prototype designs in afterburner was in excess of Mach 2.{{refn|group=N|The YF119 did not yet incorporate the design changes for increased thrust and were still 30,000-lbf thrust engines, while the YF120 did and were 35,000-lbf thrust engines; as a result, both ATF prototypes achieved better performance with the GE engines.<ref>Aronstein and Hirschberg 1998, pp. 221-222.</ref>}}<ref name="Goodall_p102-3">Goodall 1992, pp. 102–103.</ref><ref>{{cite AV media |url=https://www.youtube.com/watch?v=nY0j3dmlHec |title=YF-22 - Road to the Raptor with Tom Morgenfeld, Test Pilot |date=17 April 2022 |publisher=Western Museum of Flight |location=Torrance, California |access-date=30 June 2023 |people=Morgenfeld, Thomas A.}}</ref> Flight testing continued until December 1990 with the YF-22s accumulating 91.6 flight hours in 74 sorties while the YF-23s flew 65.2 hours in 50 sorties. Following flight testing, the contractor teams submitted their ATF full-scale development proposals on 31 December 1990. The teams' NATF designs, often referred to as "[[Lockheed YF-22#NATF-22|NATF-22]]" and "[[Northrop YF-23#NATF-23|NATF-23]]" (they were never formally designated), were included in their proposals as well.<ref name=Miller_p38-9>Miller 2005, pp. 38–39.</ref>