Editing Avro Canada CF-105 Arrow (section)

===Production===

[[File:Cf-105 Arrow002.jpg|thumb|RL-204, late 1958]]
To meet the timetable set by the RCAF, Avro decided that the Arrow program would adopt the [[Cook-Craigie plan]]. Normally a small number of prototypes of an aircraft were hand-built and flown to find problems, and when solutions were found these changes would be worked into the design. When satisfied with the results, the production line would be set up. In a Cook-Craigie system, the production line was set up first and a small number of aircraft were built as production models.<ref>Whitcomb 2002, p.&nbsp;86.</ref><ref name = 'pigott55'>Pigott 1997, p.&nbsp;55.</ref> Any changes would be incorporated into the jigs while testing continued, with full production starting when the test program was complete. As Jim Floyd noted at the time, this was a risky approach: "it was decided to take the technical risks involved to save time on the programme&nbsp;... I will not pretend that this philosophy of production type build from the outset did not cause us a lot of problems in Engineering. However, it did achieve its objective."<ref name=Floyd/>

To mitigate risks, a massive testing program was started. By mid-1954, the first production drawings were issued and wind tunnel work began, along with extensive computer simulation studies carried out both in Canada and the United States using sophisticated computer programs.<ref name = 'peden38'>Peden 2003, p.&nbsp;38.</ref> In a related program, nine instrumented free-flight models were mounted on solid fuel [[Project Nike|Nike]] rocket boosters and launched from Point Petre over Lake Ontario while two additional models were launched from the NASA facility at [[Wallops Island]], Virginia, over the Atlantic Ocean. These models were for aerodynamic drag and stability testing, flown to a maximum speed of Mach&nbsp;1.7+ before intentionally crashing into the water.<ref>Page et al. 2004, p.&nbsp;15.</ref><ref>Belleau, Naomi. [http://www.navy.forces.gc.ca/cms/4/4-a_eng.asp?category=12&id=193 "Domestic Operations: Trinity's "Fiona" takes the plunge in search of Avro Arrow"] ({{webarchive|url=https://web.archive.org/web/20110613033931/http://www.navy.forces.gc.ca/cms/4/4-a_eng.asp?category=12&id=193|date=13 June 2011 }}) ''Canadian Navy''. Retrieved: 11 September 2010.</ref>

Experiments showed the need for only a small number of design changes, mainly involving the wing profile and positioning. To improve [[angle of attack|high-alpha]] performance, the leading edge of the wing was drooped, especially on outer sections, a [[Dogtooth extension|dog-tooth]] was introduced at about half-span to control spanwise flow,<ref>Whitcomb 2002, pp.&nbsp;89–91.</ref> and the entire wing given a slight negative [[Camber (aerodynamics)|camber]] which helped control trim drag and pitch-up.<ref name = 'campagna37'>Campagna 1998, p.&nbsp;37.</ref> The [[area rule]] principle, made public in 1952, was also applied to the design. This resulted in several changes including the addition of a tailcone, sharpening the radar nose profile, thinning the intake lips, and reducing the cross-sectional area of the fuselage below the canopy.<ref name=Floyd/>

The construction of the airframe was fairly conventional, with a semi-[[monocoque]] frame and multi-spar wing. The aircraft used a measure of [[magnesium]] and [[titanium]] in the fuselage, the latter limited largely to the area around the engines and to fasteners. Titanium was still expensive and not widely used because it was difficult to machine.<ref>Whitcomb 2002, pp.&nbsp;109–110.</ref>

The Arrow's thin wing required aviation's first {{cvt|4000|psi|MPa}} hydraulic system to supply enough force to the control surfaces,{{fact|date=December 2019}} while using small actuators and piping. A rudimentary [[fly-by-wire]] system was employed, in which the pilot's input was detected by a series of pressure-sensitive transducers in the stick, and their signal was sent to an electronic control servo that operated the valves in the hydraulic system to move the various flight controls. This resulted in a lack of control feel; because the control stick input was not mechanically connected to the hydraulic system, the variations in back-pressure from the flight control surfaces that would normally be felt by the pilot could no longer be transmitted back into the stick. To re-create a sense of feel, the same electronic control box rapidly responded to the hydraulic back-pressure fluctuations and triggered actuators in the stick, making it move slightly; this system, called "artificial feel", was also a first.<ref name = 'campagna73-74'>Campagna 1998, pp.&nbsp;73–74.</ref>

In 1954, the [[Rolls-Royce RB106|RB.106]] program was cancelled, necessitating the use of the backup [[Wright J67#Curtiss-Wright Derivatives|Wright J67]] engine instead. In 1955, this engine was also cancelled, leaving the design with no engine. At this point, the [[Pratt & Whitney J75]] was selected for the initial test-flight models, while the new TR 13 engine was developed at Orenda for the production Mk 2s.<ref name = 'pigott56'>Pigott 1997, p.&nbsp;56.</ref>

After evaluating the engineering mock-ups and the full-scale wooden mock-up in February 1956, the RCAF demanded additional changes, selecting the advanced RCA-Victor ''Astra'' fire-control system firing the equally advanced [[United States Navy]] [[AIM-7 Sparrow#Sparrow II|Sparrow II]] in place of the MX-1179 and Falcon combination. Avro vocally objected on the grounds that neither of these were even in testing at that point, whereas both the MX-1179 and Falcon were almost ready for production and would have been nearly as effective for "a very large saving in cost".<ref name = 'peden46-47'>Peden 2003, pp.&nbsp;46–47.</ref> The Astra proved to be problematic as the system ran into a lengthy period of delays, and when the USN cancelled the Sparrow II in 1956, [[Canadair]] was quickly brought in to continue the Sparrow program in Canada, although they expressed grave concerns about the project as well and the move added yet more expense.<ref name = 'campagna68'>Campagna 1998, p.&nbsp;68.</ref>