Editing Lockheed SR-71 Blackbird (section)

===Propulsion system or powerplant===
====Complete powerplant====
The SR-71 used the same powerplant as the A-12 and [[YF-12]]. It consists of three main parts: inlet, J58 engine and its nacelle, and ejector nozzle. "Typical for any supersonic powerplant the engine cannot be considered separately from the rest of the powerplant. Rather, it may be regarded as the heat pump in the over-all system of inlet, engine, and nozzle. The net thrust available to propel the aircraft may be to a large extent controlled by the performance of the inlet and nozzle rather than by the physical potentialities of the engine alone."<ref>https://apps.dtic.mil/sti/citations/ADB183548,'Inlet-engine matching considerations', Obery and Stitt, NACA Conference on Turbojet Engines for Supersonic Propulsion, AD B183548, Section VII, Paper 2.</ref> This is illustrated for the Blackbird by the thrust contributions from each component at Mach 3+ with maximum afterburner: inlet 54%, engine 17.6%, ejector nozzle 28.4%.<ref>Campbell, David. [https://arc.aiaa.org/doi/10.2514/3.60402 "F-12 Series Aircraft Propulsion System Performance and Development"], "Table 1 Propulsive thrust distribution"</ref>

When stationary and at low speeds the inlet caused a loss in engine thrust. This was due to the flow restriction through the inlet when stationary. Thrust was recovered with ram pressure as flight speed increased (uninstalled thrust {{cvt|34000|lb(f)|kN}}, installed at zero airspeed {{cvt|25500|lb(f)|kN}} rising through {{cvt|30000|lb(f)|kN}} at 210&nbsp;knots, unstick speed).<ref>{{cite book |url=https://www.sr-71.org/blackbird/manual/1/1-7.php |title=SR-71 Flight Manual |via=SR-71 Online |section=Section 1 |pages=1–7}}</ref>

At supersonic speeds not all the airflow approaching the inlet capture area entered the inlet. At supersonic speeds an intake always adapts to the engine requirements, rather than forcing air into the engine, and the unwanted air flows around the outside of the cowl, causing spillage drag. More than half the air approaching the capture area had to be spilled at low supersonic speeds and the amount reduced as the design speed was approached because the inlet airflow had been designed to match the engine demand at that speed and the chosen design point ambient temperature. At this speed the spike shock touched the cowl lip and there was minimal spillage (with its attendant drag) as shown by Campbell.<ref name="Inlet development">{{cite journal |url=https://www.jstor.org/stable/44657533 |title=F-12 Inlet Development – Fig.4 – Inlet airflow |first=David H. |last=Campbell |journal=SAE Transactions |volume=83 |year=1974 |pages=2832–2840 |publisher=SAE International|jstor=44657533 }}</ref> The inlet and engine matching was also shown by Brown,<ref>Brown, William. "J58/SR-71 Propulsion Integration, attachment to CIA-RDP90B001170R000100050008-1, Fig. 3 'Inlet and engine airflow match'".</ref> who emphasized the benefit of increased engine airflow at higher Mach numbers that came with the introduction of the bleed bypass cycle. These two authors show the disparity between inlet and engine for the Blackbird in terms of airflow and it is further explained in more general terms by Oates.<ref>"Aerothermodynamics of Aircraft Gas Turbine Engines", Oates, Air Force Aero Propulsion Laboratory, Figure 13.1.17 'Elements of Inlet Airflow Supply Determination', (a) and (b).</ref>

Engine operation was adversely affected when operating behind an unstarted inlet. In this condition the inlet behaved like a subsonic inlet design (known as a pitot type) at high supersonic speeds, with very low airflow to the engine. Fuel was automatically diverted, by the fuel derich system, from the combustor to prevent turbine over-temperature.<ref>A-12 Utility Flight Manual, 15 September 1965, changed 15 June 1968, Fuel Derich System.</ref>

All three parts were linked by the secondary airflow. The inlet needed the boundary layers removed from its spike and cowl surfaces. The one with the higher pressure recovery, the cowl shock-trap bleed, was chosen as secondary air<ref name="Inlet development"/> to ventilate and cool the outside of the engine. It was assisted from the inlet by the pumping action of the engine exhaust in the ejector nozzle, cushioning the engine exhaust as it expanded over a wide range of pressure ratios which increased with flight speed.<ref>https://www.sae.org/publications/technical-papers/content/740832/, "J58/YF-12 Ejector Nozzle Performance" pp. 1, 2.</ref>

Mach&nbsp;3.2 in a [[standard day]] atmosphere was the design point for the aircraft. However, in practice the SR-71 was more efficient at even faster speeds and colder temperatures. The specific range charts showed for a standard day temperature, and a particular weight, that Mach&nbsp;3.0 cruise used {{cvt|38000|lb/h|kg/h}} of fuel. At Mach&nbsp;3.15 the fuel flow was {{cvt|36000|lb/h|kg/h}}. Flying in colder temperatures (known as temperature deviations from the standard day) would also reduce the fuel used, e.g. with a {{cvt|-10|C|F|order=flip}} temperature the fuel flow was {{cvt|35000|lb/h|kg/h}}.<ref>{{cite book|url = https://books.google.com/books?id=6svmtOFa1JIC| pages=165, 166| title=SR-71 Revealed: The Untold Story| isbn=978-1-61060-751-3| last1=Graham| first1=Richard H.| publisher=Zenith Imprint}}</ref> During one mission, SR-71 pilot [[Brian Shul]] flew faster than usual to avoid multiple interception attempts. It was discovered after the flight that this had reduced the fuel consumption.<ref name="Shul">Shul and O'Grady 1994</ref> It is possible to match the powerplant for optimum performance at only one ambient temperature because the airflows for a supersonic inlet and engine vary differently with ambient temperature. For an inlet, the airflow varies inversely with the square root of the temperature, and for the engine, it varies with the direct inverse.<ref>https://archive.org/details/sim_journal-of-aircraft_november-december-1968_5_6/mode/2up Design and Development of an Air Intake for a Supersonic Transport Aircraft, "Effect of Ambient Temperatures", p.518</ref>

<gallery widths="270px" heights="203px" mode="packed" class="center" caption="Powerplant">
File:1 - Seattle.jpg|The inlet extends from the spike tip to the four sets of three louvers that vent the spike boundary layer bleed overboard through four spike support struts. The more-forward louvers vent the forward bypass bleed. The engine extends from there to the ejector nozzle blow-in doors (shown open) and the nozzle extends from there to the ejector flaps (shown closed).
File:SR71 J58 Engine Airflow Patterns.svg|Diagrams show operation of the air inlet, flow through the engine (primary air), nacelle flow past the engine (secondary air), and flow into the ejector nozzle (primary, secondary and tertiary air).
File:Pratt & Whitney J58 ground test.jpg|This picture of an uninstalled engine being tested illustrates the need for cooling air around the exhaust duct. The engine, when installed as part of the powerplant, has secondary cooling air at {{Convert|1200|F|C|round=50}} passing over the afterburner duct which is heated internally by combustion up to {{Convert|3200|F|C|round=50}}. The heating, followed by the primary nozzle restriction, accelerates the exhaust to sonic speed as it leaves the primary nozzle (shown). The ejector nozzle (not shown) surrounds the primary exhaust with secondary and tertiary air to cushion its expansion in the ejector nozzle.
</gallery>

====Inlet====
The engine inlets needed internal supersonic diffusion since external compression used on slower aircraft caused too much drag at Blackbird speeds. Their features included a centerbody or spike, spike boundary-layer bleed slots where normal shock was located, a cowl boundary layer bleed "shock trap" entrance, streamlined bodies known as "mice", forward bypass bleed ports between the "mice", rear bypass ring, louvers on external surface for spike boundary layer overboard, and louvers on external surface for front bypass overboard. Venting this bypass overboard produced high drag: {{cvt|6000|lb(f)|kN}} at cruise with 50% door opening, compared to the total aircraft drag of {{cvt|14000|lb(f)|kN}}.<ref>https://authors.library.caltech.edu/records/6s4e6-b2j60, AE107_SR-71_Case_Study_51-120, "Drag Penalty Of Overboard Bypass At Cruise".</ref> Designer David Campbell holds a patent on the inlet's aerodynamic features and functioning,<ref>{{cite web |title=Supersonic inlet for jet engines |url=https://patents.google.com/patent/US3477455A/en |work=patents.google.com}}</ref> which are explained in the "A-12 Utility Flight Manual"<ref>A-12 Utility Flight Manual, 15 September 1965, changed 15 June 1968, Air Inlet System.</ref> and in a 2014 presentation by Lockheed Technical Fellow Emeritus Tom Anderson.<ref>Tom Anderson. "SR-71 Inlet Design Issues And Solutions Dealing With Behaviourally Challenged Supersonic Flow Systems", Lockheed Martin Corporation, presented at AEHS Convention 2014.</ref>

When an inlet was operating as an efficient supersonic compressor—a status called "started"—supersonic diffusion took place in front of the cowl and internally in a converging passage as far as a terminal shock where the passage area began to increase and subsonic diffusion takes place. An analog control system was designed to hold the terminal shock in position.

But in the early years of operation, the analog computers could not always keep up with rapidly changing inputs from the nose boom. If the duct back pressure became too great and the spike was incorrectly positioned, the shock wave become unstable and would shoot quickly forward to a stable position outside the cowl. This "inlet [[unstart]]" would often extinguish the engine's afterburner. The asymmetrical thrust from the other engine would cause the aircraft to yaw violently. [[Stability Augmentation System|SAS]], autopilot, and manual control inputs would attempt to regain controlled flight, but extreme yaw would often reduce airflow in the opposite engine and stimulate "sympathetic stalls". This generated a rapid counter-yawing, often coupled with loud "banging" noises, and a rough ride during which crews' helmets would sometimes strike their cockpit canopies.<ref>{{harvp|Crickmore|1997|pp=42-43}}</ref> One response to a single unstart was unstarting both inlets to prevent yawing, then restarting them both.<ref name="landis_jenkins_p97">{{harvp|Landis|Jenkins|2004|p=97}}</ref> After wind-tunnel testing and computer modeling by NASA Dryden test center,<ref>{{cite web |year=2004 |title=NASA Dryden Technology Facts – YF-12 Flight Research Program |url=https://www.nasa.gov/centers/dryden/about/Organizations/Technology/Facts/TF-2004-17-DFRC.html |url-status=dead |archive-url=https://web.archive.org/web/20190912045121/https://www.nasa.gov/centers/dryden/about/Organizations/Technology/Facts/TF-2004-17-DFRC.html |archive-date=12 September 2019 |access-date=9 March 2019 |publisher=NASA |location=US}}</ref> Lockheed installed an electronic control to detect unstart conditions and perform this reset action without pilot intervention.<ref>{{harvp|Rich|Janos|1994|p=221}}</ref>

During troubleshooting of the unstart issue, NASA discovered that the vortices from the nose chines were entering the engine and reducing engine efficiency. To fix this, the agency developed a computer to control the engine bypass doors. Beginning in 1980, the analog inlet control system was replaced by a digital system, Digital Automatic Flight and Inlet Control System (DAFICS),<ref>{{Cite web |title=SR-71 Online - SR-71 Flight Manual: Section appendix, Page A-2 |url=https://www.sr-71.org/blackbird/manual/appendix/a-2.php |access-date=21 August 2023 |website=www.sr-71.org}}</ref> which reduced unstarts.<ref name="landis_jenkins_p83">{{harvp|Landis|Jenkins|2004|p=83}}</ref>

<gallery widths="252px" heights="189px" mode="packed" class="center" caption="Inlet">
File:Lockheed SR-71A Blackbird at Evergreen Aviation & Space Museum (6586637283).jpg|Entry to the inlet. Behind is the outer wing and hinged portion of the nacelle that encloses the engine. The spike is shown in the forward position (for speeds below Mach 1.6). Just discernible behind the cowl lip are spike boundary layer bleed slots where the normal shock is located at higher speeds when the spike has moved rearwards, the cowl bleed "shock trap" ram intake, streamlined bodies ("mice") and, between the mice, the forward bypass door openings<ref>Tom Anderson. SR-71 Inlet Design Issues And Solutions Dealing With Behaviourally Challenged Supersonic Flow Systems, Lockheed Martin Corporation, presented at AEHS Convention 2014, 'Inlet Diffuser Photos'.</ref> that dump unwanted air externally through the front louvers and cause nacelle drag.<ref>https://authors.library.caltech.edu/records/6s4e6-b2j60, AE107_SR-71_Case_Study_51-120 "Drag Penalty Of Overboard Bypass At Cruise"</ref> When the landing gear is down, ambient air flows in reverse through the bypass to supplement the front inlet flow into the engine.
File:Lockheed SR-71A Blackbird at Evergreen Aviation & Space Museum (6586640345) (2).jpg|A rear view of the inlet where air enters the engine. Two features were added after flight testing highlighted the need: 1) "mice," visible as streamlined shapes, added to reduce the diffusion rate after pilots noted rumbling; and 2) rear bypass doors added to prevent unstarting the inlet during descents with low engine flow.<ref>{{cite conference |last=Anderson |first=J. Thomas |year=2013 |title=How Supersonic Inlets Work – Details of the Geometry and Operation of the SR-71 Mixed Compression Inlet |url=https://www.enginehistory.org/Convention/2014/SR-71Inlts/SR-71Inlts.shtml |conference=AEHS convention |access-date=October 26, 2024 |website=Engine History}}</ref> The ring of doors is at the extreme rear of the inlet as shown by their accompanying rear-turning scoop, extending from 7 o'clock to 5 o'clock, which directs the air through the nacelle to the ejector nozzle.<ref>{{cite web |url=https://ntrs.nasa.gov/citations/19750003899 |title=Continuous-output terminal-shock-position sensor for mixed-compression inlets evaluated in wind tunnel tests of YF-12 aircraft inlet |work=NASA Technical Reports Server |date=December 1974 |publisher=NASA |pages=Figures 7, 8 |last1=Dustin |first1=M. O. |last2=Cole |first2=G. L. |last3=Neiner |first3=G. H. }}</ref> The door actuator<ref>{{cite web |url=https://ntrs.nasa.gov/citations/19740030201 |title=Preliminary results of flight tests of the propulsion system of the YF-12 airplane at Mach numbers to 3.0 |work=NASA Technical Reports Server |date=November 1973 |publisher=NASA}}</ref>
File:Inlet shock waves at Mach 2.jpg|[[Schlieren]] flow visualization of shock waves for started and unstarted inlet at Mach 2
</gallery>

====Engine and nacelle====
{{main|Pratt & Whitney J58}}
The engine was an extensively re-designed version of the J58-P2, an existing supersonic engine which had run 700 development hours in support of proposals to power various aircraft for the US Navy. Only the compressor and turbine aerodynamics were retained. New design requirements for cruise at Mach 3.2 included:
* operating with very high ram temperature air entering the compressor, at {{Convert|800|F|C|sigfig=2}}
* a continuous turbine temperature capability {{Convert|450|F-change}} hotter than previous experience ([[Pratt & Whitney J75]])
* continuous use of maximum afterburning
* the use of new, more expensive, materials and fluids required to withstand unprecedented high temperatures

The engine was an afterburning turbojet for take-off and transonic flight (bleed bypass closed) and a low bypass augmented turbofan for supersonic acceleration (bleed bypass open). It approximated a ramjet during high speed supersonic cruise (with a pressure loss, compressor to exhaust, of 80% which was typical of a ramjet). It was a low bypass turbofan for subsonic loiter (bleed bypass open).<ref>https://authors.library.caltech.edu/records/6s4e6-b2j60, AE107_SR-71_Case_Study_321-450, p. 27.</ref><ref>A-12 Utility Flight Manual, 15 September 1965, changed 15 June 1968, 'Start Bleed And Bypass Valve Actuation', Figure 1-7</ref>

Analysis of the J58-P2 supersonic performance<ref name="roadrunnersinternationale.com">Bob Abernethy. https://www.roadrunnersinternationale.com/pw_tales.htm, 'More Never Told Tales of Pratt & Whitney'.</ref> showed the high compressor inlet temperature would have caused stalling, choking and blade breakages in the compressor as a result of operating at low corrected speeds on the compressor map. These problems were resolved by Pratt & Whitney engineer Robert Abernethy and are explained in his patent, "Recover Bleed Air Turbojet".<ref>{{cite web | url=https://patents.google.com/patent/US3344606A/en | title=Recover bleed air turbojet }}</ref> His solution was to 1) incorporate six air-bleed tubes, prominent on the outside of the engine, to transfer 20% of the compressor air to the afterburner, and 2) to modify the inlet guide vanes with a 2-position, trailing edge flap. The compressor bleed enabled the compressor to operate more efficiently and with the resulting increase in engine airflow matched the inlet design flow with an installed thrust increase of 47%.<ref name="roadrunnersinternationale.com"/><ref>William Brown. J58/SR-71 Propulsion Integration, attachment to CIA-RDP90B001170R000100050008-1, Fig. 3, 'Inlet and engine airflow match'.</ref> A continuous turbine temperature of {{Convert|2000|F|C}} was enabled with air-cooled first stage turbine vane and blades. Continuous operation of maximum afterburning was enabled by passing relatively cool air from the compressor along the inner surface of the duct and nozzle. Ceramic thermal barrier coatings were also used.

The secondary airflow through the nacelle comes from the cowl boundary layer bleed system which is oversized (flows more than boundary layer) to give a high enough pressure recovery to support the ejector pumping action. Additional air comes from the rear bypass doors and, for low speed operation with negligible inlet ram, from suck-in doors by the compressor case.

<gallery widths="220px" heights="165px" mode="packed" class="center" caption="Engine/nacelle">
File:Pratt & Whitney J58-JT11D-20K turbojet engine, 1962 - Lockheed SR-71A Blackbird, 1966 - Evergreen Aviation & Space Museum - McMinnville, Oregon - DSC01037.jpg|View of J58 engine shows some features required for flight at Mach 3.2: titanium inlet guide vanes and first stage compressor blades for lighter weight at high ram temperatures, transonic first stage compressor blades and low hub/tip ratio compressor entry, both scaled from the bigger Mach-3 J91 engine compressor, 2-position flaps on the inlet guide vanes and three of the six bypass tubes.
File:Pratt & Whitney J58 18.jpg|The afterburner was rated for continuous operation at {{cvt|3200|F|C|-2}} made possible with ceramic coatings (colored white) on duct liner and flame holders<ref>https://ntrs.nasa.gov/citations/20090018047, "History Of Thermal Barrier Coatings For Gas Turbine Engines: Emphasising NASA's Role from 1942 to 1990".</ref> and compressor bleed air cooling the duct and nozzle (above Mach 2.1 when the bleed was flowing). The nozzle is fully open, the maximum afterburning position. The main purpose of the variable nozzle area was to control engine operation which it did in conjunction with varying heat release in the afterburner.
File:Lockheed SR-71A Blackbird at Evergreen Aviation & Space Museum (6586637067).jpg|The inlet at left was depressed when the engine ran at high power settings with inadequate inlet ram (stationary and low flight speeds). The lower-than-ambient pressure in the inlet brought in extra air through the spike bleed and forward bypass louvers shown on the inlet external surface. Adequate secondary cooling air came in through the suck-in doors (shown open on the hinged nacelle).
</gallery>

====Ejector Nozzle====
The nozzle had to operate efficiently over a wide range of pressure ratios from low, with no inlet ram with a stationary aircraft, to 31 times the external pressure at {{Convert|80000|ft|m|abbr=on}}. A blow-in door ejector nozzle had been invented by Pratt & Whitney engineer Stuart Hamilton in the late 1950s<ref>Jack Connors. https://arc.aiaa.org/doi/book/10.2514/4.867293, "The Engines of Pratt & Whitney: A Technical History", ISBN 978 1 60086 711 8, p. 328.</ref> and described in his patent "Variable Area Exhaust Nozzle".<ref>"Variable Area Exhaust Nozzle", U.S. Patent 3,062,003</ref> In this description the nozzle is an integral part of the engine (as it was in the contemporary Mach 3 General Electric YJ93.<ref>"Variable Geometry Exhaust Nozzles and Their Effects on Airplane Performance", SAE 680295, p. 5.</ref> For the Blackbird powerplant the nozzle was more efficient structurally (lighter) by incorporating it as part of the airframe because it carried fin and wing loads through the ejector shroud. The nozzle used secondary air from two sources, the inlet cowl boundary layer and rear bypass from immediately in front of the compressor. It used external flow on the nacelle through the tertiary blow-in doors until ram closed them at Mach 1.5. Only secondary air was used at higher speeds with the blow-in doors closed.

At low flight speeds the engine exhaust pressure at the primary nozzle exit was greater than ambient so tended to over-expand to lower than ambient in the shroud causing impingement shocks. Secondary and blow-in door air surrounding the exhaust cushioned it preventing over-expansion. Inlet ram pressure increased with flight speed and the higher pressure in the exhaust system closed, first the blow-in doors and then started to open the nozzle flaps until they were fully open at Mach 2.4. The final nozzle area did not increase with further increase in flight speed (for complete expansion to ambient and greater internal thrust) because its external diameter, greater than nacelle diameter would cause too much drag.<ref>Design For Air Combat, Ray Whitford, {{ISBN|0 7106 0426 2}}, p. 203</ref>

<gallery widths="220px" heights="165px" mode="packed" class="center" caption="Ejector-nozzle">
File:Lockheed SR-71A Blackbird at Evergreen Aviation & Space Museum (6586638859) (2).jpg|Ejector nozzle at the rear of the powerplant. The engine nozzle (left) is the first component in the exhaust system, followed by the secondary and tertiary air flows and ejector nozzle. The tertiary doors are open. There is a fixed convergent/divergent shroud and the ejector nozzle trailing flaps are at their minimum area (closed). These nozzle and door positions correspond with full afterburner up to transonic speed, after which the doors close and flaps start to open. Secondary air from the inlet passes between the engine and nacelle and joins the blow-in door air to control the expansion of the engine exhaust through the shroud and trailing flaps.
File:SR-71A taking off with afterburner RAF Mildenhall 1983.JPEG|A similar viewing angle, unstick speed 210 knots, to the "exploded" view, and with the same operating configuration: afterburner nozzle open, blow-in doors open and trailing flaps closed due to low internal pressure with low speed low inlet ram. Note the dark con-di shroud. Air entering the blow-in doors joins secondary air from the inlet and flows over the fixed shroud surface and trailing flaps while surrounding the exhaust from the engine.
</gallery>