Apr 282011

The history of stealth aircraft design remains shrouded in secrecy.  Though selected details regarding the development of such successful stealth aircraft as the Lockheed Martin F-117, Northrop Grumman B-2 and Lockheed Martin F-22 have come to light in recent years, information on the unrealized projects and studies that preceded these aircraft has been less easy to come by, as most remain classified.

One study that has come to light is the McDonnell Douglas Model 226-458 Quiet Attack Aircraft (QAA) of 1974, a project sponsored by the Office of Naval Research that offers a snapshot of the state of stealth aircraft development in the early 1970’s.  The Model 226-458 is briefly mentioned in the book Advanced Tactical Fighter to F-22 Raptor: Origins of the 21st Century Air Dominance Fighter by David C. Aronstein et al., an excellent technical history of the F-22.  This article summarizes a declassified report covering certain aspects of the QAA study.


Reduction of aircraft radar cross section (RCS) became a big priority for the US military in the aftermath of the Arab-Israeli war of 1973, in which large numbers of Israeli aircraft were downed by Soviet-made surface-to-air missiles over a very short period of time.  Realizing the same fate could befall the US if it ever got into an all-out conflict with the Soviet Union, the Air Force and Navy began to invest more heavily in technologies to reduce RCS.

Initially, creating a stealthy design was a matter of trial and error.  Modeling clay was progressively applied and shaped on models, which were then sent to an RCS range for testing.  It was a slow and inexact process, but significant progress was nonetheless made.  Aircraft from this period that resulted from this approach typically featured extensive surface blending, exemplified by the portly Northrop Tacit Blue; the Model 226-458’s elegantly blended airframe suggests that it was also designed in this manner.

McDonnell Douglas was supposedly the among first US aerospace companies to arrive at a determination of what the RCS levels had to be in order for an aircraft to avoid detection while performing a useful military mission.  This was in no small part due to their ownership of the Grey Butte Microwave Measurement Facility, one of the best test ranges in existence for measuring very low RCS values.  According to Bill Sweetman, Grey Butte was originally established by McDonnell Astronautics for reentry vehicle testing. Lockheed and Northrop also conducted various tests at Grey Butte. However, unlike its rivals, McDonnell Douglas seems to have come up short in designing an aircraft that could achieve the desired RCS levels.

The QAA program was an exploratory investigation to conduct applied research on the feasibility and potential operational value of a covert, quiet, carrier-based aircraft for Navy/Marine Corps missions.  RCS reduction was achieved through the filleting of corners and blending of the wing and tail surfaces into the fuselage, employing a v-tail and providing an internal weapons bay.  At the time, previous attempts at signature reduction on military aircraft had focused on the reduction of one signature or fixes on existing aircraft, such as the application of RAM (radar absorbent material) to areas of high radar reflectivity.  The Model 226-458 was designed from the outset to reduce all signatures to a level that would permit covert operations and reduce vulnerability.

In order to develop the stealthiest aircraft possible, a baseline aircraft was required against which the impact of design changes could be compared.  McDonnell’s first study in the QAA program was the Model 226-450A, which was developed without any attempt at signature reduction, making it the smallest and lightest of the configurations studied in the program. This project was powered by one non-afterburning turbofan engine of low bypass ratio. This aircraft carried all stores externally and was in the same size and weight class of the A-4 Skyhawk.

From this initial design, the low noise and infrared signature Model 226-452A was developed.  This aircraft was powered by a high bypass tip driven fan propulsion system, sized to operate at minimum throttle setting, which was selected to minimize propulsion noise and infrared signatures at quiet speed. The study featured a wing of greater area and higher aspect ratio, which was considered the best compromise for low aerodynamic noise without excessive penalties in size, weight, and performance losses, but with enough noise reduction to permit covert observation of an enemy on the ground.

Model 226-452A evolved into model 226-453A through external shaping and gold coatings on the windows, both of which reduced the RCS characteristics. Radar Absorbing Material (RAM) was added in the inlet and exhaust ducts to further reduce the RCS, and weight was reduced by the use of graphite epoxy for some of the external skin to produce Model 226-454A.  During the initial design efforts, a scaled GE 1/10 tip-driven turbofan was used to provide a low-noise, low JR propulsion system. The initial contract was extended to permit the evaluation of existing gas generators for powering the tip-driven fans and to incorporate design refinements in the Model 226-454A.

During this contract extension period, design refinements were made, resulting in the final Model 226-458. This aircraft utilized the core of a single TF-34 turbofan engine as the gas generator for the two tip-driven fans, retaining the internal fuselage bomb bay for covert missions.  It was also designed to carry much larger bomb loads externally under the wings, had a higher maximum lift coefficient, and included design features to reduce visual signature.  In external size and appearance the Model 226-458 was almost identical to the initial Model 226-454A, and internally it included similar features to reduce signatures such as RAM, duct acoustic treatment, and plug nozzles to reduce infrared signature and RCS.  The estimated “quiet” speed of the aircraft was 115 knots, with maximum speed of 445 knots.

At the time, the QAA configuration did not readily lend itself to conventional aerodynamic analysis methods, as the necessary computing power and mathematical models were still some years away. To reduce the radar signature of the aircraft, the airplane incorporated continuously curving leading and trailing edges on the wing and tail surfaces, blending of the wing/fuselage juncture, large leading and trailing edge fillets, and v-tails.  One source claims that “Yehudi Lights,” or airframe-mounted lights which matched the ambient background lighting level, were also considered to reduce visual signature.

These modifications aided greatly in reducing the RCS of the configuration. However, they also increased the uncertainty in the previous aerodynamic analysis.

To reduce this uncertainty, McDonnell Douglas conducted low speed wind tunnel tests in their Advanced Design Wind Tunnel on the Model 226-454A configuration, which essentially had the same aerodynamic characteristics as the final Model 226-458. The test results didn’t significantly alter the original estimates of total aircraft lift and drag characteristics.  Flow visualization pictures verified the favorable aerodynamic effects of the wing fillets and body camber.  The longitudinal static stability of the QAA configuration was determined to be adequate for the anticipated range of center of gravity locations.  The elevator power achievable with the v-tail layout was judged to also be sufficient.  The trailing edge split flaps provided a moderate increase in maximum lift coefficient but also caused a substantial increase in the drag coefficient. The aircraft also was shown to have a low level of static direction stability and no tendency toward directional departures at high angles of attack.

The continuously curving planform edges of the Model 226-458 were superior to the long straight edges on other aircraft; these tended to produce higher RCS values at the aspects that were perpendicular to the edge, and RCS increases with longer edges.  Curved edges reduce radar returns, but the returns are still generally detectable, and the drawback is that there is a continuum of viewing aspects that are perpendicular to some point on the edge. The use of a single curve that wraps around an entire aircraft, as on the Model 226-458, guaranteed that there would be a detectable edge return from almost any viewing aspect.  This may have been one of the key reasons it was never built.

The requirements for low aerodynamic noise and low RCS during covert operations dictated the carriage of weapons in an internal weapons bay. While enhancing the covert capability of the aircraft, this approach severely limited its military utility. As configured, the Model 226-458 could carry two 500 lb bombs in the internal weapons bay.  The aircraft was also capable of carrying up to 12,000 pounds of external bombs, thus making it comparable to conventional attack aircraft as well as capable of covert operations.  However, the external carriage of weapons dramatically increased both the RCS and aerodynamic noise signatures.

To evaluate the impact of external carriage of stores on the frontal aspect RCS signature of the QAA, a RCS test program of a full-scale wing section, pylon and simulated MK-83 1000 lb bomb was conducted.  The RCS model used in the program was fabricated and tested at the McDonnell Douglas Microwave Test Facility located east of Palmdale, California on an abandoned airfield.  The site consisted of a square mile of relatively flat, level land including four macadam runways. The site used primarily the Ground Plane Reflection Range Technique developed at Lincoln Laboratories.

The model consisted of a full-scale wing section, a full scale MK-83 1000 pound bomb shape and an aerodynamic pylon fairing designed to envelope the recently developed Douglas l4B bomb rack.  This rack was selected because it did not have sway braces, which could have increased the radiated aerodynamic noise.  A fifteen percent thick airfoil was used on the Model 226-454A design to provide high lift at quiet operating speeds. Therefore, the NACA 0015 symmetrical airfoil section was used to simulate the QAA wing.  The pylon contour was developed to provide a streamlined shape around the Douglas l4B bomb rack and reduce the RCS signature from that of the exposed bomb rack.  The MK-83 1000 lb bomb was selected because it was typical of the low drag weapons then in operational use and was compatible with the bomb rack.  The wing, pylon and bomb were fabricated individually and then joined together for the test program.

The model’s primary wing structure was wood and the outer skin was aluminum.  The wing ribs were formed of ¾ in plywood.  Plywood spars were also installed at five chordwise stations.  All joints were glued and fastened with wood screws. The wing framework was covered with aluminum.  The same basic construction technique was used in the fabrication of the pylon and cylindrical centerbody of the MK-83 bomb shape.

The centerbody wood frame was then capped with circular endplates.  The ogive forebody and conical aftbody of the bomb shape were machined of styrofoam. The forebody and aftbody were bonded to the centerbody and the entire bomb shape sprayed with sealer. Over the sealer a coating of conductive silver paint was applied to simulate the metal casting of the bomb case. Tail fins on the bomb were fabricated of wood and also coated with conductive silver paint.

The RCS test results showed a marked increase in the median RCS values when pylons and external stores were added to a clean wing.  The application of RAM to both the wing and pylon surfaces significantly reduced the median RCS values.  However, the median RCS values of the complete configuration (wing, pylon and bomb) with RAM applied remained higher than the median RCS value of the clean wing without RAM.

The results of the QAA program show that it was possible to combine low signatures with good mission capabilities in the same aircraft.  The resulting aircraft potentially had the operational versatility to perform covert missions where low signatures were essential, in addition to conventional attack missions with higher payload and performance.  It was determined that a truly stealthy aircraft could only be obtained by planning early on for low signatures of design, rather than as an afterthought. McDonnell Douglas concluded that the major design features needed for future stealth aircraft included low wing loading for low aerodynamic noise as well as good maneuverability, a propulsion system with high bypass ratio to reduce propulsion noise and infrared signatures, a superior ratio of thrust to aircraft weight for good performance, and exterior shaping to reduce radar cross section.

If a low RCS was the only signature of interest, the Model 226-458 could have been smaller with better performance. In spite of the limitations imposed by low signatures, however, the resulting enhanced probabilities of survival and surprise attack appeared to be desirable trade-offs for this concept.  While the wind tunnel test program was of limited scope, it did provide sufficient data to substantiate the earlier QAA low speed analyses. No design problems were uncovered in the test program.  The RCS test program provided frontal aspect data on the incremental RCS increase resulting from the addition of external stores to a clean wing. In addition, data on the degree of masking obtainable with the application of RAM to the wing leading edge, lower wing surface and pylon was also collected.

From the experience gained in the Quiet Attack Aircraft program, McDonnell Douglas concluded that stealth aircraft could be developed if low signature goals were emphasized from conceptual design through production and deployment.  The program showed that aerodynamic noise reduction and infrared reduction were critical in determining the size of a stealthy aircraft, and RCS reductions could be achieved by shaping the sized aircraft.  Substantial reductions of all primary signatures in existing non-stealthy military aircraft were deemed to be impossible to obtain by modification or retrofit.


The Navy ultimately passed on building the McDonnell’s Quiet Attack Aircraft for reasons that still remain unknown.  McDonnell reworked the concept for the Air Force in the DARPA pre-XST studies of 1974-5.  In early 1975, it teamed with Teledyne Ryan, which had demonstrated expertise in designing low RCS drones.  DARPA eventually concluded that the McDonnell Douglas/Teledyne Ryan team was not focusing their work on the areas that DARPA considered to be essential and no continuation contract was awarded.

This decision was no doubt related to the revolutionary work occurring during this period at rival companies – Northrop and Lockheed.  Northrop’s stealth work in this period is still largely classified, but the events at Lockheed are more well known.  Lockheed had developed a more successful approach to designing stealth aircraft by utilizing an obscure mathematical model by the Russian Pyotr Ufimtsev to develop software called Echo 1.  This software enabled the company to reliably estimate the RCS of a design made with faceted flat panels and thus achieve virtual invisibility to radar.  Lockheed built a model based on this research called the “Hopeless Diamond,” which led to the “Have Blue” XST test bed of 1977 and ultimately the famed F-117 Nighthawk.  Years later, with dramatic increases in computing power, designers were able to return to stealth designs with surface blending that were aerodynamically (and aesthetically) superior, such as the Northrop-McDonnell Douglas YF-23, which had a somewhat similar layout to the Model 226-458.  McDonnell Douglas’ experience with the QAA no doubt contributed to the development of this innovative but ultimately unsuccessful tactical fighter.

Black and White images courtesy Naval History & Heritage Command; contemporary artist’s impression courtesy Greater St. Louis Air & Space Museum (which I highly recommend visiting and supporting); and modern CG artist’s impressions ©2011 Jared A. Zichek.


 Leave a Reply

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>



This site uses Akismet to reduce spam. Learn how your comment data is processed.