Hacker News new | past | comments | ask | show | jobs | submit login
SR-71 Disintegrates Around Pilot During Flight Test (in 1966) (alexisparkinn.com)
126 points by frisco on March 17, 2009 | hide | past | favorite | 34 comments



Amazing story.

Read the story carefully. Note the incredible details of what that aircraft could do. Now read this line again:

By far, the most memorable flight occurred on Jan. 25, 1966.

This was more than four decades ago. Can you imagine what they're doing today? In some ways, it makes me angry, because I suspect that the military solved some engineering challenges decades ago that scientists are still wrestling with today.


If you like that sentiment, you will probably like Burt Rutan's talk at Ted ( http://www.ted.com/index.php/talks/burt_rutan_sees_the_futur... ). The theme is a lack of progress in aerospace and how to resolve it.


Seconded. That's one of my favorite TED talks.


This was more than four decades ago. Can you imagine what they're doing today?

The world is a strange place.

30 years ago, you could fly from New York to Paris in 3.5 hours (Concorde). Now it takes more than 7. You'd think technology would always move forward, but apparently that's not the case. If something is good enough, that might be good enough forever. Depressing.


More than 40 years ago.. This was some of the simulator technology they were using

http://en.wikipedia.org/wiki/SR-71_Blackbird#Flight_simulato...

The Link Simulator Company's SR-71 Flight Simulator was developed during 1963 – 1965 under a deep "black" security blanket because it and the team Link assigned to it were given access to CIA OXCART and USAF R-12 / SR-71 clearances, the complete list of names of classified vendors supplying parts and software that had to be simulated, the total aircraft performance envelope data and a government-produced satellite photo montage of almost the entire continental United States to provide optical imagery for the RSO's portion of the Flight Simulator. This later capability was mounted on a separate, large, rectangular glass plate (approximately 6 feet (1.8 m) by 12 feet (3.7 m) in size) over which moved an optical sighting head that traveled at the scaled speed and direction of the Blackbird during its simulated flight. Realistic and accurate images were then displayed in the Optical View Sight and SLR RCD (Radar Correlator Display) in the RSO cockpit. Imagery was not provided to the pilot's simulator, which like the RSO simulator, had translucent window panels with varying degrees of lighting to change a simulated flight from daylight to night flying conditions


That is true - it is the same with space. They put a man on the moon 40 odd years ago and now they are still struggling to launch a shuttle without it blowing up.


The moon trips were single shot with the sole purpose of putting someone on the moon before the Russians. Additionally, only a tiny part of the rocket had to re-enter the atmosphere to return the crew.

This link http://www.metafilter.com/48654/Jan-28-1986#1189388 perhaps explains what's wrong with the shuttle better than I could.


They've likely not exceeded the SR-71 by very much since then. The last several decades have seen the trend where costs are more frequently taken into account when deciding whether or not to pursue a line of scientific research. This is akin to the new direction the lazy chip makers have taken with multicore-multichip instead of continuing to work at improving the individual chips as much.


The F-22 is pretty cool. Total overkill actually.


I am particularly amazed by the F-22 avionics. More info on this at: http://www.f22fighter.com/avionics.htm

An excerpt:

"The F-22's avionics and software system is the most advanced ever integrated into an aircraft. It is the first aircraft to use integrated avionics, where the radar, weapons management system and electronic warfare system work as one, giving the pilot unprecedented situation awareness."


The SR-71 was an amazing piece of engineering.

The panels actually didn't line up - as they expanded when it got warm. So basically at take off the aerodynamics of the plane were shot - you had to fly for a bit for it to hit normal performance (it also meant it literally leaked until it got hot enough and sealed).


Which is why they took off with a minimum of fuel, flew around to heat up, and then refueled in mid-air.


From the SR-71 wikipedia entry:

"A defensive feature of the aircraft was its high speed and operating altitude, whereby, if a surface-to-air missile launch were detected, standard evasive action was simply to accelerate."


There is a rumor going around in the South African Air Force that the South Africans with the help of the Israeli Air Force (who had big ties in the bad old days) figured out a way to shoot the SR71s down. It had to do with a technique called storm-climbing (which was novelized by Wilbur Smith in 'Eagle in the Sky') where a normal fighter aircraft would go full-tilt with maximum weight in a downward trajectory, flatten out the dive, dump excess weight, and start climbing vertically. This way the pilots would get to an altitude where the missiles could get a successful lock, which in this case was far higher than the 'normal' ceiling for these aircraft. From this height they could then fire off the missiles. I'm not at all sure if they ever attempted it, but I'm pretty sure they didn't succeed. Still in interesting thing to hear :)


That storm-climbing maneuver sounds really cool!!! Do you have more detailed info on that? Other than Wilbur Smith's novel, that is.

It reminds me of that classical optimal control problem in the 1960s: the U.S. Air Force wanted to find the fastest climb for its F-4 Phantoms (so they could reach their operational ceiling ASAP to intercept Soviet bombers). The optimal path was counter-intuitive: first climb, then dive, reach supersonic, climb again. Sounds crazy, but they could reduce the climbing time dramatically that way!


I've actually noticed that behavior myself when screwing around in X-Plane with the F-4E; I take off at low-speed, high AOA to gain a few thousand feet, and then drop the nose at full afterburner to use both the engines and gravity to go transsonic, at which point the F-4 can accelerate faster, and maintain a higher angle of climb without losing velocity, due to the "lower" drag at supersonic speeds.

I had figured it was perhaps a weird characteristic of the simulation's breakdown of airframe components, but if that's how things worked in the real F-4's, that just gives me that much more respect for the physics simulation in X-Plane. No wonder it's FAA certified :)


The seminal paper in which these findings were published:

BRYSON, A.E. and DENHAM, W.F., "A steepest-ascent method for solving optimum programming problems," Trans. ASME. J. Appl. Mechanics, June 1962, pp. 247-257.

I can't find a PDF copy anywhere. It sucks. If you want to read a non-technical paper on it, try this one: http://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=00506395


Well, I don't have any material on it but you can see that it makes sense from a mechanical physics point of view. Load up a craft to maximum possible weight (with ballast even) Then fly as high and is fast as you can (during which you build up the maximum achievable potential and kinetic energy) then during a dive build up as much speed as you can (during which you convert your potential energy from the weight and height of the craft into more kinetic energy) as you level out, you dump the excess weight and start climbing. Since you now have less weight but the same kinetic energy, you are free to take your craft higher than the height from which it started to descend.


Yes, your explanation makes sense. That's what I thought. Nonetheless, I was interested in more in-depth technical info... some technical papers / reports on it would be interesting. By the way, have you heard of the pop-up maneuver?


no! please elaborate! :)


The pop-up maneuver's goal is to minimize exposure to enemy fire. Imagine that you're flying an F-16 over enemy territory and that you want to hit a nuclear power-plant with a couple of JDAM's. Since the target is certainly protected by lots of AAA and SAM's, you want to drop the bombs as far away as possible from the target. In other words, you definitely do not want to fly over the target, as that will most likely get you shot down.

You can use the pop-up maneuver then. You fly low and fast to avoid radar detection and AAA fire. When you get close to the target you climb at 45 degrees, roll 180 degrees (you are now flying inverted), prepare the JDAM's to be deployed. You then pitch up and when you are at the top of a parabolic flight path, you release the JDAM's. The inertia build-up during the ascent will make the bombs go faster and farther. Once the bombs are deployed, you fly low again.

There are some technical papers on this. It's heavy math: optimizing flight paths is not exactly a trivial problem ;-)


This sounds very much like something the SA Air Force did with their mirages in the Bush war of the 70's and 80's.

BTW - There is a book written about the history of the Mirage in the SAAF called 'vlamgat' (lit. fire/fiery - behind/ass). Try http://www.saairforce.co.za/reviews/33/vlamgat-cd.

Anyway. In this book, which is written by an ex-pilot, a maneuver is described which sounds very similar to the pop-up. The aircraft also approaches the target at low altitude, then pitches up. As the pilots pull up on the stick they release their bombs, but continue climbing up and away. The bombs (old fashioned, dumb bombs) would then fly in a parabolic trajectory like you'd expect and land somewhere near the target ;)

If I remember correctly this tactic was actually used in combat with surprisingly effective results. I say surprising because nobody believed you could actually hit a target with a bomb that flies in a trajectory with a horizontal length of a few 100 to 1000 meters and was released on the pilot's 'gut' feeling...


The biggest problem I see with intercepting a SR-71, is not getting another plane up to the same sort of altitude but getting the plane in the right piece of the sky, taking the wikipedia stats as an example, Mach 3.2 (2,200+ mph, 3,530+ km/h,1900 knots+) at 80,000 ft (24,000 m), if your interceptor take 2 minutes to get up to 80,000 ft, even doing some sort of zoom climb your aiming at a point 73 miles in front of the SR-71 position at take off, and I would expect the real world time to be longer from detection to interceptor at altitude. So you could only do it with a large radar network to track the SR-71 plus airbase in front of the current flightpath.

Then even if you did launch a missile it would not have an easy job of hitting the target


The SR-71 used something similar to go supersonic,

'After take off and the top up the Blackbird will perform a manoeuvre called the Dipsy Doodle, here it passes mach 1 by a climb to 33,000ft followed by a sharp dive to allow the aircraft to traverse high drag transonic range without using a lot of fuel. The aircraft will then climb to at least 60,000 ft.'

http://www.historyofwar.org/articles/weapons_blackbird.html


This has been posted regularly to Digg, Reddit, etc for years. Guess it's HN's turn - though I'm surprised it hasn't come up already.



Thanks for digging, I knew I had seen this before. Amazing story anyway.


... in 1966.

Perhaps there should be an HN rule about dating articles not from this year in their titles? That said, this is a great story.


I've noticed a number of people following the convention of putting the year in parenthesis at the end of a dated article. It would be great if everyone followed this. (Edit: just noticed this article follows the convention; I assume that the title was edited after your comment)


Also take a look at "Skunk Works: A Personal Memoir of My Years of Lockheed" by Ben Rich (manager at Lockheed after Kelly Johnson). Despite the title, he also talks extensively about the U2 and SR-71 projects - Amazing pieces of engineering all.

http://www.amazon.com/Skunk-Works-Personal-Memoir-Lockheed/d...


I always thought it would be safer to stick the pilot in the center of the plane, using periscopes or (now) video to fly. But I guess you're more likely to be thrown free of a disintegrating plane if you're in the usual pilot position.


Like the pilot said, though, it was only incredible luck that left him alive. I doubt that your odds are significantly improved or decreased by where you're sitting if your supersonic airplane falls apart around you.


Awesome! I loved the SR-71 as a kid and have been wondering what the replacement (Aurora) can do. I'm sure my son (now 3) will be fascinated!


Simply awe inspiring. Bookmarked for sure.

I've recently started assembling an SR-71 kit (with an f-117 to go) because I've always wanted the models sitting on my bookshelf.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: