How SR-71 Blackbird Designers Overcame Canonical Jet Engine Limitations | Hacker News Reader

How SR-71 Blackbird Designers Overcame Canonical Jet Engine Limitations

reddit.com

151 pointssigna1154 comments

Unfortunately, while the reddit post seems authoritative, it contains a multitude of errors and ends up somewhat misleading. The most significant error is:

"What the front half of a jet engine (the intake/diffuser, and the compressor blades, i.e. all the stuff that happens before fuel is burned) does is; it heats up the air until it's hot enough for fuel to ignite" ...etc.

The heating up of air in the gas turbine cycle (Brayton cycle) is an UNDESIRABLE, but difficult to avoid consequence of the ideal gas law. A higher efficiency and power output would be possible if this heating did not occur, hence measures such as water injection or intercooling on large stationary gas turbines.

There are a number of other more subtle errors in the post.

A bunch of redditors did in fact pick up on this, but their comments are buried deep down in the pile.

This.

If you have any interest in manufacturing, IMO the turbine blades in a jet engine are incredible.

Without going into loads of detail, 2 points that blow my mind:

1) Due to compression in the intake (and as per the ideal gas law D_Alex mentions above), the temperature of the air when it reaches the turbine is higher than the melting point of the turbine blades. The workaround for this is the blades are made with hollow channels through which cold air is blown, keeping them just cool enough.

2) You may be familiar with 'creep' - where materials under stress gradually relax to relieve that stress, even when the stress is lower than the yield stress of the material. Creep is made worse by (i) repeated stress cycles and (ii) high temperatures. In a typical jet turbine engine, you have blades spinning at 10000 rpm, at high tempratures, with a few microns clearance to the edge of the turbine housing. If they were to elongate and rake the edge of the housing at that speed ...

is point 1 typical for all turbojet & turbofan engines, or is this specific to then J58?

On modern jet engines: yes. There's a documentary / video about the manufacture of the Rolls-Royce Trent engine that mentions this fact and shows a CAD drawing / part of the manufacture process of the blades right after the combustion stage.

Point #1 is pretty cool.

I worked as an IT intern for Pratt and Whitney in the Middletown, CT plant in the mid 90's. I got to see samples of the blades straight from the factory floor, and indeed they had those cool-looking grooves and channels in the them.

My occasional trips to the shop floor for some PC maintenance or another was really the coolest part of that job.

Point #2 is also pretty cool. In order to prevent overcome problems of creep, the turbine blades can be made in a single crystal [1] [2]. This prevents much/all of the creep, as creep is associated with grain boundaries. It's pretty phenomenal that these single crystals can be manufactured into jet turbine blades.

[1] http://www.appropedia.org/Single_Crystal_Turbine_Blades

[2] http://www.freepatentsonline.com/6673308.html

*edit fixed links.

ran into this graphic while installing some modeling software awhile back, shows thermal stress on the blades

http://www.comsol.com/products/multiphysics/

Nitpick: creep is different from relaxation.

Creep is when you apply a constant load on the material, and the material elongates (creeps) gradually, keeping the same stress.

Relaxation is when you apply a constant extension to the material, and the stress reduces gradually (material relaxes), keeping the same extension.

It also dances around the key innovation used to divert the flow from the turbojet core to the ramjet component: the movable (fore and aft) inlet cone. It doesn't happen automatically.

Indeed. The control of shock waves inside the inlet (also via that moving cone) is a very interesting part of the engine, too.

While it is fascinating, this is a really long-winded and sort of rambling approach to an explanation of how the J58 worked. Even the Wikipedia page does a better job of describing the engine without all the filler of the author's need to type through his "geek out"

http://en.wikipedia.org/wiki/Pratt_%26_Whitney_J58

Not to mention that picture with the metal lighting up with the heat and the afterburner on full throttle

All that rambling is just his passion about the topic :)

How is it long winded? I thought it was pretty concise. You could fill books on just that one engine.

What's long-winded is the amount of text that's not actually related to the design and operation of the engine. The information he provided was actually fairly slim but there were a lot of proverbial "oohs and ahhs."

Neat anecdote about the SR-71's speed from a former pilot http://www.jumbojoke.com/the_king_of_speed.html

(couldn't find a better link; this is an excerpt from Sled Driver: Flying the World's Fastest Jet by Brian Schul)

You'd also probably like this speed story from (I think) the same guy: http://www.planeandpilotmag.com/speed-is-life.html

ooo ooo ooo, any more stories by that chap?

Two other stories mentioned among the comments on reddit:

http://www.barthworks.com/aviation/sr71breakup.htm

http://tailspinstales.blogspot.se/2010/01/slowest-blackbird....

There is a great podcast featuring an hour long interview with an SR71 pilot and author

http://omegataupodcast.net/2012/03/91-flying-the-sr-71/

And take a look at the other omegatau podcasts if you like geeky, if you speak German there are even more

I've read a lot of the Blackbird stories, and that first one might be the best yet.

Thanks!

Damn, that's an expensive test flight.

Even more fun when you listen to it: http://soundcloud.com/torrenegra/the-king-of-speed-sr-71

(This is one of my favorite stories ever. As such, I couldn't resist to get a professional voice to record it. I have tried to contact Brian Schul to transfer the ownership of the recording to him, but I have not been able to get a reply from him)

Awesome narration! Blackbird stories are always great. There is one of a Blackbird taking damage assesment photographs of Libya in 1986. There they were outrunning a SAM missile, the pilot said that its probably the time that the Blackbird was pushed to fly the fastest....

Wow, thank you for this - one of my favourite stories too. This is worthy of its own post in my opinion (Reddit would probably love it, too)

Edit: How much did it cost to get this done, and is the same voice actor still available? He did a really good job of capturing the ATC tone! I'd love to hear more sled stories read by him.

$70 or so in VoiceBunny.com. Disclaimer: I'm it's co-founder.

Is it possible to request the same voice actor that did the one you linked?

Interestingly, right now in Oxford UK is being tested a precooler for the sabre jet engine designed to power skylon spaceplanes to Mach 6 in air breathing mode (before going to orbit once out of the atmosphere with stored oxidiser).

This precooler is a crucial and fascinating bit of technology, cooling the incoming air down by 1000Kelvin in 100ms without itself frosting up, so it can be compressed and burnt. It is designed to solve exactly the problem outlined in the reddit comment. There was a fascinating BBC article about this recently:

http://www.bbc.co.uk/news/science-environment-17864782

Looks like a startup rather than 'dinospace' too, for what that's worth round here.

Why would you cool down to later compress (and heat up again)?

Sounds like a shotgun solution to the passer by such as myself

Edit: it is a shotgunsolution... They just added an old turbine after the cooler component to test it.

At Mach 6 it's just too hot and energetic to compress and stabley combust, as I understand. Recall that stagnation temperature is a function of the square of velocity. The reddit article says the compressor 'puts heat back in' but that's a byproduct rather than the aim. The aim of the compressor in this engine is to get enough moles of oxygen in the combustion chamber with enough moles of hydrogen to combust and produce enough thrust. You couldn't compress the air sufficiently to squeeze it into the combustion chamber if you didn't cool it a bit first.

This is my understanding based on reading what's publically available on the net anyway.

They had to develope a special fuel for the SR 71. One with very high flash point. I imagine that the temperatures at mach 6 are too high for normal fuels too.

The JP7 was also used as a coolant. The fuel was pumped around the hottest part of the plane to soak up the heat before it was burnt

Yep. Skunk works describes how they pumped it around the cockpit to help cool the aviation electronics and pilot. Just to add to the scariness of flying at mach 3+, you're also surrounded by jet fuel.

Jet fuel is remarkably non-flammable as anyone who has ever tried to light a Aga will discover

You guys hijacked a thread about a jet engine that doesn't even use traditional fuel. get with the program :D

They had to develope a special fuel for the SR 71. One with very high flash point. I guess that the temperatures at mach 6 are too high for normal fuels too.

Probably because it is a rocket engine and not a jet engine?

http://en.wikipedia.org/wiki/SABRE_%28rocket_engine%29

"Why would you cool down to later compress (and heat up again)?"

Because the efficiency of any combustion engine depends on the temperature differential between the input and the output. It cannot produce more energy than that differential (second law of thermodynamics) so cooling the input allows the engine to produce more energy, assuming the output remains the same.

but the input is 1000C, then it cool down to -700C, then the turbine compress to some 100C, then it mix with fuel and burn.

why not bypass the turbine at that point and take it easy on the cooling?

If you like aerospatial history but also if you'll like to know how all this birds were made, you should read "Skunk Works". I have just finished reading it, and is a great book. It tells all the development of the U2, SR71, F117, and more.

It's also greatly recomended for entrepreneurs, they worked as an small team, a very talented and focused group of people building unbeliable planes. At the end it has a great rant against all the bourocracy that has built up alog the years at the government.

Just an amazing read!

Second Skunk Works. They really did some amazing things, and that is probably the best book on it. It reads like folklore.org, in that it's one of the principal people involved retelling the stories.

It always kind of irks me when people talk about the heat of reentry coming from friction with the air. It doesn't come from friction, it comes from compressing the air.

Nope, if it was just compression you would not leave a plasma trail that far after the shuttle. Drag aka Friction is rapidly slowing the ~100,000 kg shuttle down from Mach 22. You can get disk breaks to glow slowing down a 1,000kg car from 100mph to zero quickly this is dumping over 2.3 million times that energy mostly into a lot of vary low density air.

PS: ((6.9 km/s / 100mph)^2 * 100) = 2,382,349.16

http://www.newton.dep.anl.gov/askasci/eng99/eng99460.htm

Sure, pressure does increase the temperature, but what's happening is your increasing the pressure of significantly heated air. As I understand it the shuttles bow shock stays under 3ATM during reentry which is hardly enough to be all that interesting by it's self. But, dump enough energy to get air to ~1000c at STP and increase the pressure and you get some real heat.

Also the shuttle is descending (aka falling) from LEO which adds significant kinetic energy, but the ride get's a lot more boring at low mach numbers so when you look at the energy needs to dump in that 'glow/burn' phase it's about the same as orbital velocity.

The point was that heat of re-entry of a blunt object results from compression rather than friction (negligible part), once it reaches the atmosphere (at beginning there is hardly any air and hence temp/heat/compression is small).

See more of the discussion here: http://www.rocketryforum.com/showthread.php?31212-Reentry-he...

There is a lot of miss information on that form.

First off, both energy and velocity is conserved even though Energy is a function of velocity squared.

So when you mix air of two different velocity's you get direct heating along with the change in velocity. You also get quite a bit of thermal radiation which heat up the air before it feels the physical effects from the craft. So, while the pressure does increase the actual pressure never get's all that high despite the huge increase in temperature. For a sanity check if you assume the energy of the air starts at the equivalent of ~30c at 1ATM and 1ATM = ~15lb/square inch then pressure required to get anywhere close to the observed temperatures get's ridiculous as in 100's of g's of deceleration.

Also, the blunt surface in front of a hyper-sonic aircraft collects a pressure wave off the surface. With the air next to the skin moving at the same relative speed as the craft. So the friction heating precedes the craft, and is negligible right next to it. The aircraft is also cooling the air next to it which is what creates the cooler buffer zone. (This is why they use blunt surfaces in the first place.)

Some is from compression, some is from friction. A flat plate aligned with flow will experience heating and not compress the flow.

Has anyone thought of strapping a rocket onto the SR-71 and getting it to orbit?

(Obviously much more complicated than my whimsical comment but maybe do-able ...?)

The SR-71 is probably too small to carry a space-faring rocket, but the concept is reasonable: http://en.wikipedia.org/wiki/Air_launch_to_orbit#Supersonic_...

This splits combined-cycle propulsion into multiple vehicles: air-breathing propulsion for the aircraft, chemical rocket for the spacecraft, thereby improving the efficiency of the whole launch.

Not for orbital flight, but NASA did use one as a testbed for rocket engines, such as those for the NASP.

It's funny how the Internet works sometimes: I clicked on a link on HN to a random comment on a Reddit post, just to find out that I know the original Reddit poster (saw his pic on the post)!!! Haha :)

What is really cool is to actually see an SR-71. The Intrepid in NYC has one on deck along with the retired shuttle Enterprise. There's a Concorde and some other cool aircraft too.

Reply on news.ycombinator.com