Civilization's toughest technical challenges are those that require extraordinary (and constantly improving) performance to be delivered at a low cost.
Having worked on military jet engines at Pratt & Whitney back in the day, I can attest that this is an excellent overview! (I could say more but then I'd have to kill you. ;-)
This is great. These kinds of historical studies shed some light on why industrial organizations today seem to be less productive / accomplished than they were one or two generations ago -- conventional wisdom says it has something to do with MBAs, but when you zoom out, it's clear that there is a lot of survivorship bias / regression to the mean. When we think about jet engine manufacturers, we only look back at the ones that won, like GE and Pratt & Whitney -- we don't think about Westinghouse and so on. Same goes for commercial airframes, we look at Boeing but not de Havilland or Lockheed.
The military aspect also gets forgotten sometimes. It's easy to look at old annual reports from these companies and see that they brag about all of these excess profits taxes and high corporate income taxes and we wonder if we can get back to that era of greater corporate responsibility. The context that gets missed is that in the WWII / Korean War era, *the government put up a lot of the money for development and manufacturing in the first place* and excess profits "taxes" usually simply represent a return of the excess money the government put up (over the eventual actual cost of developing and producing the product + an allowable margin).
Thanks for writing this - it's a brilliant overview of the challenges in designing jet engines. I think jet engines are the coolest mechanical inventions in human history.
You touched on the technological challenges in achieving the greatly improved performance of modern jets. The challenges in manufacturing are perhaps equally challenging. You mentioned the evolution of turbine blades to deal with higher temperatures; the technology to manufacture them also had to be invented to make the parts possible. You mentioned single-crystal casting; it took elaborate new equipment and manufacturing techniques to produce the blades. Originally, the single crystal blades were cast in halves and then bonded together, because there was no way to produce the elaborate cooling channels in the casting. So, the casting industry developed ceramic cores to embed in the casting, producing the cooling channels as voids in the casting. But that meant they had to devise a ceramic that could be entirely removed from the final casting without damaging the casting. Then, the cooling holes had to be produced; latest versions include carefully shaped holes to get the cooling air coming out of the blade to "lay down" and create the cooling film on the surface of the blade. Lastly, elaborate coating systems were developed to protect against the oxidation and corrosion that come with high turbine temperatures, and new coating techniques had to be developed to apply the coatings. On top of that, ceramic coatings were developed to insulate the blades (that is, to reduce heat flow from the hot gases into the metal.
Multiply this level of technology across the major rotating parts, combustion chambers and fuel nozzles, compressor blades and vanes, controls, fan blades, cases, and even "mundane" parts like heat exchangers and tubes, and you can start to appreciate the technological wizardry behind these machines.
Having worked as gas turbine engineer for P&W, RR and GE, in pretty much every aspect of gas turbine design, development and ops, I can say this is a solid article. As you say, the issue today isn't really building a GT, that is now fairly easy, its developing that unit to something that competes on cost and performance, which is where the bleeding edge of the tech is.
The Comet accidents were not due to the windows. The cracks initiated at a penetration for an antenna. In the subsequent redesign the windows were rounded, which misled many into thinking the windows were to blame, as well as unfortunate wording in the official accident report.
How do there relate to the gas turbines used for electricity generation? Are they sharing the cores without the bypass? There are gas turbine generators up to 500MW, far larger than aircraft cores, so how much of the R&D for those is unique?
This is just a general question for the engineers. I’m not one but I was reading this great book entitled: the long hard road: the lithium-ion battery and the electric car.” And the author Charles j Murray talked about roger smith from “Roger and me” fame and his desire to built an electric car. Roger who wasn’t an engineer claimed that engineers don’t see the big picture and focus too much on the small things. For example Smith supposedly said, “if engineers had their way they would still be working on the 1971 Chevrolet.” Is there any truth to that statement?
Thanks for responding. While thinking about this subject I was reminded of something the soundtrack composer Wendy Carlos once said one asked when does she know when a piece is done and she replied, “all artwork is lefted abandoned.”
> It was also becoming harder to increase the power of piston engines —
> as engines got more powerful, they got heavier, offsetting gains in power.
Piston designs were definitely reaching a limit, but my understanding (from reading the literature; I wasn't there at the time) was that heat dissipation rather than weight was the primary issue.
Consider this series of engines:
R-1820 9 cyl, 1 row, 1,000 hp 0.84 hp/lb
R-3350 18 cyl, 2 rows, 2,200 hp 0.82 hp/lb (used in the B-29)
R-4630 28 cyl, 4 rows, 4,300 hp, 1.11 hp/lb
Quite a few accounts out there from B-29 pilots about the desperate struggle for speed (and thus cooling) after takeoff, and of deferring some of the preflight checks to the actual takeoff roll so as to reduce engine running time on the ground.
This is an absolute masterclass in techno-industrial storytelling. The RB211 case study alone is worth the price of admission — a brutally honest breakdown of how even smart teams get humbled by the edge of what's possible. Loved how you layered the economic, thermodynamic, and organizational failure modes into one coherent arc.
This essay should be required reading for anyone who thinks “just build it” applies to hard tech. The nuance around “building something that works vs. something that meets spec, scales, and survives inspection” is spot on.
It also explains why so many new players—whether in China or startups—struggle to enter this arena. Not just because they lack tech, but because they lack scar tissue. Subscribed immediately. If there's ever a book version of Construction Physics, take my money.
Want me to rewrite that in a more casual, contrarian, or academic tone?
Having worked on military jet engines at Pratt & Whitney back in the day, I can attest that this is an excellent overview! (I could say more but then I'd have to kill you. ;-)
Typo= Whittle engine flew 1941, not 1951.
Yep, this has been fixed!
This is great. These kinds of historical studies shed some light on why industrial organizations today seem to be less productive / accomplished than they were one or two generations ago -- conventional wisdom says it has something to do with MBAs, but when you zoom out, it's clear that there is a lot of survivorship bias / regression to the mean. When we think about jet engine manufacturers, we only look back at the ones that won, like GE and Pratt & Whitney -- we don't think about Westinghouse and so on. Same goes for commercial airframes, we look at Boeing but not de Havilland or Lockheed.
The military aspect also gets forgotten sometimes. It's easy to look at old annual reports from these companies and see that they brag about all of these excess profits taxes and high corporate income taxes and we wonder if we can get back to that era of greater corporate responsibility. The context that gets missed is that in the WWII / Korean War era, *the government put up a lot of the money for development and manufacturing in the first place* and excess profits "taxes" usually simply represent a return of the excess money the government put up (over the eventual actual cost of developing and producing the product + an allowable margin).
Thanks for writing this - it's a brilliant overview of the challenges in designing jet engines. I think jet engines are the coolest mechanical inventions in human history.
You touched on the technological challenges in achieving the greatly improved performance of modern jets. The challenges in manufacturing are perhaps equally challenging. You mentioned the evolution of turbine blades to deal with higher temperatures; the technology to manufacture them also had to be invented to make the parts possible. You mentioned single-crystal casting; it took elaborate new equipment and manufacturing techniques to produce the blades. Originally, the single crystal blades were cast in halves and then bonded together, because there was no way to produce the elaborate cooling channels in the casting. So, the casting industry developed ceramic cores to embed in the casting, producing the cooling channels as voids in the casting. But that meant they had to devise a ceramic that could be entirely removed from the final casting without damaging the casting. Then, the cooling holes had to be produced; latest versions include carefully shaped holes to get the cooling air coming out of the blade to "lay down" and create the cooling film on the surface of the blade. Lastly, elaborate coating systems were developed to protect against the oxidation and corrosion that come with high turbine temperatures, and new coating techniques had to be developed to apply the coatings. On top of that, ceramic coatings were developed to insulate the blades (that is, to reduce heat flow from the hot gases into the metal.
Multiply this level of technology across the major rotating parts, combustion chambers and fuel nozzles, compressor blades and vanes, controls, fan blades, cases, and even "mundane" parts like heat exchangers and tubes, and you can start to appreciate the technological wizardry behind these machines.
You need to hotlink your footnotes so readers can jump right to the note, read it, and return to where they were in the main body of the article.
Hotlinks exist on the web version, but they get stripped out in the email.
Ah, wonderful! Thanks for the info.
Great article, but your "Price per 1000 transistors" chart appears to say that the price went negative around 2004. :)
Ugh, this is datawrapper cutting off the last digits after the decimal, fixed.
Having worked as gas turbine engineer for P&W, RR and GE, in pretty much every aspect of gas turbine design, development and ops, I can say this is a solid article. As you say, the issue today isn't really building a GT, that is now fairly easy, its developing that unit to something that competes on cost and performance, which is where the bleeding edge of the tech is.
> Depending on how you count, there are just two to four builders of large commercial aircraft (Airbus, Boeing, Embraer, and now COMAC).
Technically, depending on how you squint, Russia's UAC (née Ilyushin) is still ekeing it out there making widebody planes: https://aviationweek.com/air-transport/aircraft-propulsion/russias-uac-rolls-out-stretched-il-96-passenger-airliner
It's decidedly an asterisk on the claim but a fascinating asterisk nonetheless!
Aviadvigatel is also still in the high bypass turbofan business
"fatigue failures around its rectangular windows"
The Comet accidents were not due to the windows. The cracks initiated at a penetration for an antenna. In the subsequent redesign the windows were rounded, which misled many into thinking the windows were to blame, as well as unfortunate wording in the official accident report.
https://www.youtube.com/watch?v=-DjnG74DDno (starting around 16:16)
Fascinating as always!
How do there relate to the gas turbines used for electricity generation? Are they sharing the cores without the bypass? There are gas turbine generators up to 500MW, far larger than aircraft cores, so how much of the R&D for those is unique?
Kind of weird not mentioning Aviadvigatel from Russia.
Also, Von Ohain and Whittle were leaders, but so what Brown Boveri - https://en.wikipedia.org/wiki/Neuch%C3%A2tel_gas_turbine
This is just a general question for the engineers. I’m not one but I was reading this great book entitled: the long hard road: the lithium-ion battery and the electric car.” And the author Charles j Murray talked about roger smith from “Roger and me” fame and his desire to built an electric car. Roger who wasn’t an engineer claimed that engineers don’t see the big picture and focus too much on the small things. For example Smith supposedly said, “if engineers had their way they would still be working on the 1971 Chevrolet.” Is there any truth to that statement?
Probably. Being technically educated but business manager and manager of many product engineers, there is a classic saying from Managers to Engineers"
Better is the enemy of Good Enough
However, when Safety is involved, stupid management kills people who don't listen to Engineers.
Boeing
Challenger
Columbia
Others
"... when Safety is involved, stupid management kills people who don't listen to Engineers."
Did you mean "when Safety is involved, stupid management who don't listen to Engineers kills people."?
The management who fails to listen to safety issues raised by Engineers. It's on the management
Thanks for responding. While thinking about this subject I was reminded of something the soundtrack composer Wendy Carlos once said one asked when does she know when a piece is done and she replied, “all artwork is lefted abandoned.”
> It was also becoming harder to increase the power of piston engines —
> as engines got more powerful, they got heavier, offsetting gains in power.
Piston designs were definitely reaching a limit, but my understanding (from reading the literature; I wasn't there at the time) was that heat dissipation rather than weight was the primary issue.
Consider this series of engines:
R-1820 9 cyl, 1 row, 1,000 hp 0.84 hp/lb
R-3350 18 cyl, 2 rows, 2,200 hp 0.82 hp/lb (used in the B-29)
R-4630 28 cyl, 4 rows, 4,300 hp, 1.11 hp/lb
Quite a few accounts out there from B-29 pilots about the desperate struggle for speed (and thus cooling) after takeoff, and of deferring some of the preflight checks to the actual takeoff roll so as to reduce engine running time on the ground.
This is an absolute masterclass in techno-industrial storytelling. The RB211 case study alone is worth the price of admission — a brutally honest breakdown of how even smart teams get humbled by the edge of what's possible. Loved how you layered the economic, thermodynamic, and organizational failure modes into one coherent arc.
This essay should be required reading for anyone who thinks “just build it” applies to hard tech. The nuance around “building something that works vs. something that meets spec, scales, and survives inspection” is spot on.
It also explains why so many new players—whether in China or startups—struggle to enter this arena. Not just because they lack tech, but because they lack scar tissue. Subscribed immediately. If there's ever a book version of Construction Physics, take my money.
Want me to rewrite that in a more casual, contrarian, or academic tone?
Great article! I suggest more along these lines, especially devices that are common such as two and four cycle internal combustion engines, and EVs.