5 ms·

The issue is that other forms of supersonic propulsion, outside of jet engines, are all still in very early experimental stages, and not proven. Engine manufac

The issue is that other forms of supersonic propulsion, outside of jet engines, are all still in very early experimental stages, and not proven. Engine manufacturers have spent a TON of R&D on fanjet engines and making them several orders of magnitude more efficient than the engines of 20 years ago even, but supersonic flight needs more than a fanjet - they need to look at turbojets, with afterburners etc. There is not much scope to play with with those unfortunately. What works at subsonic speeds at 30,000 feet won't at Mach 2+ at 80,000 feet. But there are other aspects to conserving energy. When Concorde lands, then energy expended by the braking systems on the rollout is the same as that used to power an average family house for a day. If they could improve kinetic energy recovery systems (as they do on Formula 1 cars and some electric cars), then that energy could be fed back in some way, to power on board avionics etc.

bigiain 10y ago
"making them several orders of magnitude more efficient than the engines of 20 years ago even" Surely that's an exaggeration? (If not - I'd love links to explain what's changed... 100 to 1000 times improvement since the mid '90s seems unlikely to me though...)
- david-given 10y ago
  It's an exaggeration. But aeroplanes have gotten much, much more efficient (and quieter) --- this suggests an improvement of 55% since 1950: https://en.wikipedia.org/wiki/Fuel_economy_in_aircraft https://en.wikipedia.org/wiki/Fuel_economy_in_aircraft Looking at actual figures: Concorde got about 5MJ per passenger-km. A modern 747-400 gets about 0.9MJ per passenger-km. Meanwhile trains vary enormously but US trains appear to get about 2.7MJ per passenger-km, while Portugese ones are about 0.07MJ; but I don't know if that's comparing like with like (the data is unclear). Cars seem to get about 1.2MJ per passenger-km. A big cruise ship seems to get about 5.9MJ per passenger-km, plus you need to spend lots more energy on, essentially, life support because they're so slow. These numbers were brought to you by https://en.wikipedia.org/wiki/Energy_efficiency_in_transportation https://en.wikipedia.org/wiki/Energy_efficiency_in_transport... and the 'units' command, which is awesome. Also, someone who isn't me needs to overhaul that page to use consistent units everywhere.
  - CydeWeys 10y ago
    > US trains appear to get about 2.7MJ per passenger-km, while Portugese ones are about 0.07MJ I'm with you, there has to be some flaw with differing methodologies going on here. It's way too discordant. I'm surprised at the 2.7 figure; are trains really more than twice as worse as individual passenger vehicles? I can't imagine so. Especially not the trains I use most often, electric-powered urban subways.
    - vonmoltke 10y ago
      I'm on my phone right now so I can't post nice details, but the Wikipedia number for "Portugal" is mislabeled. Per the reference, that is just for the Lisbon rail system. The US DOE report the US number was sourced from averages all passenger rail systems in the US, and specifically notes significant system-to-system variance. The biggest variance is that the Lisbon numbers are based on an average of about 400 passengers per train kilometer, while the US averages only about 40 across all systems.
    - david-given 10y ago
      Yeah, that seemed really odd to me, but I thought maybe it was due to diesel vs electric-with-regenerative-braking, which can be disturbingly efficient. Trains should be vastly more efficient than any individual transport. The only US commuter train I've seen was the SFO Caltrain, which was very not electric. Are there many electric trains there?
      - ghaff 10y ago
        >Are there many electric trains there? Yes. For example, Amtrak's Northeast Corridor is electric as is light rail for the most part (entirely?). There's also an electrification project underway for Caltrain. I'm not sure about how much commuter rail in the US is electric in general.
    - Someone 10y ago
      The higher the speed, the more important avoiding air resistance is to get high efficiency. Planes are efficient, ones they are high up in the sky because there is less air there. Unfortunately, they need lots of energy to get up in the low pressure regions, and we do not have ways to recover that energy in landing. Moving slower also increases the per mile efficiency, as air resistance grows faster than linear in speed. Net effect (highly simplified) is that trains beat high speed trains irrespective of distance, high speed trains beat planes on shorter distances (up to a few hundred miles), but planes beat high speed trains if the distance covered in one hop is large enough. Hyperloop aims to change that by being a train riding at low atmospheric pressure (planes don't only burn fuel to go from A to B, they also burn fuel to stay up; hyperloop wouldn't need to do the latter) (edit: ideally, it also would recover energy on braking)
ThenAsNow 10y ago
> Engine manufacturers have spent a TON of R&D on fanjet engines and making them several orders of magnitude more efficient than the engines of 20 years ago even, but supersonic flight needs more than a fanjet - they need to look at turbojets, with afterburners etc. There is not much scope to play with with those unfortunately. What works at subsonic speeds at 30,000 feet won't at Mach 2+ at 80,000 feet. Whether you fly subsonically or supersonically, thermodynamic efficiency will be driven by engine pressure ratio and turbine inlet temperature. The advanced turbofans flying today have very impressive gas generator cores (high-pressure compressor, low-emissions combustors, high-pressure turbines) that would be equally applicable to supersonic airframes. The low pressure part of the cycles will definitely optimize to a different point for subsonic vs. supersonic applications, with supersonic cruise preferring higher fan pressure ratios and lower bypass ratio. Like Concorde, you are going to want to size the engines such that you do not cruise in afterburner (assuming you are staying roughly below Mach 3; above 3 - 3.5, the ramjet cycle starts becoming more favorable). However, this is all within scope for what the aeropropulsion vendors know how to do. The big issue is that the specific engine configuration you are going to want for a supersonic cruiser is unique to that application. Originally, Aerion was going to use JT8Ds, which they later shifted away from. My point here is it's not principally an R&D issue, it's that if you want to minimize supersonic fuel consumption, you are going to want a specific configuration for which there will not be a huge engine market. Contrast with the market for something like a CFM56 or V2500. Being able to leverage the ADVENT three-stream/variable cycle work (and GE YF120 before that) will also help, but the point still stands.

bigiain 10y ago

"making them several orders of magnitude more efficient than the engines of 20 years ago even" Surely that's an exaggeration? (If not - I'd love links to explain what's changed... 100 to 1000 times improvement since the mid '90s seems unlikely to me though...)

david-given 10y ago
It's an exaggeration. But aeroplanes have gotten much, much more efficient (and quieter) --- this suggests an improvement of 55% since 1950: https://en.wikipedia.org/wiki/Fuel_economy_in_aircraft https://en.wikipedia.org/wiki/Fuel_economy_in_aircraft Looking at actual figures: Concorde got about 5MJ per passenger-km. A modern 747-400 gets about 0.9MJ per passenger-km. Meanwhile trains vary enormously but US trains appear to get about 2.7MJ per passenger-km, while Portugese ones are about 0.07MJ; but I don't know if that's comparing like with like (the data is unclear). Cars seem to get about 1.2MJ per passenger-km. A big cruise ship seems to get about 5.9MJ per passenger-km, plus you need to spend lots more energy on, essentially, life support because they're so slow. These numbers were brought to you by https://en.wikipedia.org/wiki/Energy_efficiency_in_transportation https://en.wikipedia.org/wiki/Energy_efficiency_in_transport... and the 'units' command, which is awesome. Also, someone who isn't me needs to overhaul that page to use consistent units everywhere.
- CydeWeys 10y ago
  > US trains appear to get about 2.7MJ per passenger-km, while Portugese ones are about 0.07MJ I'm with you, there has to be some flaw with differing methodologies going on here. It's way too discordant. I'm surprised at the 2.7 figure; are trains really more than twice as worse as individual passenger vehicles? I can't imagine so. Especially not the trains I use most often, electric-powered urban subways.
  - vonmoltke 10y ago
    I'm on my phone right now so I can't post nice details, but the Wikipedia number for "Portugal" is mislabeled. Per the reference, that is just for the Lisbon rail system. The US DOE report the US number was sourced from averages all passenger rail systems in the US, and specifically notes significant system-to-system variance. The biggest variance is that the Lisbon numbers are based on an average of about 400 passengers per train kilometer, while the US averages only about 40 across all systems.
  - david-given 10y ago
    Yeah, that seemed really odd to me, but I thought maybe it was due to diesel vs electric-with-regenerative-braking, which can be disturbingly efficient. Trains should be vastly more efficient than any individual transport. The only US commuter train I've seen was the SFO Caltrain, which was very not electric. Are there many electric trains there?
    - ghaff 10y ago
      >Are there many electric trains there? Yes. For example, Amtrak's Northeast Corridor is electric as is light rail for the most part (entirely?). There's also an electrification project underway for Caltrain. I'm not sure about how much commuter rail in the US is electric in general.
  - Someone 10y ago
    The higher the speed, the more important avoiding air resistance is to get high efficiency. Planes are efficient, ones they are high up in the sky because there is less air there. Unfortunately, they need lots of energy to get up in the low pressure regions, and we do not have ways to recover that energy in landing. Moving slower also increases the per mile efficiency, as air resistance grows faster than linear in speed. Net effect (highly simplified) is that trains beat high speed trains irrespective of distance, high speed trains beat planes on shorter distances (up to a few hundred miles), but planes beat high speed trains if the distance covered in one hop is large enough. Hyperloop aims to change that by being a train riding at low atmospheric pressure (planes don't only burn fuel to go from A to B, they also burn fuel to stay up; hyperloop wouldn't need to do the latter) (edit: ideally, it also would recover energy on braking)

ThenAsNow 10y ago

> Engine manufacturers have spent a TON of R&D on fanjet engines and making them several orders of magnitude more efficient than the engines of 20 years ago even, but supersonic flight needs more than a fanjet - they need to look at turbojets, with afterburners etc. There is not much scope to play with with those unfortunately. What works at subsonic speeds at 30,000 feet won't at Mach 2+ at 80,000 feet. Whether you fly subsonically or supersonically, thermodynamic efficiency will be driven by engine pressure ratio and turbine inlet temperature. The advanced turbofans flying today have very impressive gas generator cores (high-pressure compressor, low-emissions combustors, high-pressure turbines) that would be equally applicable to supersonic airframes. The low pressure part of the cycles will definitely optimize to a different point for subsonic vs. supersonic applications, with supersonic cruise preferring higher fan pressure ratios and lower bypass ratio. Like Concorde, you are going to want to size the engines such that you do not cruise in afterburner (assuming you are staying roughly below Mach 3; above 3 - 3.5, the ramjet cycle starts becoming more favorable). However, this is all within scope for what the aeropropulsion vendors know how to do. The big issue is that the specific engine configuration you are going to want for a supersonic cruiser is unique to that application. Originally, Aerion was going to use JT8Ds, which they later shifted away from. My point here is it's not principally an R&D issue, it's that if you want to minimize supersonic fuel consumption, you are going to want a specific configuration for which there will not be a huge engine market. Contrast with the market for something like a CFM56 or V2500. Being able to leverage the ADVENT three-stream/variable cycle work (and GE YF120 before that) will also help, but the point still stands.