Bjorn’s Corner: Supersonic transport revival, Part 7

By Bjorn Fehrm

September 21, 2018, ©. Leeham News: In the last Corner we looked at the heating of the aircraft when cruising at high Mach.

Now we will address the biggest problem for supersonic airliners, the engines and nacelles.

Figure 1. The Concorde engine and nacelle. Source: Google images.

The Supersonic powerplant

The subsonic airline engines and their nacelles are technology which has been perfected over the last six decades. The engines have morphed from straight jets to high bypass turbofans and the nacelles are lightweight designs with only the thrust reversers as movable parts.

This changes dramatically when we talk about engines for Mach 2 airliners. To understand why, we need to look into some fundamentals on how engine thrust is generated and the role of the nacelle.

Engine thrust

A jet or turbofan engine is an air pump. It develops thrust by taking in air at the front and kicking it out faster at the back. The difference in the speed of the air entering the engine at the front and going out the back generates thrust:

                           Engine thrust = air mass flow * overspeed

If our engines would work internally like this (simply increasing the speed of the air) we wouldn’t have a big problem with engines for supersonic flight.

Unfortunately, for jet engines, there is “a candle in the wind” problem and a few more which complicate things.

The “candle in the wind” problem is the air can’t pass the engine’s combustor faster than Mach 0.2. Otherwise, the flame will blow out on the flame holders.

Additionally, fan and compressor stages don’t work well if the air passes them faster than Mach 0.5 (with Mach 0.5 axial velocity the blade tip airspeed is well over Mach 1. Any faster axial flow and the whole blade is supersonic, which spoils efficiency).

So for a Mach 2.2 SST the air molecules, hovering in the air at zero speed before entering the inlet, must accelerate to Mach 1.7 in the inlet (Mach 0.5 relative speed to the fan/compressor stages) and then to Mach 2.0 for the part which passes through the core.

Air weighs 1.2kg/m3 at ground level and about 0.2kg/m3 at SST cruise altitudes. With a Boom SST engine swallowing between 250m3 to 500m3 of air per second dependant on bypass ratio, a mass of 50kg to 100kg must be accelerated every second from zero to Mach 1.7 for the bypass flow and to Mach 2.0 for the core flow.

Ram air drag

The acceleration of the air is done with a force generated by the inlet. The resultant opposing force on the nacelle and by it the airframe is called “Ram air drag” as it has the same direction as airframe drag.

Additional aircraft drag means more thrust to keep the aircraft speed. The thrust to compensate for the airframe drag and the Ram air drag is created by the engine burning fuel in the combustors to heat the air which is then ejected at high speed out the back (the heated air has increased pressure and this is used in a nozzle to convert pressure to airspeed).

We need to take the core air from Mach 2.0 and the bypass air from Mach 1.7 relative to the surrounding air to about Mach -1 (the minus showing the acceleration is in the negative flight direction) to get thrust when the air exits the engine.

Once again we are changing the speed of the air we have going through the engine. This time, the change in speed is Mach -2.5 (bypass air) to -2.8 (core air). The change of speed of the air weighing 0.2kg/m3 costs energy, energy which is taken from the fuel.

A normal airliner engine must also accelerate the air in the inlet and then brake it to zero and kick it out the back at about Mach -0.5. But now the bypass air, which is 90% of all air, is only speeded up from zero to Mach 0.35 (Boeing 787, Airbus A350 cruising at Mach 0.85) and the total change of speed before the air goes out the back is less than Mach -1.

The Ram air drag for a subsonic airliner’s turbofans is modest compared to an SST engine’s Ram air drag.

Fuel consumption

The process described above is the reason a Mach 2 SST engine consumes twice the fuel per produced thrust unit compared to a normal airliner engine if all processes have the same efficiency. This is also a challenge as we will see.

One of the challenges involves the acceleration of the air in the inlet to Mach 1.7 and then kicking it out the back at Mach -1 (all relative to the surrounding air). This is not so easy as we will see in the next Corner.

24 Comments on “Bjorn’s Corner: Supersonic transport revival, Part 7

  1. It is the compessor that shall work from standstill to M2.0 that is the main reason of reducing the speed of the air sucked into the Engine. The combustion can nowdays be done supersonic (SCRAMJET), the turbine can work supersonic axial flow (with less efficincy) and the exhaust is no problem to accelerate supersonically. If you use rocket assiset launch to M1.7 a cheap ramjet like on some Air to Air missiles can drive the Aircraft in supersonic Cruise, but for Commercial operations at Airports with option for TOGA you need the rotating blades+vanes compressor and it does not like supersonic axial flow.

    • Thanks Claes,

      haven’t studied Scramjets, so good to know. Re Ramjet engines, what is the internal speed of the air through the combustion process? What I’ve read it’s around M 0.2 as well (for the same reasons as the jet engine’s) so we have the acceleration, braking and negative acceleration there as well. = Ram air drag.

      • The main attraction or ram engines is that they do not have a rotating compressor, they take a few oblique shocks then a normal chock to go subsonic and further compress the air thru a diffusor and work subsonically in combustion. The disadvantage is that they only work supersonically, the faster the better. The scramjets do without the noraml chock and sometimes use oblique chocks in the burner as flameholders. Some AA missiles use a rocket in the exhaust to accelerate to supersonic speed and then have the heat of compression to warm up a solid fuel to emit combustional gases that mix with air and drive it in ram jet mode.
        In theory you could do the same with an SSBJ, use a combination of catapult and rocket assisted take off to go supersonic then run it in ramjet mode, works also with JET-A1 (or JP-7 if you fly very high).
        You need a “spare charge” for a one time “Take Off Go Around”. Like using the take-off assisted rockets on a fighter for an engine out situation to reach the runway in a dead stick landing or a one time go around.

  2. Why dont they have a top mounted conventional jet engine. And 2 wing mounted supersonic engines. And switch back and fourth

    • Top engines are ruled out because of increased maintenance issues and rotorbust requirements. Also, more engines is more weight and more drag and more cost. None of it is impossible, just not even close to worthwhile for a commercial product.

    • variable geometry inlets are not “mandatory” for supersonic flight. for instance the F-16 and (I believe) F-35 use fixed geometry inlets. they XF-32 only had a variable geometry intake on the STOVL version for increased airflow at hover, but was intended to be fixed in forward flight.

      IIRC even the YF-23, a M2.2 aircraft with M1.6+ supercruise had a fixed geometry inlet.

    • @Billbo, the article is password protected, can you synopsize the findings a bit more than your one-liner, please?

      • sorry, I forget that I get awin through work…

        revised IP turbine has been rolled out to 60% of the “affected” fleet (1000Bs and Cs)
        new blade base material and coatings to fix erosion problem
        redesigned compressor coming
        testing with cracked compressors ongoing in hopes of convincing authorities to extend inspection intervals
        Ten and T7000 compressors “haven’t shown” the cracking problem yet, but have the same eigenmode in the same performance window, so they are planning to roll the same updates out to Ten and T7000s

        • Well that fully confirm my suspicions.

          Looks like maybe in time (stay tuned) to avoid a massive fleet failure syndrome ala ANA.

          Its no longer a bug, its a feature.

          We design them to crack so you don’t have to.

          Lot of bucks tied up in that mess.

        • should have mentioned, the failure mode for the first 2 stages of the IP compressor is harmonic resonance caused by the differential rotation speeds of the LP and IP compressors. fix is basically reshaping the blades to change the natural harmonics of the blades to take their resonance frequency outside the range experienced in normal operation.

          something that you would think would be a pretty basic analysis performed during the design stage.

          • It is sometimes not that easy to calculate an exact Life of these blades as you have a variation in blade shape and eigenfrequencies you have a variation in dry film durability and a variation in friction in the blade attachment with time. The the first few blade stages transmit some of the disk hoop stress thru the blade root flanks lowering the stress in the disk slot.
            Hence always good to have a “suffcient margin” of induced harmonics at high rpms.
            One old trick is to use prime numbers of blades and vanes hence reducing the chance of a rotor giving pressure pulses that match the other rotors per rev harmonics.

        • Haselbach is quoted as saying of the TEN “So even if it takes 3-4 years for the problem to show up, now is the time to tackle it. It might never show, but that’s what we are doing.” Would be interesting to know what % likelihood he apportions to the different outcomes absent the changes.

          Also re the erosion problem being down to high sulfur pollution levels around certain APAC airports, would be interesting to know whether this was a case of encountering an unexpected performance boundary that simply happened to be with the T1000 or instead if it was not unexpected and was down to design failure.

          • Bilbo:

            Interesting I came across a WWII reference to that in the ME262 engine and the same fix, reshape the blade.

            Its extremely puzzling that a mfg of jet engine would have missed it at all let alone for so long.
            I continue to wonder if arrogance is involved. We did this a million time, we don’t need to do no stinking tests, we are RR! (shades of Cat) – people and organization do get full of themselves.

            Now the core (pun intended) is sulfur, before it was salt in the air.

            How much time do NZ jets spend in Asia air though? They got his as bad as anyone (two shutdowns).

            I find it extremely worrying that rather than being prudent, they are using models to forecast behavior so they can try to remove the restrictions. Fix the problem and you don’t have those restriction do you? Put your time and effort into a fix not some kind of iffy work around.

            We have seen failures of testing to prove out engines before, not something you should take a chance on.

            Terms like very little are best replaced with NONE.

  3. The more you read the more amazing Concorde looks! No surprises the soviets actually asked the UK (in the Cold War) for help getting their engines (on the TU-144 ) to work better, or why the engine ‘computers’ that adjust the engine ramps were removed when the Concordes were retired as they are a ‘state secret’.

    • I am truly in sceptisim that any 60s era tech is considered a state secret.

      Urban Legend strike again.

    • Theres an interesting story on the engines for the Soviet Tu-144. The only available engine with the required thrust for the initial prototypes was the Kuznetsov NK-144 turbofan .
      Clearly a turbofan wasnt the answer but it took some time for the Kolesov RD-36-51 turbo jets to be produced and satisfactory for use. This of course produced much improved range. An unusual feature of the engines was ‘translating plug’ in the exhaust to create a variable area nozzle.
      https://en.wikipedia.org/wiki/Kolesov_RD-36-51

      • What I agree with is the Concorde was an amazing execution.

        Technically it had few flaws.

        Almost unheard of for a jump in tech like that to come off so well.

Leave a Reply

Your email address will not be published. Required fields are marked *