Why does Harrier have split hot exhaust nozzles?

Question

This is a repost after the original questions were deemed too complicated.

Hawker Sea Hawk shot by Smudge 9000 under CC-SA 2.0 licence.

Harriers seem to inherit elements from Sea Hawk, that the exhaust from a single engine is diverted to two side exhausts.

@Peter Kämpf answered my previous question that Harrier was basically built upon Bristol/RR Pegasus engine, which drew inspiration from Michel Wibault's Gyroptere VTOL assembly.

From "S.A. WIBAULT FRENCH-GERMAN SOUCOUPE 'GYROTER'?" (1944) By Rob Arndt

Layout of the Pegasus. Public domain.

However, I noticed the contrast between two designs. Gyroter is more of a shaft-driven lifter like (X)F-35, as contrasted to direct-lift of Harriers. The hot exhaust of Gyroter directly exits at the aft in a single circular nozzle. The cold exhaust nozzles could have further forward positions than Harriers, allowing the placement of the core engine to the further aft, instead of closely around CoM, while maintaining thrust balance.

@Carlo Felicione , in his answer to Why does the Harrier jet have four landing gears? , said that the two aft nozzles forced Harriers to adapt bicycle landing gears.

Close-up of P.1127 Kestrel (Harrier prototype) shot by Alan Wilson under CC-SA 2.0 licence.

You can see the limited space for rear landing gear bay. Also the width of the tail assembly is directly limited by the separation of rear nozzles. It needs extra heat protection as well.

Also in the question Is it possible to modify an AV-8B Harrier II to a supersonic aircraft by applying some modification? @Aeroailas has shown the BS100 engine, designed for the Harriers' supersonic successor P.1154, adding aftburners to the cold bypasses (PCB) for supersonic flight.

Source: aeromodelbasic.blogspot.com

And I found P.1216, a improved decendant of P.1154. In Wikipedia the change is explained:

The problem with using PCB was that its thrust developed considerably more heat, noise and thrust induced vibrations compared with that produced in the Harrier. As a result, the rear fuselage configuration employed on the Harrier was not suitable. Beginning in 1980 the design team began investigating possible solutions. By removing as much of the rear of the airframe as proposed in the preceding P.1212 and P.1214 designs they developed a small short fuselage with twin booms which removed the tailplanes from being from being directly in the path of the engine's exhaust. A similar solution to the three-nozzle problem was adopted in the later YAK-141 prototype supersonic VTOL interceptor.

The configuration allowed for two nozzles at the front and a single 90-degree rotating rear exhaust which was more efficient than having to split it into the twin rear nozzles used on the Harrier. The booms provided a low drag location for the carrying of cannons in the front as well as missiles and avionics, while also shielding the jet exhausts from heat seeking missiles. Instead of the Pegasus engine whose design dated back to the 1950s the design used the Rolls-Royce RB.422.48 with PCB fitted only on the front nozzles. The design allowed the engine to be removed without needing to remove the wing as on the Harrier.

Source: Buttler, Tony (2017). British Secret Projects : Jet Fighters since 1950 ( 2nd edition) (Hardback). Manchester: Crecy Publishing. ISBN 978-1-910-80905-1.

BAE's ASTOVL concepts. Public domain.

So here's my question: Why does Pegasus engine have two widely seperated aft (hot) nozzles instead of one straight TVN? Why Harriers place their tail assembly between such nozzles?

Wouldn't it be better to have aft exhausts:

1. At bottom-sides in the further aft like Yak-38?

Uploaded by Tosaka under CC-A 3.0 licence.

2. Between two tail booms like P.1216, Yak-43/141 and (X)F-35?

Archived in Roy Cochrun Collection

3. Sheathed in the exhaust duct like X-32?

Power system of Boeing X-32, archived by globalsecurity.org.

4. Or elevate the tail assembly like Phantoms?

F-4K using British Spey turbofans. Public domain.

Any of them will eliminate the hurdles of strength requirement, heat protection, bicycle landing gears and thrust loss.

Answer 1

The big problem faced by VTOL is getting the thrust-to-weight ratio above unity. So anything that reduces thrust or adds weight is bad. The P.1127/Harrier, being the first practicable solution, wwas uncomfortably close to the borderline.

Long tailpipes waste energy and reduce thrust, so it had to be short. But a long rear fuselage then gets in the way. De Havilland's solution with the conventional Vampire had been to split the tail, and this would one day also be adopted for the Hawker P.1216 VTOL project and today's F-35 Lightning II. But Sidney Camm at Hawker worried at the time that it would add too much weight. So his solution for the Hawker Sea Hawk had been to split the tailpipe. It had worked very well, perhaps better than expected. Meanwhile a complex multi-ring system for thrust vectoring, such as used on the F-35, was untried and posed a significant technological risk to the project. The split nozzles would need only a single swivelling joint each, much reducing the risk. So he wanted to keep them for his VTOL design.

The layout also had the advantage of a "four-poster" powered lift airstream, which helped to make the hover more stable and controllable - another vital feature in those days of VTOL pioneering ignorance. Rolls-Royce built their "flying bedstead" free-flying hover test rig to develop the necessary control systems.

It all worked. The P.1127 became the world's first practical VTOL jet.

Answer 2

Why does Harrier have split hot exhaust nozzles?

Because a central tail exhaust is too far from the two front exhausts which are fed by bypass air. The split exhaust makes all ducting mercifully short, so the airframe can be lightweight and compact. Don't forget that pressurized gas cannot be ducted in air conditioning pipes; it needs heavy pressure pipes to do so. As I said in this answer, the P1127 was barely able to hover with 700 lbs of equipment stripped and fuel for 3 minutes. A long and heavy duct for pressurized gas was completely out of the question.

Regarding the reheat of the P1154: This was always a pipedream and the airplane would had melted its tires well before it would had been in the air. Next, fireproofing the fuselage structure would had required a titanium belly and screens of pressurized air to limit structural heating. This was demonstrated by the VJ-101 X2 (which had afterburners in its podded engines) where heat loads in hover with reheat were unmanageable.

A central tail exhaust is only feasible with lots of lift close to the fuselage tip. Note that the F-35B creates more lift with its forward fan than with its rear nozzle. But this kind of complexity is necessary if reheat is to be used for supersonic flight. And using two engines will carry the risk of an unrecoverable roll on engine failure. When the engine fails in a single VTOL jet during hover, the pilot can still safely eject. Not so if two engines are used.