Bill Gunston's conceptual stealth aircraft designs

[Mossie]

You can only see a small amount of the fan face from directly head on, which is the way the airflow is going.  Most of the air is flowing by the curved surfaces in the ducting & is following a flat S as it goes through the engine.

[PR19_Kit]

Mmm, while I agree with your characterisation of what air likes to do, I don't with your characterisation of the inlet path of the Harrier.  If you can see the fan face there doesn't appear IMHO to be much of an "S" bend there.

OK, try standing directly in front of a Harrier and see how much of the fan is visible. [see abpve, handily posted while I was writing the previous, thanks Mossie]

Most photos that show the fan are taken from alongside the cockpit and are looking down the inlet duct at an angle. The fan is very close to the front end of the duct, so much so that it almost, but not quite, touches the rear of the cockpit tub.

[rickshaw]

I'm sorry, that picture merely confirms that you can see the fan face.  No way that can be described as an "S" path.   An "S" path I believe indicates that the airflow has to actually turn corners.  In that photo it is obvious that it does not.  If it did, you wouldn't be able to see the fan face.

[Mossie]

But how much of the fan can you see?  Very little.  The only airflow that might go straight into the fan is a little bit of boundary flow along the fuselage side & IIRC, the intake narrows in close to the fan & creates a shallow stylised S itself.  Most flows with the ducting as I mentioned above. 

[Hobbes]

I'm sorry, that picture merely confirms that you can see the fan face.  No way that can be described as an "S" path.   An "S" path I believe indicates that the airflow has to actually turn corners.  In that photo it is obvious that it does not.  If it did, you wouldn't be able to see the fan face.

I'd call it an S path. Sure, there are no 90 degree bends, but the top view of the air path is clearly an S shape. Similar shapes in car racing circuits are definitely called an S.

[rickshaw]

I suspect we will have to agree to disagree.  I would call it at best a very shallow "Y", rather than an "S".   

[Weaver]

I checked the source again last night (Harrier by Francis K.Mason) and I did remember it correctly: the swerve in the Harrier's intake trunking (whatever you choose to call it) is already at the limit of efficiency, which is why a side-by-side two-seater was ruled out.

Anyway, back to the plot: my original point was that Bill Gunston's design is unbuildable, because, if you look at the position of the front nozzle, consider where the fan casing must be in relation to it, and how big the fan casing must be relative to the nozzle size (by looking at a Pegasus), you can see that the dorsal intake is too far back: it's on top of the fan, rather then in front of it.

Another criticism is balance. VTOL designs of this ilk have to have their engines in the middle so that their front and aft nozzles straddle the centre of gravity, and this is the case in BG's design. However, BG's design mates this to a tailless delta wing more typical of rear-engined designs, whose centre of lift must therefore be FAR aft of it's CofG. It would need large, constant tail-down trim to fly straight....

[tahsin]

One can almost sense the mood against sensationalist pressman getting into What if , of all places . THAT real life companies failed to build a hipersonic manned reconnaissance / strike complex should not be used be against Bill Gunston . He was on talking terms with Russian designers , he did fly Spitfires just at the beginning of the Cold War . How can he be really informed about the projects ? Or with all that Englishness , would he ever divulge anything he knew ?

Bill Gunston's biggest failing was his dogged effort to sell Harrier  the dog to US , specifically to Navy or USAF just in the way RAF was practically forced to buy the plane . I am for one who grew up reading Gunston and do not regret a single moment of it .

[Weaver]

I've nothing against Bill Gunston in general, and I grew up reading him too. I do feel though, that I didn't get an impartial or fully rounded view of the Harrier from him, and in the context of what were supposed to be factual reference books, his polemicising became tiresome: that sort of material is more suitable for magazine articles.

[Mossie]

I suspect we will have to agree to disagree.  I would call it at best a very shallow "Y", rather than an "S".   

The kink doesn't have to be much, as long the output flow is offset from the input.  In fluid dynamics, it's termed as an S curve, even though it doesn't actually resemble a classic S shape.  It's also sometimes called a Sigmoid (after the Greek letter Sigma, but it resembles this even less), although this is more of a mathematical term.

[PR19_Kit]

I suspect we will have to agree to disagree.  I would call it at best a very shallow "Y", rather than an "S".   

I know I have a spare pair of glasses somewhere...........

[martinbayer]

I have to admit I'm a bit curious about the exact definition of a 'limit of efficiency' regarding inlet flow redirection in the Harrier, especially since there has been at least one design by a major aerospace company with a far more radical inlet flow path - the Lockheed Reverse Installation Vectored Engine Thrust (RIVET): http://www.aircraftdesign.com/rivet3vw.jpg

For related performance and thrust loss questions, I was in email contact with its designer, Dan Raymer, about 10 years ago, and he told me the following: "The inlet 180 degree bend was perceived as a bigger problem and studied much more [than the exhaust flow, for which Lockheed had Rolls-Royce, who build the Harrier engine, do the nozzle design and analysis using Harrier data and analysis methods]. We did CFD analysis and also built and tested a blow-down duct model (about 1/10 scale as I recall). We needed to use internal turning vanes to keep the flow attached, but were able to do that and get acceptable front face distortion with, again, roughly a 3-5% total loss in the duct. This matches the 100-year-old data we found for flow through pipe bends in old mechanical engineering texts."

Martin

[elmayerle]

Given the work that's been done on fans capable of tolerating higher levels of inlet distortion (differences, for example,  between the fan on the F118 and the fan on the F110), I don't see that RIVET would have that much of a problem as long as they got the inlet to work reasonably well.

[martinbayer]

That's exactly the point. And since, according to Raymer, the inlet *could* actually be made to work pretty well in the RIVET arrangement, it raises the interesting question whether Gunston's V/STOL stealth fighter could be based on such a reverse engine installation as well. Also, a central weapons bay could be accommodated, if two smaller diameter side by side engines with only one forward and one aft articulated asymmetric but mirrored thrust nozzle each were installed instead of one large central engine with four nozzles a la Pegasus. One indication for this could be the fact that the nozzles of the Gunston design are clearly well below the aircraft centerline, while for the Pegaus they are on either side of the engine central axis. Both facts could be reconciled if the engine diameter fit in the lower half of the fuselage body, allowing the upper fuselage half behind the inlet to accommodate the reversing air duct and the room between the engines to host ordnance. One crucial question would of course be the minimum achievable engine diameter for the required thrust, but at least in principle such a layout would appear feasible. Plus, it would completely shield the engine fans from being visible from the front, thus helping to reduce the radar cross section and improve stealth characteristics.

Martin

[Weaver]

Thanks for the RIVET drawing Martin - I've never seen that before. I'm not surprised that it's feasible though, because the duct is relatively long, giving the flow time to settle down and space for corrective measures like vanes etc... to be fitted. The inlet duct in the Harrier is both kinked and short, which means there's no time for settling/correction before the disturbed flow hits the engine face.

I always assumed that the Gunston design was twin-engined (I think he says so in the accompanying text). The fact that the nozzles are on the lower corners of the fuselage doesn't automatically argue against a single engine though, since just about ALL advanced Pegasus schemes had such nozzles, in order to avoid the under-fuselage upflow which is such a feature of the Harrier.

I agree that a reverse-flow setup could be accomodated in the Gunston design. It's still hard to reconcile the size of the fuselage with the size of the nozzles and the implied size of the engines though: if they use a Pegasus-like design, i.e. a high-bypass turbofan with it's bypass air ducted into two of the four nozzles, then the fan diameter has to be WAY bigger than the core diameter, because it has to generate 50% of the total thrust with a much lower energy exhaust.