Hurricane wing

There’s a 1940 report on the performance of the Hurricane versus the Spitfire which was prompted by questions regarding the difference in speed between the two. The original was never published but is available as DSIR 23/784. I have not actually read the document but a summary is included in the paper on the Spitfire’s wing here https://www.aerosociety.com/news-expertise/journals-papers/papers-of-the-journal-of-aeronautical-history/. I guess the phenomenon you mention is just one more factor to add to the list. Without wanting to re-open the whole debate on the wing thickness it does appear that Camm’s attention to detail was somewhat less than that of Mitchell, reflected in the number of man-hours required to build each type.

Give me a day or two and I’ll scan the papers

edit…..Oh, hang about, wing-body interference was one of the factors noted in the report.

Thanks!

Yes, that ‘expansion’ problem (and how it got worse the ‘better’ the aerodynamic form of the fuselage taken in isolation) is the basis of my next TAH brain-dump / article.

Only thing is, the Hurricane could, I believe, have done with a bigger (or at least better-shaped) fillet. The full size 24-ft tunnel tests identified a thickened boundary layer over the root which got worse with the propeller on (higher free steam V). The wing’s high camber thick wing up against the slightly longitudinally convex ‘wall’ of the fuselage was enough to cause significant separation even at the very low speeds of the test. 9 lbs extra drag down to root interference at zero degrees AoA – the Whirlwind, with less upper surface curvature and a properly flat slab side, produced 1.8 lbs under the same conditions (a gentle breeze) and at plus 2 degrees.

I would be very interested indeed in the 1935 document. I have the work of Ower, but that was earlier, and all of the 1930’s American work (Klein et. al.) plus some Canadian (another Klein) and German (Muttray) material, but nothing British from 1935 apart from the known Supermarine stuff.

OK, I’ll send you copies. Of course the Hurricane did not require much in the way of a fillet given the fuselage structure and shape at wing root, as was well known at the time. There was only really an issue for some semi-monoque types where a circular/elliptical shape was considered optimal and hence ‘expansion’ of the airflow at the fuselage-wing joint in a low wing type needed to be managed. I’ll dig out the R&M from around 1935 that summarised the state of knowledge on the subject. A fair bit of the more detailed understanding was a spin-off (pun intended) of the copious amount of research carried out by the NPL and RAE on spinning and spin recovery that had been ongoing since the mid 1920s.

Apologies, I was thinking of RM1826 and 1937. Sorry – reading it back I got the wrong end of the stick thinking it was Hurricane not Hawcon, but it is still just as relevant really. I am still working on something on wing root and interference drag, and the more material and evidence I can bring to it the better!

Erm………..what experiment would that be Matt? I can’t see any mention of that in this thread, or have I missed something?

Suggest `The incomplete guide to airfoil usage`=google.or;m-selig.ae.illinois.edu.ads.html, might be useful for props and airfoils/aerofoils..

Ralph, please can you give me the short version of the Hurricane fillet experiments, perhaps by PM? Very useful for something I’m putting together.

Yes.

The efficiency of a wing was and is what mattered – that is the lift/drag ratio. There has been a lot of criticism with hindsight of designers thick wing choices, eg. the P-38, which completely misses the point. It is not just about thickness creating more drag, as if you are also creating more lift then you are efficiently using the energy of the aeroplane, which is the point of a good wing design.

The problem, as addressed by the R&M Schneiderman mentioned, was flow breakdown which generally occurred more readily on thick sections (though not necessarily) and which, when it did occur, usually happened at the root due to the influence of the fuselage over a thicker wing. Then you got drag but no commensurate lift.

Later, local acceleration to the speed of sound caused shockwaves on the upper surface of thicker wings, and so thinner was indeed better (less local acceleration over maximum camber point). But in subsonic terms, there was nothing so awful about a 19% wing.

Hurri RV7 – as was normal, t/c ratio went down from root to tip. There was a tiny change in the aerofoil, but don’t be misled by the tip sections, which were just ’rounded’ with smooth curves. The true aerofoil ends at the outboard rib. The other thing to be aware of is that the centre section gives a false impression of camber, t/c and angle of attack. You need to look at that seperately from the wing proper – ie outboard of the undercarriage. I built a Hurricane wing, and got horribly caught up in the un-necessarily complex geometries.

Excuse me if I’ve missed something in the above posts, but all the talk seems to be about the drag characteristics of the wing rather than it’s lift properties. Perhaps I’ve been mislead, but I was always given to understand that the thick wing was an intentional choice as it gave plenty of lift at high angles of attack, thus allowing the very tight turning radius that the Hurri is famous for. The argument goes that with British fighter design just coming out the the bi-plane era, it was fine to have a wing that perhaps wasn’t the slickest if it in turn gave something like bi-plane manoeuvrability, a “best of both worlds” scenario.

I’ve put a Clarke YH 19% profile over a side view from the center section drawing, they agreed very nicely!

I’m interested in the tip airfoil, which apparently is still a Clarke YH airfoil but with a different % thickness.

I guess it should come as no surprise to hear that the official reports, R&M 1826 and 1937, draw no such conclusions. The emphasis is on drag and its relationship to the transition point (were flow breaks down). Secondary discussion on the full-scale results versus model tests and on interference at the root, with experiments on alternative wing fillets. No doubt the results could be extrapolated to draw conclusions on what t/c is acceptable up to what speed, but it is not part of the conclusions. The comment about laminar flow is probably his interpretation of the discussion regarding the shift in transition point between what was seen with the polished smooth metal surface of the test wing panel and the situation with a typical ‘real world’ wing.

Sorry, should have read the book again instead of quoting from memory!

Actually that is not what it says. Quoting from the book…
“For aeroplanes designed to operate up to speeds of 200mph it would seem that the wing thickness should not exceed 20 – 25 per cent. Later, with the advent of laminar flow aerofoils, it was found that for aircraft operating at speeds up to 400 mph, the wing thickness should not exceed 12 – 15 per cent”
I’ll check the actual RAE report later to see how they presented the results.

Miles did some experimental work for the RAE with the M6 Hawcon, in which various wings could be fitted with thicknesses ranging from 30% to 15%. It flew initially in 1935. Results showed that for speeds up to 300 mph 20% thickness gave a reasonable compromise between strength & drag, remembering of course that monoplane design was in its infancy strength/weight ratio wise. For speeds above 400 mph then 15% was considered the maximum. Miles did though experience difficulty in producing a wing strong enough in wood to create the 15% wing. Data from Miles Aircraft since 1925.

Yes, but bear in mind that no aerodynamic expert would state that making the wing thicker would not appreciably add to drag. There were as many publications stating the opposite available then as there are now.

Camm wanted an efficient lifting foil for his interceptor, and he was reassured correctly that it could be thick enough to have sufficient strengh as well as being efficient. It would drag more and lift more than a thin foil, but have more or less the same efficiency. Claims that the NPL said it wouldn’t drag more ignore all the evidence presented at the time by bodies including the NPL.

Not that the extra drag of the wing by itself was that big a factor. If your wing is lifting more you fly at a reduced angle of attack for the same amount of lift. Reduced angle of attack means less drag. Not entirely self cancelling, but it’s another reason why efficiency of a wing as a lifting device was considered more important to an aeroplane’s performance than the absolute drag of the foil.

As it hàppens there were boundary layer thickening and flow seperation problems near the root that would have been lessened by either using a thinner wing, properly flat, vertical sides or a big expanding fillet – but all were to do with fuselage interaction not wing profile or thickness alone.

The NPL couldn’t warn about these because it would have taken tests on a complete Hurricane to find them (interference theory wasn’t generally up to speed in the UK, despite Ernest Ower’s best efforts). When this was done and these issues were found in late 39 by the RAE (real-world aeroplanes not aerofoil sections were more their bag), it was too late.

It may have been not being warned of ‘bad junction’ interference drag being exacerbated by thick wings that Camm was complaining about. Or it might just be that the story is entirely apocryphal.

I’m inclined to believe the account in “Sydney Camm and the Hurricane”, Hawkers were moving from externally braced biplanes to monoplanes and would have been focused on how to make the wing strong enough.

If a aerodynamic expert had said making the wing thicker does not appreciably add to the drag I would have considered this an acceptable compromise.

Yes it impacted the top speed, but it also made the Hurricane the best multi-role aircraft during the early parts of the war IMO.

To revisit this – I have been looking into this in some depth for a forthcoming article in The Aviation Historian.

There are a number of clarifications that are necessary. The first is that if one is looking for the NPL research on the Clark Y or YH it will be just that. Not the Hurricane, or ‘High Speed Fighter’ or whatever. Aerofoils were considered in isolation, and when they were tested in tunnels, they were hung in tunnels as simple solitary units.

The other important point is that when drag was discussed in relation to aerofoils in the 1930’s it was almost always in relation to lift. Absolute drag was very rarely even specifically mentioned, just published as supporting data behind the lift/drag curves! If there was such a conversation it would, in all probability, have been about the lift/drag curve of the Clark Y or any other aerofoil, not its drag. It was all about the efficiency of the foil – and that’s a legacy of much of the foil work being related to propeller design. Sure enough, the efficiency of a 19% Clark Y is more or less the same as that of a 13% one (below compressibility speeds).

Finally, Camm would have had access to piles and piles of data from all around the world on aerofoils all of which – whether conducted in a turbulent or laminar flow stream – show fat foils drag (and lift) more than thin ones. It is unlikely that he or anyone involved would construe the word of an NPL scientist to mean that all this (and common sense) was wrong.

Anyway, look out for the October TAH, folks. I’m in the one out this Friday, too!

I’ll add two pence as well. We could get rich.

I worked at Vickers/British Shipbuilders towing tank at St Albans in the early eighties (440ft x 20 x 10). Our bread and butter was testing models of cargo ships, but we had interesting work from time to time. This included semi-submersible drill rigs, submarines and patrol boats.

A fundamental limitation in a tank is that each test must be followed by ten minutes to allow the water to stop moving. A healthy order book dictated that we ran twenty four hours a day and the result of that was that there was no time to play. Even then, each run must have been worth hundreds of Pounds.

A relevant example was patrol boats. At the time there was a new Dutch design which was ‘short and fat’, and was fundamentally different to our own increasingly long and thin types. We decided to ‘play’, and devoted some time to testing a short fat boat, which proved to be inferior to our long thins. Ordinarily that would have ended research, but a couple of us thought that a quick and one-off mod to the stern might be worth a look. It cut 15% from the high speed drag! Congratulations came there none, and it was never pursued. The Hurricane wing question got me thinking about this, and I realised that if a customer had asked about a short/fat boat at the time, senior management would have said ‘that won’t work, forget it’.

My office had all the records of the management of the facility since it’s inception in the twenties, and fascinating reading it was. In hindsight I can see that the overriding motivation had always been the improvement of corroboration of model/full-size results. That continued in my time as computers were being introduced to the modelling (early CFD). Despite the fact that we were among the most able and qualified people to come up with new hydrodynamic thinking, we knew that there was neither time or money to allow it. Hard to believe really, when I tell you that the overall boss was the man who invented the bulbous bow. And I’ll bet he did that in ‘playtime’.

It is also commonsense that no national government would want to support publishing a paper in 1935 which set out the latest in aeronautical thought to a world starting to rattle its sabres….

Actually a surprising amount of fundamental aeronautical research was published by the ARC in Britain and NACA in the USA over many years. The characteristics of Clark YH was published by NACA in 1926 and by 1934, when the apparent advice on drag was given to Hawker, many similar reviews had been published, including a key paper describing the attributes of an aerofoil with various thickness/chord ratios. All of this was readily available to the aircraft industry

Getting into drag

It has been said that Camm selected the ClarkYH profile, with thickness of 19% at the root, for the Hurricane after the RAE had advised that there was no appreciable drag advantage in using thinner profiles, based on research carried out in their new compressed air windtunnel. Can anyone confirm that this RAE report was produced, if it is available or is the story apocryphal?

My tuppence :

In 1935, within the literature, such as Air Annual of the British Empire, Aircraft Engineering etc, there seems to be a live debate between designers on matters that were turning accepted wisdom upside down, eg stressed skin fuselages versus girder construction and monoplanes versus biplanes. I do not think there would be one defining document that represents the Eureka moment implied in the selection of aerofoil for the Hurricane.

One realization that comes out of the literature on various designs is increased drag using a deeper section monoplane aerofoil, but decreased drag in comparison to a biplane with thinner profile, but added overall drag caused by flying wires, struts.

In the context of Hawker thinking, the PV 3 & PV 4 biplanes immediately preceded the dramatic step to a monoplane design. There is a PV4 wing pictured on hawkerhind.com which shows a larger aerofoil in comparison to the Hart family biplanes, which travelled over to the first fabric covered Hurricane wing design. You can almost see the design thinking evolving.

A lot of Clark YH aerofoil work was done by Bristols; the Bulldog biplane had it which caused so much angst to Hawkers when they lost the contract to Bristols for RAF fighter in the late 20’s, no doubt Camm would have studied this competing design intensively and would have known its characteristics, if not the astonishing progress of American designs like the DC-2 which gave British design pause after the 1934 London – Melbourne Air Race.

No doubt Camm would have needed ammunition to go to the Hawker Board to persuade them to back a radical design change. No doubt ‘what was in the air’ at the time in terms of design thought and progress was driving him to reconsider his approach with the PV4. It would be natural that an independent third party verification of this design change was prudent technically and politically, and where better than from the RAE. It is also commonsense that no national government would want to support publishing a paper in 1935 which set out the latest in aeronautical thought to a world starting to rattle its sabres.

These designs were the product of an orchestra : mathemeticians, stressmen, businessmen, scientists etc all bringing something to the party,and the conductor was the chief designer. I doubt if a single document really would have such a profound impact, unless you live in a degraded age where one report on WMDs is enough to go to war over…