Canopy aerodynamics

[elbmc1969]

Which features influence the aerodynamics of fighter canopies and to what extent?

Of course there is the angled windshield and the flushness of struts. But what about curved windows?
Which fighters do you think have especially bad or good canopy aerodynamics?

There are pressure diagrams of some canopies either on a previous thread here or on YouTube. As I recall, the was a bubble canopy P-51 and either an earlier P-51 or a P-57 razorback.

Angular windscreens will see tremendous turbulence at the edges where the airflow spills over the side or the top. I haven't seen a pressure map, but I expect that the Bf109 is terrible. Curved surfaces allow something along the lines of laminar flow. Even curving the forward corners is far better.

 

[elbmc1969]

It doesn't get much worse than this:View attachment 776663 Blackburn Skua

I think that the Brits wanted an optically flat windscreen exactly perpendicular to the pilot's line of sight to eliminate distortion while dive bombing. I've spent a lot of time studying the Skua and it's the only thing that makes sense.

Keep in mind that the Skua was a dive bomber with a secondary role as a fighter.

 

[bf109xxl]

Aerodynamically perfect canopy. Surprisingly, the pilots didn't appreciate it.

 

[spicmart]

Aerodynamically perfect canopy. Surprisingly, the pilots didn't appreciate it.
View attachment 777991

Jedi fighter. Use the Force.

 

[bf109xxl]

Jedi fighter. Use the Force.

This guy didn't stop there. Here's one of his latest true Jedi projects.

 

[spicmart]

This guy didn't stop there. Here's one of his latest true Jedi projects.
View attachment 778043

It looks SR-71-ish.

 

[spicmart]

As aerodynamic as it gets.

 

[bf109xxl]

It looks SR-71-ish.

However, it is a Mach2 front-line bomber with a 1,500 kg bomb load and ski landing gear for landing on unpaved airfields according to a design proposed in 1959.

 

[OldGeezer]

One thing that's often overlooked is that the top of the canopy, where the curve reverses from up to down, is often a contributor to critical Mach number, even on propeller planes. In the process of going up and over and back down, the local airflow undergoes quite a bit of acceleration and compression shock waves can form there. It's an interesting subject to read up on.

 

[pigeon]

One thing that's often overlooked is that the top of the canopy, where the curve reverses from up to down, is often a contributor to critical Mach number, even on propeller planes. In the process of going up and over and back down, the local airflow undergoes quite a bit of acceleration and compression shock waves can form there. It's an interesting subject to read up on.

Would that be the reason for presumable dive instability in late P47's vs the razorbacks?

 

[Simon Thomas]

Edit: I think it is working.
It's ARC R&M 1811 from the 1937 v. 51 p. 225.

If the pdf doesn't load, use this link and go to page 225.

Technical report v.51 1937.

Hathitrust has all the rest of the ARC R&M's if you have a bit of a search.

 

[tomo pauk]

Some light reading from 1937:

I have problems with downloading the doc.

 

[GregP]

me, too.

 

[Aeroweanie]

Edit: I think it is working.
It's ARC R&M 1811 from the 1937 v. 51 p. 225.

If the pdf doesn't load, use this link and go to page 225.

Technical report v.51 1937.

Hathitrust has all the rest of the ARC R&M's if you have a bit of a search.

Unfortunately, Hathitrust doesn't have all of them. It looks like someone got tired of scanning and started saying "not scanned because of copyright issues". I don't find any beyond R&M 2169

I have a colleague digging some of the missing ones out of the University of Michigan library for me.

 

[Simon Thomas]

It might have been removed since I looked through there a few years ago. I found report 13 (1918) to report 84 (1951). Roughly R&M 360 to R&M 3127.
PM me if you need any between those.
Edit: Found the report I was missing.

 

[Aeroweanie]

R&M 2175 seems to be the upper limit. For the ones beyond this, I see: This item is not available online — Limited - search only — due to copyright restrictions.

Also, 2126-2150 is blocked.

Where did you find the ones further in the series?

 

[Jabberwocky]

http://www.wwiiaircraftperformance.org/mustang/Lednicer_Fighter_Aerodynamics.pdf

Above report (CFD analysis of the P51B/D, Spitfire IX and FW-190 A-8 and D-9) specifically notes that there were aerodynamics issues with the Spitfire's forward windscreen.

Wartime solutions (tested but not implemented) to the windscreen/canopy offered improvements of 7-12 mph in top speed.

 

[Aeroweanie]

http://www.wwiiaircraftperformance.org/mustang/Lednicer_Fighter_Aerodynamics.pdf

Above report (CFD analysis of the P51B/D, Spitfire IX and FW-190 A-8 and D-9) specifically notes that there were aerodynamics issues with the Spitfire's forward windscreen.

Wartime solutions (tested but not implemented) to the windscreen/canopy offered improvements of 7-12 mph in top speed.

Thanks - I wrote it!

 

[Jabberwocky]

Remember that it was difficult to make curved glass/perspex that was optically flat

Thanks - I wrote it!

I though your profile pic looked familiar - probably what triggered the memory of the paper.

Ever given any thoughts of updating it? I frequent F1 forums and the aero chaps/chappettes they are all fired up about the open-source aerodynamics analysis tools that have been developed in the last 5-10 years.

 

[Holtzauge]

Thanks - I wrote it!

Read that paper with pleasure. Found the analysis on the Spitfire's too steep windscreen especially interesting.

Another thing that interests me with regards to canopies is the leakage factor: AFAIK, the Bf-109 got a substantial gain in speed when they experimented with reducing the leakage in the engine cowling.

Given the pressure differences between the inside and outside on the canopies of typical WW2 fighters with sliding canopies, I would assume that that would have an impact here as well?

This would of course have shown up in the testing NACA did in the full-scale tunnels but is lost in wind tunnel modelling and I suspect also in CFD? Or is there a good way to account for it?