Favourite Naval Fighter

[ThomasP]

In addition to what DarrinW mentions above, I have run across a couple of factors that might have influenced the difference(s) in loss rates. As of early- to mid-1944 and later, not necessarily in order of importance:

1. In mid-1944 the F6F-3 had more armour (212 lbs) than the F4U-1 (150 lbs). Part of the F4U-1 armour was the 'skid' plate over the fuel tank on the top of the fuselage in front of the cockpit - which was obviously important for it to even exist, but for which there was no equivalent deemed necessary on the F6F-3. On the other hand, the F6F had a skid plate on the bottom front of the fuselage (which the F4U did not have) extending from the just behind the lip of the engine cowl to a little bit behind the wing LE. See the ACP sheets below.

The armour on the F6F-3 decreased from 212 lbs in mid-1944 to 200 lbs by Oct'45 - due to the reduction in oil cooler armour by 12 lbs.

The armour on the F4U-1 series increased from 150 lbs total in mid-1944 to 215 lbs in the F4U-1C/D by Aug'45 - with a 5 lb increase in the pilot's forward armour (from 40 lbs to 45 lbs) and a whopping increase in pilot's rear armour of 92 lbs (from 63 lbs to 155 lbs), and the removal of the 32 lb oil tank armour plate.

The quality of thin bullet proof armour used in US aircraft did not increase to any great degree during the war, so was the above due to operational research? If so the F4U units may have seen a serious need for up-armouring the pilot protection - whereas the F6F units did not? How much did the difference in armour contribute to aircraft losses?


2. The F6F had 3 separate internal self-sealing fuel tanks while the F4U had only 1 self-sealing internal tank (the 1 normally used on combat sorties). Getting home after having your fuel tank holed is always a serious problem, but in the PTO even getting home without holes in your fuel tank seemed to be a significantly greater problem than in the ETO - if for no other reason than the difficulties in navigation over open ocean.

In mid-1944 the F4U-1 had 177 lbs of SSFT, while in Aug'45 it had 179 lbs - so basically no change.

In 1944 the F6F had 244 lbs of SSFT, while in Oct'45 it had 190 lbs. Why the change? Maybe better SSFT material/design, or due to operational research, maybe both?

I think the materials and designs of SSFT did increase in effectiveness considerably during the early war period, but they were (I think) already near the best achieved during the war by the time the F6F and F4U became operational in any numbers. Maybe not?


3. The F4U performed more (% wise, I think) ground attack/close support missions than the F6F. I would assume that the PTO was similar to the ETO in that in general more fighter aircraft were lost per sortie to ground fire on ground attack/close support sorties than lost to enemy aircraft on Air-Air sorties.

The ACP sheets below are from 1945 and are the best readable quality I could find, so I typed in the appropriate 1944 values for the armour and SSFT weights on the respective sheets.

 

[EwenS]

It also needs to be borne in mind that for the USN while the F6F & the F4U entered combat from land bases in Aug and Feb 1943 respectively, it was the F4U that was engaged in those to a far greater extent through to the end of WW2 courtesy of USMC squadrons.

The position on the carriers was different. The F6F was the principle carrier fighter from its first combat at the end of Aug 1943 until Dec 1944. The only F4U on the fast carriers in the Pacific were a handful of F4U-2 nightfighters.

Only in Dec 1944, with the rise of the kamikaze menace and the need for more fighters on the carriers did the F4U begin to appear in the CAGs. Initially that was through the allocation of available USMC squadrons. Only in 1945 did USN F4U squadrons begin to appear on the carriers.

In the RN the first operational sorties with the F4U were in late Feb 1944 on Illustrious in the IO, but it was mid-April before her squadrons saw combat. Meanwhile in Europe the F4U first entered combat during Operation Tungsten against the Tirpitz on 3rd April 1944. The F6F combat debut in the RN was also during Operation Tungsten on 3rd April 1944.

 

[DarrenW]

I am curious, is your data set exclusively from the USN, or did you include operational experience from the Commonwealth?

The report only includes data pertaining to US aircraft assigned to both Navy and Marine air groups.

I can look for data concerning the Commonwealth air units but I don't have it readily available at the moment. Maybe someone here can provide this data perhaps??

 

[Geoffrey Sinclair]

The US did a lot more publishing of statistical data in detail then the British. There is no RAF equivalent of the USAAF Statistical Digest or RN equivalent of the USN Air Combat Statistics.

Expanding the USN numbers a bit. A big problem with the USN reporting is the use of action sorties, where one or more of the formation shot at or was shot at, which leaves a lot of sorties uncounted. That is a big chunk of the data needed to determine loss rates. Table 9 for carriers in the Pacific 1944/45 says CV 88,335 action and about 120,815 non action sorties, CVL 20,679 action and about 48,595 non action sorties, CVE 29,744 action and about 79,331 non action sorties

Accident rates tended to go up early in the war then decline, not much of a problem comparing F4U and F6F.

There are the different accident rates of fleet versus light versus escort carriers which is a problem when comparing the F4U and F6F.

The operational loss rates on action sorties for carriers, F6F 321 losses from 62,240 sorties, F4U 182 losses from 6,488 sorties, land is F6F 19 losses from 4,116 sorties, F4U 161 losses from 54,121 sorties (ignoring a few service unknown sorties, no losses). The land based F6F needs around 12 losses to match the F4U rate, the losses are small enough that one mini disaster could have made a big difference. However the non action land based sorties losses are 463 F4U to 48 F6F suggesting the F4U did have the lower non combat loss rate.

Operational loss rates on carrier action sorties,

By time, 1942 to 1945, 2.82%, 0.86%, 0.71%, 0.56 %, (land based 1.21%, 0.49%, 0.17%, 0.23%.)
by carrier type, CV 0.76%, CVL 0.61%, CVE 0.49%.
CV by year 2.46%, 0.92%, 0.71%, 0.70%
CVL by year, n/a, 0.75%, 0.66%, 0.54%
CVE by year 3.10% (1942/43 and only 419 action sorties), 0.76%, 0.29%

F4U flew 443 action sorties from CVE, no operational losses, the rest of the sorties and all operational losses from CV. F6F flew 41,715 action sorties from CV, 15,099 from CVL and 5,572 from CVE.

F6F by carrier type, CV 0.51%, CVL 0.6%, CVE 0.32%.
TBF/M by carrier type, CV 0.81%, CVL 0.63%, CVE 0.39%.

Table 9 is 1944/45 loss rates for action, non action sorties for CV, F6F 0.5%, 0.7%, Navy F4U 0.74%, 1.15%, Marine F4U 0.79%, 0.86%. As much as people talk about carrier based Marine F4U units the USN says 5,489 action and around 15,777 non action sorties by Navy units compared with 2,650 action and around 4,904 non action sorties by Marine units. It means the Marine units are close to where 1 or 2 losses makes a big difference to their casualty rates.

The action sortie figures have a certain symmetry, on CV F6F had 2/3 the loss rate as F4U, land based it was F4U had 2/3 the loss rate of F6F. For 1944/45 CV where you have an idea of non action sorties, F6F operational loss rate around 0.63%, F4U around 0.98% of all sorties.

Table 15, assuming Own Losses are operational only, land based 1944/45 F6F 0.29% of all sorties, F4U 0.25%. The Navy F6F had 12 operational losses from 1,868 action sorties, the Marines 3 losses from 1,646 action sorties.

 

[PAT303]

Looking at NACS table 1 on page 14, I found that Hellcats flew 95 percent of their total wartime action sorties from carriers. In contrast, Corsairs flew only 15 percent of their sorties from carriers. It is a well established fact that operating from carriers is far more hazardous than from land.

That's why the FAA pilots preferred the Hellcat, the Corsair had better performance but the Hellcat didn't try and kill you every time to took off and landed.
Drachinifel has some excellent YouTube video's on the subject as does armored Carriers.

 

[Admiral Beez]

I have to give a nod to the Fulmar. In the entire history of the FAA no other fighter shot down more enemy aircraft. Not bad considering it served only three years (May 1940-June 1943). But that aside, my pick is the Sea Hurricane as the only naval fighter made in Canada.

 

[Thumpalumpacus]

That's why the FAA pilots preferred the Hellcat, the Corsair had better performance but the Hellcat didn't try and kill you every time to took off and landed.

... and then there's the Seafire ...

 

[GrauGeist]

What, no love for the Airabonita?

 

[buffnut453]

What, no love for the Airabonita?
View attachment 873316

No, no, and thrice NO!!

 

[Thumpalumpacus]

What, no love for the Airabonita?
View attachment 873316

See what happens when you remove the nose armor.

 

[pinehilljoe]

What, no love for the Airabonita?
View attachment 873316

she does look sweet with Navy Yellow

 

[Admiral Beez]

Had the carrier Aquila entered service, its Reggiane Re.2001 fighters would have been a match for the FAA's Fulmars and Sea Hurricanes.

 

[GrauGeist]

she does look sweet with Navy Yellow

View attachment 873368

To be 100% honest, the P-39 and the XFL-1 were good looking aircraft.

It's a shame that looks did not reflect performance.

 

[Admiral Beez]

she does look sweet with Navy Yellow

View attachment 873368

I never understood why they switched to taildragger configuration. The Fleet Air Arm landed a conventional tricycle P-39 onto an escort carrier.


View: https://www.reddit.com/r/WWIIplanes/comments/zbsj91/p39_airacobra_landing_on_aircraft_carrier_the/

 

[Shortround6]

I never understood why they switched to taildragger configuration. The Fleet Air Arm landed a conventional tricycle P-39 onto an escort carrier.


View: https://www.reddit.com/r/WWIIplanes/comments/zbsj91/p39_airacobra_landing_on_aircraft_carrier_the/

Yes and...................................NO.
It was done by Captain Eric Brown om April 4, 1945. just short of 5 years after Airabonita first flew.
Not only was Captain Brown a very experienced test pilot this as a very carefully arranged 'stunt' to get Captain Brown into the record books one more time.
The P-39 was being used a test mule for a proposed flexible fight deck. Brown and the Aircobra did a number of close approaches (simulated landings) without actually touching down. At which point Brown declared an "emergence" and the Carrier's Captain gave him permission to land (this had been arranged beforehand). The P-39 was running light. Brown had flown over a dozen approaches. The P-39 landed safely, the mechanics fixed the "problem" (and/or added a bit more fuel) and Brown executed a take-off from the HMS Pretoria Castle which was an 18.5 knot ship that was about 100 ft longer than most escort carriers, or about 40-50 longer than the Sangamon and Commencement Bay classes.

The take-off was more a problem than the landing.
Landing a normal P-39 with a normal pilot would have been a much more hazardous undertaking.
Getting off a 1-2 knot slower ship with a shorter flight deck in rougher seas was probably not going to end well.

 

[GrauGeist]

I never understood why they switched to taildragger configuration. The Fleet Air Arm landed a conventional tricycle P-39 onto an escort carrier.

At the time the XFL-1 was being developed, conventional aircraft were the preferred types for carrier operations.

One of the main reasons, would be the tail-hook arrangement. You may notice the the XF5F, which was being developed at the same time as the XFL-1, also had a conventional design while it's USAAF cousin, the XP-50, had a tricycle undercarriage.

There was also the consideration of the nose-gear's durability, which could be problematic on land-based aircraft and would certainly take a beating when arresting aboard a carrier.

 

[EwenS]

The P-39 might have been the first aircraft with a retractable undercarriage to land on a carrier, but it was not the first tricycle aircraft to do so.

The first aircraft with a tricycle undercarriage to land and take off from an aircraft carrier was the Lockheed XJO-3, a one off version of the twin engined Lockheed 12 small airliner with a fixed undercarriage. That was achieved on the USS Lexington on 30 Aug 1939.

View: https://www.reddit.com/r/WeirdWings/comments/saiqa0/lockheed_xjo3_first_photo_is_of_her_undergoing/

Edit. There appears to be a conflation of two elements of carrier development here.

1. With the advent of tricycle undercarriaged jet aircraft, the RAE at Farnborough had been carrying out research as to how these might be operated off a carrier. One issue was how they might be adapted for tail down catapult launch and the problems that might be encountered during that as well as on approach to the flight deck (different aircraft attitude required for example) and how the long stroke nose wheel might react amongst the arrester gear. Airacobra AH574 was adapted for the shore based trials in 1944. There are photos of it on one of the test catapults at Farnborough in July 1944. Brown's "emergency landing" on 5th April 1945 was an extension of those early trials, that had included ADDLs into the arrester gear at Farnborough, and a precursor to him taking the DH Vampire aboard a carrier for the first time in Dec 1945. The early fears about the undercarriage arrangements proved unfounded.

1945 also saw a Meteor hoisted onto Pretoria Castle to check on its deck handling capabilities (no landings or take offs at that time).

2. The very first proposal of the flexible deck was at a meeting on 11 Jan 1945 at Farnborough where Brown was present. But that discussion was about how undercarriageless aircraft might be landed. Doing away with the weight of the undercarriage was the whole point of the flexible deck. Model experiments were carried out in 1945. Jan 1946 saw a small test rig at Farnborough for static drops and then weighted gliders. A full sized experimental flexible deck was not available until Dec 1947. Brown did use the Airacobra to familiarise himself with the low level runs required to the experimental flexible deck before making the first landing on it in a modified Sea Vampire.

See "Farnborough and the Fleet Air Arm" by Geoffrey Cooper.

 

[R Leonard]

F7F and PBJ both landing on and launching from USS Shangri La on 15 November 1944 had retractable tricycle landing gear.

File:Grumman F7F-1 Tigercat landing aboard USS Shangri-La (CV-38) on 15 November 1944.jpg - Wikimedia Commons

commons.wikimedia.org

File:North American PBJ-1 aboard USS Shangri-La (CV-38), circa in November 1944.jpg - Wikimedia Commons

commons.wikimedia.org

 

[don4331]

And of course 18/April/'42, the tricycle gear B-25s took off from USS Hornet. (Aside: I have 22 September 1936 for Potez 565 off NM Bearn as the 1st twin to land and take off from a carrier).

Tail dragger aircraft advantages:

a. Lighter - Stoddard-Hamilton Glasair II TD (Tail Dragger) was ~100 lbs. lighter than the FT (Fixed Tricycle)*​

b. More Robust - even in 2026, bush planes are mostly tail draggers.​

Tail dragger aircraft disadvantages:

c. Poor forward visibility when in take off position for single engine.​

d. Propensity for ground looping - as the center of gravity is behind the main gear, when landing in cross wind, there is danger of tail wanting to pass the nose.**​

e. Requires thrust of engines over horizontal stabilizer/elevators to allow the tail to be lifted (tail needs to be lifted, to allow the wings to be at low drag angle of attack allowing acceptable acceleration). Which is why tail dragger largely doesn't work for jets - the limited area (hot) thrust from a jet doesn't provide adequate airflow over the tail.​

In the both advantage and disadvantage category, the tail dragger design is "self limiting" on maximum" weight.

Weight needs to be centered at the corrected position at flying altitude (slightly ahead of the center of lift). When the plane is on ground, the center of weight is naturally behind the main gear (keeps the tail wheel on the ground). There will be a limit to how much weight you can add at the correct position for flying that the tail will be able to lift off the runway to take off. <You can cheat a little with extended tailwheels as they decrease the distance the c.g. is behind the main gear which makes it easier to lift the tail. The extreme is the FiSk 199 which lifts to ideal overload take-off angle>.​

*The Glasair FT was slower too, but having the nose gear in the middle of the thrust from the engine is largely responsible. For the fighters in question here, its not relevant.

**But carrier aircraft can't ground loop (well, its extremely difficult) - most of the time, the ship is pointed directly into the wind (no cross wind to speak of) and the arresting gear ensures the tail can't pass the front...

From what I've learned about the Allison V-1710 in the adding 2 speed discussion, I wouldn't have fewer concerns about taking off in P-39 than a Spitfire.

Remove the boost controls***, fill up with 100/130 fuel and the Airacobra will be accelerating down the deck with almost 1,600 enthusiastic American Quarter horses powering me on (4 blade propeller might be of advantage ala carrier Mossie). Tricycle gear will allow me to clearly see when to pull up.​

​

If you add a catapult/ramp even fewer concerns. Airacobra has lots of issues, but it can be made to take off in short distance.​

​

*** Better still boost controls with a takeoff (?1 minute?)setting, so I can focus on everything else right after takeoff, without worrying about the engine blowing up.

 

[R Leonard]

(Aside: I have 22 September 1936 for Potez 565 off NM Bearn as the 1st twin to land and take off from a carrier).

But a tail dragger

And a tail dragger landing on a carrier, generally, unless someone screws the pooch, has little propensity for ground looping. Ashore is another story. The F4F/FM series was notorious for ground loops when operating from a field.