FW-190 - How Good Was It, Really?

[swampyankee]

Drgondog and Swampyankee, were there any areas of design that weren't known to designers of WW2 aircraft that led them up blind alleys looking back?

This is something I'd have to give a lot of thought to. I don't think there were too many blind alleys of aircraft that entered WW2 service, although there were a couple of aircraft that probably would not have been accepted into service had they been developed in peacetime, one being the Curtiss SB2C.

I would think that two big areas would be ergonomics and damage tolerance. The efficiency of the pilots of the F4U and P-38 were compromised by cockpit design.

 

[pinehilljoe]

There were several encounters with FW 190 and Bf 109 and F6F - Operation Dragoon in Med and IIRC in North Sea area.

do you know of any books with accounts of the encounters?

 

[DarrenW]

I suppose it could be called living a dream. But, I don't get to fly them, and rightly so since most of my flight time is in Cessnas and Pipers, with a smattering of odd time in a few diverse types.

I DO get to help fabricate restoration and repair parts, and get in a lot of riveting and general fabrication, and I never thought I'd get to do that. Volunteering at a museum that flies warbirds DOES help a lot. Anyone can do that if they are lucky enough to live near such a museum. As it happens, I do.

You might try volunteering when you get near retirement. I learned my sheet metal skills within a couple of years of starting, and have been doing it now for about 10 years. I DID get to make a few parts for some rather interesting restorations. I'm really looking forward to seeing our Bell YP-59A fly! It has a lot of my own work in it, along with a LOT of work from about 100 other guys, including real restoration experts who work for and with Steve Hinton.

I like the idea of doing some of those same things when I reach retirement age. There's a few museums within reach of me, one of them is a flying museum as well. Just being around that kind of living history on a semi-regular basis fascinates me greatly!

 

[DarrenW]

I would think that two big areas would be ergonomics and damage tolerance. The efficiency of the pilots of the F4U and P-38 were compromised by cockpit design.

Speaking of cockpit design, I have read that the FW-190 instruments and controls were very well thought out and because of this most pilots could instinctually find their way around the cockpit with little to no trouble. I never compared it's cockpit to that of the Bf-109 series, but was curious about what details of it's layout were considered an improvement over the Messerschmitt fighter? Also what aspects, if any, weren't as good?

 

[drgondog]

Bluff bodies frequently have reduced drag coefficients as Reynolds' number increases; laminar flow is more prone to early separation. Of course, one wants to avoid separation.

One cannot avoid transition from laminar to turbulent flow at very low RN relative to normal flight speeds. The Base components of Parasite Drag are in turbulent flow. Separation due to adverse pressure gradients is a different discussion.

As an aside, do remember that 1) aerodynamicists are all going to be working from the same basic assumptions 2) NACA went to a great deal of effort and expense to make wind tunnels, such as the variable density wind tunnel, to achieve full-scale Reynolds' numbers and subsonic Mach numbers and 3) many of the reported Cd0 values are from flight test data. Or, largely bad, from handbook data.

Agreed - That said, even within the same company the referenced baseline Parasite drag components as function of RN, varied. For example, The P-51B-1 Performance Calcs began with a reference base point of CD vs RN at 1.84x10^6. The Basic Drag build up for the P-51D- was at 2.0x10^6and the P-51H Base Drag build up was at 9x10^6. Even within the NAA Mustang group you had to be a little careful in 'selecting' a CDo for quote.

All of the above Reports picked drag data and components from NACA wind tunnel testing results. A compilation of a range of both airframe low speed drag coefficient and individual component Cd for P-51, 38, 39, 40, F4U, F6F, etc can be found NACA Wartime Report L5A30.. The Wind tunnel airspeed was a uniform 100mph, then each RN was calculated as f(mean aero chord).. All of these were production airframes, not scale models.

As to methodology (at least for NAA) "Aerodynamics of the Airplane" by Clark R. Milliken, John Wiley and Sons, (1941) was the foundation for Schmeud's team

 

[Ivan1GFP]

Agreed - That said, even within the same company the referenced baseline Parasite drag components as function of RN, varied. For example, The P-51B-1 Performance Calcs began with a reference base point of CD vs RN at 1.84x10^6. The Basic Drag build up for the P-51D- was at 2.0x10^6and the P-51H Base Drag build up was at 9x10^6. Even within the NAA Mustang group you had to be a little careful in 'selecting' a CDo for quote.

This business of different CDs even for the "same" aeroplane is exactly why I was asking for the equivalent flat plate area.
It is a much simpler concept and beats trying to figure out whether everyone is using the same reference area.
Here is a good description from a name that seems to come up a lot:


8/26/94

Analytical Methods {NWNet}


The concept of "equivalent flate plate area" comes from noting
that while:

Drag Where: Drag = resistance force
CD = ------------------ rho = density
.5 rho Vel^2 Sref Vel = Velocity
Sref = Ref Area

is nondimensional, the Sref is awkward, as how the reference area is
chosen can differ. So if we instead define:

Drag
f = -------------
.5 rho Vel^2

the Sref no longer appears, but f has the units of length squared. If
English units are used, you get the equivalent flate plate area (f) in
units of square feet. Note, this does not mean that a flate plate of
the same area as f would have the same drag - a flate plate has a CD of
roughly 1.17 (in 3-D flow, according to Hoerner and 1.98 in 2-D flow),
not 1 as implied in the equation! We can see this from:

Drag = CD .5 rho Vel^2 Sref

where Sref is the frontal area of the plate. Hence, the equivalent
flate plate area is:

CD .5 rho Vel^2 Sref
f = ----------------------
.5 rho Vel^2

So you can see that the flate plate of area Sref has an equivalent flat
plate area 1.17 times its true area (in 3-D flow)!


-Dave Lednicer
Analytical Methods, Inc.

All of the above Reports picked drag data and components from NACA wind tunnel testing results. A compilation of a range of both airframe low speed drag coefficient and individual component Cd for P-51, 38, 39, 40, F4U, F6F, etc can be found NACA Wartime Report L5A30.. The Wind tunnel airspeed was a uniform 100mph, then each RN was calculated as f(mean aero chord).. All of these were production airframes, not scale models.

NACA Report L5A30 can be found here:
https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19930092668.pdf

By the way when saving a copy locally, don't do as I did a few years back and save it under its original name.
You will never find it again! Name it something appropriate.

- Ivan.

 

[drgondog]

This business of different CDs even for the "same" aeroplane is exactly why I was asking for the equivalent flat plate area.
It is a much simpler concept and beats trying to figure out whether everyone is using the same reference area.
Here is a good description from a name that seems to come up a lot:


8/26/94

Analytical Methods {NWNet}


The concept of "equivalent flate plate area" comes from noting
that while:

Drag Where: Drag = resistance force
CD = ------------------ rho = density
.5 rho Vel^2 Sref Vel = Velocity
Sref = Ref Area

is nondimensional, the Sref is awkward, as how the reference area is
chosen can differ. So if we instead define:

Drag
f = -------------
.5 rho Vel^2

the Sref no longer appears, but f has the units of length squared. If
English units are used, you get the equivalent flate plate area (f) in
units of square feet. Note, this does not mean that a flate plate of
the same area as f would have the same drag - a flate plate has a CD of
roughly 1.17 (in 3-D flow, according to Hoerner and 1.98 in 2-D flow),
not 1 as implied in the equation! We can see this from:

Drag = CD .5 rho Vel^2 Sref

where Sref is the frontal area of the plate. Hence, the equivalent
flate plate area is:

CD .5 rho Vel^2 Sref
f = ----------------------
.5 rho Vel^2

So you can see that the flate plate of area Sref has an equivalent flat
plate area 1.17 times its true area (in 3-D flow)!


-Dave Lednicer
Analytical Methods, Inc.




NACA Report L5A30 can be found here:
https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19930092668.pdf

By the way when saving a copy locally, don't do as I did a few years back and save it under its original name.
You will never find it again! Name it something appropriate.

- Ivan.

Ivan - Flat plate area is a useful tool to compare Total Drag of one airframe vs another for the same airspeed and altitude. The issue is that multiple assumptions must be made to derive Total Thrust of the system as well as some precision on the 'assembled parasite drag components' and some assumptions must me made regarding Induced Drag (namely Oswald efficiency) for a degree of precision there.

Chapter 14 of Fluid Dynamic Drag by Hoerner supplies a nice 'Kentucky windage' model to drive to approximate Propeller thrust as f(HPxefficiciencyxK/V), then adds a general approximation for Exhaust Thrust (which varies as function of Boost/mass flow rate of combusted products through known annular area of multiple stacks, and know external pressure at the calculated altitude).

He then slides to T= Drag of airframe (including ALL components of Parasite Drag including zero lift, but also form drag as f(angle of attack, external stores, etc) multiplied by a compressibility factor), Plus Induced Drag

Yielding T= THP + Tet = Drag (parasite and pressure) of airframe + Induced Drag=Total Drag= D. CD follows for Flat Plate and q=1/2 x rho x V^^2 and S=Sref of Wing Area.

Leading to Flat Plate Drag = D/q = Cd/S

Dean also used this method to try to bundle an array of fighter Drag characteristics and further extracted Parasite Drag as one lump value for 250mph at 10,000 feet by subtracting Induced Drag from Thrust to get to lump sum "All other Drag".

To make them somewhat useful, he had to assume no form/pressure drag, no cooling drag and god know what for exhaust thrust for all the fighter variants.

So, where does that leave us relative to accurate Zero Lift Drag that we believe in for each separate case? Swamp Yankee is correct regarding looking to NACA (or equivalent) drag values extracted by careful methodology? Unfortunately those values are hard to find for the universe of non US aircraft via a consistent and repeatable methodology -------------> leading to Hoerner for Kentucky windage.

 

[Ivan1GFP]

Hello Drgondog,

Thanks for the reference to Hoerner. I found a PDF of the book a few years ago but don't use it much because I get lost pretty quick when trying to read it. Chapter 14 looks not to be so bad and even uses a Messerschmitt 109G as an example! I actually do have a few questions about what you wrote but I suppose I should read the chapter first.

Back to the case of the FW 190A, Corsair, and Hellcat:
With a comparison between Corsair and Hellcat, I still believe that Equivalent Flat Plate Area is a valid method (even if assumptions must be made) because of the great similarities in size, performance, airfoils, weights and propulsion. I believe that total drag is more important than aerodynamic shape in this case because that is what actually affects performance.
When the FW 190A is added, I really have no idea where any significant amount of data can be found other than for performance and the numbers there do not tend to be very consistent (maybe because I am relying on mostly English sources).

Thanks for giving some background and more proper methods for doing this type of calculation.

- Ivan.

 

[Vincenzo]

Time ago posted from JoeB
"There was only one, May 8 1944 between 800 Sdn FAA Hellcat I's (ie. F6F-3's) off HMS Emperor v. Bf109's of 8 and 10./JG5, off Norway. The Hellcats were initially surprised but 2 Hellcats and 3 Bf109's (a G6 and 2 G2's) were lost per each side's loss accounts."

"All USN F6F victories in Europe were against German bomber and transport types in the invasion of Southern France, flying from CVE's Tulagi and Kasaan Bay. They spotted German fighters at long range once, but the enemy declined to give combat. Those were F6F-5's."

 

[Ivan1GFP]

Thanks for the reference to Hoerner. I found a PDF of the book a few years ago but don't use it much because I get lost pretty quick when trying to read it. Chapter 14 looks not to be so bad and even uses a Messerschmitt 109G as an example! I actually do have a few questions about what you wrote but I suppose I should read the chapter first.
.

Hello Drgondog,
I just skimmed through the process that Hoerner used to calculate the drag of the "Me 109G" and the process and estimates / formulas he uses are actually fairly simple (mostly).
I also found a couple interesting things:
He seems to be using equivalent flat plate area in quite a few places.
His information on the Me 109G is not terribly accurate. It is probably closer to a F than a G.
He compared his numbers to a French Wind Tunnel test of a Me 109 in Paris in 1941 but the 109G did not come out until 1942 and did not appear with the Beule until the G-5 and G-6 series in 1943.

Could the differences in numbers be because it really was a Me 109E that was tested by the French?
It will take me a while to digest just this one chapter.

- Ivan.

 

[drgondog]

Hi Ivan - as noted, Flat Plate Drag is an approximation based on loose assumptions that are valid ranging from 'purty near' to 'some of the time'

Your comments that deriving and comparing flat plate drag calc (per Hoerner approach) between F6F and F4U has some validity, and to a degree include the FW 190 but it will be less 'close'. The reason is that the Wing Airfoil and Oswald factor are close and the wing for a reasonably clean airframe is a dominant component of Parasite and, Form Drag due to lift. That said, although the root section of the FW 190/ F6F, F4U are all NACA 230xx and 'close' between each airframe in Max T/C, the wing area of the FW 190 is much less than either the F4F and F4U. This is reason the F6F and F4U low speed Parasite Drag CD values are approximately 15-20% higher than the FW 190.

In most discussions about Drag, "the Wing is the Thing'"..

Hoerner is more rigorous in the follow on paragraphs where he breaks down individual airframe components for a Drag build up of individual dtag components including friction, of wing, fuselage, exhaust stacks, protuberances, gaps, empennage, cockpit enclosure, mast, etc - and refers to previous chapters.

Your comments about the data points and reference to earlier wind tunnel are valid, but perhaps recognize that except for struts for Horizontal stab, very few items in 109 series IMPROVED until G-10/K over 109E/F

 

[Ivan1GFP]

Hello Drgondog,

While I agree with you that the more rigorous approach is the "correct" one, sometimes the additional data that can be generated doesn't add to the comparison even if it helps us better understand how the aeroplane behaves in something other than straight and level flight.
In straight and level flight such as for a maximum speed run, you are mostly interested in thrust and drag.
Lift is whatever it takes to maintain level flight and induced drag is just drag.
Induced drag may vary by quite a lot with speed and altitude, but with these three aeroplanes, they all reach their maximum speeds at nearly the same altitude and even the speeds are not that different.
Does it really matter if one has a much more induced drag while the other has more parasite drag?
That would make a difference in a climb or in transient maneuvers but not so much in a maximum speed run.

The point in asking for equivalent flat plate area was to get away from the coefficient of drag which I believe is somewhat misleading.
Using Hoerner's example with the Me 109G with a 172 square foot wing as the reference area gives a certain CD.
What happens if we were to do nothing more than increase the wing area by 15% or so?
We would probably lose a few MPH, but imagine the improved CD we would be getting!
The point is that in this hypothetical case, the total drag would most likely increase and that the equivalent flat plate area would be larger and this would tell us more than just a misleading lower CD value.

Regarding the Me 109 E to F to G to K changes, I will have to disagree with you.
The change from the E to F was actually fairly substantial:
Radiators, Propeller, the location of the Stabilizer, the Fin cross section, obviously the spinner and cowling shape, slats, wing armament, Wing planform, supercharger intake, retractable tail wheel, and probably a few more items I can't remember at the moment.
On the same power, (Early F models used the same engine as the late E (E-7?)) the F was faster.
On the other hand, after the F, the aerodynamics did not change much except to get gradually worse as more bumps and bulges appeared and radiator scoops grew larger. Of course power also increased which is why performance continued to improve.

- Ivan.

 

[tomo pauk]

I'll continue the Fw 190 discussion here, from another thread:

Hello Tomo Pauk,
Are you certain that the F-8 used external air intakes? There is a beautiful FW 190F-8 in the Udvar Hazy Smithsonian which appears to have normal intakes. Why would extra armour, cowl guns, and extra internal fuel tanks (behind cockpit?) be mutually exclusive?

One thing that hasn't been mentioned thus far is that the F and G series were originally just factory Umrustbausatz special equipment versions of standard A series fighters.
The first F series fighters were just redesignations of ground attack versions of the A-4 series and continued through to factory modifications of the A-9 series
The first G series I believe began as a renaming of ground attack A-5 series.
The FW 190A-5/U8 that was in the USN test would be designated a FW 190G-2 later in the war.
The powerplants of these Jabo were calibrated differently and may have had different equipment but so far I have not been able to find anything that does a direct comparison between a fighter and a ground attack engine. Perhaps the A-5 in the test lacked a pressurized ignition system as I have seen described in other forums. That would explain the engine cut during service ceiling test.

- Ivan.

Both F-2 and F-8 were specified with external air intakes. That was an easy expedient, tested also (but without filters) on fighters. The cowling with external and internal intakes could be switched between the two in field depots. 190Fs were from get-go outfitted with improved armor ( 'item' 1) on the doc, external intake (item 2)) and ETC 501 as standard (item 3)):



Please note that for tropical use, the distance of the oil cooler cowl vs. engine cowl was to be increased from 10 to 20 mm, with associated loss in speed by 15-20 km/h.

But yes, the same production line could be shared between the fighter and F-B versions of the Fw 190, te initial F-B versions were just fighters that carried bombs when required.

The engines were the same, I don't know what will mean that they were calibrated differently? However, the Jabos were 1st to use over-boosting in service, 1st by use excess fuel injection as ant-detonant ('C3 einspritz'), and quickly after that the 'simple' overboosting. IIRC there was no difference in pressurizing of the ignition system, perhaps it was just the case of sub-standard engine, a thing that popped out sometimes with other people's engines.

 

[Shortround6]

Drgondog and Swampyankee, were there any areas of design that weren't known to designers of WW2 aircraft that led them up blind alleys looking back?

I think the rapid rise of drag in the Mach .6-.8 region caught a lot of people by surprise. There were a lot of requirements issued for aircraft flying 450-500mph at 15-20,000ft that never came close to being fulfilled. Not to mention the prop tips going supersonic.

 

[swampyankee]

Speaking of cockpit design, I have read that the FW-190 instruments and controls were very well thought out and because of this most pilots could instinctually find their way around the cockpit with little to no trouble. I never compared it's cockpit to that of the Bf-109 series, but was curious about what details of it's layout were considered an improvement over the Messerschmitt fighter? Also what aspects, if any, weren't as good?

Well, the F4U cockpit seemed to be designed for people over 6 ft tall, which is not particularly sensible when median height of white, American males was 5 ft 8 in, and fewer than 20% were over 6 ft. The P-38 cockpit was so cold, pilots had to deal with frostbite. In others, instruments and controls were placed haphazardly, and without consideration of the difficulty of retraining pilots to a new type or even to their use with heavy clothing and a life vest.

 

[DarrenW]

do you know of any books with accounts of the encounters?

Time ago posted from JoeB
"There was only one, May 8 1944 between 800 Sdn FAA Hellcat I's (ie. F6F-3's) off HMS Emperor v. Bf109's of 8 and 10./JG5, off Norway. The Hellcats were initially surprised but 2 Hellcats and 3 Bf109's (a G6 and 2 G2's) were lost per each side's loss accounts."

"All USN F6F victories in Europe were against German bomber and transport types in the invasion of Southern France, flying from CVE's Tulagi and Kasaan Bay. They spotted German fighters at long range once, but the enemy declined to give combat. Those were F6F-5's."

Just to add to this information, during the war there were only a total of five German aircraft claimed as destroyed by FAA pilots while flying the Hellcat. Besides the three fighters mentioned in Vincenzo's post, there were also two He 115 float planes claimed as brought down less than a week later, on May 14th. One of these was awarded to 800 Sdn, while the other was shared by this same unit with 804 Sqn, also flying Hellcats. These were the final aerial victories scored in Atlantic waters by British Hellcat pilots, and by the fall of 1944 the aircraft were sent to the Far East for operations against the Japanese, were they eventually earned the lion's share of their 52 total aerial victories.

During the invasion of southern France, the US Navy credited F6F pilots with a total of eight German aircraft destroyed. On August 19th, VF-74 scored victories over a Ju-88 and Do-217, while in the evening VOF-1 was credited with the destruction of three He-111s. Two days later, on August 21st, VOF-1 brought down an additional three Ju-52s.

The source of the above information is Hellcat Aces of World War II by Barrett Tillman.

In total than, there were thirteen German aircraft awarded to pilots while flying the Hellcat. Of the three fighters claimed, two were Bf-109s and one a FW-190.The Germans in turn claimed three Hellcats shot down during the May 8th encounter, while the British only acknowledge the loss of two. There is confusion as to how the Hellcats were lost however, as one account states the two Hellcats collided in mid-air during the fight, while another says that one was hit by AAA fire. No American flown Hellcats were lost to Luftwaffe aircraft during the operations in southern France.

 

[DarrenW]

Well, the F4U cockpit seemed to be designed for people over 6 ft tall, which is not particularly sensible when median height of white, American males was 5 ft 8 in, and fewer than 20% were over 6 ft. The P-38 cockpit was so cold, pilots had to deal with frostbite. In others, instruments and controls were placed haphazardly, and without consideration of the difficulty of retraining pilots to a new type or even to their use with heavy clothing and a life vest.

Interesting. In contrast I have heard that the P-47 had a rather cramped cockpit. Any truth to this?

 

[DarrenW]

I'll continue the Fw 190 discussion here, from another thread:

Thanks, I was going to mention that as I inadvertently picked the wrong thread to ask my questions about the FW-190....

 

[drgondog]

Interesting. In contrast I have heard that the P-47 had a rather cramped cockpit. Any truth to this?

I have been in P-47, MiG-15, Bf 109G, FW 190, P-51, P-40 and P-39. P-47 roomiest, MiG-15/Bf 109 cramped but MiG 15 had great visibility. I fit well in the Mig 15 at Eglin in 1954 at age 9. Ditto Bf 109 at age 11.

 

[DarrenW]

I have been in P-47, MiG-15, Bf 109G, FW 190, P-51, P-40 and P-39. P-47 roomiest, MiG-15/Bf 109 cramped but MiG 15 had great visibility. I fit well in the Mig 15 at Eglin in 1954 at age 9. Ditto Bf 109 at age 11.

Thanks for confirming this first hand drgondog, I had a feeling that what I heard about the P-47 cockpit was untrue.