Bf-109 vs Spitfire vs Fw-190 vs P-51

[Shortround6]

The merlin had a swept volume of 27 Litres while the Me 109 had a swept volume of 34 to 36 litres It wasnt until the 37 litre griffon variants came in to sevice that the engines of both were of approximately the same size.

The swept volume of the engine, while interesting to the engine designer, is of no value to the airframe designer.
The airframe designer is much more interested in power per unit of weight of the engine and in the physical size of the power plant.
The DB engines were designed to run at lower rpm than the Merlin and actually weighed about the same (not including superchargers) as the engine blocks, crankcase could be made lighter. Height and width of the Merlin and the DB engines were very close also, much closer than the difference in swept volume might lead one to believe.

 

[tail end charlie]

The swept volume of the engine, while interesting to the engine designer, is of no value to the airframe designer.
The airframe designer is much more interested in power per unit of weight of the engine and in the physical size of the power plant.
The DB engines were designed to run at lower rpm than the Merlin and actually weighed about the same (not including superchargers) as the engine blocks, crankcase could be made lighter. Height and width of the Merlin and the DB engines were very close also, much closer than the difference in swept volume might lead one to believe.

I was just mentioning it as the swept volume is an important factor in the power output of an engine and the power output is an important factor in the speed of an airplane. There are many other things of course.

 

[Kurfürst]

Hi Bill and Shortround,

I think that somethings may be being confused here.

OVERALL drag between a Mustang and a 109 may very well be similar.
We would have to compare specific models at the same altitude and know the power output for both engines at that specific altitude.

However, even if the 109 is equal or even slightly faster than the Mustang using the same power it is still a "dragger" airframe because it is a smaller airframe. The Mustang with it's bigger wing and larger fuselage is more streamlined for its size. THe larger size allows it to carry the extra fuel and heavier weight of armament. I don't want this to veer off into tangent but I think we can all agree that that the Mustangs .50 cal MGs and ammo weighed more than the 109 normal internal armament.
So you have a small but higher drag airframe (the 109) vs a larger but lower drag airframe (Mustang) giving the same TOTAL drag.

I understand the difference between absolute drag values (ie. equivalent flat plate area) and relative drag (usually expressed as a coefficient related to some other area value, ie. total surface area or more often, related to wing area only). My problems with the statement are really the following:

Generally I don't believe expressing aerodynamic 'cleaness' with a wing area related coefficient gives you realistic comparison values, since in this case we are comparing the total drag of the aircraft (fuselage, radiators, antenna etc) to a single element of the drag, that is responsible only for the fraction of the total drag, ie. Hoerner assumes that the 109 wing assembly was responsible for 37.5% of the total drag. Such comparison, by its nature always gives better coefficients for aircraft with larger wing areas.

For a simplified example, lets assume that the total drag of the 109 is 100, and its wing has an area of 16. The wing thus is responsible for 37 units of drag (see above); the coefficient would be 100 drag / 13 wing area, ie. a drag coefficient of 6.25.
Now, double the wing area of the 109, to 32 units, and leave everything else alone. The wing's drag component is now 74 (ie. 2x37), the remainder is still the same as with the original wing, 63 (ie. 100-37), a total of 137 units of total drag, but the coefficient now is 4.28, almost 50% 'better' than the original 109... even though there were absolutely no improvements to skin friction, the fuselage or anything else... its exactly the same aircraft, just with oversized wings. I understand that it may be somewhat more complicated in practice, but you get the point I guess: It would be nonsense to state that the same 109 with bigger wings is cleaner than a 109 with smaller wings, and its only seems so because our basis of relation (wing area) takes account only 1/3 of the total picture.

Wing area related coefficients of drag are NOT a measure of quality of aerodynamic cleaness (I believe stated many times by NACA papers as well), but just that, a mere coefficient for engineers. I believe from factor is probably better expressed in front area related form factor coefficients, not to mention, given an aircrafts multitude of flying conditions (near stall, fast flight, climb, landing, flaps/slats deployed etc.) there's a myriad of drag coefficients for each condition..

From Hoerner: At 380mph, 22,000 feet at 6700 pounds gross weight, 1200 Hp:

Building up all the drag factors, applying Thrust calculated with 140 pounds of exhaust thrust at 380mph and propeller efficiency of .85 and getting total drag yielded:

The Drag Area of the 109G is approximately 6.2 sq ft with Cd of .036 at that speed and altitude.

The total wetted area of the 109G, according to Hoerner, is 590 square feet. Also according to the example calculations using the 109G Hoerner's calculated CDwet = .0105. (Divide Drag Area by Wetted Area). Compare against the following values derived from Flight and Wind Tunnel tests:

The tables in Lednicer's WWII Fighter Aerodynamics present the following at ~360kts, 15,000 feet.
1. Spit IX Drag Area = 5.4 sq Ft, Wetted Area = 831.2 sq ft, CDwet = .0065
2. P-51B Drag Area = 4.61 sq ft Wetted Area = 874..0 sq ft, CDwet = .0053
3. P-51D Drag Area = 4.65 sq ft Wetted Area = 882.2 sq ft, CDwet = .0053
4. Fw 190A-8 Drag Area=5.22 sq ft, Wetted Area = 735.0 sq ft, CDwet = .0071
4. Fw 190D-9 Drag Area=4.71 sq ft, Wetted Area = 761.6 sq ft, CDwet = .0063

I believe Hoerner's study was discussed a number of times. I don't believe for a minute that Hoerner's method would be unprofessional in many details, in fact I am pretty sure his study is as perfect as it can be. However I believe its purpose is misunderstood - its certainly not a serious study specifically on the 109s aerodynamics, rather than a generic example provided how to calculate such things, and what factors are present in drag, and how it can be improved. Also, the following has to be taken into account:

Any calculation is just as good as the basic numbers it uses. Hoerner's study however uses ad hoc number s for its calculation, the - rather optimisitic imho - 85% propeller efficency is an assumption, with no actual data on the actual VDM propellers effiency curves - and we know too well that some of these were optimised for climb speeds, some for high speed, high altitude flight (this is obvious looking at climb curves for normal altitude G-14 and high altitude G-14/AS with equivalent power levels, but different props).
Exhaust thrust is yet another assumption with a some 20% margin of error in what actual thrust is available for the aircraft, and this foundamentally effects the accuracy of the calculations. In both propeller and exhaust thrust, I believe Hoerner's guess are a bit on the optimistic side with regards of the actual thrust available.

Secondly, he assumes a top speed of 380 mph (612 km/h) for his '109G', which is waaay lower than what was actually measured on the 109G at these power levels (1200 HP at altitude relates to 1.3ata Kampfleistung, ie. 30 min ratings). The actual measured/nominal speeds were, at this rating, atltidue:

G-1 through G-4: 660 km/h / 410 mph
G-5/G-6 : 630 km/h / 391 mph

Needless to say, underestimating top speed by as much 50 km/h or 30 mph (!!!), or 10% will yield drastically worser drag values, since power requirements increase on the cube for higher top speed. Ie. Hoerner's fantasy like 109G does 380 mph at 1200 HP; the real one did 410 mph on the same 1200 HP; the get Hoerner's 109G from 380 mph to 410 mph, ie. the speed actually achieved by the real one, it would take an 1500 HP output engine at that altitude, ie. in practice, the most powerful engine in the 109K.

So I believe Hoerner's study, while theoretically perfect, in practice are flawed by his wrong base data (the weight figures BTW are also curious, 6700 lbs would be a correct weight for an early 109G-2, but the G-6 was some 200 lbs heavier, yet still faster than Hoerner's example, unless it had gunpods... which would add another 500 lbs.... so, what is this Franken109 of Hoerner's really is?)

They are also at odds with the drag coefficients and polars stated by the wartime Messerschmitt AG papers, which all state a drag coefficient of 0.023 for the Bf 109F/early G, which I posted on this board quite a few times btw..

The Drag Area of the 109G is approximately 6.2 sq ft with Cd of .036 at that speed and altitude.

Moreover if we take Hoerner's calculated drag area for the '109' seriously, its also at odds with common sense and reasoning: simply an aircraft with higher drag and lower power output can't be as fast or faster as another, ie.

Compare against the following values derived from Flight and Wind Tunnel tests:

The tables in Lednicer's WWII Fighter Aerodynamics present the following at ~360kts, 15,000 feet.
1. Spit IX Drag Area = 5.4 sq Ft, Wetted Area = 831.2 sq ft, CDwet = .0065

OK, so we have Spit IX with drag Area = 5.4 sq Ft, 1595 HP at critical altitude, doing 404 mph at critical altitude (21 k feet).

Please explain, how its possible, that a 109G, which according to Hoerner, has a drag area 6.2 sq ft (but 4 sq. ft according to Messerschmitt AG...), manage to do about 405 mph at 21 feet, with an 1200 HP powerplant... while being draggier, and having about 2-300 HP less. I would say its either a miracle, or Hoerners drag coeff is flat out wrong, as is the base date he used.

Same comparison BTW at Sea levels, ie. compare Spit IX F/LF speeds to 109G-2/G-6 speeds and the corresponding powers.. the 109G was considerably faster (by ca 30 km/h if we look at nominal specs, ie. ) near the ground than the Merlin 61 powered Marks, despit having slightly less power available. And it was about as fast as the Mark IX LF, which had considerably more (by some 300 HP!) available...

 

[Kurfürst]

continued...

4. Fw 190A-8 Drag Area=5.22 sq ft, Wetted Area = 735.0 sq ft, CDwet = .0071
4. Fw 190D-9 Drag Area=4.71 sq ft, Wetted Area = 761.6 sq ft, CDwet = .0063

Here's another great opportunity to make a comparison, because the FW 190A-8 and the G-14 (which was aerodynamically identical to G-6, and considerably worse than the F-x/G-2) had exactly the same engine output at SL: 1800 PS.

Corresponding speed specifications were:

A-8, at 1.42ata/1800 PS, without ETC 501 rack (otherwise standard fitting I believe, -12 km/h at SL): 545 km/h
G-14, at 1.7ata/1800 PS, clean: 568 km/h.

Again explain how it is possible... supposedly larger drag, equal power, yet faster.. with Hoerner's 6.2 sq ft with Cd of .036 for the 109, it is impossible when the Fw 190A-8 has a calculated drag Area=5.22 sq ft. However if we take the actual Messerschmitt drag figures in account (ie. coeff of 0.023 = 4.002 sq. ft. for F-4/G-2, and taking into account that the G-6/G-14 is quite a bit worser, since the thing was some 20-30 km/h slower due to negative aerodynamic changes.. so lets assume say 4.5 sq. ft.), it begins to make sense.

Fine discussion BTW!

 

[drgondog]

continued...



Here's another great opportunity to make a comparison, because the FW 190A-8 and the G-14 (which was aerodynamically identical to G-6, and considerably worse than the F-x/G-2) had exactly the same engine output at SL: 1800 PS.

Corresponding speed specifications were:

A-8, at 1.42ata/1800 PS, without ETC 501 rack (otherwise standard fitting I believe, -12 km/h at SL): 545 km/h
G-14, at 1.7ata/1800 PS, clean: 568 km/h.

Again explain how it is possible... supposedly larger drag, equal power, yet faster.. with Hoerner's 6.2 sq ft with Cd of .036 for the 109, it is impossible when the Fw 190A-8 has a calculated drag Area=5.22 sq ft. However if we take the actual Messerschmitt drag figures in account (ie. coeff of 0.023 = 4.002 sq. ft. for F-4/G-2, and taking into account that the G-6/G-14 is quite a bit worser, since the thing was some 20-30 km/h slower due to negative aerodynamic changes.. so lets assume say 4.5 sq. ft.), it begins to make sense.

Fine discussion BTW!

Kurfurst - the questions you posed have too much to respond to in a single pass.

First three comments.

Lednicer's values are all from flight tests and full scale wind tunnel tests. Lednicer in turn performed a VSAERO analysis with sophisticated routines to model pressure gradient build ups and associated form drag due to boundary layer separation - which in turn combined to yield values very close to the wind tunnel data. I will see if Paul placed the Lednicer Fighter Comparison on his Performance thread and upload it if not.

Second, If you have the Hoerner analysis, look to the discussion of Aerodynamic Efficiency at the end of Chapter XIV. It is an excellent breakdown of the various components of drag on the Me 109G. The key subtitles include Wing , Fuselage, Radiator, Tail and Induced. For each of these he breaks out the total drag as a function of Smooth Skin Friction, Forced Turbulence, Surface Imperfections and "Additional Parts (of drag components)". For the latter (Additional parts) the Radiator is signigicant - nealy 40% of the total wing drag.

He went further to delineate the main components into types of drag in aggragate - namely Skin Friction Drag (33%), surface roughness (15%), exposed parts, especially of the engine (33%), interference drag (6%), influence of compressibility (at M=.55 ---> 6%), INDUCED DRAG (7%).

Third - you are absolutely correct that Hoerner's analysis is not the last workd on Me 109 aerodynamics - but it is an excellent tutorial regarding the effects of various forces applied to an airframe which are loosely sorted into those due to lift and those due to interference/turbulence and those due to 'form'. As I stated before Hoerner's Fluiud Dynamic Drag was one of my texts in undergraduate school and was my primary source in different airframe design courses later.

The very last paragraph on efficiency summarizes the findings and states that more than half of the drag could be reduced based on much better manufacturing techniques (skin gaps, rivet heads, surface imperfections) and ruthless clean aerodynamic design (i.e Implied much better design of radiator, landing gear doors, etc) that the efficiency woul be raised from 40% to a much higher value - and corresponding speed for the same engine.

Maybe take this offline??

Regards,

Bill

 

[Timppa]

One interesting tidbit:

When a captured Spitfire VB (EN830 of 131 Squadron) was fitted with DB605A, it achieved 300mph (483km/h) at sea level, compared to 316mph (509 km/h) for the Bf109G.

One may argue that the front fuselage contour is not the same, but I believe that the difference was negligible.

 

[drgondog]

One interesting tidbit:

When a captured Spitfire VB (EN830 of 131 Squadron) was fitted with DB605A, it achieved 300mph (483km/h) at sea level, compared to 316mph (509 km/h) for the Bf109G.

One may argue that the front fuselage contour is not the same, but I believe that the difference was negligible.

Timppa - that is interesting. Any more details like 'did they measure performance of the Spit VB with the Merlin prior to the installation for the DB605A?, etc, etc to provide more insight to comparisons?

 

[RCAFson]

Timppa - that is interesting. Any more details like 'did they measure performance of the Spit VB with the Merlin prior to the installation for the DB605A?, etc, etc to provide more insight to comparisons?

By definition a Spitfire VB is a very early production aircraft (probably 1941) with a low performance, by later standards, engine while a 109G can refer to aircraft built up to late 1944. Using 18lb boost and a Merlin50. the VB could do 325mph at 0ft:
http://www.spitfireperformance.com/w3228speed.gif

while a standard VB with 9lb boost could do 280mph, so with 12lb-16lb or boost we would expect the VB to achieve 300-320 mph:
http://www.spitfireperformance.com/spitfire-V-raechart.jpg
and in the above example it achieved 302 and 317mph at 12 and 16lb boost.

The 109 was refitted with the Merlin and built under license with that engine in Spain:
Hispano Aviación HA-1112 - Wikipedia, the free encyclopedia

 

[Timppa]

Any more details like 'did they measure performance of the Spit VB with the Merlin prior to the installation for the DB605A?, etc, etc to provide more insight to comparisons?

The plane crash-landed in France during a fighter sweep on November 18, 1943. It may have flown before it was handed over to Daimler-Benz AG. The object of the engine modification was to test the cooling system of the Mk V and so the original radiators under the wing were retained, although all the armament were removed. It was flown against the 109G to produce comparison of performance at different heights.

(Source: FlyPast -Magazine)

 

[Shortround6]

I would note that the difference between the re-engined Spit and the 109 seems to be about 5%.
Granted there is no armament drag on the Spit but given it's larger wing and and wetted surface the results are not too surprising.
Given that the Spit had more room/volume inside the airframe for armament and fuel it doesn't seem that the 109 airframe is any more sophisticated than the Spitfire, just smaller.

I will grant that the Spitfire did not often use it's extra volume for fuel but the potential was there as shown by the leading edge tanks on the MK VIIIs and photo recon planes as well as the aft fuselage tanks. We can argue about the effectiveness of center line guns vs wing mounted guns but with many Spitfires carrying 2 20mm guns a 4 RCMGs vs the one 20mm and 2 RCMGs of many 109s the Spit had to big enough to carry twice the armament internally, or on the original models to carry eight guns instead of two.

I would note that an old book on basic aircraft design (very basic) gives a rule of thumb that a 25% reduction in wing area is good for a 3% increase in speed. Given the slightly more than 25% reduction in area from the Spitfires wing to the 109s wing that would seem to cover over half the difference assuming that everything else was the same on the two planes.
By the way, it also gives a rule of thumb that a 25% reduction in wing area (everything else staying the same, which it never does) will increase the minimum speed (stalling speed) by 15%. I am assuming that this is for the plain wing with no flaps or slats deployed.

 

[Kurfürst]

The "Messerspitt" speed curves, but I have no time for many comment as I head out for a big AS event (350+ participants 8). A few: engine cowliong was from Bf 110, take note that the take off weight of Spit V is rather light (removed armament iirc). The drag of the wing armament on Spit was considerable, around 8 mph. For comparison purposes (I dont think direct comparison was made) speed of "series production 109G" is shown. Given the date of paper (May 1944) and weight of 109G, its probably an earlier G-6 test, tad bit slower than nominal values (630kph), but is impossible to identify which 109G test flight/aircrafts' data was used. The original M45 powered Spit V speed however seem to match up with nominal speed values for the +9 lbs Spit.

Of interest is the decreasing advantage of 109G over altitude, but this may be also due to lower weight of the DB-Spit (less AoA required, dominant induced drag reduced at altitudes).. its odd that this doesn't show on the M45 powered Spit's curve.

 

[NZTyphoon]

An interesting test on a Spitfire V was run at Farnborough in 1943: (Spitfire Story Alfred Price)
A standard Mk VB was tested and developed a maximum speed of 357 mph (height, engine type not given)
Multi stack exhausts fitted = gain of 7 mph
Removal of carburettor intake ice-guard = gain of 8 mph
Fitting of faired rear view mirror = gain of 3 mph
Whip aerial in place of mast = gain of .5 mph
Cutting cartridge case and link ejector chutes level with wing = gain of 1 mph
Sealing, rubbing down painting and polishing wing leading edge = gain of 6 mph
Polishing remainder of airframe = gain of 3 mph

Top speed now = 385.5 mph

The greatest single increase - 8 mph - was made by removing the ice-guard to the carburettor air intake thus increasing the speed and mass of air entering the carby. Replacing the fish-tail exhausts with similar stacks to the Mk IX added 7 mph = a 15mph gain.

Filling and polishing the leading edge of the wing added 6 mph, and polishing the rest of the airframe 3 mph = 9 mph gain.

The cumulative effects of changes to the engine's ability to breath - ie: increased air pressure and mass-flow in, plus less backpressure on exhaust gases, was more important than induced drag (even on the critical wing leading edges) to the performance of the Spitfire V used in these tests.

According to Price a Spitfire VB W3228 with a Merlin 50M with a cropped supercharger and standard wings was tested with a top speed of 333.5 mph (537 Km/h) at 2,000 feet (609.6 m) , rate of climb 4,720 ft/min, engine generating 1,585 hp at +18 lbs. : loaded weight = 6,450 lbs (2,925 kg) cf DB601 Spitfire V 502 Km/h (312 mph) at same altitude. The Spitfire V fitted with a DB 605 weighed 2,730 Kg = 6,018 lbs.

 

[The Basket]

Has there been a study on the range of bhp of the same model of engine and the differences between a good one and a bad one....I can imagine in the primitive days of 1940s...scope for a engine to be slightly not up to spec.

Example is a motorcycle magazine saying they have tested Honda Fireblade engines and the highest difference was 9 bhp between standard engines.

This is with the latest production technology and it is Honda....

So I'm always wary when graphs are brought out.

 

[Njaco]

Price also gave comparison details in his book "Fw 190 At War" with tests between the Fw, Spit and I believe P-51. Too wordy but 'll try to scan the page for reading.

 

[Milosh]

An interesting test on a Spitfire V was run at Farnborough in 1943: (Spitfire Story Alfred Price)
The Spitfire V fitted with a DB 605 weighed 2,730 Kg = 6,018 lbs.

NZTyphoon, the graph says mit (with) DB605A, G=2930kg and mit (with) Merlin45, G=3030kg.

Something is amiss.

 

[drgondog]

NZTyphoon, the graph says mit (with) DB605A, G=2930kg and mit (with) Merlin45, G=3030kg.

Something is amiss.

Milosh - was the armament of the Spit removed?

 

[Shortround6]

Has there been a study on the range of bhp of the same model of engine and the differences between a good one and a bad one....I can imagine in the primitive days of 1940s...scope for a engine to be slightly not up to spec.

Example is a motorcycle magazine saying they have tested Honda Fireblade engines and the highest difference was 9 bhp between standard engines.

This is with the latest production technology and it is Honda....

So I'm always wary when graphs are brought out.

There is always some variation but each engine (at least in the west) was placed on a test stand and run for several hours before shipment from the factory. During this time the engine was connected to a load and power measurements were taken. Any engine not meeting specifications (plus or minus so many percent or a minimum HP ) would be rejected for refitting/repair.
That being said there were variations in service. Rings and valves could wear, linkage adjustments could move a bit, Condition of sparkplugs, etc.
Some times an engine that seemed to work fine at sea level was well down on power at altitude. Air is an insulator and in the thinner air at high altitudes some magnetos and wiring harnesses shorted out or cross fired.

 

[NZTyphoon]

NZTyphoon, the graph says mit (with) DB605A, G=2930kg and mit (with) Merlin45, G=3030kg.

Something is amiss.

No, read the chart carefully it says Spitfire V mit DB605 A (G=2,730kg) ie 2[European 7 with a bar] 30 kg ) = 6,018 lbs.

BTW the Me109 Serie weighed 3100 kg = 6,834 lbs

Price went on to observe that relatively small deteriorations in the airframe could reduce the performance of a Spitfire in the same way that the RAE experiment increased its performance: "oil leaked onto the outside of the airframe which then picked up dust or sand; dents or scratches on the aircraft (particularly to the leading edge of the wing); repaired battle damage etc..."

Taking a look at this 222 Sqn Spitfire VB I can see an airframe that has seen better days; paint chipping off the leading edge of the wing, a raised, square patch on the wing, mismatched paint, oil streaking on the wingroot..

Looking at the captured Spitfire VB with DB605 - I wonder how well the airframe was looked after at Rechlin?

 

[Milosh]

No, read the chart carefully it says Spitfire V mit DB605 A (G=2,730kg) ie 2[European 7 with a bar] 30 kg ) = 6,018 lbs.

BTW the Me109 Serie weighed 3100 kg = 6,834 lbs

That darn bar.

 

[Kurfürst]

Has there been a study on the range of bhp of the same model of engine and the differences between a good one and a bad one....I can imagine in the primitive days of 1940s...scope for a engine to be slightly not up to spec.

Good point! There were certainly a lot of variance with engines as well, although its rarely covered in tests, the used engines were seldom bench tested. One example that gives you some idea, where bench testing was done is found below - in this case the DB 601 used in the Emil prototype was 45 horsepower down in power, corresponding (postive) corrections were made for the flight measured performance in the report.

Kurfrst - Meprotokoll vom 26.4.38, Geschwindigkeit Bf 109 V15a

This is with the latest production technology and it is Honda....

Perhaps even a better point - Honda rules for reliability. Get one and you never want to change for another brand. 8)