20/20 Hindsight - different armament?

[Jank]

I think the loss of velocity that Delcyros listed at 300 meters is frankly too little. A start of 880ms and still travelling at at 847ms at 300 meters? The .50had a great BC but there would be more velocity loss. That much I know.

The differences in striking velocity at 300 meters between the data that Delcyros used and Tony used is as follows:

For the .50, Delcyros used 847ms and Tony used 740ms

For the 20mm, Delcyros used 741ms and Tony used 655ms.

Tony said,"The .50 Cal was measured against homogenous plate of 350 BHN. At 300m this gave the following figures:
Striking velocity (from aircraft gun: 36" barrel) = 740 m/s.
Penetration at 0 degrees = 22 mm

The 20mm was fired against plates of various hardnesses. Against 350 BHN plate, the following figures are given:
Striking velocity at 300m = 655 m/s
Penetration at 0 degrees = 39 mm"

I take it that we have closure now. The 20mm has a penetration superiority on the order of 77%. At angles of deflection, that superiority would increase.

It is simply not possible that, pursuant to Delcyros' assertion, increasing the velocity by 107ms for the .50 and 86ms for the 20mm (a 20ms difference in increase between the two) would then result in the .50 now outpenetrating the 20mm.

Besides, "Lunatic" would disagree with you. That settles it.

[Jank]

Squeelig said, "Wouldn't the effect of yawing have a greater effect on the fifty's ability to penetrate than the twenty milimeter?"

Good point. Yes. The .50 has greater sectional density than the 20mm and given its comparatively low weight, relies on that for penetration so yawing would indeed be more detrimental to the penetration of the .50 than the 20mm.

You will also notice that the .50 is affected by angles of deflection much more than the 20mm.

I think that ideally, comparing perpendicular hits (90 degrees or 0 degrees depending on how you figure it) is unrealistic. All comparisons should be at 45 degrees. Comparing penetration at 90/0degrees gives a false boost to the .50 cal's armor penetration abilities under circumstances that would be a rarity in the real world of combat. Even Delcyros' data wouldn't support an outcome where the .50 is superior to the 20mm at 45 degrees.

For an illustration of the effect on penetration due to angle of deflection between the .50 and 20mm, look below. Looking at the Mk I vs. .50 tests, we see 100% parity at 90/0 degrees, then at 20 degrees, the .50 is down to 70% and the 20mm is down to 95%, then at 40 degrees, the .50 is down to 40% and the 20mm is down to 75% of it's ability to penetrate. The 20mm retains penetrative ability at angled strikes far better than the .50 does. (This is Mk I and not the Mk II)

And why just examine armor penetration? Look at the penetration through sand. The .50 at 200 yards penetrates 15" while the 20mm Mk I penetrates 24" (a 60% increase over the 50's ability to penetrate). (This is Mk I and not the Mk II)

[Jank]

Delcyros,

As I said above, I think the loss of velocity that Delcyros listed at 300 meters is frankly too little. A start of 880ms and still travelling at at 847ms at 300 meters? The .50had a great BC but there would be more velocity loss. That much I know.

From "Lunatic's" website at Lunatic's WWII Aircraft Gun Ballistics Page

Note that the .50 BMG is listed as having a velocity of 736ms at 300 meters and not the 847ms figure you are using.

As you well know, Lunatic knows what he's talking about which, of course, necessarily means that you don't know what you're talking about.

Lastly, it just so happens that Tony's data matches Lunatic's. This makes perfect sense because Lunatic bases his information on Tony's data. Tony lists 740ms at 300 meters.

Again, Tony's data states:

The .50 Cal was measured against homogenous plate of 350 BHN. At 300m this gave the following figures:
Striking velocity (from aircraft gun: 36" barrel) = 740 m/s.
Penetration at 0 degrees = 22 mm

The 20mm was fired against plates of various hardnesses. Against 350 BHN plate, the following figures are given:
Striking velocity at 300m = 655 m/s
Penetration at 0 degrees = 39 mm

Even if the data above was realized with Mk. II as opposed to Mk. V, (difference at the muzzle of less than 35ms) that wouldn't diminish the penetration at 300m by more than 17 mm (a reduction of 17mm is 2/3 of an inch and would represent a 43% decrease) which would be required in order to put it at below the .50's ability to penetrate 22mm.

[delcyros]

Originally Posted by Jank

Delcyros,

As I said above, I think the loss of velocity that Delcyros listed at 300 meters is frankly too little. A start of 880ms and still travelling at at 847ms at 300 meters? The .50had a great BC but there would be more velocity loss. That much I know.
Note that the .50 BMG is listed as having a velocity of 736ms at 300 meters and not the 847ms figure you are using.

The 20mm was fired against plates of various hardnesses. Against 350 BHN plate, the following figures are given:
Striking velocity at 300m = 655 m/s
Penetration at 0 degrees = 39 mm[/b]

Even if the data above was realized with Mk. II as opposed to Mk. V, (difference at the muzzle of less than 35ms) that wouldn't diminish the penetration at 300m by more than 17 mm (a reduction of 17mm is 2/3 of an inch and would represent a 43% decrease) which would be required in order to be below the .50's ability to penetrate 22mm.

Dear Jank, dear Tony,

The impact velocities are correct, as I stated above:
"*) calculated with a vector speed of 400 km/h relative to target (111 m/s)"
so I used Lunatics values for 300 m (for example: 20mm MK II: 673 + 111 = 784 m/s; .50cal: 736 + 111 = 847 m/s) to get a true impact velocity (including plane vectoring speed, the gun is not fired with speed zero). Altough all will be a bit lower due to the higher initial rate of deceleration. Note that the 20mm round even benefits from this (!) and puts the .50 cal to a disadvantage, factoring the higher BC (= less deceleration) of the 0.50 cal rounds. I would be glad to have a computing system for deceleration rates of small calibre rounds, but sadly I do only have them for medium to large calibre rounds (50 mm and up), so these are approximations but very close ones, relied on Lunatics page (whom´s calculation I trust unless You present better results).

Tony, can You give me more datas or a copy about this specific test. All I have studied so far implies that 39 mm @ 300 m / 0 deg with a plate hardeness of 350 BRH is totally contradicting even british datas and sources (not to speak of other US and german tests). A testplate of 39 mm is quite difficult to process with a specification calling for 350 BRH, hence specifications do call for 240 BRH for these thickness and US tests. I know there are plates of this hardeness used by germans but for the US I know only two experimental plates, which match those requirements, and none of them actually was tested against 20mm rounds. I am not aware of other special plates used for .50 cal tests on prooving grounds (it may have happened), so this source would interest me in particular. Indeed such a figure would surprise not only me but also friend and expert Dr. Wegener. In case this plate indeed has 39 mm and 350 BRH and has no major internal flaws (bubbles, pipings (long tube-like bubbles), cracks, laminations, and any other irregularities, this would exceed even the special treated plates with about 300 Brinell (32 Rockwell "C") hardness for plates under 51mm thickness!
This penetration value is also totally off-line if I use the official Ministery of Defense british penetration formula for homogenious armour, announced 1942:
B0 = 500 - (160)(0.434294482)[LOGe(D / 1.5648 )]

B1 = 54000 / (B0 - BHN)

FOB1 = 11800 / (65o - OB)

FOB2 = (43.4)(PQF)(T/D)[SQRT(BHN)]/COS[(1.5)(OB)]

FOB3 = 929 - B1 - FOB1 + FOB2

BRITNBL = FOB3 / SQRT[W/D3]

But it fits very well with a plate hardeness of 250 BRH (.8% difference).
This would indeed imply an actual energy density above 65000 ft-lbs/in^3, which is hard to believe (impact force necessary to cause such energy densities would even be higher than a 37 mm NS37 impact from this distance!).

Thanks in advance,

[Jank]

"The impact velocities are correct, as I stated above:
"*) calculated with a vector speed of 400 km/h relative to target (111 m/s)"
so I used Lunatics values for 300 m (for example: 20mm MK II: 673 + 111 = 784 m/s; .50cal: 736 + 111 = 847 m/s) to get a true impact velocity (including plane vectoring speed, the gun is not fired with speed zero)."

You are correct and I am an ***.

I am still not convinced that the .50 cal would out penetrate a Mk V 20mm at 300 meters though. Perhaps Tony can give his thoughts.

And what about at a 20 degree angle? What about at a 40 degree angle? As I indicated above, pependicular strikes would be rare. Tony indicated that after striking the aluminum skin, the bullet yaws thereby not striking with the maximum ability to bring its sectional density to bear. And lastly, the penetrative ability of the .50 would be affected more by both yaw and angled strikes than the 20mm.

Additionally, when the USN determined that ther 20mm was 2.5 times as destructive at longer ranges, how exactly did they define destructive? Even the Mk II doesn't have 2.5 times the armor penetration ability so I believe that were referring to effects beyond mere armor piercing ability. Whatever effects they include, they would favor the 20mm over the .50.

And finally, the 6 x Mk V setup can still be swapped out for a 6 x Mk II setup at a total increase in weight of 150lbs for all six guns. As you may recall, I originally advanced the idea of using Mk V instead of Mk II because you were concerned with weight degrading the flight performance of the Thunderbolt. I always maintained that it wouldn't make a hill of beans difference, especially in the ground attack role.

[delcyros]

I actually liked the idea with the MK V due to their lower recoil , weight & higher rof. The long barreled MK II with recoil-damping gunmount might offset the larger brutto recoil as I learned from Jank, but it still is a very bulky weapon.

Regarding higher impact obliquities, the 0.50 penetration curves drops with a steady rate, much steeper than the 20 mm. This may be partly caused by the thickness / diameter relation and mostly because of the blunter nose of 20 mm projectiles. I can perform calculations, if You desire (all factors beeing the same as above=300m, 220 BRH):
20mm MK II: 41,6mm(20deg), 37,5mm(40deg), 33,4 mm (45 deg)
Note that it takes only a loss of 4 mm between 20 and 40 deg and further 4mm between 40 and 45 deg. impact obliquity for the 20mm AP MK I fired from 20mm HS MK II.
0.50 cal AP: 39,1mm(20deg), 33,0mm(40deg), 28,9mm (45deg)
Note that the .50 cal projectile loses 6 mm penetration between 20 and 40 deg. impact obliquity and further 6 mm for 40-45 deg. impact obliquity.
Loss is relative to armour quality. The 220 BRH are too low, as we recognized, so I estimate that the .50 cal round would have more detremental losses with higher grade armour material than this softer "comparison steel". Also, the less pointed nose of the 20mm round will perform better at high impact obliquities (program credits only for a standart middle shaped nose), which is not reflected by the calculation above (the significance rises with impact obliquity).
Yawing effects are not that important for ground attacks, since few vehicles (unlike planes!) had layered armour layout (or inner mounted plates). They are not quantifyable according to recent knowledge. They do happen and this is expressed with an "estimated percentage of Yaw" rather than with "a loss of penetration" -figure, altough Tonys infos covering this aspect are highly interesting. Too much factors play a role: impact velocity, plate 1 thickness, plate 2 (..3...4) thickness, angles, spacial distribution (the larger the distance beween the more probable are yawing effects), spin of projectile and weight (=stability border to induce yaw) and much others. Yawing effects can be considerable but more "stable" projectiles (not necessarely heavier ones) need more inducing force for them.
There is no doubt that the 20mm round carries more HE charge, is heavier, has more source for fragmentation and therefore will cause more damage against air targets and "soft" ground targets.

[Jank]

What would the armor penetration be for a .50 vs a 20mm fired from a Mk V at 300 meters at 20 degrees deflection? At 40 degrees deflection?

[delcyros]

...38,0mm at 20 deg. with a similar 4mm dropping rate to the MKII with 33,9mm at 40 deg and 29,8mm at 45 deg. (true values should be slightly better due to less pointed nose with higher impact obliquity, note). 0.50cal as above, so there is parity in penetration at around 30 deg (probably more close to 25 deg.) and beyond this the 20mm MK V, despite the lower muzzle velocity, has a better penetration according to M79 APCLC.

[Tony Williams]

A few comments:

The USN penetration graphs to which I referred I have only in the form of poor-quality photocopies. I do not know how the results were obtained, but I can hardly imagine that they had a huge variety of plates of different thicknesses and fired at each one at every range and striking angle - I expect that they carried out a few actual tests at critical points and interpolated the rest (that would be the only practical way to do it IMO).

I should emphasis that the 20mm AP tested was the US M75, which was different from any British rounds. It was solid steel, weighed 165 g and was fired at 775 m/s. The British developed several Marks of 20mm AP, but the RAF appeared to make relatively little use of them; the most common seems to have been the Mk II, which had a hollow cavity and weighed 141 g (I don't have a muzzle velocity figure, but it was presumably around 820-850 m/s from the Mk II).

The penetration figures most often quoted for the British 20mm refer to the SAPI, which was the hollow HE shell filled with incendiary material and fitted with a hardened steel nose-cap instead of a fuze. This had the advantage of matching the trajectory of the HEI, and the two types of round were usually mixed in equal numbers in the ammo belt. Obviously, the penetration of the SAPI would not have been as good as a solid AP. I don't have directly comparable figures, but the penetration of the SAPI at around 20-25mm was similar to that of the .50 AP - i.e., good enough to deal with aircraft armour. Incidentally, I have never seen any penetration claims for the .50 AP which exceed 25mm armour plate, at any distance or striking angle.

I also have penetration graphs for the German 20mm AP (which also had small cavities, often filled with incendiary material), these show that the low-velocity MG-FF (117 g at 585 m/s) would penetrate 18mm at 300m (against 150 kg/mm2 armour), reducing to 11mm after penetrating a 3mm dural skin at 70 degrees. The MG 151/20, which fired the same proj at just over 700 m/s (v around 530 m/s at 300m), penetrated 23mm (or 12mm after passing through the dural). Obviously, with similar projectiles the Hispano would penetrate significantly better than this, as the cartridge was far more powerful.

The USN charts for the fall-off with striking angle show that the .50 AP (which gives 22mm/300m/0 degrees) sees a fall-off to 17mm at 20º, 13mm at 30º, 11mm at 40º, 9mm at 45º, 7mm at 50º and 4.5mm at 60º.

The USN chart for the 20mm AP is less helpful, because although it shows penetration at 0º for 300, 350 and 400 BHN plate (46mm, 39mm and 32mm respectively), it only gives one figure for other angles. However, it is reasonable to assume that these figures probably relate to 350 BHN plate as that is the middle one of the three, and the same used for the .50. Anyway, they show a sharp initial fall-off to 21mm/20º, before levelling off at 19mm/30º, 18mm/40º, 17mm/45º and 10mm/60º.

Tony Williams: Military gun and ammunition website and discussion forum

[Jank]

Tony said, "I should emphasis that the 20mm AP tested was the US M75, which was different from any British rounds. It was solid steel, weighed 165 g and was fired at 775 m/s."

Geez, that's like 6 ounces!

[delcyros]

Originally Posted by Tony Williams

I expect that they carried out a few actual tests at critical points and interpolated the rest (that would be the only practical way to do it IMO).
(...)I have never seen any penetration claims for the .50 AP which exceed 25mm armour plate, at any distance or striking angle.

It´s not surprisive. With a plate QF:1.0 and BRH 220 there will be no target plates of such thin thicknesses to test vs. .50cal impacts. A BRH of 300-400 is more plausible, ergo a plate QF of between 1.2 an 1.4. For such properties and under assumption that the gun does not move (unlike an airplane), I get fairly comparable results for the .50cal: 21,9-25,8mm at 300m/ 0 deg.
USN penetration specifications are usually very reliable sources. I understand that plates have been penetrated under controlled circumstances and most records do also give a remaining striking vel. for the projectile (I suspect but am not sure, that this was recalculated via distance travelled behind the plate either in air or through sand). So actually there was no 39 mm plate with 350 BRH but (probably) a usual 1" plate of 350 BRH. The penetration at 300m distance was recorded and the remaining striking velocity was calculated back to reach 39mm with the M75 20mm AP for Navy ballistic limit of full penetration. This was common practice also for larger projectiles, which are more familar to me.
With the update in the datas (heavier projectile, striking vel), I still have problems to recall this result via M79 APCLC:
M75 20mm AP(165 g plain steel), impact vel.: 655m/s, plate QF: 1.1 (equals 280 BRH), obliquity: 0 deg; max. penetration: 1.41" (35,8mm)
plate QF: 1.3 (equals 350 BRH): max penetration: 1.19" (30,2 mm), so either the projectile performs much better than estimated or the program has severe failures or the plate does not fit to the specification calling for 350 BRH. There is still a significant gap in energy density, something to keep in mind.

[schwarzpanzer]

I think that for a fighterbomber or heavy fighter, HE and AP performance is vital.

I have often thought that the P47 should have had 4 Hispano 20mm's instead of 8 .50's. Instead of another pair of 20mm's, I'd have extra ammo or fuel.

Would it be possible to have a 40mm, or even 57mm, under each wing of a P47?

I know America couldn't make the Hispano work though, so it's pie in the sky.

[Tony Williams]

...to the debate on gun recoil any more, but I came across some new and relevant information while looking for something else (which is usually the way )

Tim Mason's book "The Secret Years - Flight Testing at Boscome Down 1939-45" is concerned with the official tests of new or modified aircraft to ensure that they were ready for service. Boscome Down was the home of the "Aeroplane and Armament Experimental Establishment" (or A&AEE), which tested every aspect of the planes' performance, handling, and armament. The book summarises the results of the tests, picking out points of interest.

One of the planes tested was Hurricane Mk 1 L1750, which in late 1939 was fitted with two Hispano cannon in underwing gun pods (obviously a less aerodynamic solution than the later in-wing installation). The book contains this comment: "Aircraft handling remained unchanged, although firing one gun induced slight yaw; performance suffered only marginally."

Several Hurricane Mk IIC were tested, as they tried to 'debug' the cannon installation. There were only two complaints concerning the cannon: that the ammo capacity of the original drum-fed version was inadequate, and that the guns froze up at altitude - which proved difficult to solve as they couldn't get enough hot air to them.

Some Spitfires with four 20mm cannon were tested. The installation was regarded as satisfactory, the main comment being that the installation added 200 lb (91 kg) and shifted the CG rearwards, required a 7.5 lb inertia weight. Four cannon also proved more difficult to keep warm than two plus MGs (I have read before that this was a key reason for not fitting Spitfires with four cannon - the guns froze at altitude). This problem took until mid-1944 to solve - although it then returned with the Griffon-engined versions as the engine ran much cooler than the Merlin so didn't provide as much gun heating. It is stated that the Hispano+Browning combo was preferred because the Browning was much less prone to freezing up.

The Spitfire XXI seems to have resolved these problems, the only comment about the four 20mm cannon being that they gave "accurate shooting from the steady platform".

To sum up, there is not a single reference to the recoil of the cannon causing any issues with either the Hurricane or the Spitfire, except for the comment that firing just one gun in the Hurricane caused "slight yaw". The main reason for the RAF avoiding the four-cannon layout was the gun freezing problem at altitude, which was tolerable in the Hurricane as it soon became used for ground attack anyway. It is probably significant that the only reference I can recall to the use of four-cannon Spitfires was in the Mediterranean theatre, where ambient temperatures were higher.

Tony Williams: Military gun and ammunition website and discussion forum

[Jank]

Thanks Tony. These reports on four 20mm installations in the Hurricane and Spitfire indicate that a six 20mm installation in a P-47 would not have presented recoil related issues of any significance.

Thanls for the follow up.

[merlin]

To my mind the question is round the wrong way. The US managed perfectly well with the 0.5" machine gun till the korean war. Perhaps we should be wondering if the RAF got it right with opting for eight 0.303"? Wouldn't four or six 0.5" have been better? Come the Battle of Britain, such puny rifle calibre guns were only saved by the Dixon (usually mistakenly referred to as De Wilde) ammunition.