Home Forums WWII How effective was AA?

Viewing 25 posts - 1 through 25 (of 25 total)
  • Author
    Posts
  • #92542

    I’m currently re-basing my WW2 figures and, finally, putting my previously un-based vehicles on bases. Big job. The figures are 1/72 & 20mm and are late Europe US, British & German. I use Blitz Krieg Commander rules.

    As usual with such a task, it gives one the opportunity to look at what you have, what needs improving, adding and culling.

    I’ve decided I want a bespoke AA unit base for each field force. Each force already has access to a single ground attack/ fighter-bomber & the rules allow the possibility of calling in an air strike each turn.

    AA fire is fairly abstract & the purpose of the proposed AA models are to give substance to this aspect (I also have field gun bases to represent the off-table artillery).

    This leads me to the question posed in the thread title. I’m quite happy about the mechanism for calculating a successful attack but think more thought on the AA point may be useful.

    I would not mind tweaking the existing abstract rule in light of any comments & further research made. What is the chance of hitting a low flying attacking aircraft? Is it more likely that AA fire would just disrupt the attack?

     

     

    donald

    #92555
    deephorse
    Participant

    First of all, what are the chances of AA hitting/driving off attacking aircraft in your rules?

    Less enthusiasm, please. This is Britain.

    #92571
    John D Salt
    Participant

    What is the chance of hitting a low flying attacking aircraft? Is it more likely that AA fire would just disrupt the attack?

    I’m going to say “low”, and “yes”.

    People who dislike tables of semi-digested numbers should look away now.

    WO 291/1108 "Performance of the Bofors 40mm gun during operational engagements 1942–1944"
    
    This report gives figures for the "rounds per bird" found to be needed to down aircraft 
    targets over the following periods:
    
    Period			    Rounds per bird
    27 Mar 1942 to 30 Sep 1942	 880
    01 Oct 1942 to 31 Mar 1943	 760
    01 Apr 1943 to 6 Jun 1943	 220
    07 Jun 1943 to 31 Mar 1944	1320
    
    These summary figures include all fire-control methods.  Using the Stiffkey stick 
    (sights correctional Mk IV) was found slightly superior to predictor control.  
    Other methods are "materially less accurate".
    For single-engine fighters below 1000 feet in daylight, over the period 01 Apr 1943 
    to 6 Jun 1943, 100 rounds per bird was achieved.  This was thought to represent the 
    best possible performance for the Bofors.
    There was insufficient data to determine effectiveness at night.  However, the 
    marked increase in rounds per bird in the last period was thought due to the facts 
    that most targets engaged at night at heights over 100 feet, and that the inherently 
    wasteful tactic of barrage fire was introduced, accounting for 50% of the ammunition 
    expended.
    
    WO 291/8 "Analysis of AA gunfire 1 Mar to 5 Apr 1941"
    
    The following figures are given for rounds per casualty, with the caution that 
    "The number of casualties involved is still too small for reliable conclusions to be drawn".
    
    March
    Consolidated figures			4.5"	3.7"	3"	4.5"	3.7"	3"
    Independent fire, target seen		 149	-	 195	 211	 818	 606
    Independent fire, target unseen		4001   14693	-	4792	4924	3472
    Indirect fire, target unseen		5145	5886	4953	3991	5497	4419
    
    WO 291/94 "The 40mm Bofors and the 20mm Oerlikon, a Comparison of Hitting Value and Lethal Effect"
     
    The results given here, from calculations, are to be regarded as comparative rather than absolute. 
    The target is taken to be a Junkers 88, as this is the aircraft about which the most is known of 
    its vulnerable areas.
    
    Categories of damage used in this report are:
    A       Immediate break-off
    B       Delayed crash
    C       Crash within 2 minutes of strike
    
    Types of fire control compared are:
    SS      Sights correctional (Stiffkey stick)
    PC      No. 3 Predictor control
    NES     Eye shooting
    
    Percentage chances of a successful engagement are given as:
    
    .                       Prevention of           Failure to return
    .                       attack (Cat A)          (Cats B + C)	
    
    Weapon	FC              Crossing  Direct        Crossing  Direct 
    .                       target    approacher    target    approacher	
    
    20mm    Single NES        0.6%     3.5%           2.5%     8.5%   				
    .       Triple Gyro       3.4%    25.0%          13.5%    51.0%
    40mm    NES               7.5%     8.6%          10.5%    11.4%
    .       SS               13.8%    16.3%          19.2%    21.3%
    .       PC               14.5%    17.0%          20.3%    22.1%
    
    "...it is unlikely that these low flying targets would be seen at ranges over 1,500x , 
    thus no use can be made of the greater effective range of the 40mm."
    
    An analysis of engagements of  low-flying coastal raiders in ADGB gives the following 
    average numbers of rounds per engagement:
    
    40mm    PC              7 rounds
    .       SS(NES)         8 rounds
    20mm    NES            50 rounds
    
    Time of engagement, in seconds:
    
    .                First seen to first rd    First rd to last rd   Overall time
    40mm PC                  7.5                       6                13.5
    40mm NES                 6                         7                13
    20mm NES                 4                         9                13
    
    
    WO 291/671 "Effect of Enemy LAA on British Fighter-Bombers"
    
    This is a theoretical paper, and as usual in such cases the figures should be taken 
    as comparative rather than absolute.  It seems to indicate that the hazards calculated 
    here are if anything too great, but represent a "reasonable operational figure".
    Four types of attack are considered; a 60º dive, releasing RP at a slant range of 
    1700 yards; dives at 30º and 15º, releasing RP at 1000 yards; and low-level bombing 
    from 90 feet.
    
    20mm guns are assumed to have an effective range of 1000 yards, 40mm guns 2000 yards, 
    and 37mm guns to be "inappreciably different" from 40mm guns.  It is assumed that 4 guns 
    are deployed in a ring of radius 400 yards around the target.  Each 40mm barrel is assumed 
    to fire 1 round per second, each 20mm barrel 4 rounds per second.  On the assumptions used, 
    either type of weapon will have about the same lethal effect per unit time the target is 
    in range.  It is noted that a multi-barrelled weapon with N barrels is less than N times 
    as effective as a single barrel.
    
    The probabilities of a single attacking aircraft being lost are calculated as:
    
    Attack       Firing up to crossing   Firing whole time       Time aircraft in range 
    .            point only              target in range         (seconds)
    
    .            40mm        20mm         40mm       20mm         40mm      20mm
    60º dive     1.29%        0%          1.43%       0%          13         2
    30º dive     1.82%       1.08%        2.48%      1.08%        20         8
    15º dive     2.54%       1.87%        3.42%      1.9%         24        11
    90ft level  10.3%        9.4%        11.1%       9.7%         28        14
    
    The report concludes that, against 40mm LAA, the 90-ft low-level attack is eight times, 
    and the 15º dive attack twice, as dangerous as the 60º dive.  Against 20mm LAA, the 
    60º dive stays practically out of range; the 30º and 15º attacks are about half as 
    vulnerable as against 40mm; and the low-level attack is almost as dangerous.
    It is pointed out that terrain masking is ignored, so the danger of the low-level attack 
    is overstated unless the aircraft can be seen all the way from maximum range.
    
    

    While two of the studies are model-based, and so not to be regarded as absolute numbers, and the historical survey of HAA points out it has too small a sample to be trustworthy, I think something useful might be done for LAA by taking the rounds-per-bird figure from WO 291/1108 combined with the engagement times from WO 291/94, modified by the relative vulnerability of dfferent attack profiles indicated in WO 291/671.

    All the best,

    John.

    #92572
    Etranger
    Participant

    I would not mind tweaking the existing abstract rule in light of any comments & further research made. What is the chance of hitting a low flying attacking aircraft? Is it more likely that AA fire would just disrupt the attack?

    That’s going to have the same effect in practical terms, at least using BKC. An aborted attack is an unsuccessful one, regardless of material damage to the aircraft. Actually shooting down the ‘plane makes no difference to the possibility of another air attack in subsequent turns.

    Most of my references are packed away, & I’m on the wrong computer for most of my links! Thus I’ve only made brief comments below.

    Well organised AA defences could be lethal to ground attack aircraft & many (most) aircraft losses in 2TAF and 9TAC 0in NW Europe were due to FlaK damage of various sorts. Here’s a contemporary US viewpoint http://www.lonesentry.com/articles/defga/index.html

    By 1944 in NW Europe there was a lot of AA around, although the western allies found that there generally wasn’t much call for it, due to the virtual disappearance of the Luftwaffe. This article has 73%+ of allied air losses in Normandy being due to ground fire of various forms. http://www.dtic.mil/dtic/tr/fulltext/u2/a432943.pdf

    http://scholars.wlu.ca/cgi/viewcontent.cgi?article=1130&context=cmh has an interesting view on the role of air power in Normandy, albeit not directly related to your question.

    • This reply was modified 3 years ago by Etranger.
    • This reply was modified 3 years ago by Etranger.
    #92576

    It’s probably needless to mention I don’t really know much about WW2 – ask me about the Bronze Age & I’m a bear!

    John’s interminable table made for really interesting reading & don’t, all in all, contradict the abstract results that BKC use. Airpower, on my BKC tabletop, is not guaranteed & often not successful but the threat can keep players “honest” in terms of not bunching your troops & seeking cover (not unlike the effects of off-table artillery, really)

    I’ll thank ET also for his links also. I’ve bookmarked the first two & will read later.

    The final article was fascinating. It does point out that indirectly, Allied airpower ensures that there aren’t a dozen Tigers, fully fuelled and ammo’d, on my tabletop (also good because I only own 2 Tiger models).

    #92610
    deephorse
    Participant

    Are you going to set out what the abstract AA rule is in BKC so that we can judge how accurate/inaccurate it is?

    Less enthusiasm, please. This is Britain.

    #92621
    John D Salt
    Participant

    Having, inevitably, whipped up a weeny spreadsheet to do a few simple sums on the stuff I have just posted, I thought I should comment on a handy simplification that would let you get a quick idea of the likely order of kill probability for light AA.

    Glommimg up a time-weighted aggregate of the historical “rounds per bird” for Bofors shooting, and discounting half the rounds expended in the last period as having been wasted on barrage fire, gives an average of 701 rounds per bird over the entire period. Considering also that the average time spent firing in eye-shooting engagements was 7 seconds, and the rate of fire of the Bofors is taken as being one round a second, this gives an expected p(kill) of 1% per gun per engagement, as near as makes no difference.

    If you want to characterise the different methods of air attack by exposure time, the times given in WO 291/671 less the reaction time for 40mm weapons would justify numbers very close to 7 seconds for a high diving attack, 14 seconds for a shallow dive, or 21 seconds for a laydown attack, corresponding to nice round p(kills) of 1%, 2% and 3% per gun respectively. A similar exercise using the exposure and reaction times for the 20mm, taking rounds per bird as 3500 and shots per second as 5 to give the same expectation of a kill per second’s firing, as the study suggests, makes steep dives invulnerable, and if we have shallow dives and laydowns exposed for 3.5 and 7 seconds respectively gives p(kills) of 0.5% and 1% per gun respectively. With a few assumptions or additional data, this method can easily be extended to different rates of fire, or to smaller calibres (quicker reaction, shorter range, more rounds needed for effect, more rounds fired per unit time), although I think by 1945 it was obvious that 20mm weapons were pretty marginal for air defence, and 0.5s or rifle-calibre MGs of little use other than to cheer up the blokes firing them.

    The real trick, with AA guns dotted about the place and a number of possible approach routes for attacking aircraft, which also have a free choice of attacking speed and altitude, is to estimate how much engagement time each gun or battery will get at a given aircraft. Consideration of these factors could also show the importance of the surprise element, as referred to in the ground attack pilot’s motto “one pass, haul ass”. If we showed the benefit to the air defender of a second pass by ignoring the initial reaction time, it being assumed that everyone is now fully on the ball and the guns traversing fast enough to follow the aircraft, the p(kills) for a 40mm would go up by a percentage point per gun, and 20mm by half a percentage point, which might not sound a lot, but is in some cases a doubling of their effectiveness.

    As always, the figures available are never quite the figures we want, and all this says nothing about the (probably much more important) effect of putting the ground attack pilot off his stroke. I rather fancy the idea of trying to develop a procedure whereby the attacker decides how much time he is prepared to spend on the attack (assuming that slower, lower and gentler attacks are most accurate), knowing that this must be traded against the air defender’s fire effectiveness. Obviously, a target that has no air defence can be attacked with maximum accuracy, and you can go round again for as many passes as you like in perfect safety.

    All the best,

    John.

    #92627

    Thank you, John, for your knowledge & its application to my question but the wargaming tail is starting to wag the dog.

     

    donald

    #92638
    Etranger
    Participant

    As always, the figures available are never quite the figures we want, and all this says nothing about the (probably much more important) effect of putting the ground attack pilot off his stroke. I rather fancy the idea of trying to develop a procedure whereby the attacker decides how much time he is prepared to spend on the attack (assuming that slower, lower and gentler attacks are most accurate), knowing that this must be traded against the air defender’s fire effectiveness. Obviously, a target that has no air defence can be attacked with maximum accuracy, and you can go round again for as many passes as you like in perfect safety.

    All the best,

    John.

    Not quite perfect safety, other aircraft, electrical and telephone cables & the ground still took their toll. Look how close some of the pilots get to each other & to obstacles https://www.youtube.com/watch?v=XsqtItEnGUs After watching that it isn’t hard to work out why the Germans were so worried by the jabos.

    • This reply was modified 3 years ago by Etranger.
    #92640
    MartinR
    Participant

    Tbh for wargaming purposes I just borrow the mechanisms used in Spearhead. Light Flak has virtually no chance of downing an attacking “plane” (which actually represents a flight of four) but any air attacks within 1000m of an unsuppressed Flak unit is subject to a penalty.

    Nice and simple.

    "Mistakes in the initial deployment cannot be rectified" - Helmuth von Moltke

    #92645

    Are you going to set out what the abstract AA rule is in BKC so that we can judge how accurate/inaccurate it is?

    My apologies for taking so long to get back to you.

    I will pretend you know nothing about BKC.

    Air support needs a designated FAC with a force. Each turn, this “unit” rolls a modified dice to see if air support will come and attack the designated target.

    There is a process using directional dice to see where the aircraft will unload bombs or strafe (ie be careful at choosing targets anywhere near your own units).

    Command units (considered to include AA units) & AA units within range can fire back.

    The number of “attacks” these each have are pre-determined in a Stats table. Any “hits” may be saved depending on the aircraft (ie saving rolls).

    If sufficient hits” are made (ie equal or exceed the aircraft’s “hits” total) it is downed. If not, the defender rolls a dice for each “hit” & any 6s will abort the attack. NB each aircraft type may have differing stats in the Table.

    If the attack is successful, a similar process determines the damage wrought.

    There are a few nuances to this but I hope I’ve made the process clear.

     

    donald

     

    #92656
    Etranger
    Participant

    Although it sounds simple, the mechanism does allow for amusing (to the opponent) events, such as bombing your own HQ units.

    #92659
    deephorse
    Participant

    I will pretend you know nothing about BKC.

    And you would be correct.

    I’m currently reading “Air Power at the Battlefront : Allied close air support in Europe 1943-45” by Ian Gooderson.  Obviously it has a lot to say about the question you asked, and I will try to distill a few helpful points from it.

    Less enthusiasm, please. This is Britain.

    #92664

    I’m currently reading “Air Power at the Battlefront : Allied close air support in Europe 1943-45” by Ian Gooderson. Obviously it has a lot to say about the question you asked, and I will try to distill a few helpful points from it.

     

    I look forward to them.

    In case you’re interested, here’s an overview of the rules:

    Actually, it’s for BKC 11 & I use the first version but not a lot of difference.

     

    donald

    #92666
    willz
    Participant

    AA weapons are very effective against infantry and light trucks/jeeps/cars.

     

    #92667
    Whirlwind
    Participant

    As always, the figures available are never quite the figures we want, and all this says nothing about the (probably much more important) effect of putting the ground attack pilot off his stroke.

    Just thinking out loud, but I wonder if this could be done by comparing air attacks on ships as an easier dataset to compile?  The difference in effectiveness between attacking unarmed ships / surprised ships and defended ships might at least give an order of magnitude for the effect.

    https://hereticalgaming.blogspot.co.uk/

    #92711
    John D Salt
    Participant

    Just thinking out loud, but I wonder if this could be done by comparing air attacks on ships as an easier dataset to compile? The difference in effectiveness between attacking unarmed ships / surprised ships and defended ships might at least give an order of magnitude for the effect.

    Unfortunately I have a frightful dearth of naval AA effectiveness data, too. The only thing that springs to mind is a USN document I have tucked away somewhere that gives rounds per bird for different naval guns in the context of anti-Kamikaze shooting. The availability of VT fuzes and radar control makes them a rather different proposition from the sort of thing one would need for general terrestrial WW2 ack-ackiness.

    If anyone knows any good sources of such data, point me at ’em… especially anything on the reduction in aircraft weapons delivery accuracy, rather than the more obvious but less useful rounds per bird.

    As a final remark on the Bofors data I posted and lightly analysed, I observe that if we take the (historically correct on average) figures of 700 rounds per bird and 7 seconds’ shooting per engagement, and if we treat guns in troops or batteries of 4, and aircraft in flights of 4, then rolling 1d6 per battery/flight engagement and downing one aircraft on a roll of 6 is highly defensible on grounds of numerical accuracy.

    From Ochoin’s description, it sounds as if the rules he is using allow quite nicely for the reduction in attacker effectivness and the risk of fratricide, which is always good to see.

    All the best,

    John.

    #92716
    John D Salt
    Participant

    Oooh, and another number I’ve discovered, from the OR classic Morse & Kimball —

    WW2 British merchantmen were equippped with AA guns, and an assessment made of their effectiveness. The assessment showed that only 4% of attacking aircraft were shot down. This was considered insufficient to justify the effort, and it was proposed that the provision of AA armament be discontinued. Fortunately, someone then pointed out that this was the wrong measure of effectiveness. When the success of the air attacks was considered, it became apparent that 10% of defended ships were sunk when attacked, as against 25% of undefended ships.

    As 4% of attackers shot down is the same number I have proposed for a defending Bofors battery, it does not seem unreasonable to suggest that it would also result in a 60% reduction in attack effectiveness.

    Mind, 4% seems to me like a fairly brutal chop rate, as, if it were maintained, it would mean that only 36% of crews would complete a 25-mission tour.

    All the best,

    John.

    #92724
    Etranger
    Participant

    …..Mind, 4% seems to me like a fairly brutal chop rate, as, if it were maintained, it would mean that only 36% of crews would complete a 25-mission tour. All the best, John.

    That would be strike missions against targets though, not all missions would have been so perilous. That said, the various RAF Strike Wings seemed to have some very heavy losses from time to time. There’s not much on line.

    #92725

    May I interrupt the discussion to say that John is a delight?

    Personally, I’m not much of a statistical fan but I’ve both learnt from the ones Mr Salt has provided & enjoyed his obvious joy in the facts & figures.

    If Michael was into giving out awards, I’d suggest “Credit to the Forum”.

     

    donald

    #92751
    John D Salt
    Participant

    …..Mind, 4% seems to me like a fairly brutal chop rate, as, if it were maintained, it would mean that only 36% of crews would complete a 25-mission tour. All the best, John.

    That would be strike missions against targets though, not all missions would have been so perilous. That said, the various RAF Strike Wings seemed to have some very heavy losses from time to time. There’s not much on line.

    Indeed. What I think are the final dregs of numerical snippetry lurking in my collection related to this topic come from WO 232/24, “German weapon development”, and record the experience reported by the Luftwaffe’s II SG 1 in 1942 on the Eastern front. From this it is easy to calculate the chop rate they were facing:

    Aircraft type  Sorties  Losses  Loss rate    Chance of completing a 25-mission tour
    Hs 129	        3138      20      0.64%        85%
    Hs 123	        1532       5      0.33%        92%
    Bf 109E	        1838      16      0.87%        80%
    

    …not that the Luftwaffe did 25-mission tours, but it’s a handy standard of comparison. Quite why the Hs 123 should be so relatively safe is a bit of a mystery to me. Presumably the 109E suffers from having an in-line engine raher than a radial.

    May I interrupt the discussion to say that John is a delight?

    Personally, I’m not much of a statistical fan but I’ve both learnt from the ones Mr Salt has provided & enjoyed his obvious joy in the facts & figures.

    If Michael was into giving out awards, I’d suggest “Credit to the Forum”.

    COUGH *blush*

    All the best,

    John.

    #92763
    Don Glewwe
    Participant

    COUGH *blush* All the best, John.

     

    You forgot *turns head*  

    https://brawlfactory.net/

    #92770
    Whirlwind
    Participant

    Mind, 4% seems to me like a fairly brutal chop rate, as, if it were maintained, it would mean that only 36% of crews would complete a 25-mission tour.

    I haven’t got it with me, but IIRC John Terraine, quoting a late 1942 study (16 Nov 1942, AIR20/2859?), said that the actual survival rate for torpedo bomber crews on a 25-mission tour was 17.5%.

    https://hereticalgaming.blogspot.co.uk/

    #92775
    John D Salt
    Participant

    I haven’t got it with me, but IIRC John Terraine, quoting a late 1942 study (16 Nov 1942, AIR20/2859?), said that the actual survival rate for torpedo bomber crews on a 25-mission tour was 17.5%.

    That would imply a chop rate of about 6.7% per mission, which is steely stuff. Naval avaitors probably see things just as bad, but there’s a fair chance of getting the blokes back if you write off the aircraft in a deck landing (“They say in the Air Force a landing’s okay/If the pilot and aircrew can still walk away/But in the Fleet Air Arm the prospects are grim/If the landing’s piss-poor and the pilot can’t swim”).

    I vaguely seem to recall that tour lengths were set with the idea that crews should stand a more or less even chance of getting to the end of them. There’s a cheery thought.

    Compound interest always seems surprising. I remember when I was at Hunting Engineering and someone decided to try taking combat attrition into account in our fleet operational availability models (help the Treasury save money by not buying spares for dead helos). There were two surprises; first, that availability rates took an initial jump upwards, and, second, that mission achievement sank to near zero in under a week. The first surprise was because all our models included three spare airframes in the squadron. As aircraft were never destroyed in our usual simulations, they never did anything, and people forgot about them. Then we started killing aircraft, so knackered machines carrying the maximum number of non-critical defects were splashed and replaced by pristine defect-free aircraft, at least for the first three to be shot down. This meant less fixing, and more availability. The second problem was easily explained by the fact that the loss rate per mission had been set by a lunatic, or, at least, someone so pessimistic as to make Corporal Fraser look like a Prozac-fuelled Polyanna. Based on numbers from Westland System Assessment Ltd (WSAL, pronounced “weasel”) the attrition rate had been set at something like 20% per mission — every one of them worse than Schweinfurt. I’m sure the numbers must have had some caveat about “for comparative purposes only” on them, but when I pointed out that then-current doctrine said all offensive air (and presumably aviation) operations would be suspended if attrition rates reached 3% per mission, they decided a re-think was needed. I’m sure we came up with some sensible numbers in the end. Oh, and it turns out that availability rates *do* take a sudden jump upwards in real life when live operations start; but that’s because the war emergency spares packs become available, instead of having to make do with peacetime Treasury parsimony.

    All the best,

    John.

    #92783
    Etranger
    Participant

    Indeed. What I think are the final dregs of numerical snippetry lurking in my collection related to this topic come from WO 232/24, “German weapon development”, and record the experience reported by the Luftwaffe’s II SG 1 in 1942 on the Eastern front. From this it is easy to calculate the chop rate they were facing:

    Aircraft type  Sorties  Losses  Loss rate    Chance of completing a 25-mission tour
    Hs 129	        3138      20      0.64%        85%
    Hs 123	        1532       5      0.33%        92%
    Bf 109E	        1838      16      0.87%        80%
    

    …not that the Luftwaffe did 25-mission tours, but it’s a handy standard of comparison. Quite why the Hs 123 should be so relatively safe is a bit of a mystery to me. Presumably the 109E suffers from having an in-line engine raher than a radial.

    John.

    The Hs123 may have been used more at night, & the faster Me109 used for higher risk missions. The Me’s inline engine did make it more vulnerable & The Hs 123 was particularly noted for its rugged reliability by the pilots. There also weren’t that many HS123 in service by 1942 ~ around 6 with II/SchlG1 in mid 1942 according to this site http://www.historyofwar.org/articles/weapons_henschel_hs_123.html but numbers seemed to go up and down quite regularly.

    I once looked after a Luftwaffe ground attack pilot. This was 20 years ago so he’s probably long gone. Anyway Erwin flew Me 109s & Stukas on the Eastern Front, until getting shot down and seriously injured on a mission. It was German FlaK that got him, not Russian.

    • This reply was modified 3 years ago by Etranger.
Viewing 25 posts - 1 through 25 (of 25 total)
  • You must be logged in to reply to this topic.