British 4.5"/45 (11.4 cm) vs American 5"/38 (12.7 cm) Mark 12

[wmh829386]

RPK is a terrible metric.
1. Long range will always have inflated RPK by its very nature, which regardless of the efficiency of the system.

2. It depends on types of attack, how determined is the attack pushed, pilot tactics, and of course type of aircraft
(E.g. With Mk4 radar ranging, the difference in Mk37 engaging
A. sea skimming approach
B. 1942 style IJN TB approach
C. German/Italian pilot that turns away when burst is close
D. Fw 200 straying into range without realising
E. Approach with "corridor" laid
F. Kamikaze with various tactics
G. Mass Ju87 dive bombing squadron
H. Costal low level approach in landing operation
varied massively.)
This means RPK depends on what type of engagement the fleet saw as much as (or even more so than) the FC system itself.

3. Over-claiming. Sorting out the kills in one engagement is hard enough. Attempts to verify claims in mid to late war is boarderline impossible. Making hand-waving arguments and throwing factors like 2x, 10x, 0.5, 0.25, whatever on claims is mostly counter productive. I understand the temptation of using 1942 data and generalise to the rest of the war, but with the changes overall tactic and technology, it's not reliable.

4. For bomber formation, scattering the formation is as important as scoring kills because concentration allow bombers to saturate FC channels of LAA in a particular sector.

I hope it is clear why the RPK argument is going to be futile.

[Steve Crandell]

FC shouldn't be degraded by very much during a ship maneuver, assuming the director and guns can traverse fast enough to stay on the target. Of course, that isn't always the case with ships using HACS and/or guns with manual traverse.

[Steve Crandell]

I think the biggest problem with RPK as a metric is it always goes up as the number of ships firing goes up. There is also the question of whether ships are encouraged to begin firing as soon as possible. But we are repeating ourselves.

[dunmunro]

The effect of wind is to modify the velocity and drift of the projectile. Both HACS and FKC can compensate for the effect of wind via spot correction inputs or via the APV and Drift Correction inputs. However wind also effects aerial targets directly, unlike surface targets, and that can create rather complex target motion.

1. Use of fictitious speed and heading is well documents.
2. No, target motion in the wind is not complex. Especially if you are using GRUB with radar ranging. It will look complicated if you estimate heading using optical means.
3. There's a reason to separate all the variables rather than bundling them to spotting correction jist like surface fire control.

This is from Pout:

Target Behaviour
This is not under the control of the weapon-system designer, and is not
even determined by the characteristics of the target alone . The pilot can
introduce considerable uncertainties , which the fire-control system can
only deal with up to a point. The errors of the weapon system will always
be greater if the target is manoeuvering, for whatever reason, since
changing accelerations are introduced and these can only be determined
crudely by the radar tracker, and then only if smoothing times are not
increased (they are in the VST case of Table 2.6). The Manoeuvre Predictor
case was selected to give some impression of the effect of target
behaviour in degrading performance despite using such a predictor, but
deliberate random manoeuvre can always degrade a gunnery system that
depends on prediction. The only counter to this is the argument that the
pilot's aim is also degraded by this behaviour, but in the long run the
solution had to be a system not involving future prediction. Most
weapon-system development for anything but relatively short ranges of a
few thousand yards has been based on guided projectiles, of which many
kinds now exist.
Table 2.9 summarises the equivalent effects of various assumptions
about the target on the future-position estimate. Roughness of flight is
defined as manoeuvering of the target that is not under the control of the
pilot, and which results from atmospheric effects such as wind and
turbulence on typical aircraft of the period.

HACS and FKC could and did compensate for observed wind by modifying the ballistic inputs (as per Mk 1 computer) but Mk1, like HACS and FKC could not compensate for wind induced target motion except by introducing artificial components into the computer's target motion modelling.

Again, you're making statements specific to individual engagements and then generalizing them to all engagements - in effect you're trying prove that HACS/FKC could never successfully engage aerial targets when we know that this isn't true.

[dunmunro]

RPK is a terrible metric.
1. Long range will always have inflated RPK by its very nature, which regardless of the efficiency of the system.

2. It depends on types of attack, how determined is the attack pushed, pilot tactics, and of course type of aircraft
(E.g. With Mk4 radar ranging, the difference in Mk37 engaging
A. sea skimming approach
B. 1942 style IJN TB approach
C. German/Italian pilot that turns away when burst is close
D. Fw 200 straying into range without realising
E. Approach with "corridor" laid
F. Kamikaze with various tactics
G. Mass Ju87 dive bombing squadron
H. Costal low level approach in landing operation
varied massively.)
This means RPK depends on what type of engagement the fleet saw as much as (or even more so than) the FC system itself.

3. Over-claiming. Sorting out the kills in one engagement is hard enough. Attempts to verify claims in mid to late war is boarderline impossible. Making hand-waving arguments and throwing factors like 2x, 10x, 0.5, 0.25, whatever on claims is mostly counter productive. I understand the temptation of using 1942 data and generalise to the rest of the war, but with the changes overall tactic and technology, it's not reliable.

4. For bomber formation, scattering the formation is as important as scoring kills because concentration allow bombers to saturate FC channels of LAA in a particular sector.

I hope it is clear why the RPK argument is going to be futile.

It doesn't matter what you think about RPK as a metric. What matters is that Marland's article uses USN wartime estimates of RPK that we can confidently assume overestimate 5in/38 performance by a factor of 2X or more, and compares those numbers to Pout's analysis of HACS RPK and from that states that HACS was 1/3 to 1/2 as effective.

[wmh829386]

We have data for Mk37 against manuvering target. Is there similar data for HACS using GRUB? All we have is that GRUDOU is much quicker to get a solution. And even bomber formations are less compliant than sleeves. So I see very little reason to claim GRUB can achieve similar performance given its difference from mk37 and assumptions.

[wmh829386]

It doesn't matter what you think about RPK as a metric. What matters is that Marland's article uses USN wartime estimates of RPK that we can confidently assume overestimate 5in/38 performance by a factor of 2X or more, and compares those numbers to Pout's analysis of HACS RPK and from that states that HACS was 1/3 to 1/2 as effective.

What matters is that I have little regards over claims made base on RPK alone, and I encourage more to take the same view.

[wmh829386]

HACS and FKC could and did compensate for observed wind by modifying the ballistic inputs (as per Mk 1 computer) but Mk1, like HACS and FKC could not compensate for wind induced target motion except by introducing artificial components into the computer's target motion modelling.

I am just talking about wind correction. The effect of wind on target motion should never, ever required fictitious input. The toilet training for pilot navigation is to do simple vector addition-
Velocity relative to air + wind velocity = velocity over ground
(Note that vector (arrows) are added not speeds)

For any measurements from ships, either the velocity over ground is measured (Mk37/GRUB if height is constant) or rate relative to ship is measured. In either case velocity relative to air is never measured and there is no reason to add the wind velocity to target motion.

The problem is HACS does not have dedicated wind correction for ballistics. When enter manually via spotting correction, ballistic of wind changes with range. Apparently fictitious speed and inclination is the common solution.

I am not saying HACS never hit anything. I am suggesting it doesn't perform as well as Mk37, especially when using VT fuse and radar, when it can match performance of later systems down to 2000 yards.

[dunmunro]

HACS and FKC could and did compensate for observed wind by modifying the ballistic inputs (as per Mk 1 computer) but Mk1, like HACS and FKC could not compensate for wind induced target motion except by introducing artificial components into the computer's target motion modelling.

I am just talking about wind correction. The effect of wind on target motion should never, ever required fictitious input. The toilet training for pilot navigation is to do simple vector addition-
Velocity relative to air + wind velocity = velocity over ground
(Note that vector (arrows) are added not speeds)

For any measurements from ships, either the velocity over ground is measured (Mk37/GRUB if height is constant) or rate relative to ship is measured. In either case velocity relative to air is never measured and there is no reason to add the wind velocity to target motion.

The problem is HACS does not have dedicated wind correction for ballistics. When enter manually via spotting correction, ballistic of wind changes with range. Apparently fictitious speed and inclination is the common solution.

FKC fed it's generated elevation and training data into an AFCC (or an FCB) and the wind component was added to generated gun elevation and deflection data and then sent to the guns:

You can see the data transmission in the Tribal class FC circuits diagram in my post of Thu Jan 13, 2022 9:17 pm and this is described in the AFCC handbook: http://archive.hnsa.org/doc/afcc/index.htm

[wmh829386]

The diagram shows FKC directly produce HA elevation without wind correction, while Training does go through AFCC. That means the elevation component of deflection and range are not corrected. Hence spotting correction/fictitious speed is used.

[dunmunro]

The diagram shows FKC directly produce HA elevation without wind correction, while Training does go through AFCC. That means the elevation component of deflection and range are not corrected. Hence spotting correction/fictitious speed is used.

Ballistic modification for wind was made.


Pout:

Gun orders These are the settings for the fuze and the elevation and bearing to
be set in the gun mounting to achieve the future position required. Although the
range data and the horizontal and vertical deviations from the sightline are
directly obtained from the HACS Table, many factors have to be applied to these
to convert them into gun orders that will achieve the desired future position: the
aiming data must be converted to the axis system of the gun-bearing relative to
the ship's head and elevation above the ship's deck ; the effects of wind, gravity drop
and gyro on the spinning shell causing sideways drift have to be allowed
for, as well as factors - such as charge temperature - that affect the performance of
the shell before it is in flight; displacement of the gun mounting from the director,
and own-ship movements.

[Byron Angel]

Wind correction would have been a "best guesstimate". Wind velocity alone will differ dramatically even a couple of hundred feet above the surface. And that would mean spot and correct. Three or four miles away, exact wind direction will be uncertain as well.

B

[wmh829386]

The diagram shows FKC directly produce HA elevation without wind correction, while Training does go through AFCC. That means the elevation component of deflection and range are not corrected. Hence spotting correction/fictitious speed is used.

Ballistic modification for wind was made.


Pout:

Gun orders These are the settings for the fuze and the elevation and bearing to
be set in the gun mounting to achieve the future position required. Although the
range data and the horizontal and vertical deviations from the sightline are
directly obtained from the HACS Table, many factors have to be applied to these
to convert them into gun orders that will achieve the desired future position: the
aiming data must be converted to the axis system of the gun-bearing relative to
the ship's head and elevation above the ship's deck ; the effects of wind, gravity drop
and gyro on the spinning shell causing sideways drift have to be allowed
for, as well as factors - such as charge temperature - that affect the performance of
the shell before it is in flight; displacement of the gun mounting from the director,
and own-ship movements.

There's no wind correction in the input for FKC. And the diagram shows elevation doesn't go through AFCC. So the wind correction is never incorporated. Are you suggesting the schematic you've shown is mistaken?

Wind correction and assumption of level flight are often cited as short commings of HACS compared to Mk37. Of course wind speed and direction depends on altitude, but in engaging TB or Glide bombing approach, it can still reduce the spotting correction needed. My whole point is that the difference betwenn Mk37 and HACS is so big that claims of "they just will do equally good/bad" doesn't make sense

[dunmunro]

The diagram shows FKC directly produce HA elevation without wind correction, while Training does go through AFCC. That means the elevation component of deflection and range are not corrected. Hence spotting correction/fictitious speed is used.

Ballistic modification for wind was made.


Pout:

Gun orders These are the settings for the fuze and the elevation and bearing to
be set in the gun mounting to achieve the future position required. Although the
range data and the horizontal and vertical deviations from the sightline are
directly obtained from the HACS Table, many factors have to be applied to these
to convert them into gun orders that will achieve the desired future position: the
aiming data must be converted to the axis system of the gun-bearing relative to
the ship's head and elevation above the ship's deck ; the effects of wind, gravity drop
and gyro on the spinning shell causing sideways drift have to be allowed
for, as well as factors - such as charge temperature - that affect the performance of
the shell before it is in flight; displacement of the gun mounting from the director,
and own-ship movements.

There's no wind correction in the input for FKC. And the diagram shows elevation doesn't go through AFCC. So the wind correction is never incorporated. Are you suggesting the schematic you've shown is mistaken?

Wind correction and assumption of level flight are often cited as short commings of HACS compared to Mk37. Of course wind speed and direction depends on altitude, but in engaging TB or Glide bombing approach, it can still reduce the spotting correction needed. My whole point is that the difference betwenn Mk37 and HACS is so big that claims of "they just will do equally good/bad" doesn't make sense

For FKC Total Training Correction incorporates wind correction via the AFCC and wind correction is also applied to gun range and gun range is used to modify gun elevation at the gun mount.

The pointer is arranged so that once set to the direction of the wind, the own course motor and the L.S.T. motor will keep it set to that direction.

12. Own, enemy and wide speeds are set by means of the butterfly heads on the top of the clock.

13. The effects of applying the six settings, i.e., L.S.T. and own speed, inclination and enemy speed, and wind direction and wind speed are:-

(i) Rate of change of range is set on the range-rate clock, the amount thus set being indicated on the range-rate scale.
(ii) The range. corrections for enemy and wind speeds along are set on the range-spotting dial.

(iii) The deflections due to enemy and wind are set on the inner of the three pointers (pointer "A") on the deflection dial.

(iv) Deflection due to own ship's speed across is supplied in the reverse direction to the outer pointer (pointer "C") of the deflection dial and to the datum deflection transmitter.

Line of Sight Training (L.S.T.).
15. L.S.T. is received in the clock by both synchronous and step-by-step transmission from the D.C.T. The use of the step-by-step is described in paragraph 65. The synchronous besides being used as in paragraph 66, where it passes through the training clutch, has added to its cross levelling correction, gun deflection and drift to become gun training. (See paragraph 86.)

70. These speed settings (purple) are together with table training (yellow) enemy compass course (green) and the wind direction by gyro (orange) resolved to produce own, enemy and wind speed across and along the line of sight.

Own and enemy speeds along are added to produce range rate (see Plates 10 and 11.)

Enemy and wind speeds along are added and used in the calculation of range correction (see Plates 10 and 12). Own speed across is required in the calculation of gun deflection (see Plates 5 and 6).

Enemy and wind speeds across are required in the calculation of both gun deflection and datum deflection (see Plates 5 and 6).

78. The amount of deflection required to allow for a certain enemy or wind speed across depends on the range. The number of minutes of deflection per knot of "speed across" at any range (known as the "deflection factor") is shown graphically in Diagram 8; the graph is explained in Chatter VII.

85. In H.A. fire the cross levelling synchronous circuit is utilised to transmit total training correction (T.T.C.) from the F.K.C. to the A.F.C.C. where it is added to L.S.T. to produce gun training. This change is achieved by the H.A./L.A. change-over switch and in H.A., the cross levelling clutch must be to Normal.

It took some reading to figure all this out.

Anyways, the only reference that I can find to false setups is Friedman's description of HACS 1 from 1930 and at least some of this was due to the problems in the HACT 1's optical display which was largely corrected in HACT3.



Neither Marland or Pout mention issues with wind or false setups.

Both the RN and USN seemed to do equally poorly against pre-war drone targets. Once a target departs from straightline motion then any advantage MK33/37 might have is lost.

BTW, I found some excerpts from an incomplete FKC manual in my files:

'The deflection calculating system used in the fuze keeping clock is based on the latter circle, known
as the "deflection circle". - It must be appreciated that the projectile's time of flight must be the time to
the future position of the target, Le., to the centre of the circle.

It can also be seen that the deflection circle need not be in a horizontal plane, and that with gliding
targets it will be inclined at the angle of dive.

In the deflection calculating mechanism, horizontal target flight is assumed in that the angle between
the line of sight and the plane of the deflection circle is taken as being the same as the angle of sight, i.e.,
the line of intersection of the vertical sight plane with the plane of the deflection circle is parallel to the
horizontal plane through the binoculars of the director sight.

Mechanism based on this assumption assumes level flight at constant speed and course, when dealing
with directly approaching targets, but a reasonably correct answer will be obtained for a gliding crossing
target.

So some ability to predict against targets in a shallow dive.

The manual also confirmed that ballistic corrections were introduced by the AFCC or FCB via Training Correction and Gun Range dials (at the actual gun). This is further confirmed by Bromley, British Gunnery Computers:

Section 4.1 The Fuze Keeping Clock...

The addition of the total deflections to the Director
Training and Director Setting to give the Gun Training and
Gun Elevation was generally performed in a Fire Control Box
or Fire Control Table which was used with the same director
for surface qunnery.

[wmh829386]

For FKC Total Training Correction incorporates wind correction via the AFCC and wind correction is also applied to gun range and gun range is used to modify gun elevation at the gun mount.

The pointer is arranged so that once set to the direction of the wind, the own course motor and the L.S.T. motor will keep it set to that direction.

12. Own, enemy and wide speeds are set by means of the butterfly heads on the top of the clock.

13. The effects of applying the six settings, i.e., L.S.T. and own speed, inclination and enemy speed, and wind direction and wind speed are:-

(i) Rate of change of range is set on the range-rate clock, the amount thus set being indicated on the range-rate scale.
(ii) The range. corrections for enemy and wind speeds along are set on the range-spotting dial.

(iii) The deflections due to enemy and wind are set on the inner of the three pointers (pointer "A") on the deflection dial.

(iv) Deflection due to own ship's speed across is supplied in the reverse direction to the outer pointer (pointer "C") of the deflection dial and to the datum deflection transmitter.

Line of Sight Training (L.S.T.).
15. L.S.T. is received in the clock by both synchronous and step-by-step transmission from the D.C.T. The use of the step-by-step is described in paragraph 65. The synchronous besides being used as in paragraph 66, where it passes through the training clutch, has added to its cross levelling correction, gun deflection and drift to become gun training. (See paragraph 86.)

70. These speed settings (purple) are together with table training (yellow) enemy compass course (green) and the wind direction by gyro (orange) resolved to produce own, enemy and wind speed across and along the line of sight.

Own and enemy speeds along are added to produce range rate (see Plates 10 and 11.)

Enemy and wind speeds along are added and used in the calculation of range correction (see Plates 10 and 12). Own speed across is required in the calculation of gun deflection (see Plates 5 and 6).

Enemy and wind speeds across are required in the calculation of both gun deflection and datum deflection (see Plates 5 and 6).

78. The amount of deflection required to allow for a certain enemy or wind speed across depends on the range. The number of minutes of deflection per knot of "speed across" at any range (known as the "deflection factor") is shown graphically in Diagram 8; the graph is explained in Chatter VII.

85. In H.A. fire the cross levelling synchronous circuit is utilised to transmit total training correction (T.T.C.) from the F.K.C. to the A.F.C.C. where it is added to L.S.T. to produce gun training. This change is achieved by the H.A./L.A. change-over switch and in H.A., the cross levelling clutch must be to Normal.

You are quoting AFCC manual, but the data flow diagram from Tribals shows the FKC directly generate gun elevation order, bypassing AFCC. There is no wind input at the FKC so elevation is not corrected. You might want to check it again.
https://www.kbismarck.org/forum/downloa ... hp?id=3921