nother question that the splash itself has unstable form while for correct focusing is required clearly expressed stable border.
If one were trying to range on a splash using a coincidence range finder, this would be true, as one had to have a clearly defined object. With stereo, the operator keeps on the target and uses the short and long diamond grid pattern of the reticle to estimate lateral and range errors. He really doesn't need a clearly defined image on which to range. The operator did not make a cut--that is range on the splash and send it mechanically to the range plot. Instead he verbally sent the estimate verbally to the artillery officer. The spotting officer did likewise with his estimate, and ultimately the artillery officer used his throat microphone to send the final fall of shot estimate to the computer room.
The main problem when visually attempting to estimate fall of shot and determine MPI is that at very long ranges, the splashes are only visible for few seconds at best, and he must use his experience to feel the range rather than actually accurately measure fall of shot placement.
Their shoulders held the sky suspended;
They stood and Earth's foundations stay;
What God abandoned these defended;
And saved the sum of things for pay.
I have had the great pleasure of corresponding with Master Chief Skelley over the past few years. He was instrumental in our research and was great assistance for our Fast Battleship gunnery article that was in WI a couple issues ago. The letter to the editor that MC Skelley wrote was during his early career on Iowa. By 1992 he had become more proficient in stereospotting at extended ranges. Stereospotting requires extensive training and MC Skelley, having fired over 3,500 16-inch rounds, was perhaps among the most experienced in the big gun era.
During WWII, the USN considered 20,000yds the maximum effective range for direct spotting; i.e. stereospotting. Above this bracket and halve technique was desirable and was a particularly useful technique when coupled to radar ranges under certain scenarios such as when firing beyond radar spotting ranges, MB radars disabled but other radars available etc.
A unit of error for a rangefinder is the average error from a trained operator and corresponds to a minimum detectable angle of 12’. It is not fixed but rather is a function of range and can be calculated by this formula:
(58.2 × R²)/(B × M)
Where:
R = Range in thousands of yards B = Baselength in yards M = Magnification
The 26.5’ Mk 48 at 20Kyds:
58.2 × 20² / 8.8 × 25 = ± 105yds
This error figure doesn’t represent the actual error that might be expected in operational conditions. You also have fixed calibration and set up errors and over a ranging series a variable range-to-range error. Additionally you have transmission and plotting errors. The total systematic mean range error for an Iowa class battleship at 20kyds might be around ± 172yds (two 26.5ft Mk 48, two 46ft Mk 53 and one 46ft Mk 52), 269yds at 30,000yds.
Direct stereospotting accuracy can be assessed against the baseline instrument unit of error. Generally 1.5 E is consistent from a trained spotter. 158yds at 20K, 355yds at 30Ky for the Mk 48 rangefinder. Note the much larger error at long range and this is why doctrine called for bracket and halve technique for adjusting fire via optics.
Thanks for the detailed reply. Here's another unit of error table for the 46ft USS New Jersey unit
Unit of Error Table - 46 ft./25x Rangefinder Range in yards 1 Unit of error in yards
5000 3.8
6000 5.5
8000 9.7
10000 15.2
12000 21.9
14000 29.8
16000 38.9
18000 49.2
20000 60.7
25000 94.9
―Most of this was extracted from OP-1572 - Rangefinders MK52, Mod 0 and 1...‖
Optical Characteristics: Stereoscopic Rangefinder MK 52, Mod 1
Optical Baselength: 46 ft Optical Length: 48 ft Diameter (Average): 25.3 in Weight: 11,975 lbs Magnification: 25 X Stereo Acuity (compared to unaided human eyes): 5,520 x 1 Improvement (As a point of reference, a pair of 7 x 50 binoculars are a 14 x 1 Improvement.) Range Limits: 4,500 to 70,000 yards at http://www.ussnewjersey.com/vol6-02.pdf
All the Best
wadinga
"There seems to be something wrong with our bloody ships today!"
Both our Unit of Error tables appear to be extrapolated values derived through the magnification factors etc. Presumeably experiments were carried out to support this assumption. Because performance varies from individual to individual and within the same individual over time due to fatigue or stress (like concern over being hit in the chest with a 15" projectile), these have a variable relationship with reality. Using your equation on the 20Ky I get 60.86 -close to 60.7 in the 46ft table. What is the origin of the 58.2 constant?
total systematic mean range error for an Iowa class battleship at 20kyds might be around ± 172yds
means averaged results from a total of five rangefinders? Do you know if this is an experimental result?
Can you expand on
Generally 1.5 E is consistent from a trained spotter. 158yds at 20K
Is the E the Euler number so 1.5 times is the centre of a normal distribution curve? This appears to be an entirely theoretical value multiplied by a fudge factor of 1.5 to give a value which is somewhat lower than the average for five rangefinders working together.
None of this is mean to be critical in any way but I am struggling to get to grips with the concepts.
All the Best
wadinga
"There seems to be something wrong with our bloody ships today!"
Both our Unit of Error tables appear to be extrapolated values derived through the magnification factors etc. Presumeably experiments were carried out to support this assumption. Because performance varies from individual to individual and within the same individual over time due to fatigue or stress (like concern over being hit in the chest with a 15" projectile), these have a variable relationship with reality. Using your equation on the 20Ky I get 60.86 -close to 60.7 in the 46ft table. What is the origin of the 58.2 constant?
I do not think they are extrapolations. I am not certain how they derived that equation, certainly by testing, but I do know that 1 unit of error is representative of the mean error that a trained range taker would be expected to obtain on that particular instrument under ideal conditions. Of course this error value will vary from take to take and will certainly change as the conditions change. I think that stress/fatigue of range taking from stereo units has been a bit overplayed. 1980s testing and even the combat deployments to ODS showed that operators could still deliver good ranges over extended periods although to be fair they weren’t being shot at either.
total systematic mean range error for an Iowa class battleship at 20kyds might be around ± 172yds
means averaged results from a total of five rangefinders? Do you know if this is an experimental result?
That is representative of an error budget for an Iowa class battleship based on fleet experience and includes all 5 RF and is based on trained RF operators
Can you expand on
Generally 1.5 E is consistent from a trained spotter. 158yds at 20K
Is the E the Euler number so 1.5 times is the centre of a normal distribution curve? This appears to be an entirely theoretical value multiplied by a fudge factor of 1.5 to give a value which is somewhat lower than the average for five rangefinders working together.
None of this is mean to be critical in any way but I am struggling to get to grips with the concepts.
All the Best
wadinga
No “E” should have really been lower case ‘e’ and is the symbol for 1 unit of error. In this case I was referring to the spotting error not the ranging error. Generally the mean spotting error is between 1.2-1.5 e. Ex. For the 46ft Mk 52 at 20Kyds, the average spotting error for an experienced spotter might be about +/- 82yds. This value does not hold constant however and certainly changes as the target disappears over the horizon.
The 50cm radars were first proposed in Dec 1939. An expermental set was installed on Nelson by summer 1940 but this was actually the proto 282 for pom pom direction. It used two small yagi aerials (nicknamed fishbones), one for send, and one for recieve. The beam width was 37*. After this, the equipment was converted to a proto 284 by intalling two 1/2 cylinder antennas to the main director. This antenna array recieved the nickname; "the pig trough aerial". The early version used two pig troughs, one for send and one for recieve. The range performance obtained with this set was never again matched by a production version 284. This lashed up set was removed from Nelson in Nov 1940, and Nelson was equiped with a regular production 284 during 1941.
KGV became the first warship to recieve the production Type 284 in Dec 1940. The early version type 284 did not use lobe switching and it's best bearing accuracy was 0.75* using max signal. Range resolution was 300 meters, and range accuracy was +/- 150 yards. Tests on with the KGV set revealed the following range performance:
It was not tested vs a battleship until it was used vs the Bismarck, but the range listed is commonly 20,000 yards. KGV's 284 picked up Bismarck intermittantly at about 25,000 yards, but could not be used for spotting or consistant ranging until closer to 20,000 yards. The Suffolk's 284 was able to track Bismarck to a max of 26,000 yards. One of the interesting things about the 284 was that it had to be shut down every so often to rest the transmitting triodes. A noted point of anoyance was that there was a prolonged warmup period when re-starting the radar.
The improved 284 came available in early 1942. The new version was called Type 284I ashore, but a set modified to the new specs was called Type 284M. The new model was only first installed on new construction or on warships not already equiped with a "radio range finder". Those warships already equiped with the early model 284 had to make due until early 1943. The new model had a range resolution of 150 meters, but the Royal Navy wasn't satisfied with the range accuracy, because it deteroriated with increasing range, although it was intially better than the early model's, and far superior to the optics. The new version used lobe switching and a common mode antenna. Both the pig trough and an array of 6 yagi "fishbone" antennas could be used depending on the exact installation. The fishbones could produce a 1/2 power beam width of 4.5*. Specs for the pig trough antenna vary from source to source from 3* to 5*. Bearing accuracy using lobe switching could be as good as 0.08*, but inconsistancies in the director itself could diminish this attainment. The effective range, battleship to battleship, is listed as 29,000 yards. Historically, the Duke of York's 284M picked up the Scharnhorst intermittantly at about 35,000 yards, but the amplitude of the pip(s) was too small to fix the bearing until the range had closed to 26,000 yards.
Entering a night sea battle is an awesome business.The enveloping darkness, hiding the enemy's.. seems a living thing, malignant and oppressive.Swishing water at the bow and stern mark an inexorable advance toward an unknown destiny.
