Torpedos, extreme range and accuracy

[Von Due]

I am looking at the possibility of shots at extreme range, 5,000-10,000 metres with a reasonable probability of a hit and have done some calculations and looked at methods for getting target speed and heading, setting up for a plain 90* shots.

Needless to say, the accuracy needed is really touching on the limits of what can be achieved but here are some thoughts so far. Please chime in if you spot flaws etc!

Speed is if possible even more crucial to get just right and the 3.15 rule is, imo, too rough to give that exact speed for those ranges. I am looking at 730 metres per knot per 23 minutes 39 seconds. The time and distance per knot would have sufficient resolution to determine speed with a high level of accuracy, even in rough sea where target speed is yoyo-ing over short distances.

About target heading. I have looked at speeds of 7.5 knot and 12.5 knots range to impact point 10,000 metres, torpedo speed 30 knots. The way I reasoned and do correct me here if I'm wrong:

First, I calculated the time T in seconds it takes the 30 kts torpedo to travel 10,000 metres to the ideal impact point P. This is the ideal situation where I somehow managed to get target heading accurate and am on a perfect perpendicular course.

For a 12.5 kts target starting at torpedo fireing point A, in that same time, it will travel T x 12.5 x 1.852 / 3.6 metres.

I now have a triangle with two sides of 10,000 and T x 12.5 x 1.852 / 3.6 metres.

Now if I got the target heading wrong by +- 1*, that would mean the distance from P to true impact point P+ or P- is
T x 12.5 x 1.852 / 3.6 x tan 1*and the true distance traveled is
10,000 +- T x 12.5 x 1.852 / 3.6 x tan 1*

I now need the time it takes the torpedo to travel those distances, T+ and T- and I calculate how far the ship will travel in the same time.

By comparing true impact points to where the ship would actually be, I find that I will miss the aiming point by some 30 metres. The actual distance would vary depending on whether I missed by -1* or +1* but roughly 30 metres will do.

That means I should not attempt long range shots at targets at around 60 metres in length or less but for a ship or 100 metres or more, I should have a good chance of hitting if I aim at center.

For target speed of 7.5 knots, the possible error would be around +- 11 metres, meaning if the target is moving slow, I should be able to hit a 50 metres or more long target if I aim at center.

All this of course rely on accurate heading and speed calculations but the bottom line is, even if I get the heading wrong by 1 degree either way, I still stand a good chance of hitting at 10,000 metres.

Any inputs and corrections greatly appreciated!

[Red October1984]

Wow....I am sorry i cant help you...but i dont understand a bit of that. I wish i did. I wish i could figure out all of that. But, I dont. I just line up and shoot...:O:

It'd be cool to one day, understand all that. :up:

I have, on a few occasions, hit with torpedoes at 6000-7000 meters.

[Von Due]

The problem with long range shots was something I pondered when I played SH2. I was looking for a way to get the targets without having Superman, the X Men and the A Team onboard those Marvel Comics destroyers racing towards me wherever I was within medium range. IIRC, the graphics didn't allow for much sniping but now that I am able to identify ships at greater ranges I might have a chance.

I have only succeeded one or two times before in SH3 at 10,000 metres or so and only at high resolution but I had no good idea of why it couldn't be done a bit more consistently. A few things are certain. At those ranges I'm dealing with fractions of degrees and 1/10th's of a knot but it can be sorted to a pretty good level.

A rough sketch to accompany the first post
http://i189.photobucket.com/albums/z193/WhoCaresWho/torp.jpg

[Red October1984]

I close to about 2000-4000 before i fire. When i do fire, I set it up so that i either

1. Take out as many targets as possible

2. Make sure that i take out my one target

3. Can disable the most dangerous target

I have disabled several destroyers and then followed the High Value Target ships a good while and sink them. But, the third option is a bit harder than the first two.

A torpedo to the screws can do a lot to a destroyer. :up:

[derrinurban]

To me the hardest part about extreme range attacks was the ship speed and actually firing the torpedo. The math is the easy part. You can get the ship speed close. You will probably never be 100% accurate but very close. Then you need to time you shot perfectly. Hitting the fire button precisely at the right time is almost impossible. Again you can be close but not perfect. Those fractions of seconds between when you are supposed to hit the button and do and the second for the torpedo to clear the tube make hits at that range lucky at best. IMHO

[HW3]

That plus if you are using steam torpedoes, the wake might be spotted and evasive maneuvers taken by the ship, which will also cause a miss.

[Shkval]

Well for me the classic stationary 4 bearing (i.e. three bearing... I use fourth bearing only in low visibility) + ship length x 1.943/ seconds from bow to stern, works pretty good...
Except for:
1. Outgoing targets (it doesn't work at all... any solutions? except the answer "don't go after the outgoing ships")
2. Changing course targets.
3. And those... I call them "dead angle" targets (targets who quickly enter into "constant distance" state i.e. too wide range of possible solutions for AOB... and you get equal degrees between bearings... bad karma).
For me usual distance is 3 to 6 MAYBE 7 km (depends on weather, type of a torpedo, time of a day/night, visibility and generally combination of all these factors).
Everything above 7 km is a shot with 50% of pure luck if you are firing single torpedo... salvo is a different story... chance of a hit grows exponentially with every additional torpedo. But that's a waste of ammo in my humble opinion...
I did scored hits at 11 km... but that was clear, perfectly calm sea at night with moonlight and the large target (HMS Hood)... task force was announced trough contact report so I did have time to position myself nicely and prepare the attack... there was a few more cases but that was pure luck...

[Von Due]

Thanks a ton for the input, you all!

This is an attempt to find out if it is at all possible to increase the probability, to find out how much margin for error I actually have, theoretically up against the game and what it allows.
Questions are, how much inaccuracy for a 100+ metres ship? How much error can be allowed to hit within 40 metres to either side of centre? And how accurate does the game allow me to be?

So far I have found the stationary 4 bearing method by far the best but it does have limitations on accuracy if I disable the hydrophone lines and even they are not telling the true story, sometimes pointing near the engine room, other times near the screws. Without the lines I have to allow for 1 degree error on each bearing and that could mean getting the target course wrong by as much as 10 degrees or even more and reading speed is no better off when I need that much higher accuracy. I am looking at the possibility of using optics at closer range to get more exact bearings but unsure if it can be done within the game, in particular the fact that I can only read a reasonably accurate bearing via the TDC and need to jump to nav map while time is ticking.

As for timing the fireing, at those ranges it's impossible to be exact, resolution and how much each pixel is in ship length means I have to eyeball it so there is inaccuracy there.

I already have found that surface attacks are right out. Only by staying submerged can I keep the course at a constant heading and even then it takes a bit of rudder to keep it at an exact heading. Submerged, it's manageable but I don't want it worse.

Steam torpedos and wake, that is indeed a challenge. What I wonder is how to decrease the probability of them spotting it. Lighting, weather and their crew experience matter but what can be done to maximize the chances there? Reason I wonder is I had ships taking evasive maneuvers at night when my steam torpedos were some 1500 metres away. Just bad luck and the A Team on the bridge? I really would like to know.

Outgoing targets, I'm sorry but what do you mean by that? Targets past the fireing angle?

I depend on target keeping a constant heading and speed. Without that, I don't think I could guaranty a hit at 400 metres :) let alone at 10,000 metres.

[Hinrich Schwab]

I rely on the Fast-90 method for all of my shots. However, the range comes into play at point-blank (< 500m/yds) and at extreme range (> 7000m/yds). I did some long calcs a while back and the float is only +/- 1.5 degrees. No big deal at point-blank, but the extra 1.5 degrees does matter at extreme range. Using ATAN (Target Speed/Torp Speed) to determine lead angle, subtract the 1.5 degree float for starboard-to-port targets and add the 1.5 float for port-to-starboard targets. My hit rate is around 75%, taking into account dud rates, possible aspect changes from the target and general weirdness. My longest range hit and kill is around 9km.

[CCIP]

My best shot was two separately-aimed torpedoes that hit a tanker from 7200m. It was part of a convoy that I beat up pretty savagely over a couple of days and had got kind of disorganized, with escorts out of sight. It was sailing on the edge of the convoy, with a freighter ahead of it, and no escorts anywhere. I went up on the surface, far enough for them not to see me, used both ships as reference to match their course and speed, and got a decent range reading. I then scooted slightly ahead, got into position to get a dead-on 90 degree shot, and waited for the leading ship to get into a 90 AoB position. I then simply shifted my sights to the 2nd ship (which gave me its exact AoB, because it was following the same course), took a range estimate, fired a fish, took another range estimate, fired another one on it. About 7 minutes later, both of them hit. It was pretty satisfying!

Other than that, I rarely hit anything at more than 2k. Unless they're convoy targets or especially valuable ships (big tankers, big warships), I don't even really try if they're at over 1km.

[Von Due]

Not something I expected to find out but nevertheless:

I was wondering about the 4 bearing method and how much error would be introduced in each step, no matter how meticulous I was there would always be a chance of errors in each step so I was wondering if one could chop off even one. Turned out I could.

First, this requires a stationary plotting. I haven't looked into plotting while moving so thus far the geometric method is the only way I know of there but:

Stationary plotting.
Get the 3 bearings down from a common point. Now choose a distance on the 2nd bearing that can easily be halved. I choose 2,000 metres to keep the compass circle within 1000 m for highest resolution.

Copy either of the other two bearings with the compass

Let this copy run straight through the , in my case, 2,000 metres mark on the 2nd bearing so it intersects the last bearing.

From this intersection, draw a line across the 2nd bearing at the half mark, in my case the 1,000 m mark. This new line is the target's heading.

EDIT: Instead of geometry, I went for vectors. If we call the common point U for u boat, the target heading line's intersections with the 1st bearing at point A, 2nd bearing at V and 3rd bearing at B, then 1/2(UA + UB) = UV and UA + UB = 2UV, thus by copying say, the 3rd bearing and let it run through the 2,000 mark, you get vector UB and it meets the 1st bearing at A. I will look for a way to use vectors for plotting while moving but invite anyone to beat me to it :D

[Shkval]

Outgoing targets, I'm sorry but what do you mean by that? Targets past the fireing angle?
Moving away targets.

[TorpX]

Von Due,

I mostly play SH4 and do not make many very long range shots, but I would say if you can get even a few hits at these ranges, without game exploits, you are doing quite well.

In any case you might be interested in what the USN came up with, in regard to the optimum torpedo track angle. They graphed out the deflection angle versus the track angle and found the optimum TTA is not 90 deg., as might be thought. The optimum angle occurs where the deflection angle reaches a maximum and therefore changes only slightly as the TTA changes.

The optimum TTA can be computed by:

Optimum TTA = 90 deg. + Arcsine (target speed / torpedo speed)



Here is the USN torpedo fire control manual. The relevant diagrams are plate I, figure 2, and plates XVII, and XVIII.

http://hnsa.org/doc/attack/index.htm

[desertstriker]

personally i get up close and personal THE longest shot i will fire from is around 3km and the closest is 100 meters with an impact fuse (magnetic take aprox 300m to arm). i have just found that anything over 3km has a higher chance for the fish to not arrive at the target due to early detonation or just died ran out of juice.

PS i use ONLY electric torpedoes.