The convergence problem

[Hitman]

I have opened this thread as a continuation of the discussion initiated in the "How do you shoot your convoys", as I did not want to hijack it :oops:

I thought the convergence problem would be of interest for anyone doing manual targeting, so here goes a brief explanation:

When you calculate your firing solutions, you do it from either scope or UZO, but those instruments are not above the torpedo but instead some 35 metres behind, in the conning tower. Thus, if you fail in determining the distance to the target, you could have problems with angled shots, the more noticeable when the Gyro angles are high. In the following image you can see and understand where the problem is: Ship A and sihp B have exactly the same bearing from the periscope/uzo, however ship A is closer than ship B. Thus, if you miscalculated the distance target, the convergence between the bearing line (green) and the torpedo trajectory will produce itself in the wrong place. In ship A distance is correct, and the torpedo run intercepts the bearing to target (Line of sight) exactly where the target is. That's an impact!:up: However, in case B a large error in distance coupled with a large Gyro Angle will cause the torpedo to miss the target astern, even if AOB and speed were correctly calculated.:down:

http://img37.imageshack.us/img37/64/convergence.jpg

Thus, keep this in mind:

-The larger the distance to target, the more likely a convergence error
-The worse the distance to target estimating, the bigger the convergence error
-The larger the Gyro Angle, the more likely a convergence error will have big effects
-The TDC applies the proper convergence correction (Try this: Go to F6 screen and with stationary submarine, set manually your target as 90º left or right of you, speed zero. The green line will converge with the line of sight from your scope at the distance to target you did input. Play a bit with the distance dial and you will see what I mean)

Hope it helps :up:

EDIT:

You can check this in your TDC (F6) screen:

Set the TDC to manual input. Move the bearing to target dial to 90º left , leave AOB and speed at zero. That would represent a target being stationary left of your submarine, periscope/uzo bearing 270º (90º red side). Note that the green line for the torpedo is curved, making the torpedo run match the 270º bearing from your boat exactly at the distance you told the TDC that the target is. Play a bit with the distance dial and you will understand what I mean.

[TarJak]

Nice clear explanation Hitman:up:

[Heibges]

That is why I never use, what the Uboat Commanders Handbook calls "The 90° shot". :D

And this is another reason, although the TDC doesn't require it, that you use the good perpendicular approach course. It makes determining range, and estimating AoB much easier in my opinion.

[joegrundman]

The larger the distance to target, the more likely a convergence error


Of course, although there is more likelihood of generating range errors at longer ranges, I think the size of the convergance error is somewhat smaller as the angle of difference between the "true" and "false" positions reduces with increasing range.

I suspect the peak of range related error is somewhere in the medium range. Watch this space and I'll prove it ;)

[looney]

Hmm no proof yet I see ;)

[Von Due]

I know this is a very old topic but I am looking into non-zero gyro angle shots and I have made a few finds that might be of interest to anyone else looking into this.

Tools used here included the attack map, pato's bearing/range tool, draggable chronometer as well as a draggable watch with a seconds hand. All torpedos at fast speed setting, tubes opened well before launch.

I set up a small series of test shots where I aimed at a fixed point relative to the sub's orientation, that means a fixed distance (d in meters) from the center of the uboat at a fixed bearing b in degrees, noting down gyro angle (g in degrees) and total run time (t in seconds measured with the chronometer) and time since launch order (tl also in seconds measured with the watch). Also noting down the range setting on the TDC (r) where the estimated time shown on the attack map (F6) would agree with the watch (not the chronometer).

The sub was stationary and submerged in calm weather for max stability and predictability.

I would press the launch button the moment the seconds hand on the watch reached 0/12. When the torpedo reached the designated point, I paused the sim and noted down the data collected.

First a straight run shot at b = 0 for d = 3000. Results:
g = 0 (as expected), r = 3000, t = 131s, tl = 132. Note the 1 second difference between the chronometer (t) and the watch (tl).

Next I set up for b = 270, d = 3000. Results:
g = 267.4, t = 135.5, tl = 136, r = 3100. Note the 4 seconds increase in tl from the straight shot. This is due to the time for turning as well as the slightly greater distance of travel. Also note the TDC range setting r = 3100. This is necessary to get the correct tl = 136. Also note an indication that the chronometer will only lag behind 0.5 seconds and not 1 as in the other shots. This could be a fluke but 3 test shots gave the same result, a lag of 0.5s.

Then a shot for b = 315, d = 3000. Results:
g = 313.95, t = 132, tl = 133, r = 2700. Here tl is 1 second greater than with the straight shot, and r has to be set to 300 meters short of 3000 to get the correct tl = 133.

Next was a shot at b = 315, d = 2000. Results:
g = 313.4, t = 87.5, tl = 88, r = 1800. Here r is 200 short of 2000 to get the correct tl.

Then for b = 315, d = 1000, I got:
g = 311.5, t = 43.5, tl = 44, r = 900. Again, r is 90% of d.

Finally I set up for b = 45, d = 1000. The results here were surprising:
g = 48.65, t = 43.5, tl = 44, r = 1025. For 45 degrees starboard, the range had to be increased by about 25 meters, while 45 degrees port would require r to be about 10% lower than d. The gyro angle off 0 bearing is also different. 48.65 vs 48.5 for the port side shot.

The results were the same for all 4 bow tubes on the VIIB with no noticable difference for port and starboard tubes. I have no explaination for the difference in range setting for the 315 and 45 shots at 1000 meters. Will continue the tests.


EDIT. Test 2
Found that the prediced time would consistently drop for clockwise gyro angle up to 90 degrees, from 130 at 0 deg. to 121 at 90 deg, and from 130 to 134 seconds at 270 deg. for counter-clockwise gyro angle. It is not about port vs starboard but clockwise vs counter-clockwise as tube 5 at the stern showed the same peculiarities.