I think that TMA is only possible when you can correctly estimate the target's speed. Then you compare multiple bearings over a period of time, and 'fit' both data (bearings and speed) into one coherent result. If you only have the bearings, his range (and relative course too) are impossible to estimate.

So the thing is to estimate speed via some reliable source. The way I see it, the mode of choice in the last decades is TPK (turns per knot), a thing you only can use if at the same time you can also classify the target, all of that presuming you know the TPK ratio for it- a not so easy thing to do with 1940s sonar gear and intel.

You would be correct for single leg TMA. To get a single solution for speed you would do multi-leg TMA.

The Eckelund method does not require any estimate of target speed, it provides range only.

I am not sure when the specific techniques I previously was familiar with were developed, but I agree they would be tough in WW II era subs.

I have only used active sonar to lure destroyers away from convoys. If I ping them from a 90 degree bearing then dive and go ahead flank they are more apt to center in on my last position where the ping came from while I angle in towards the body of the convoy. Imagine a T where the bottom of the letter is my position the far right point of the top cross bar is the merchants and once I pined them I headed for the top left point of the cross bar to intercept.

It really work well once in shallow water I'd say 100 feet with a rough surface. The destroyers stay in the area I pinged from while I intercepted and fired torpedoes at the merchants. By the time they turned and came back I had sank three ships and was heading away from them. They depth charged the area they thought I was in but I was a good 2000 yards away and they never approached me.

[QUOTE=groomsie;1879706]You would be correct for single leg TMA. To get a single solution for speed you would do multi-leg TMA.

/QUOTE]

You mean, collecting a series of bearings, then turning your own course and then collect a few more?

I've been trying to learn how to use different attack methods including radar only and sonar only (active and passive and passive only).

I have a practice mission (one merchant ship, dark night) that I use. I've been pretty successful using radar only (with the OTC mod) and find that the radar range agrees closely to the chart plot ranges (with the radar and sonar only attacks, I do check once I've fired the torpedoes).

Today I tried the 4 bearing passive sonar method to get course, bearing, and target speed (3 bearings with sub stationary + one more after moving to get range.) I had the sub at 70 ft so I wouldn't get any cues besides the sonar trace which I used for the bearing measurements. I did check range with active sonar once I had a plot running, but the range was more than double the predicted (and later verified range.) So far I have not had any success with using active sonar to predict range (always over estimates.)

As the passive only attack evolved, I chickened out and decided that since the range was the most questionable thing I would use a constant bearing attack (once set up I did a second 3-bearing check on target course, but was running out of time and couldn't make the run to get a check on range)

I choose a 60 degree attack, and loaded the aim bearing, AOB, and speed into the TDC. I used the attack map which showed the sonar bearing and when the sonar trace crossed the 353 degree bearing (my aim bearing) I fired one torpedo. Once I launched I rose to P depth and raised the scope so that the ship would show up on the attack map. I had calculated the ship would be at 1600 yrds when it crossed my zero bearing, but it actually measured out to 1950 yrds. The course (and actual AOB) were accurate. The torpedo hit.

Biggest issue is that it took me a long time to collect the bearings (I used 5 minutes between bearings), and a very long time to do the graphic construction (zoom in to the chart to locate the points accurately, and zoom out to extend the lines.) Thank goodness for "pause game". I bet the WWII submariners really appreciated their pause buttons :DL And there was not enough time to recheck range.

I am surprised that my active sonar ranges are so far off.

Tom

Now THAT is a great tactic! I love it!:rock:

Wow, lotta incorrect information here. I've got a US Navy maneuvering board textbook that I found online dated 1941.

Page 26:

(copy pasted)

Case XI

TO DETERMINE COURSE, SPEED, AND RELATIVE POSITION OF A TARGET BY BEARINGS ALONE


GIVEN: TWO SETS OF THREE OR MORE TIMED BEARINGS, EACH TAKEN ON A TARGET BY AN OBSERVING
UNIT WHICH CHANGES ITS OWN KNOWN COURSE OR SPEED OR BOTH BETWEEN SETS OF
BEARINGS.
TO DETERMINE: COURSE, SPEED, AND RELATIVE POSITION OF THE TARGET.
Example.—An observing vessel G, while on course 110°, speed 15.0 knots, obtains radio bearings on target vessel Mas
follows:
Time 0800. Bearing of M 000°.
Time 0900. Bearing of M 349°.
Time 1000. Bearing of M 336°.
At 1015 the observing vessel changes course to 045° and increases speed to 20.0 knots. Bearings are next received as
follows:
Time 1030. Bearing of M 327°.
Time 1130. Bearing of M 302°.
Time 1230. Bearing of M 273°.
Required.—(a) Course and speed of M. (b) Relative position of M at 1230. (See fig. 15.)
Procedure.—Plot position of observing ship at any convenient point G, and lay out the 0800 bearing line as G . . . . bu
the 0900 bearing as G .... 62, and the 1000 bearing as G .... 63. By any of the methods shown for case IX, draw a slope
P .... Q .... R across these bearing lines so inclined that the intercepts P .... Q and Q . . . . R are proportional to the
time intervals between bearings.
In a similar manner lay out the second set of bearings, G . . . . bit G . . . . b5 , and G . . . . &6 - Draw the
slope T .... U .... V so inclined that the intercepts T . . . . U and 17 .... V are proportional to the time between
these bearings.
From any point e, lay out the first vector of G as e . . . . gu and transfer the slope P .... Q .... R to gi. Draw the
second vector of G, e . . . . g2 , and transfer the slope T . . . . U .... V to g^. This intercepts the slope from gi at zn.
e . . . . m represents the course and speed of M.
Determine the Relative Speed g2 . . . . m and by means of the Logarithmic Scale find the Relative Distance travelled
by M between the 1030 and the 1230 bearings. Lay off this distance, T . . . . W, and by completion of the parallelogram
locate the position M at 1230. T' .... M is equal to T .... W and is the Line of Relative Movement between 1030 and
1230. M's bearing and distance from G at 1230 is G .... M.
Answer.—(a) Course 164°, speed 12.6 knots, (b) 57.5 miles bearing 273° from observing vessel.
NOTE.—Solution by this method will not be obtained if the second vector of G should lie along the transferred slope P .... Q .... R.
For this reason this slope is transferred to g\ before G changes either course or speed or both.
If the bearing does not change in either set, a solution is still obtainable. The slope in this case is a constant bearing and is laid off in both
directions from the head of the vector concerned.
In case G makes a change of course when the last bearing of the first set is obtained, this bearing may be used as the first bearing of the
second set.
When bearings alone are available, the results should be considered as approximations only. This is occasioned by the fact that a small error
in one or more bearings will change the inclination of the slopes to be transferred, and this in turn will change the final results. If sufficient time
is available, a third set of bearings will act to check the course and speed of the target.

So Sanders, is that "multi-leg TMA"?

GIVEN: TWO SETS OF THREE OR MORE TIMED BEARINGS, EACH TAKEN ON A TARGET BY AN OBSERVING
UNIT WHICH CHANGES ITS OWN KNOWN COURSE OR SPEED OR BOTH BETWEEN SETS OF
BEARINGS.


So.. Yes. The observing ship must change course or speed or both between bearings 3 and 4.

:salute:

I would resist any passive only TMA for the reason that it just wasn't done in WWII. I've set up a practice mission and tried it out. It works, but is just so time consuming and multi-stepped that error is almost inevitable in the game. With a fast-moving target forget it.:haha:

So you won't see me handing out passive only TMA techniques on Subsim. That being said, I use passive all the time to fire torpedoes when the target course and speed is already determined, and by superimposing the bearing on the target track I can derive the exact position. If I'm working with position keeper on, I can compare actual with projected bearing to detect target course changes.

That way I can use visual, radar, active sonar and passive sonar techniques any way I want to mix up a lethal brew of Frankenstein targeting. Usually I'm using all that with a constant bearing attack to mitigate any range errors and make setup something I can do half an hour before I shoot. I hate working under pressure. People get hurt that way.....:har:

I'm still looking for the incorrect information......... I can't seem to find it. Did someone actually utter a falsity? I just can't find a consequential misstatement, much less a "lotta."

The incorrect information:

The ability to determine a targets course range and speed from bearings alone did not exist in WWII.

That you have to be able to know your own position with certainty to do a bearing only plot.

The game tools are inadequate for the technique.

That TMA is only possible when you already know the target's course and speed.

There's too many totally different answers as far as targets being able to hear your active pings. Someone really needs to do some testing.

I also saw an assumption that merchants don't have sonar, I know of at least 2 that do (nippon and akita maru), there's probably more.

I challenge everyone to produce a single instance of an American, British, Dutch, German, Italian or Japanese submarine doing a wartime passive sonar TMA on any target. I can't say absolutely that it didn't happen, but it didn't happen.:har: I sure hope I'm wrong but I'm not holding my breath.

I don't remember seeing that claim. If it's there, you're right.

Now cut that out! If passive TMA is using bearings only it's fundamental that taking bearings is something our game does too well and TMA is possible. Who said that? Again, I didn't recall anyone making that argument.

Daniel Prates said "I think that TMA is only possible when you can correctly estimate the target's speed." He did not mention course at all! And groomsie (who shares your experience in having performed TMA himself) corrected immediately that Dan's statement is true only with single leg TMA, and that double leg TMA makes estimated speed unnecessary.

That's a possiblilty. My information that no targets respond to active sonar pings comes from Ducimus, who has torn the AI totally apart and jabbed it unmercifully to see how it yowls. If there exists any reliable source of information Duci is it.

However, I don't recall any claims to infallibility on his part. When he posted his findings I just quit pinging warships, as the results of that action would be unrealistic if Ducimus is correct. I wouldn't stake my life on it, but I have a high degree of confidence that he knows what he is talking about and is not in the habit of making unsubstantiated claims.

Let's set up a practice mission and play with it. It's been awhile since the AI was tortured and poked and it's high time we did some of that!:rock:


Are you talking active sonar or just passive? How can you tell? Again, I've never had any reaction from any merchant at any time of the war to being pinged. Heck, I pinged the USS Essex repeatedly when.......never mind.:88) Let's just say her reaction was to sink rapidly when I put three or four torpedoes into her. I got no subsequent chances to test Japanese merchant reactions to pings...:oops:

I've pinged DDs far away, and up close and personal. The only way they know I'm there, is if they hear boat noise, or spot my scope.

Not once have they charged in after pinging them. So I quit doing it, because it semed .......................................... well..................just wrong to do it.

Good deal. Now some of us need to do that under more controlled conditions before we totally declare Ducimus correct. Thanks for the input.