TMA and range determination

[Troels K]

Hi :)

I’m new to Dangerous waters.

I’m having some difficulties determining the correct range to a contact. :damn:

At the moment I’m using active sonar to find the range. However, it would be nice if the range could be found without active sonar.

Therefore, how do you guys determine a suitable range to a possible target?

I know the basic TMA like lead/lag course, but the manual says that two or tree legs will result in a single possible solution (page 5-22), these two or tree maneuvers confuses me. :-?

This is probably a huge subject, however, some short guidelines would also be appreciated. :up:

Thanks in advance.

Troels

[MaHuJa]

The easy way out, is to use the autotma. Given enough time, it'll give you a dead-on solution.

Manually, things are a bit more complicated. But all in all, the extra legs are pretty much a requirement. It's a bit hard to explain, at least without illustrations, but there is a saying of sorts; that a change in bearing rate is needed for a good solution. By changing your own course or speed, you're forcing that bearing rate change. The more radical the change, the better.

Personally, I only exceptionally start working on a solution before well into a second leg.

[Sonoboy]

If you can, assign trackers to that contact on both the towed and spherical arrays. Then, in TMA, where the two different lines converge is a good approximation on the range of the contact.

[Troels K]

Thanks for the replies :up:

MaHuJa:
The thing I don’t get is, how do I use a change in bearing rate to estimate the range? :roll:


Sonoboy:
Good advise I will try that. :)

Troels.

[Henson]

If you understand the difference between overlead and lead lines of sight, turning from an overlead to a lag or vice verse will provide you with both a minimum and maximum range.

[Henson]

MaHuJa:
The thing I don’t get is, how do I use a change in bearing rate to estimate the range? :roll:




A higher bearing rate will produce a greater bearing 'fan' which will limit the ranges based on an assumed speed. The lower the bearing rate, the greater number of possible rangees for a given speed (ie max range is greater).

There is also some funky math you can do to get range from ownship maneuvers, but I'm not sure I can post that here. I'll check.

[SKeeM]

If you can, assign trackers to that contact on both the towed and spherical arrays. Then, in TMA, where the two different lines converge is a good approximation on the range of the contact.

This is the best way to get range. If you can pick up your target on more than 1 sensor and then merg them to form a master contact. At that point you have a good range.

[Molon Labe]

Thanks for the replies :up:

MaHuJa:
The thing I don’t get is, how do I use a change in bearing rate to estimate the range? :roll:


You'll notice the difference at the TMA stations as you attempt to find a solution. On the first leg, the bearing fan created is next to worthless because you can find so many vastly different solutions that fit the data. When you make a second leg, your ships contribution to the bearing rate is different, which means that most of the solutions that you looked at during the first leg will not fit the data from the second leg, leaving you with a much smaller set of possibilities. If you can learn, or at least reasonably estimate, the target's speed, the number of valid solutions begins to approach 1.

[Troels K]

Once again, thanks for the replies, my TMA has improved significantly. ;)


There is also some funky math you can do to get range from ownship maneuvers, but I'm not sure I can post that here. I'll check.
Sounds good, I hope it's possible.


Best regards.

Troels

[Henson]

Google Ekelund range. You can find this on the internet, but you'll have to dig.

Considering where I got this formula I'm still not comfortable with repeating it in an internet forum, but that seemed a good compromise. There a re a few technical papers that wil take you through the process. It basically uses bearing rate and a change in the line of sight to determine a best estimate of range.

[OKO]

You'll notice the difference at the TMA stations as you attempt to find a solution. On the first leg, the bearing fan created is next to worthless because you can find so many vastly different solutions that fit the data. When you make a second leg, your ships contribution to the bearing rate is different, which means that most of the solutions that you looked at during the first leg will not fit the data from the second leg, leaving you with a much smaller set of possibilities. If you can learn, or at least reasonably estimate, the target's speed, the number of valid solutions begins to approach 1.

you have all in that short but accurate description

I just mention one more thing : during the record of the LOBs (line of bearing) your sub must NOT change depth/speed/course, or you will generate corrupted LOBS

so, in resume, you will have a first "pure" leg of 3 or 4 LOBs (preferably 4), thenyou will change your course and you will have 1 or 2 corrupted LOBs (NOT to use in the dot stack because they are corrupted) then again 3 or 4 LOBs on the second leg, after your course change.
then you will align the first 4 and last 4 LOBs on the dot stack, and if you have a good speed estimation of the target, the TMA become really accurate and quite easy.

The main and common problem is to start to work at TMA when you don't have enought information there.
It's totally useless to do that because you could have more than one solution and you could be desesperate to be unable to make the TMA.

So you have to wait the RIGHT TIME to start to work at this station
and the right time, if you make a TMA with a single array, is when you finished to record your 2 legs of data, NEVER before.
This "small" problem cause many people to think TMA is difficult when it is just not.

Basically, you should spend only some minutes per hour on your TMA station, and certainly not big time there.
But you have to start to work ONLY when it's time to do and certainly not earlier.

[Wim Libaers]

It would have been convenient to have a (time corrected) Ekelund or Spiess line calculator integrated into the TMA screen. You can do it now, but you need some extra tools (writing down bearings and rates often, programmable calculator), while all the necessary information is already in the TMA screen.

[SeaQueen]

It would have been convenient to have a (time corrected) Ekelund or Spiess line calculator integrated into the TMA screen. You can do it now, but you need some extra tools (writing down bearings and rates often, programmable calculator), while all the necessary information is already in the TMA screen.


Oooo.... another good thing to stick in my spreadsheet...

[MaHuJa]

Getting the bearing rates in the sonar stations on the subs alone would help a lot there.


Anyway, the thing I said applies for general solutions as well - it's through the effect the second leg has that you can vastly reduce the number of solutions.

[sonar732]

I can still remember ultilizing the "wheel" for basic TMA while in sonar A school. ;)

Place data #1 in outside wheel...turn for data #2....mmmm...Look at your crude drawing of LOS...

[TopTorp '92]

Hi :)

I’m new to Dangerous waters.

I’m having some difficulties determining the correct range to a contact. :damn:

At the moment I’m using active sonar to find the range. However, it would be nice if the range could be found without active sonar.

Therefore, how do you guys determine a suitable range to a possible target?

I know the basic TMA like lead/lag course, but the manual says that two or tree legs will result in a single possible solution (page 5-22), these two or tree maneuvers confuses me. :-?

This is probably a huge subject, however, some short guidelines would also be appreciated. :up:

Thanks in advance.

Troels

Let the auto TMA do do the work. Turn the boat 60 degrees every 20 to 30 min or so. Stay in a lag line of sight. Drive behind the target. Shoot. Clear datum. Then do it all over again tomorrow with another target.

Worry not about the complex math. It's too tedious, time intensive and it will warp your sense of judegement. Watch the sonar broadband display for obvious changes in bearing rate and reposition ownship accordingly. Don't forget to stay in a lag LOS. Drive behind the target. Close the range. Shoot. Clear datum . . . and do it all over again tomorrow . . . with a different target.

Keep a snapshot ready just in case one sneaks up on you.

Top Torp

[OKO]

Let the auto TMA do do the work. Turn the boat 60 degrees every 20 to 30 min or so.

With manual TMA, you need 4 LOBs on the first leg + 2 corrupted LOBs (course change) + 4 LOBs on the second leg => 10 LOBs

10X2 minutes => in 20 minutes from the second LOBs (first LOB is 4 minutes and not 2 minutes and you better not use it), AND if you respected the record pattern (no depth/course/speed change except beetween the 1st and second leg) you'll have a very accurate and quite "easy to find" solution.

You don't need any math or calculation here, you just need to respect the pattern and start to work when you have enought data.
Then you could use the dot stack with maximum efficiency, for very accurate (less than 5% of range error) solutions, in 2 minutes or less.

Even on a target changing course (but more difficult if the target change her speed), you could follow her with a quite good accuracy, but you will need at least the last 3 or 4 LOBS recorded with a straight path for the target or you will have trouble to determine the exact new course of the target.
Once you have 4 + 4 LOBs on 2 legs, range is not anymore a problem, only target course is still a problem if the target often change her course.

[TopTorp '92]

You know, this range issue can be reasonably resolved through an exercise with DW's ability to create scenarios. I recommend that one of our full time gammers do this and make a new thread.

Make a series of scenarios with two vessels in the game: own ship and a target. Place the target at a known bearing and range for each scenario and give it zero speed. Place OS on a known course that is perpendicular to the target. You now know what the bearing rate is for a stationary target for the canned range. You can also measure the bearing rate from the broadband display using an index card as a ruler. The bearing line will be less slanted at narrow aspects and at farther distances. Doing this will build judgement about bearing rate behavior and target range given OS speeds and various LOSs.

The next time you do battle with an on-line gammer, you will have the ability to estimate with reasonable precision target range simply based on the bearing rate and the LOS. ADCAP needs no additional information to find its target.

If all you have is a bearing and no bearing rate or even a bearing drift (you don't know yet whether the target draws left or right) then you can snapshot the target using the bearing you have and enable the torpedo at the minimum range. You still have a wire to the weapon that can update its course and other parameters.

Don't waste too much time polishing the TMA cannon ball thinking you're gonna get the best firing solution available or even get a decent range. You only need to get the best bearing available and use it. This is the way it was done in WWII, during the Cold War and I tell you it is the way it can still be done in this computer game.

[OKO]

Make a series of scenarios with two vessels in the game: own ship and a target. Place the target at a known bearing and range for each scenario and give it zero speed. Place OS on a known course that is perpendicular to the target. You now know what the bearing rate is for a stationary target for the canned range. You can also measure the bearing rate from the broadband display using an index card as a ruler. The bearing line will be less slanted at narrow aspects and at farther distances. Doing this will build judgement about bearing rate behavior and target range given OS speeds and various LOSs.

I'm sorry but that's really not a good way to proceed =>
you will have near the same bearing rate for a ship going 20 knts at 20 miles and a ship going 10 knts at 10 miles, and also same bearing rate for a ship with a 45° course to you at 20knts and 10 miles.

You definitly CAN'T have a range with this method
And this is why Target Motion Analysis ... was created ...

The next time you do battle with an on-line gammer, you will have the ability to estimate with reasonable precision target range simply based on the bearing rate and the LOS. ADCAP needs no additional information to find its target.

If all you have is a bearing and no bearing rate or even a bearing drift (you don't know yet whether the target draws left or right) then you can snapshot the target using the bearing you have and enable the torpedo at the minimum range.

1 rule of torpedo use : Never ever open the seeker to early or your target will fly away.
Certainly one of the biggest mistake to do : nothing easier than to evade a torp pinging you at long range.

You still have a wire to the weapon that can update its course and other parameters.

If you don't know the target range, this is completly useless to try to guide a torp.
Is the target at 5 miles or 15 miles ? when do I have to open the seeker to lock the target ?
You talk me about a blind trying to throw the ball into a basket ...

And defender will have all time needed, because you gave him informations about torpedo position and threat, to built a big a efficient wall of CM that will make your torps completly crazy !
You even better open the seeker to late, and then turn the torp to 180° than to open your seeker to early.
because this way, you passed the CM wall (any confirmed skipper should know how to build a good CM wall) and your torp could come back from behind.
So, NEVER open the seeker to early.

Don't waste too much time polishing the TMA cannon ball thinking you're gonna get the best firing solution available or even get a decent range.

ok, you probably must be joking : I told you just above
1) just work at TMA when it's time to do and not before (read above...), it takes 2 minutes to make a VERY accurate TMA if you do it at the right time, after you have made a good recording patterns
2) if you apply the method described above, you will have a range at less than 5 % of the real one, say 2% when you are trained.

You only need to get the best bearing available and use it. This is the way it was done in WWII, during the Cold War and I tell you it is the way it can still be done in this computer game.

We are not anymore at WWII
torps are now wire guided, are going more than 10 times (for diesel) and 25 times (for nuke) farther than WWII torps, and your sub have much better detection capabilities

On what you said, I can tell you never had a decent solution on a manual TMA. You told me about snapshoters story here.
And a snapshoter against a confirmed skipper, with manual TMA, is just dead meat :roll:

So I just couldn't agree at all to any of your statments above :nope:

[TopTorp '92]

All good and pure criticisms. However, in the snapshot scenario I envisioned earlier ownship's position had already been compromised by the fact that Troel has used his active sonar and I assumed the range to the target had already diminished well inside say 6,000 yds. The point of the exercise I described was to construct a range of possibilities by which we can reasonably estimate target range given only a few moments of bearing rate data. Combined with other clues from sonar (like getting target blade data), this estimate can help you make tactical choices in an environment where time is the commodity. Combine bearing rate from broadband display along with your LOS, use speed from DEMON and you will have a reasonable range estimate to the target using the position-keeping software that is supplied in the game.

In the scenario you mentioned about the same bearing rate for a target doing 20kts at 40kyds (20nm) as a target doing 10kts at 20kyds (10nm) the best way to get a change in bearing rate without turning ownship is to speed up or slow down. Remember, there are two ways to zig: either turn or change speed. Both take time and both will have an impact on sonar's performance.

Also, for the ranges you cited earlier, although this is a game and DW sonar doesn't come with D/E capability (meaning you won't get a feel about range closure), sonar will still be able to provide SNR data for any contact. So, that 40kyd target will sound different from the 20kyd target based on SNR. Sure, my recommended method will not suffice in isolation. But, nobody makes tactical choices in isolation. All available data must be examined, considered and a tactical picture painted on the CEP.

[Doc Savage]

I have a bit of a specific question.
I understand most of the logic described in the above posts and have been putting it in practice pretty accurately. However I almost always have problems when the contact is travelling almost directly towards or away from me (something like <20 degress and >160 degrees to the LOB). In these cases my solution is pretty much way off.

What I currently try to do, is get some kind of range by increasing speed to 10knts+ perpendicular to the last LOB and using the point where the bearings cross as the range. This gets me some kind of accuracy if the contact is close (like 20-30 kyrds). The only trouble is that now that I've hit the throttle so close to the contact, he knows I'm there and tries to hit me with something.
Also, at longer ranges, the point where the LOBs cross becomes a bit "mushy" and I frequently end up with a 20-30kyrd stretch of water where the contact might be.

The other problem I face with the above is that I usually cannot figure out whether the contact is coming towards me or going away.

Does anybody have any advice on how to deal with this kind of situation?

[OKO]

I have a bit of a specific question.
I understand most of the logic described in the above posts and have been putting it in practice pretty accurately. However I almost always have problems when the contact is travelling almost directly towards or away from me (something like <20 degress and >160 degrees to the LOB). In these cases my solution is pretty much way off.

What I currently try to do, is get some kind of range by increasing speed to 10knts+ perpendicular to the last LOB and using the point where the bearings cross as the range. This gets me some kind of accuracy if the contact is close (like 20-30 kyrds). The only trouble is that now that I've hit the throttle so close to the contact, he knows I'm there and tries to hit me with something.
Also, at longer ranges, the point where the LOBs cross becomes a bit "mushy" and I frequently end up with a 20-30kyrd stretch of water where the contact might be.



The other problem I face with the above is that I usually cannot figure out whether the contact is coming towards me or going away.

Does anybody have any advice on how to deal with this kind of situation?

Your sonar tell you if its going away or forward
it's easier with a waterfall of course.
using history will tell you quite well if the contact close or not.

If you already ensured all LOBs were very close from each others (contact closing or moving away), after 3 or 4 confirmed LOBs :
take a perpendicular course as you said, and watch carefully the sonar

If signal increase, the contact is closing you
In this case, stay in the same layer : the ennemy sub could only see you with the spherical, and go 120° from the contact, both to keep distance (and so avoid detection) and to refine TMA.

If ennemy is going away (sonar signal decrease), be carefull => ennemy could catch you with TA if you increase your speed => change layer, then go to the ennemy bearing at combat speed in the other layer than the contact.
regularly (depends on situations), change your course to 60° from the previous contact, go slow and very quiet (you are the hunter as long as not detected), and change layer to catch again the contact (with short TA -1/3- to accelerate the process).
Once you've found a good position to work, stay at the same layer than the contact, at slow speed, quietly, to make a TMA.

When ennemy is going away, it's harder than when it close, because you need to go after him.
When he come right to you, it's the best way for you to calculate a nice TMA on him when he don't even know you are around !

That's why subs never go straight :roll:

[SeaQueen]

I can still remember ultilizing the "wheel" for basic TMA while in sonar A school. ;)

Place data #1 in outside wheel...turn for data #2....mmmm...Look at your crude drawing of LOS...

Are there any good tricks in DW that you use for TMA that wouldn't immediately figure out?

With my background working in OR, I tend to think of everything as a math problem and try to construct a toy tactical decision aid. While that has some usefulness, sometimes I wonder if I'm neglecting the simple answer. :-)

[Bellman]

I wonder how many 'tricks' past and present Fleet members will contribute ? Promotion and places on advanced
courses have to be earnt. That such information is regarded as a golden key conferring an edge is a fact of life !
One Fleet had a very bright serving sonar officer i/c an advanced course imparting tricks of the RL trade !
I regret that there may be some reticence in answering this excellent question.

To what extent past and present individuals with RL experience are free to expand on TMA material, much of which,
is in the public domain is unclear. However I join SeaQueen in hoping that sources from any legitimate
quarter may expand the general understanding of some techniques !

For instance does anyone feel able to expand on the subject of anchor lines ? A topic which should
be expected to develop naturaly from remarks in this thread about manouvering relative to a target.
Or will this remain an ace up the sleeve ?

I hope I share with many the desire to progress within fair competition and not by means of using
information unavailable to new players ! Liberty, Equality, Fraternity !