How can i effectively measure bearing rate?

[phil21]

Hello guys,

here is another typical "noob-question" :
How do i get a precise bearing rate for a contact? I know in the Navmap you can find one, but this one only appears after i've done my TMA and its also influenced by it, so its useless for what i had in mind.
I also know that you can just sit there with a stopwatch and calculate it yourself but i would like to know if there's another, faster (and perhaps more reliable) way...

I'm asking this in a new thread because the threads i find with 'Search' do not really answer the question(s) i have. Right now i'm working on a simple program to calculate Target range (and maybe more later) via´Ekelund with OwnShip data plus LOB (Bearing Rate). So if theres another way to do this (getting a good TMA with (as few as possible) bearings only) this would be also of interest for me...especially if theres already a program for it.

Phil

[jmr]

Great question. It's a shame you DW doesn't give you a bearing rate on the subs sonar displays but you can find it manually.

One method involves using a utitlity called Mobo in which you overlay it on top of the TMA screen and plot a bearing line over the bearing in question to get it's measurement. You'll have to run DW in a window if you wish to overlay Mobo on top of your DW screen otherwise if you run DW in fullscreen you'll have to grab a screenshot, alt-tab to Mobo and dump the screenshot into it.

I can post a screenshot later if what I said didn't make sense.

Keep in mind Mobo won't give you the bearing rate, it only allows you to get a more precise bearing reading (to how many decimal places you'd like) than you would get by simply looking at your bearing display on the sonar screens. Once you have your bearings you can find the rate manually or dump into Excell and find it using the "data analysis ----> regression" tool.

[phil21]

Thank you,

I use Mobo then if theres no other precise method.
My wish for the next Patch/Addon/Game/whatever: Give More Sonardata like Bearingrate etc...and make it accessable for TMA. So we can skip all these additional programs...

Maybe i should also work on a Excell table for most of these 'basic' things...

[jmr]

A nice Excell sheet would be pretty cool to have. Do you need the Ekelund range formula?

[phil21]

I just finished the Program for calculating range with Ekelund (double-leg). It only needs to be tested, i think i do that tomorrow...if its ok i polish it one last and maybe i upload it somewhere (if its needed). Perhaps i also add the other stuff for a full TMA...but this will need some more time.

After this the next thing to do will be an Excell sheet with all this stuff for the differen Platforms plus the easy calculating stuff....

@jmr: you integrated it in Excell? Or just the plain formulas? I still need the Formulas for "full TMA" (Speed & Course) in combination with Ekelund range....i have seen them in this forum a while ago but i cant find them right now. Would be nice if you have those.

Phil

[phil21]

I've done a few tests with the Ekelund calculator and the results are disappointing: First of all, it seems to work as it should, but you need very precise data for it. It seems that you need at least two decimal places for bearing rate and one decimal place for all the other data for this formula to make it work reliable.
This makes the use of this method unpractical, since you don't have a reliable Bearing rate indicator in DW (e.g. using Mobo helps alot, but its hardly acurate enough for two decimal places) nor the possibility to process these data in real time.

But theres also one other possibility: i've got the wrong formula for this. I dont think that this is the point but i keep looking. Maybe i find something more "sturdy" formula that can use more unprecise data to produce at least a limited solution...

Phil

[jmr]

I had to dig around in the 2002 Subsim forum archive and found this info regarding Ekelund range formula:

Here's the full thread link http://www.subsim.com/phpBB_2002/vie...c=4518&forum=2

Brief summary:

Single leg

xDMHr
------- all that times K.
DBy inst


Double leg

xDMHo 1 xDMHo 2
-------- +/- -------- all that times K.
DBy 1 DBy 2

K = 1.934 (2 is good enough for government work)

xDMHo = ownship speed across LOS
DBy = bearing rate
xDMHr = relative speed across the LOS

TopTorp '92 gives an example:

Given:
Ownships course: 075
Speed: 8
True Target Bearing: 120
Bearing rate (Dby): 1.5 degrees per minute

Compute:
Relative bearing = 120-075=045
xDmho

sin(045)X8kts = 5.7kts across the LOS.
Range = 5.7/1.5 = 3.8KYDS

Initial detection ranges yield bearing rates frequently less than one, like .5. So, when OS changes course across the LOS and yields a change in bearing rate less than one, that mean the range to target is likely beyond 10KYDS.

Although not a part of the Ekelund formula, you can compute ownship speed IN the LOS and estimate a rate of closure, normally expressed in yards per minute.

For this example, that means you take the cosine of the angle and multiply that by OS speed. In this case the sine and cosine of a 45 degree angle is the same. So, OS speed in the LOS - expressed as yDmho using Treeders notation - is computed as 5.7kts.

If you convert the NM (nautical miles) to yards (there are 2000 yards per NM) and the time from hours to minutes, you get a closure rate of 190 yards per minute.

When I was a Fire Control Technician, our skipper didn't feel comfortable tracking another sub with closure rates above 300 yards per minute. Some skippers and XO's design a trip wire. That is, the FT is responsible for making a verbal report to the OOD (that's Officer of the Deck for you non-bubble heads) to that fact. In such a case, a reasessment was in order.

I hope I havent thoroughly confused the community with this thread. But the delicacies of TMA are indeed delicate as well as intricate. But if you are really motivated in TMA & Sub Command, you can perfect these techniques and be sure to stay outside of counterdetection range. Of course, it's all fun and games on the computer, but this is the way sub business is done.

[jmr]

There's also this thread:

http://www.subsim.com/radioroom/show...=ekelund+range

Originally Posted by compressioncut

single leg Ekelund

Range in nautical miles = (RSA x 0.955)/B*

where RSA is the resultant speed across the line of sound, and B* is the bearing rate.

Since you need to have a very good idea as to what the target's speed and course are to find RSA, the single leg Ekelund is pretty much useless, except as a training tool to confirm your understanding of the LOS relationships.

In other words even though it's pretty simple, I wouldn't really worry about it.

A double leg Ekelund is more useful.

Range in nautical miles = (change in OSA x 0.955)/change in B*

OSA being ownship speed across the line of sound, which is sin(OAA) x OS. OAA is ownship aspect angle, which is own course +/- the line of sound (i.e. target bearing).

The caveats with that ranging method are that each leg can't exceed 20 minutes, nor can the bearing rate change more than 15*.

Also, if one of your change in B* ends up at exactly 1*/min, it will not work.



That image has a LOS diagram, which should help a lot.

I'm not sure if the frequency rate Ekelund would work in game, it's been a while since I took a look at the Doppler effects. I think it would if you log the bearings and frequencies carefully, although that would be very tedious.

[PeriscopeDepth]

I wonder what ever happened to compressioncut, a very helpful fella.

PD

[SeaQueen]

Originally Posted by phil21

I've done a few tests with the Ekelund calculator and the results are disappointing: First of all, it seems to work as it should, but you need very precise data for it. It seems that you need at least two decimal places for bearing rate and one decimal place for all the other data for this formula to make it work reliable.
This makes the use of this method unpractical, since you don't have a reliable Bearing rate indicator in DW (e.g. using Mobo helps alot, but its hardly acurate enough for two decimal places) nor the possibility to process these data in real time.

I've had pretty good luck with using Ekelund ranging, particularly with time correction. In order to compensate for the relatively imprecise bearing measurements, you can just take longer legs. I usually use 5-10 minute legs. The longer your legs, the smaller the error will be provided the target doesn't maneuver to throw off your solution.

[PeriscopeDepth]

Originally Posted by SeaQueen

I've had pretty good luck with using Ekelund ranging, particularly with time correction. In order to compensate for the relatively imprecise bearing measurements, you can just take longer legs. I usually use 5-10 minute legs. The longer your legs, the smaller the error will be provided the target doesn't maneuver to throw off your solution.

That's a very good idea. I guess I'll have to put Ekelund back into my programmable calculator (god, I'm such a nerd).

PD

[phil21]

@jmr: Thanks for the formulas, i work them over tomorrow. Especially your first quote is interesting since it seems to be different from the formulas i found. The second one is exactly the algorithm i use for my calculator. But theirs still the Problem with the bearing rates...maybe with a longer leg like Sea Queen suggested it might work.

@Sea Queen: My idea of using Ekelund was to minimize the time you need to get a good range for your target, so using a long leg would be against that. However, since you need atleast one course change this point is no longer of concerne (coursechanges with TA take a while...). Second point is since htere is no way to determin Bearing rate in DW you need e.g. Mobo to do this which also takes time...But maybe i should make it work first and then think about how to use it :hmm:

Phil

[Ramius]

Originally Posted by PeriscopeDepth

That's a very good idea. I guess I'll have to put Ekelund back into my programmable calculator (god, I'm such a nerd).

PD

Why bother, just do it in your head

[SeaQueen]

Originally Posted by phil21

@Sea Queen: My idea of using Ekelund was to minimize the time you need to get a good range for your target, so using a long leg would be against that. However, since you need atleast one course change this point is no longer of concerne (coursechanges with TA take a while...). Second point is since htere is no way to determin Bearing rate in DW you need e.g. Mobo to do this which also takes time...But maybe i should make it work first and then think about how to use it :hmm:

How much time do you think you need? In a realistically scaled scenario, a few 5-10 minutes isn't a lot of time.

Also, if your goal is to minimize the amount of time needed to get an accurate range then the towed array is not the best sensor for Ekelund ranging. One of the assumptions of Ekelund ranging is that the turn you make is instantaneous. Of course, no turn is truely instanteous. In most cases, though, it's a sufficient approximation so long as the time it takes to change direction is small relative to the length of the legs. The towed array takes a long time to finish it's turn so reasonably accurate Ekelund ranging necessitates very long legs. I prefer to use Ekelund ranging for the sphere, the wide aperture array, and active intercept.

Regarding measuring bearing rate, there's several ways to do it easily. If you have broadband on a target, you can read it directly off the screen. The average bearing rate is equal to the sonar trace. Active intercept also can give you a bearing rate if you pay attention, since it has the clock and keeps track of the interval between pings.

Personally, I prefer to just use a piece of paper and a calculator to get bearing rate.

[Bellman]

Before you waltz, consider anchor ranges:-
http://www.subsim.com/radioroom/showthread.php?t=103810

The Navy recognises the need for rapid responses:-
http://www.patentstorm.us/patents/7020046.html