sonar readings are inconsistent

[I'm goin' down]

My sonar man reads differnet ranges each time I ask him for range and bearing information. Is he stupid? Is sonar inaccurate by nature? What is the story and solution, if any.

[CapnScurvy]

I know I've answered this before; when asked to give a sonar report of a target, the sonarman will only give you an "estimate" of range. Depending on his skill level will determine the "estimates" accuracy (which is never accurate, some just happen to be more accurate than others).

If you want an accurate sonar finding, you have to take the sonar reading yourself. Once the target is "pinged"; a return ping is heard; the reading is "Sent" to the TDC. The accurate range to target is displayed in the Range readout of the Position Keeper.

[I'm goin' down]

I do not think so. In my experience, when I ping the target several times, the reading are off by hundreds of yards, sometime by over one thousand yards. Someone else advised me to ping the target and wait for the return ping, but the results were as I already noted. I'll keep trying.

I cannot finish the mission in late '41 to drop the agent in the Bungo Strait. Tokkyo's Revenge with map contacts disabled. The dd and its companion, a small split freighter, lurk nearby and find my boat every time. You know what that means....!:wah:

[magic452]

IGD If you want a good sonar ping you must be sure the sonar is pointed at the center of the ship. Move the sonar left and right and get the center of when the light goes on. That's about as accurate as you're going to get in this game. It's pretty close if you do it right not so good if you don't, just like real.

Magic

[I'm goin' down]

Majic, I try it. I have not been able to get an accurate course using sonar readings. With map contacts disabled, lack of an accurate target course makes the attacks much more difficult.

[Rockin Robbins]

IGD If you want a good sonar ping you must be sure the sonar is pointed at the center of the ship. Move the sonar left and right and get the center of when the light goes on. That's about as accurate as you're going to get in this game. It's pretty close if you do it right not so good if you don't, just like real.

Magic

And interestingly, this center slop on the sonar bearing is less than the real sonar, which was even more uncertain in direction. Actually the width of the sound was dependent on sea conditions, something not modeled in SH4. We can be pretty sure (okay, POSITIVE) that the width of that audible sound is 10º. In real life it could be variable from day to day and position to position.

So we can carefully note the bearing where we begin to hear the sound on both sides, and surprisingly (irony...) note that the signal is 10º wide. Just add or subtract 5º from either of the extremes and you have your bearing.

Of course, our sonar man, regardless of ability always can tell the exact bearing of the target. Just set your sonar on that bearing and ping.

Actually, in real life, knowing the bearing of the target isn't all that you need to know to ping him, because you want the signal to hit the MOT (middle of target) when it gets there. So you have to anticipate where the target will be then, after the ping travels at 1100' per second (more or less depending on water density and temperature!). This is not modeled in the game.

Even using the sonar man's bearing, successive pings will come back with a variance in distance. It's best to take four or so and average them, as shown in WernerSobe's sonar targeting video. The active sonar ranging may give you a variance in range but it is plenty good enough to put the enemy on the bottom, either with WernerSobe's technique or the Dick O'Kane Sonar Only technique.

My feeling is that our sonar is too accurate compared to the real thing.

At the same time it lacks subtlety. No attempt was made to have different ships make different sounds, aside from merchant vs warship. The motor sounds do not change with throttle setting, there are only three sounds, slow, medium and fast for each type. Therefore screw counts are impossible, which is good because screw counts couldn't be used in WWII to deduce target speed. There is no variance in sonar conditions with temperature and density of water, and there is no background noise from the life in the water. We do not have the filters the real sonar men had to separate the sounds we are interested in from the sounds we are not. We can't vary the volume either. The only control we have is when to ping and in what direction. When we ping, we can get the range before the return ping is received! This is a very small subset of the function and environment of the real sonar systems.

[Diopos]

You're so obsessed with "range". I'd trade a good target course and speed estimate with laser precision range anytime. WHAT?!?. Actually I'm trying to improve my "Bearings-Only" techniques. They come really handy in the pre-radar phase of the war and in the "final approach" where you'd probably wouldn't be able to use the radar to confirm and refine the target's data. Eventually, if you're close enough, 0-gyro angle torp shots remove the necessity for extreme range accuracy (bit still you need a good course and even more a speed estimate). Active pinging is not included in my gaming style at the moment but I'll be sticking to "Contacts Off" permanently.
My (limited) experience with active sonar is that if the overall range to the target is "small" the readings are not bad. My feeling is that the game may have problems handling relative motion, Try multiple pings with the sub stationary and see what happens. Even better, try a "mission" where both target and sub are stationary at a known distance and see what's the active pinging result. Maybe even "calibrate" the thing at varius predetermined ranges and see what happens.That would be really useful to all of us.:yep:

.

[Rockin Robbins]

It does tend to be more accurate the closer you are. When you are at 5k yards you aren't looking for perfect accuracy anyway. You just want to work out a rough course and speed for intercept.

When I'm close enough to shoot the sonar, active and passive is plenty good enough to explode three of three torpedoes. If the range is off by 20 to 50 yards, who cares?

Think I'll take the sub out to my anchored target shooting range and do some pinging to see what I come up with. Should be interesting.

I don't use range to shoot usually, because I'm a fixed bearing guy most of the time. But you do have to have a decent range figure to do target course and speed.

[CapnScurvy]

Even better, try a "mission" where both target and sub are stationary at a known distance and see what's the active pinging result. Maybe even "calibrate" the thing at varius predetermined ranges and see what happens.That would be really useful to all of us.:yep:

Been done. The sonar returns exact range when you do it yourself. Too exact to suit me.

Like Magic452 said, the sonar target bearing must be right on the target. A degree or two off at 2000 plus yards will not return a ping. Two ways of knowing whether the sent ping hit the target or not. 1. You'll hear a return ping after one is sent (I like using headphones when at the sonar station) 2. Watch the range dial (on the left), if the ping hits the target the dial will stop at the estimated range. If the range dial makes a complete revolution and you don't hear a return ping, reset the directional bearing a degree or two and try again.

Asking the sonar man to give you a result for a target will provide an "estimate" that can be off quite a bit depending on the sonar mans skill level.

[I'm goin' down]

You're so obsessed with "range".

I am obsessed with the Three Graces!:D

[Diopos]

I am obsessed with the Three Graces!:D

You mean Aglaia, Euphrosyne, and Thalia or Cource, Speed and Luck ?

:D

.

[Arlo]

You mean Aglaia, Euphrosyne, and Thalia or Cource, Speed and Luck ?

:D

.

Grace Kelly, Grace Moore .... and Grace Jones?

http://mysite.verizon.net/res0qaye/sitebuildercontent/sitebuilderpictures/grace_kelly_white.jpg

http://farm3.static.flickr.com/2165/1924311550_6092b85103_o.jpg

http://1.bp.blogspot.com/_YA5UWBtYqsk/SnsZF154cNI/AAAAAAAAARk/9R4elGq7pbU/s400/Grace%2BJones%2B-%2BFame%2B-%2BBack.jpg

[I'm goin' down]

http://i553.photobucket.com/albums/jj369/wsm24/imagesCAMAI9DX.jpg

The Three Graces -- Muses with a sense of humor.

[Rockin Robbins]

Okay, let's load up the Shooting Gallery Test Range. Time for some fiddling around!
http://i196.photobucket.com/albums/aa293/RockinRobbins13/Silent%20Hunter%204/Sonar%20Study/SH4Img2011-10-07_183149_484.jpg
For those who don't know, the Shooting Gallery Test range is an anchored array of targets in two arcs, approx 3350 and 7800 yards range. Originally designed to test accuracy of torpedoes fired with TDC inputs of zero speed, bearing from periscope, it's been useful in lots of unforeseen uses and in fact has been one of my most used test setups.

http://i196.photobucket.com/albums/aa293/RockinRobbins13/Silent%20Hunter%204/Sonar%20Study/SH4Img2011-10-07_185029_718.jpg
Here's the overall view. Here I'm getting the true range to a target but you can't read the bearing because the screenshot is lousy. Welcome to the range. Today we will be measuring distances between a stationary submarine and stationary targets. Then we will submerge to periscope depth and compare the ranges. We'll ping four or five times per target to get variances in sonar readings from successive pings.

http://i196.photobucket.com/albums/aa293/RockinRobbins13/Silent%20Hunter%204/Sonar%20Study/SH4Img2011-10-07_183457_796.jpg
True range to target bearing 000 is 3350 yards.

http://i196.photobucket.com/albums/aa293/RockinRobbins13/Silent%20Hunter%204/Sonar%20Study/SH4Img2011-10-07_183513_500.jpg
True range to the target bearing 310 is 3250 yards.

http://i196.photobucket.com/albums/aa293/RockinRobbins13/Silent%20Hunter%204/Sonar%20Study/SH4Img2011-10-07_184959_343.jpg
True range of target bearing 305 is 7750 yards.

http://i196.photobucket.com/albums/aa293/RockinRobbins13/Silent%20Hunter%204/Sonar%20Study/SH4Img2011-10-07_184836_281.jpg
True range of the target bearing 035 is 7600 yards.

Take her down! Let's get some sonar readings and put this puppy to rest. (blub, blub, blub)

http://i196.photobucket.com/albums/aa293/RockinRobbins13/Silent%20Hunter%204/Sonar%20Study/SH4Img2011-10-07_184503_234.jpg
Pinging target bearing 000 four times. Result is 3350, 3350, 3350, 3350! No variance at all. That oughta shut up that loudmouth Rockin Robbins who stated based on the WernerSobe videos that there would be variance. No dice on that speculation! This target also returned the exact same distance at any bearing which returned a ping. 359, 360, 000 and 001 all returned 3350 yards. Targets in the front arc return an echo with a +-1º error. 2º off and you get no echo.

Yes our sonar thinks 000 and 360 are a degree apart, or maybe that they are the same but two different bearings. Maybe some kind of alternate universe in involved. Gives me a headache...

http://i196.photobucket.com/albums/aa293/RockinRobbins13/Silent%20Hunter%204/Sonar%20Study/SH4Img2011-10-07_183747_703.jpg
Pinging target bearing 310 consistently returns 3270 yards. Same variance with pointing as target 000.

http://i196.photobucket.com/albums/aa293/RockinRobbins13/Silent%20Hunter%204/Sonar%20Study/SH4Img2011-10-07_183927_265.jpg
Pinging target bearing 305 yields a consistent range of 7770 yards. At this range you must be on the perfect bearing to get a return echo. Nothing comes back 1º off.

http://i196.photobucket.com/albums/aa293/RockinRobbins13/Silent%20Hunter%204/Sonar%20Study/SH4Img2011-10-07_184033_828.jpg
Pinging target bearing 035 yields 7590 yards every time. We expect that now...

So lets compare numbers!

Target........True Range..........Sonar Range
000..............3350...................3350
035..............7600...................7590
305..............7750...................7770
310..............3250...................3270

Our sonar is perfectly consistent it its readings for successive pings. Straight ahead it seems to return exact distances. When pinging on an angle a slight error creeps in. The maximum error, at bearing 310 is only .6%, close enough to precise to be better than slide rule calculations. That's perfect for targeting purposes! These errors are so close that they could be measuring errors on my part from being a pixel off with the compass! We'll call sonar ranges perfect.

If you get a return echo from hitting a target anywhere other than MOT (middle of target) you still get an accurate reading. Targets are very narrow to the sonar and very little tolerance exists at ranges 3000 yards and greater. Your bearing can be trusted completely if you get an echo.

Sh4 with TMO/RSRDC. Shooting Gallery Test Range is available in the Subsim download area. That is all!:salute:

[I'm goin' down]

You should put that in the skipper's bag of tricks. It is a great tutorial.:yeah:

[CapnScurvy]

Yea, what I said. :O: (Well, nearly to what I said. A couple of yards off, here or there.)

I'm wondering Double R if you used the same target ship in all the positions? Just curious.

Also, I'm wondering if the difference in the exact range and sonar has something to do with the angle of the targets. Are they presenting an A-Beam position (AoB 90 degrees) in all positions, or are they randomly angled? When I've set up my target tests I usually have them angled to give me as large a silhouette as possible, just to be consistent.

I know the sonar has the modded feature of not picking up the rear 60 degree arc (between 150 and 210 degree positions). Just wondering if having the sub face East would make a difference in the targets found range at the 305 degree position when facing West (when facing East, the same target should be found at the 125 degree position)?

[Rockin Robbins]

Yea, what I said. :O: (Well, nearly to what I said. A couple of yards off, here or there.)

I'm wondering Double R if you used the same target ship in all the positions? Just curious.

Also, I'm wondering if the difference in the exact range and sonar has something to do with the angle of the targets. Are they presenting an A-Beam position (AoB 90 degrees) in all positions, or are they randomly angled? When I've set up my target tests I usually have them angled to give me as large a silhouette as possible, just to be consistent.

I know the sonar has the modded feature of not picking up the rear 60 degree arc (between 150 and 210 degree positions). Just wondering if having the sub face East would make a difference in the targets found range at the 305 degree position when facing West (when facing East, the same target should be found at the 125 degree position)?

You must have been typing while I was testing! It was good to do the tests anyway.

My initial reaction was that the angle of the target was related to the error. But you know how initial reactions can be. After looking at the screenshots closely, it looks to me that because I used the compass for the distance measurements, and did it at the magnification where I could see target and sub at the same time, a one pixel error in positioning the circle would easily account for the errors in all cases. I believe that the sonar is perfect and my measurements, because I did not zoom in to precisely position the circle, were the variance.

The targets are at different AoBs. The differential pinging on the closest targets, which were the only ones subtending more than the exact bearing to MOT, showed the same range for bow, MOT and stern, even though the target was not at AoB 90. Important detail and good catch!

I'll rotate the sub and check out active sonar distance with the same target, but I'm just about positive that repeating the original with more precision on the measurement will result in the conclusion that the sonar is always exact distance for MOT regardless of angle subtended by the target or AoB, depressing as that is.....

Didn't one of say that nobody had ever complained about the sonar?:salute:

[CapnScurvy]

I understand. The ruler isn't very precise giving only 50 yard increments.

I know you should get a true distance reading if you run the test in Auto Targeting mode. The targets true AoB should show in the Attack Data Tool Aob Dial. The true range should show on the Position Keeper. That's usually how I checked my true range & AoB before going to manual targeting for checking the Stadimeter or Omnimeter.

I'm thinking the 3000 yard targets will produce good results with using Auto Targeting, but I'm not sure the game will produce a range for you out in the 7000 yard distance. I don't think it will.

As far as the AoB. It makes sense that it doesn't really matter what the target position is, least not with the sonar. Although the radar does have a parameter that is set for determining a spike depending on what size of profile the target is in. The sonar does not.





Yeah, who cares about the sonar?! :88)