Nebelwerfer

[JonS]

Originally posted by MajorBooBoo:

http://www.carleton.ca/Capital_News/05021999/f2.htm

Uses three mikes for the horizontal plane, and two(+) for the vertical plane.

[JonS]

Originally posted by MajorBooBoo:

http://www.tno.nl/instit/fel/museum/en/earlyeq.html

I gotta get me to the Museum Waalsdorp..

Seems they are finding elevation here from two mikes (read the paragraph thouroughly please and dont be misled because theres four mics, two are used for elevation, two are used for angle).

They got a horzontal bearing using the special case of two detectors hearing the sound at the some time. Same for elevation. There is no indication that they could determine distance with this equipment.

Neither this nor the canadian sniper detector supports your argument that you only need two detectors to determine distance.

[MajorBooBoo]

Originally posted by karch:

Answer these questions given the speed of sound at 340M/S and the 2 listening posts are 340 meters apart.

What will the sound delay be at listening post A if the gun fires...

1) 1 foot from mike B

2) 680 meters from A and 340 Meters from B.

If you are 680 meters from A and 340 meters from B, and they are seperated by 340 meters, you are all in line? yes or no?

I am not answering anymore questions, till we can at LEAST agree on this.

[Michael Dorosh]

JonS, as you no doubt already know - don't believe everything you read about defence projects, especially admittedly overbudget ones that haven't been proven yet. I can't believe the Canadian government would spend 2 million dollars on this project when there is a deficiency in basic items of kit for our overseas troops, and an army-wide lack of reliable transport.

Protecting isolated peacekeeping posts is admirable, but I wonder - what good is a sound detector for snipers? Once the shot has left the gun, the sniper is going to hotfoot it away, and the damage has been done - a good sniper won't fire unless sure of a kill, making a second shot unnecessary.

And even then, the rules of engagement usually don't allow for retaliation. So even if you could pinpoint the sniper's position, without having a positive ID on what nationality the guy was, what good is the information? Even if he was still there?

Too silly for words.

And we all know that defence contractors never lie, right, so this little marvel obviously will work 100 percent of the time. yeah, right. Just like the Patriot missile scam during the Gulf War.

The article also admits that the set is only really good up to 500 metres (didn't Hathcock routinely take longer shots than that?) and there is no indication how effectively it is able to block out other noises, etc.

[ February 16, 2002, 03:21 PM: Message edited by: Michael Dorosh ]

[JonS]

Originally posted by MajorBooBoo:

I am not answering anymore questions, till we can at LEAST agree on this.

Huzzah! Nobody comment on this, and that's the last we'll hear from Booboo!

[Caesar]

A simple question to clear up the 2, 3 or 4 sensor debate - Where's my gun. Listening post A is 700m from listening post B. Assuming that sound travels at 350m/s and that Listening post A is coordinate 0,0 and B is 700,0 then what are the coordinates of my guns if B recieves the sound signal 0.371668s after A. Sorry about the numbers but I just chose a position and calculated the lengths to the listening posts using pythagorus and have left it as accurate as possible (WWII technology measuring to a millionth of a second accuracy - Hmmmm). If Major Boo Boo can calculate it correctly then he is right, if he doesn't/can't then his detractors are right.

So - where are my guns???

Now to my real question. I have read here lots of talk about directional mikes being used. How accurate (in terms of direction) were they. Did they ever just use straight triangulation. The maths involved to time differentials is a real sod, while I was 14 when taught the maths for triangulation. I also have some doubts about the accuracy using time diffentials. The listening posts would have to be a very long way apart to achieve big enough time differences. A relatively slight change in distance forwards or backwards (along the y axis if the x axis runs through the listening posts) makes very little difference to the time difference. Does anybody know how exactly they got the time difference so accurately. If you add or subtract 0.1 seconds to the above example, it makes quite a difference.

[JonS]

Originally posted by Caesar:

... Does anybody know how exactly they got the time difference so accurately ...

With the equipment available in WWII the RA managed to acheive an accuracy of 25m. Considering that a bty posn could be spread out over 200m x 200m, that's pretty reasonable.

[Redwolf]

Guys, two sensors for pure sound contact form a line for the origin of the sound known as a Hyperbel (named so in German anyway, I hope the English term is the same), which is a line with a bend.

Three sensors give you two hyperbels which will cross at two points. One of the crosspoints is the battery battery position, you will usually be able to tell since with a somewhat straight sensor position you will have the second crosspoint behind your own lines and infiltration with artillery is uncommon

The actual formulars of crossing hyperbels are more complicated than I would like to solve. I can give the raw formulars to anyone interested, but you would have to resolve them to the X and the Y position of the guns to make them useful.

[Caesar]

25m accuracy - darn that's good. Do any of the grogs here know where there is a web site that explains exactly how they did the measurements that accurately? I guess that is why they used 4 sensors - to get extra accuracy.

Redwolf - Hyperbola, yeah that's what I came up with too. I steered clear of the maths though and just plugged a whole range of numbers into excel. Going by the graph, I figure their distances apart must been quite large relative to the distance to the guns to achieve the 25m accuracy.

Major Boo Boo - Found my guns yet?

[Thomm]

Originally posted by redwolf:

Guys, two sensors for pure sound contact form a line for the origin of the sound known as a Hyperbel (named so in German anyway, I hope the English term is the same), which is a line with a bend.

My God, redwolf, you are the man! Finally the *true* solution. I had an eerie feeling (because there was no abs() or if() function involved) about the two line segments, but did not bother to dive into it deeper. Now that you stated the real solution it seems almost ... trivial, and at the same time so much harder to find the actual solution. Too sad that basic stuff like the definition of a hyperbola is so easily forgotten!

Auf jeden Fall: Glueckwunsch!

Thomm

[MajorBooBoo]

http://iva.uni-ulm.de/physik/repetitorium/MATHEMATIK/7/07_05.html

y=a/x

I am not surprised that people havent responded to my simple question. Its clear that the greatness of Newton, Hamilton, Taylor and Leibniz has been washed out over the generations. Newton was a nasty individual on a personal level, so maybe that has been handed down.

I am quite prepared to respond to all these new findings but I fear that people here would rather bask in the fuzzy contentment of believing they are right. It will be a very harsh reflection on the people who have committed themselves here so maybe you should all be like mice and stay quiet?

I will use my well recieved Q&A style.

[Asok]

I am quite prepared to respond to all these new findings but I fear that people here would rather bask in the fuzzy contentment of believing they are right. It will be a very harsh reflection on the people who have committed themselves here so maybe you should all be like mice and stay quiet?

I will use my well recieved Q&A style.

Please do.

May I suggest that you answer this question first: What confusion will arise if I assume that all triangles are right triangles?

I believe your mistaken ideas about geometry are due to that simple error.

regards,

Asok

[Sledge59]

Hello there Boohoo. You seem to have your tongue firmly planted in the cheek. Here's a question for you: If a grognard farts in the forest will anyone care?

I'm a bit of a mathematical midget, but it would seem that 2 pickup points would give 2 possible points of origin. GPS apparently needs 3 points for an accurate result. The same is true for earthquake epicenter calculations. What gives?

[Cpl Carrot]

Originally posted by Sledge59:

I'm a bit of a mathematical midget, but it would seem that 2 pickup points would give 2 possible points of origin. GPS apparently needs 3 points for an accurate result. The same is true for earthquake epicenter calculations. What gives?

You need three points because the world is a sphere (well close enough). Two points gives you a point on a plane (a flat surface). I guess if you believe the world is flat then you will need only two.

[Tero]

Originally posted by Cpl Carrot:

</font><blockquote>quote:</font><hr />Originally posted by Sledge59:

I'm a bit of a mathematical midget, but it would seem that 2 pickup points would give 2 possible points of origin. GPS apparently needs 3 points for an accurate result. The same is true for earthquake epicenter calculations. What gives?

You need three points because the world is a sphere (well close enough). Two points gives you a point on a plane (a flat surface). I guess if you believe the world is flat then you will need only two. </font>

[Cpl Carrot]

Originally posted by tero:

[surface). Depends how well you have your cartography down. If you have the landscape properly mapped and coordinates in place then you really need only two observation posts to get a fix on the sound contact. Since all this relates to counterbattery procedures there is no point in gathering sound data if you do not have the necessary topographical data to deliver the counterbattery fire to the location you've just fixed.

I was generally talking about GPS. However even on the ground you need the third to tell you whether the sound was in front or behind (of course the sound should be front or their might be some upset arty boys )

[Caesar]

Come on BooBoo, answer the question and end the debate, where are my guns?

[Tero]

Originally posted by Cpl Carrot:

I was generally talking about GPS.

Which is basically a triangulation affair through and through at 300km's rather than a (basically) a simple excersise in trigonometry at sea level.

However even on the ground you need the third to tell you whether the sound was in front or behind (of course the sound should be front or their might be some upset arty boys )

Not to mention some rather PO'd senior officers.

That is the basic assumption. And that is why you can get the job done with two listening stations.

And I think all the nice graphs and calculations are a bit amiss because without flash ranging you can only pinpoint the sound at a 45º arc (human sense of hearing being what it is).

And this being the case you can not backtrack the start of the sound because you can not get a reference on the timeframe.

This is why the guys doing the actual ranging have to use simple trigonometry rather than applied maths as seen here. And for that you need only two stations.

Lets say station A gets an audio at 1200hrs 30secs from (say) direction 13-00 and station B at 12hrs 33secs from (say) direction 11-30. With a 3 sec difference (assuming the base line between the two stations is wide enough and the speed of the sound is constant and there are no impediments and no echo) a third station plots the reports on the map (the plane as we call it in the coastal arty ). As more stations report the audio in the better but you can get a fix on the sound source with just two LP's.

QED, I think.

[Thomm]

I do not understand your post, tero! To many points mixed up.

If you claim that two listening posts (without any direction finding) can locate a sound source purely by measuring a sound delay, then you are, unfortunately, wrong.

Re-reading your post I see that you mention direction finding! How is this supposed to work? Does it not contradict to your 45 degree perception statement?!

Regards,

Thomm

[ February 18, 2002, 12:00 PM: Message edited by: Rollstoy ]

[Redwolf]

MajorBooBoo, I really don't like your attitude.

Rollstoy, the math is actually moderately complicated, you need two formular conversions and they do not resolve to simple expressions as people always hope for. Since you have to insert the solution of one of the formulas into the other (to reduce one of them from two variables to one to get X), they become rather huge. Not overly complex, but huge. I'm not going to do this given that noone else besides BooBoo showed an interest.

[Michael Dorosh]

Originally posted by redwolf:

MajorBooBoo, I really don't like your attitude.

Rollstoy, the math is actually moderately complicated, you need two formular conversions and they do not resolve to simple expressions as people always hope for. Since you have to insert the solution of one of the formulas into the other (to reduce one of them from two variables to one to get X), they become rather huge. Not overly complex, but huge. I'm not going to do this given that noone else besides BooBoo showed an interest.

redwolf, I know a forum where actual Gunners go; if there is something you really want to know, I can try and get in touch with some of them. Email me with the question (I stopped having the foggiest three pages ago at least) and I can pass it along.

[Thomm]

Originally posted by redwolf:

Not overly complex, but huge ...

To give an impression, here is the solution after a transformation of the coordinate frame to achieve x1=0, y1=0, y2=0. There are two solutions, one of which is:

x = dr13**2*x2**3 + dr12**3*dr13*x3 - x2**3*y3**2 + dr12*dr13*x2*(dr13**2 - x2*x3 - x3**2 - y3**2) + Sqrt(dr12**2*(dr12**2 - x2**2)*y3**2*(dr12**2 - 2*dr12*dr13 + dr13**2 - x2**2 + 2*x2*x3 - x3**2 - y3**2)* (-dr13**2 + x3**2 + y3**2)) + dr12**2*(x3**3 - dr13**2*(x2 + x3) + x2*y3**2 + x3*y3**2))/ (2.*(dr13**2*x2**2 - 2*dr12*dr13*x2*x3 - x2**2*y3**2 + dr12**2*(x3**2 + y3**2))))

y = dr12**4*dr13*y3**2 - dr12**2*dr13*x2**2*y3**2 + dr13*x2*Sqrt(dr12**2*(dr12**2 - x2**2)*y3**2* (dr12**2 - 2*dr12*dr13 + dr13**2 - x2**2 + 2*x2*x3 - x3**2 - y3**2)* (-dr13**2 + x3**2 + y3**2)) + dr12**3*y3**2* (-dr13**2 - x2*x3 + x3**2 + y3**2) + dr12*(dr13**2*x2**2*y3**2 + x2**3*x3*y3**2 - x2**2*y3**2*(x3**2 + y3**2) - x3*Sqrt(dr12**2*(dr12**2 - x2**2)*y3**2* (dr12**2 - 2*dr12*dr13 + dr13**2 - x2**2 + 2*x2*x3 - x3**2 - y3**2)* (-dr13**2 + x3**2 + y3**2))))/ (2.*dr12*y3*(dr13**2*x2**2 - 2*dr12*dr13*x2*x3 - x2**2*y3**2 + dr12**2*(x3**2 + y3**2))))

The other solution is of similar complexity.

Regards,

Thomm

[Tero]

Originally posted by Rollstoy:

I do not understand your post, tero! To many points mixed up.

That is RL arty field procedures for you.

If you claim that two listening posts (without any direction finding) can locate a sound source purely by measuring a sound delay, then you are, unfortunately, wrong.

The 13-00 is the Finnish convention for expressing direction.

And as it happens I did try to point out you can not measure delays as such and alone, especially if there are multiple sound contacts being plotted. It is fairly easy to plot a single gun firing a single round but try to keep up with multiple batteries doing max ROF all over the horizon.

Re-reading your post I see that you mention direction finding!

Not direction finding as such. The directions are already known, the observer notes them when he does the observation. Station A: "Gunfire, multiple shots at 13-00 (I know, the direction given is too precisise but lets ride with it) 1200,30hrs". Station B: "Gunfire, multiple shots at 11-30, 1200,33hrs"

How is this supposed to work? Does it not contradict to your 45 degree perception statement?!

Perhaps the 45º arc is too wide. But given such things as echo, athmospherics etc that is about as wide I personally would allow for as being realistically accurate.

You can give only a general direction of the sound you hear if you can not actually see the source (a fact of life I think). The fine thingies displayed were for listening to approaching aircraft and to pinpoint their location. No such luxuries for the arty pukes at the front line.

Any FO (well, überFinnish anyway) had the landscape in front of him ID'd and directions pre-noted in the 00-00 convention for arty direction purposes. LP's and flash ranging stations had the same data. If you note a sound coming from a certain quarter at XX:XX,xx hrs you can do a simplish trigonometry calculation from two points, provided you are certain the reports are of the same sound.

[MajorBooBoo]

It would appear that you have painted yourself in a corner. Is it possible to locate an enemy gun using just two sensor mics that can record a time differential?

Yes, under real conditions. It has limitations just as three or more mics has limitations. It also has benefits that the 3 or more mic solutions doesn’t have.

What is the answer then? How can this be done?

As outlined previously, the system as used in WWII (maybe till 1944) needed an LP post. This is the human that initiates turn on the system. Due to technical limitations already explained, he must participate and can’t be ignored. Sorry if anyone considers this cheating but its real world and life’s a bitch. I gave enough clues and no discussion came of my previous Q&A threads where I spelled out that he was there. Just goes to show that reading AND math comprehension isnt what it used to be.

How does that change anything?

Humans have binaural hearing. That is the missing link in the puzzle. The human also knows where he is on the X-Y plane along with the mics. He is situated between the mics and forward. The triangle that is formed has a 1000 meter length and two other sides of equal length.

How does the hearing play into this? Please demonstrate?

The CB process plays out as follows:

Gun fires

LP hears and signals “turn on” first to CP

He then gets an angular estimate/fix and reports this to CP

CP computes raw time differential

Quick-aids, like clear acetate with black hyperbels that correspond to different time differentials and 1000 meter constant separation (or whatever the true separation is) of mics limitation is placed over map and an angle drawn from LP report. This quickly gives the intersection of the raw immediate data collected. The time to get to this stage could be in the range of a minute or two. This is the major advantage of this system. Quick retaliation and not prone to long calculating time and calculation errors.

What are some of the disadvantages?

Its range limited. The further the range, the greater the error. The accuracy of the LP report decreases with range. His report will have an angular center and angular error. This angular error then manifests itself as an arc length error along the hyperbola solution from the time differential. The further away, the longer the arc length. If it’s possible, the CB artillery should hammer along the length of the arc (the gun is along the arc with some precision so hammering from both end of the arc inwards ain’t such a bad idea). The obvious use of a technique like this is for counter mortar operations. The logic is the mortars are close and mobile. A quick retaliation is much better than a more precise late retaliation being the thinking. The obvious disadvantage is the scenario of the sound coming from a perpendicular angle to the layout of the two mics. The time differential is very low and the locus of points along the asymptotes giving too many solutions. The overall system is range limited no matter how many mics/LPs you use. I believe that a 6000-meter range limitation was thrown out in another post. Sound takes 17.5 seconds to go 6000 meters. In terrain like hedgerows, sound ranging would be highly questionable.

What would you say are some real world parameters? What would start happening after experience builds in the units?

The LPs would get good after awhile. They could start recognizing sounds and associating them with ranges too (from feedback from the CP) and the weapons that produced them. They could choose positions that would maximize the angle of most effectiveness. A guess would be that his angle of effectiveness is between 45-50 degrees at most. The whole two-mic/LP system then wants to point itself so that the “deaf” perpendicular zone is not pointed at a likely enemy weapon location line. As the LP can estimate better, he can increase the angle of effectiveness better. Since Arty/mortars are not just explosive events (which ARE hard to binaurally locate) they can be fixed to within an angle of perhaps 10-15 degrees and maybe better. The better matched a persons hearing is between his two ears, the better a receptor he is. Young guys would be chosen and tested for this job. I make no apologys here. the human is in the system and is not a one eyed, one legged, one earred fool.

What do you think of the Neanderthal like math demonstrated in this thread?

It’s not the math as much as the narrow mindedness/pissy attitudes. I don’t believe that it took as many pages as it did before anyone could express what they thought was the math behind this. Its laughable that someone would say “Oh, its trivial”, and then 4 days later he has the “answer” (but I liked that guys idea about the unspotted arty animations way back when , so I will not name names). Someone came up with a high school level solution but decided that was the end of it. I would like to expand a little on the math and describe a technique. This technique would have revealed the “answer” they all sought easily. I will describe it and if needed will post a follow up to clear up any confusion. I am doing this in case there are silent followers of this thread that could learn something. The vociferous elements in this thread are clearly beyond learning anything.

When you have X amount of equations and Y unknowns (Y=X-1), the attempt should be made to find an iterative solution.

Here is the data:

The knowns are:

Speed of Sound 343 m/sec

1372 meters between recievers.

2 second time lag from sound (dr2)

r2=r1+686 meters

exact cartesian coords of recievers

taken as:

"left" reciever will be expressed as origin (0,0)

"right" reciever will be expressed as (1372,0)

If you have two mikes (x1, y1) and (x2, y2) and a time difference dr12 then you have three equations

r1^2 = (x-x1)^2 + (y-y1)^2

r2^2 = (x-x2)^2 + (y-y2)^2

r2 = r1 + dr12

To be clear, we are solving for x and y (the enemy gun location). r1 is the distance between the source and the "left" receiver. dr12 is known and 686 meters. r2=r1 +686.

x1=0, y1=0

x2=1372, y2=0

substituting in these values yields:

A. r1^2=x^2+y^2

B. (r1+686)^2=(x-1372)^2+y^2

expanding and subtracting one equation from the other yields:

C. 1372*r1+686^2=-2744*x+1372^2

Here we have the heart of the iterative technique:

1. Guess an R

2. Calculate a X that fits this R from the equation above C. to get a Y value

3. Use that X and Y to solve both of the equations A. and B.

4. Compare the R’s obtained. They will not be the same and a midpoint value chosen between them, Use that as a second guess and go back to step 1. Congrats, you are iterating.

5. If the R converges to a single value, you have a solution. If the R diverges, then the equations cannot be solved via this method. This is a very powerful technique in real world applications. Experience also could have told you that since X and Y are squared, there are a range of solutions. I knew this but wanted to see what discussion/proof would come up.

Anyone with any education beyond a high school level should have done this way back when Thomm posted the high school solutions. Anyone beyond a BS degree makes me wonder about the education institutions nowadays.

Someone asked “why are you so persistent when “everyone” thinks you are wrong?”

The only thing I concentrated on (besides throwing people off) was the math and the post by Slapdragon about what the US guys were actually doing. I have been in enough technical situations to know that you should listen to the techs, the engineers and then the science eggheads. No one here bothered answering my question about how long would the math take for a 3-mike solution. They clearly didn’t want to do it or address it. But that’s typical of the threads around here. The UK-Euro/Mapleleaf/Aussie/NZ Mafia continue to be rabble-rousing barflys.

This same Einstein that said this also made a clueless remark about the very interesting momentum thread (with the tank engine being knocked out the back). I take back anything positive I may have said about this person. He clearly did not understand what I was doing in that thread either. Evidently, thinking irritates certain people around here. I will try to avoid making them think in threads but these same people are in every thread. Says something about them I guess.

What do you think of the Uber-Canadians Finn-like use of WWI CB techniques?

Please. That’s like saying that the first guy who mounted a MG in a WWI airplane and went up and shot down a dozen planes is an ace. Any new introduction of technology can catch an enemy unaware and wreak havoc (unless you are the Brits misapplying tanks in WWI and throwing that advantage away).

I bet the German guns were a lot closer and not as well dug in as WWII arty. They didn’t move as much because they usually did not have to and were shot up like ducks in a barrel. The Canadians took their time and located them all and devastated them. Good for the Canadians. Its what I would have done.

[karch]

tero,

You are absolutely right with your calculations based on your assumptions. You are assuming they CAN get a relatively close angle from both listening posts. I am assuming you CAN'T.

If you can get the direction of the sound accurately, then you do only need 2 listening posts. All my drawings are based on my assumption that you can't get an accurate enough bearing on where the sound is coming from to pinpoint the gun. I'm not disputing your way of "Finding the Gun" just guessing direction is much harder to measure so they would have gone for the 3 point approach without worring about direction.

Not to be a pain, but could we continue this thread on New Gun Ranging thread . This thread is getting long enough to be locked soon and it really doesn't have anything to do with Neblewerfers...