Nebelwerfer

[Tero]

Originally posted by Sir Uber General:

Did either side play around with faked arty sounds to throw off enemy CB?

You can use screens that deflect the sound to so that they appear to generate from a diffrent location.

[Thomm]

Originally posted by Andreas:

All this calculating stuff is really not very meaningful in this context unless you look at how the real life system looked like...

I respectfully disagree.

It is "all this calculating stuff" that makes the system work. You could describe the problem to a mathematician and he would tell you that you need at least 3 microphones to solve the problem without ever having heard about CB.

In fact the mathematician, therefore, will tell you how the real life system HAS to look like to actually work.

Regards,

Thomm

[Tero]

Originally posted by Andreas:

As for the number of mikes - three are really the minimum, based on empirical data (although I still have to have a look at the link Brian kindly provided). The whole system consisted of four-six for a battery, spaced well apart. All this calculating stuff is really not very meaningful in this context unless you look at how the real life system looked like...

What about flash ranging ? IIRC sound and flash ranging was done simultaneously whenever possible. If you have two mikes and one or two flash ranging posts I think there is no need for a third mike.

[Asok]

Originally posted by tero:

What about flash ranging ? IIRC sound and flash ranging was done simultaneously whenever possible. If you have two mikes and one or two flash ranging posts I think there is no need for a third mike.

Look at the figure on page 7 I drew. Flash ranging is the obvious way to determine the distance between the gun and a microphone.

Place an observer at one of the microphones. Have him check for flashes. The time between flash and the sound will give you the distance between the gun and the observer. Then all you need is one other microphone or observer.

That's the mathematical minimum, of course. In practice, you want as many data points as possible, since even one corrupted measure will otherwise throw you off.

I have no knowledge whatsoever regarding RL CB procedures. This is just simple applied mathematics, as Rollstoy already pointed out.

regards,

Asok

[Andreas]

Originally posted by Rollstoy:

In fact the mathematician, therefore, will tell you how the real life system HAS to look like to actually work.

Alternatively, the mathematician could just look at how the stuff was used, since there is empirical data, and would notice that two mikes were never used. Which may save you the trouble of calculating it, unless you want to know why. What I am saying is not that mathematics is useless (read my post again time for you, maybe), but that if you look at the historical context, you get some ideas of which bits of math you need to do. Hint - no need to start with a two-mike solution. Also, if you know that a system of 4-6 posts covered 7-10km, it gives you an idea of the distances between posts, further eliminating the need for some of the math.

Simultaneous sound and flash ranging. I am not sure about simultaneously. According to my grandfather, they used flash at night, and sound during the day, and shared the post. Other units may have done it differently, but with a finite number of men, and the need to get some sleep for what is very stressful work, with a need for high levels of concentration, I don't think it would have been possible all the time.

[Michael Dorosh]

Originally posted by tero:

Originally posted by Andreas:

As for the number of mikes - three are really the minimum, based on empirical data (although I still have to have a look at the link Brian kindly provided). The whole system consisted of four-six for a battery, spaced well apart. All this calculating stuff is really not very meaningful in this context unless you look at how the real life system looked like...

What about flash ranging ? IIRC sound and flash ranging was done simultaneously whenever possible. If you have two mikes and one or two flash ranging posts I think there is no need for a third mike.

This is the point I raised and was apparently buried - I asked Andreas if they really only did flash spotting at night and sound ranging during the day - I had always thought the two were done simultaneously when at all possible.

[MajorBooBoo]

Originally posted by Rollstoy:

I respectfully disagree.

It is "all this calculating stuff" that makes the system work.

In fact the mathematician, therefore, will tell you how the real life system HAS to look like to actually work.

Regards,

Thomm

I agree with Thomm.

Its also that an Engineer can tell how you how it might NOT work due to real life glitches/foibles/human errors/etc. A technician can also show you a sneaky trick he discovered because he doesnt want to crank numbers all day. This is how things work in the real world.

We take for granted the technology that we have today. Digital electronics, memory and sensor technology make everything inexpensive and easy. Leaving stuff on all the time is no big deal cause it uses hardly any power and doesnt affect the performance of the technology.

This wasnt the case for 1940s technology or especially WWI technology.

Nowadays, error reduction can make people assume that the technology acts very close to the ideal physical principles. A simple example is the gyrostabilized gun technology. People read how it was supposed to work, keeping the gun level and on target, and start thinking/modeling it as such is easy. In reality, it would be very hard to pull this off and KEEP IT WORKING under real conditions. Nowadays, its almost trivial.

So I think this thread is enlightening because it shows that there can be intelligent discussion here, people can learn and appreciate the mechanisms behind things, theres different levels to discussing things, and that Iron Chef is a piece of work.

BTW. Asok. You never responded to my answer/questions regarding your diagram. Can you , perhaps, go back and think about some issues I brought up?

[ February 14, 2002, 10:16 AM: Message edited by: MajorBooBoo ]

[mensch]

All this math grog stuff makes my head hurt. I like Nebelwerfers cuz the make cool *Foosh Foosh* sounds and then land totally randomly on the map hitting mainly your own troops

[Thomm]

Originally posted by Andreas:

Alternatively, the mathematician could just look at how the stuff was used, since there is empirical data, and would notice that two mikes were never used. Which may save you the trouble of calculating it, unless you want to know why. What I am saying is not that mathematics is useless (read my post again time for you, maybe), but that if you look at the historical context, you get some ideas of which bits of math you need to do. Hint - no need to start with a two-mike solution. Also, if you know that a system of 4-6 posts covered 7-10km, it gives you an idea of the distances between posts, further eliminating the need for some of the math.

I will not comment on these statements (some of which I consider bizarre, some of which I consider insulting), because it will certainly lead nowhere.

But I cannot help but wonder why you try to devaluate posts which SUPPORT your statements with mathematical proof with an "I told you so" attitude!

Regards,

Thomm

[Michael Dorosh]

Originally posted by MajorBooBoo:

So I think this thread is enlightening because it shows that there can be intelligent discussion here,

But only if you personally are directing the conversation in a sufficiently arcane topic of your choosing in which you can come off as the master to our pupils.

Piece of work indeed.

[Asok]

Originally posted by MajorBooBoo:

BTW. Asok. You never responded to my answer/questions regarding your diagram. Can you , perhaps, go back and think about some issues I brought up?

Originally posted by MajorBooBoo:

How many solutions are there?

In this ideal case, there are two solutions.

No. There is an infinite number of solutions:

The solutions include the "mirror" (X4,X5) mentioned by MBB, but they also include every other point on the curve (could be two lines, I'm too tired to think straight ). They even include a point directly between the two microhones (X3; 300 m from M1, 700 m from M2).

Originally posted by MajorBooBoo:

Could sound (just think of the circles emanating from the two seperate source points X1 and X2 in ASOk's diagram (thank you for the visual aid by the way) really have the same time differential? Does a further gun produce a shorter time differential or a greater one? does it depend on the layout? Is it possible the guns could be on some line of possible answers? A curve/spiral of possible answers?

There are no "issues" here. This is not hard. The time differential comes from the distance between the gun and the mikes. That's why X1 and X2 are both possible: X2 is just 400 m farther away from both microphones. If we can hear it, X6 (not shown) 20000 m from M1 and 20400 m from M2 is also one of the possible gun locations.

The "issues" of distance between and the layout of the guns and sound ranging equipment are, however, interesting because of RL limitations: sound can only reach so far and we very likely want to play with directions as well as distances (possible dirty trick of MBB's technician) These limitations still have no effect on the basic math.

regards,

Asok

[MajorBooBoo]

What is needed to know about Sound that makes sound ranging and direction critical?

Sound is a form of energy that travels through a medium. The medium is air in this case. Sound is made up of frequencys. frequencys travel through air and lose energy at different rates. The frequencys all can be said to travel at the same rate for our discussion.

What "sounds" does a gun make when it goes off?

The propellent detonating and the gun barrel ringing are the two major noise events. The propellent is easily understood as a quick burning explosive. The gun barrel is like a tuned bell that has been struck, the shell accelerating and twisting through it, will strike (ring) it out. The ringing can then be a signature event for that particular gun barrel diameter and length and muzzle break (if it has one). The explosion has a greater amplitude because it masks the instantaneous ringing. After the explosion, the ringing continues and can be discerned.

Explain how the series of events between firing of the enemy weapon and initiating, perception, recording and computation and relaying of data occurs in overview form.

The enemy weapon is fired at some unknown distance and direction from friendly lines. The firing event is from a flashless weapon in daytime and is also behind a small hill. It is not observed visually. The sound travels outward from the fired weapon like a constantly increasing half-sphere. If looking strait down from above, this would appear as a constantly increasing circle who's radius rate is the speed of sound.

The sound arrives at a forward LP (listening post). A person there is situated close enough to the front lines and far enough in front of the recieving stations (mics) that he can percieve the sound arriving over his position, he then signals (radio or some other communications method) the CP (command post that can turn on all the MICs), so that the mics are "online" when the sound wave will start to come over them and the recorder(s) are also turned on.

The MICs then are transducers that respond to sound energy and produce an electrical signal from it. This signal (it could be amplified/filtered at this stage) is sent down a wire to the CP which has data recorders. The data recorders respond to the signals and scribe a measure of the intensity of the signal and the time event at which it occurs on paper that is moving past the pen.

Computation can be made from a comparison of when the events are recorded. The Goal is the enemy gun location within an acceptable radius of error. This information is then relayed to a firing battery so that it can take the enemy gun under fire with the intent of destroying/neutralizing or forcing him to move.

Why not leave this system on all the time?

The need for "early warning" comes from the fact that the recorders are probably based on a paper/ink/deflected pen type technology (maybe till 1944). The paper is on rolls that are accelerated to a high speed (since the time precision is critical) and these cant be left on all the time. The take up reel would just fill up too soon or some other intefering mechanical problem/formality would erupt. The ink in the pen would run out from drawing a strait line. The electronics of the day, tube based, were susceptible to burning out (like cheap light bulbs) and drifting from calibration.

What problem could there be at the LP?

The LP might be human based. If the LP has a battle raging in front of him and the distant battery is behind this battle, he would be inhibited in his ability to discern the guns firing behind this "wall of sound". Likewise, any noise to his sides or above, a friendly battery firing, a column of vehicles passing, flights of planes overhead, etc. would make his job harder. The LP then, has to be far enough back that these events are minimized. Likewise, this pushes the MIC line back. And in turn, this limits the range of the overall system.

How could this have been solved/improved?

The LP would need its own frequency sensitive alert system. A MIC at his post that could aid him in the presence of the higher frequency ringing that the guns give off (it could beep lets say in harmony). He could then "see" through the rumble of battle to know that there was some guns firing somewhere that fit the frequency of an enemy battery.

What is the limitation of this?

Explosions are actually lower frequency and low frequencys travel better than higher frequencys (ringing). They can be heard at longer ranges. They arent attenuated by ground effects as much. Higher frequencys would drop off quicker. The LP then would have to be closer to the front with this freq sensitive alert system.

Is there a need to make all the MIC cables back to the CP equal length?

The sound, once converted to an electrical signal, travels at 3/4 the speed of light or so and any error this introduces is negligible.

What are the time lags between the enemy firing and the counter fire?

The sound has to travel to the friendly mics which is range dependant. The recorders have to be stopped once there is enough data, the paper taken off and information read off the ink marks. The computations have to be made (this is still a topic of interesting discussion and wont be layed out here) and the information radio/communicated to a counter battery. It could vary greatly depending on the skills of the people involved. 15 minutes has been cited in the German case.

Heres some more in depth information...

http://www.sfu.ca/sonic-studio/handbook/Sound_Propagation.html

[ February 15, 2002, 12:02 AM: Message edited by: MajorBooBoo ]

[MajorBooBoo]

Originally posted by Michael Dorosh:

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

So I think this thread is enlightening because it shows that there can be intelligent discussion here,

But only if you personally are directing the conversation in a sufficiently arcane topic of your choosing in which you can come off as the master to our pupils.

</font>

[MajorBooBoo]

No. There is an infinite number of solutions:

If there WERE an infinite number of solutions, wouldnt we need an infinite number of sensors?

All kidding aside, you really are not correct. Anyone agree with him?

The fact is, that for a sound wave to pass over two points in a X-Y plane, and there IS a time differential measured between the passing of the wave over those two points, THEN there are only two solutions IN THAT PLANE that can be resolved for that circumstance. If we extend to X-Y-Z volume space, then a circle of points perpendicular to the plane, that includes the two resolved previous points in the X-Y plane, are the solution. Thats it.

Take the problem to any real math/physics teacher and he/she should tell you it is so (in a ideal geometric world). Are you a teacher Asok?

I will give this another day or so. It is getting to page 9 and some people are intent on derailing it. It IS very interesting. The material discussed and the behaviour exhibited.

[Redwolf]

Originally posted by MajorBooBoo:

[QB] </font><blockquote>quote:</font><hr />No. There is an infinite number of solutions:

All kidding aside, you really are not correct. Anyone agree with him?

The fact is, that for a sound wave to pass over two points in a X-Y plane, and there IS a time differential measured between the passing of the wave over those two points, THEN there are only two solutions IN THAT PLANE that can be resolved for that circumstance.

</font>

[MajorBooBoo]

Indeed. I really only give you the benefit of the doubt, your tone is not optimal

Hows my timbre?

[JonS]

Originally posted by MajorBooBoo:

Hows my timbre?

Wooden.

I agree with redwolf (and ?asok?), the two sensing point solution is a line. And even a short line contains infinite discrete points and therefore infinite solutions.

I also agree with Andreas - the pure maths on display here, while interesting, isn't really IMHO that illuminating. The guys who first thought this out presumably had a few clues between them and figured out the necesary equations. From my POV, the more interesing bit is the application of the theory, a'la Brians link, Cheers Bri. Still, all this delving into equations does have the side benefir of allowing us to watch Booboo make more booboos

Regards

JonS

[edward_n_kelly]

As a matter of interest, sound-ranging was developed by the father/son combination of William Alfred and William Lawrence Bragg.

While the concept of sound-ranging was known, the Braggs brought to a practical solution.

(William A was also instrumental in the development of ASDIC)

It was not all back-room stuff. Lawrence Bragg was awarded the MC in 1918.

The Braggs are unique in that they are the only father-son combination to be awarded a Nobel Prize (Physics 1915) and Lawrence is the yougest ever to be awrded the prize (and the one of the few to have ever delivered a Golden Jubilee (of his award) oration at the Nobel Institute).

Both "did time" in Australia; the father as Professor of Mathematics at Adelaide University and the son was born and studied there until moving to Cambridge with the family in 1909.

Edward

[MajorBooBoo]

Is that everyone? Good to see JonS committing his chips into the pot. I think that Thomm might be smart enough to actually come up with another way. He found the only technically correct solution that was demonstrated, but did not really provide an answer if he thought the math was doable. I wonder if he agrees with asok? can the math be done in a reasonable amount of time Thomm?

What happened to Brian? Last I read, he was supplying wrong information regarding Nebelwerfers placement of the warhead or some such. Has he posted in the last 4 pages?

I will post friday. I will give some clues now.

Ask yourself:

1. what logical information could they start with? would they have a reasonable idea of where the enemy was? the range limitation of their own equipment?

2. having Thomm's equations what would you do? they are the right equations and he should be commended. but theres just one more step...

3. can asok possibly be right? does it CHECK? try checking it, if it doesnt check then he isnt right. Christ, start with a known triangle like a 3-4-5 and that might help you. make all the distances and times known. It could give you aa clue.

I always believe that theres a silent majority of people 2-3 times the number of posters in the current thread that follow the thread. If they would, please speak up and vote one way or another. Throw out some ideas also if you like.

I was kind of hoping Charles would throw his chips in too!!!!

[JonS]

Originally posted by MajorBooBoo:

... I always believe that theres a silent majority ... If they would, please speak up and vote one way or another ...

Vote? What on earth has voting got to do with it? Maths isn't a democracy.

[MajorBooBoo]

Originally posted by Asok:

regards,

Asok

Why are circles emanating from the receptors? Is this an active sensing scheme you are proposing?

Sorry but you are a pied piper and have led the flock astray (ummmmmm shheeeeeeeeep).

Think of a dartboard (a snooker board?) and the noise at the bullseye. Drink alot of Pints. The mics are on the outer edges of the rings. Think alot.

[JonS]

One point sensor tells you a gun has fired

Two point sensors tells you the gun is located somewhere along one of two lines (one in enemy territory, one in your own)

Three point sensors tells you the gun is located at one of two point locations (one in enemy territory, one in your own)

Four point sensors gives you a unique target location.

Edit: Sometimes three could be enough for a unique location, depending on the geometry. Sometimes four won't give you a unique location, again depending on geometry.

Edit 2: there is nothing wrong with Asoks diagram, its merely another way of looking at the same problem. Consider it to be two known noise sources and a single unknown detector - in terms of working out the problem it makes no difference.

Edit 3: blimmin typos

[ February 14, 2002, 09:52 PM: Message edited by: JonS ]

[Redwolf]

Originally posted by JonS:

Two point sensors tells you the gun is located somewhere along one of two lines (one in enemy territory, one in your own)

I think it is rather one line leading from your to enemy territory.

Is anyone else besides Majordingsda interested in the actual formulas, effect of mismeasurement and how fast they can be calculated with WW2 pocket calculators?

[karch]

You can not do it with 2 microphones. You would be able to get a possible line, but not a specific point. Think about this logically.

You don't know how far away the gun was fired. If the time differential was 1 one second, that tells you that the distance of the firing gun is 1 second further away from the gun. The speed of sound at sea level is 1115.7 (1100 for math purposes) feet per second.

If the gun was one foot away from microphone A and 1100 feet from microphone B you would get a 1 second delay.

If the gun were 1100 feet from microphone A and 2200 feet from microphone B you would get a 1 second delay.

If the gun were 11,000 feet from microphone A and 12100 from microphone B (a 1,100 ft difference) you would have exactly a 1 second delay/differential.

Multiply do the math as much as you want, but there are an infinite number of points. I just graphed the formula X=A+1100 and it is a straight line offset away from A.

You NEED a third microphone at a bare minimum. By using the same concept as the one above, you can see that the time delay from microphone B to microphone C will ALSO be a straight line. It WILL intersect the line between line A and B, and that is where the gun is.

You would use as many microphones as possible to get a better average intersection of the points.

I can't show you the math, but by using simple logic and a graph, you can see that 2 microphones will get you a line, and 3 microphones can get you 2 intersecting lines that pinpoint the gun.

Unfortunatly my simple logic is often simply wrong. Anyone agree or dissagree with the above hypothesis?

Scott

[karch]

Originally posted by redwolf:

Is anyone else besides Majordingsda interested in the actual formulas, effect of mismeasurement and how fast they can be calculated with WW2 pocket calculators?[/QB]

If my previous post is correct, all you would need to know is the speed of sound at a given altitude and temperature, that would be an easy table to carry around. Then by measuring the time differential of the different microphones, you multiply the differential (1 second) by the speed of sound, that gives you a distance differential, and draw your lines.

M

_ X S = M

S

(distance/time times time = distance)

I've always wondered how they did this. I wonder how close my layman's idea is.