Ultrasonic Range & Penetration

[cyprio7]

hi ppl hows it goin.. i had a look on google and the forums but was havin trouble finding d info i was after...

basically, i want to transmit an ultrasonic signal, through air. i was wondering what the maximum range/power i can get on these things? obviously i dont want the device to be too big.. the smaller the better, as small as possible to be honest...but yes i was wondering on the greatest range i can get and what type of power consumption it would need for that range approximately. im hoping for max 100m range .... minimum.. 25m range... but a 50m range would be ideal... is this possible do u think and if u know of any device that can produce this thats online and u kno where i could find it could u let me know please... i would also like it to be omnidirectional if possible.

i was also wondering how penetrative this signal at high power can be, for example, if it was to be transmitted inside a greenhouse... could it get through the walls and be able to potentially reach 50m range?

iv been having a look around but i cant find specific info on this..

i dont need to kno exactly but if anyone knows if this is feasable or approx max range i can get an wot penetration/power consumption/cost wud b like for the ideal range of 50m etc i would appreciate it a lot.

thanks

 

[dknguyen]

Sonar

Air sonar cannot go that far. About 6meters max in calm air. To do this, the frequency required makes a very large beam width (about 30 degrees deviation from the center line)

Devantech sells some for about $50. Its 5cm x 3cm PCB board.

Sonar also cannot go through solid objects at all (it's sound)- that defeats its purpose in detecting objects. The only exception to this is when a membrane is used to transmit the signal.

For your range- radar or laser and even laser might not be able to penetrate the greenhouse cleanly enough to not see it. The cost is at least $2500 and the only device I have seen at that price has a range of 4m. The next cheapest devices I have seen are at least double the price.

Probably not what you wanted to hear, sorry.

 

[philba]

if you are signalling, why not use RF? there are lots of pre-made units that aren't that expensive.

 

[cyprio7]

i cant use RF. this is because i want to triangulate the signal to find what direction its coming from, and because RF travels too fast, i cant build a system accurate enough to calculate the time difference delay etc of the recieved RF. ultrasound is a lot slower speed and it can be done, but i really really need range of 25m+

 

[dknguyen]

Signal

Oh you aren't trying to rangefind? Well, then sonar, radar, and laser aren't for you.

Indoors, you would use 3 beacons each would transmit a light and sound pulse at the same time and you could use the arrival time difference between the light and sound to calculate your position in the room...but outside...sonar won't work at all beyond 10m...its would require too much power (potentially damaging power) and there's just too much noise and power dissipation with ultrasonic frequencies.

The sound frequencies that would travel the ranges you want would probably be audio frequencies, and not necessarily high pitched frequencies. The farther the distance, the lower the frequency and the more difficult it is to pick out the sound from all the noise, and a high-pitched frequency of sufficient power would be very damaging to the ear. It would also possibly break some noise laws.

RF is about the only thing unforunately that can reach those ranges and is omnidirectional. What's wrong with using GPS on the remote station and then transmitting the GPS coordinates to a basestation? If it is outside GPS is probably king.

You can use a Yagi directional anntennae to find which direction the signal is strongest which usually translates to the direction the signal is coming from. You could triangulate it that way perhaps. It is limited by mechanical precision (which costs a lot) since you would need to rotate move the antennas which introduces a mechanical aspect to the problem, and it is not necessarily reliable.

 

[cyprio7]

im trying to find the direction and the distance away. cant i get a special speaker and just like... pump up d signal an really beef it up with loads of power, then isnt it possible to make it go far. and as for the yagi uda its not practical.. i need it to be small.... not a great big tv antena lol

 

[dknguyen]

Sound wont work

You can get 60 cm long Yagis...

But ultrasonics and sound, it just won't work at those distances. You will not find an ultrasonic sensor that can take that kind of power. Ultrasonic frequencies just dissipate too quickly. Audio frequencies would travel farther for a given amount of power but you would still need something the size of a concert speaker...or bigger. It is very easy to produce noises at audio frequencies, and that means it will be very easy to introduce noise between transmitter and receiver. The result is that the speaker has to be loud enough to be more than just heard at the receiving end- it has to be loud enough to dominate the noise at the receiving end. Which means it has to be exponentially louder with distance.

It would be like a rock band thumping outside your house the entire time and would be way worse than a car with the bass turned all the way up driving by.

Also, remember, wind will greatly reduce the accuracy of sonar, even when it does work.

Again, what is wrong with GPS? It is smaller, cheaper, easier, and more reliable with a much better range. It's only about $100.

 

[cyprio7]

... i dont want to by a big hand held global positioning system... i want very small.. ideally match box size...(that i can integrate into my own system) that can triangulate where a signal is coming from and its distance..... why cant i use RF to triangulate if i use a clock speed in like the Mhz or GHz range is it stil impossible....? is there any other methods/types of signal other than ultrasonic that would allow me to triangulate? how bout audio frequency.. but like at 22khz or something (or even possible very low audio frequency that we cant hear)..... like which is just outside of human hearing so that humans cant hear it..... but then u can blast it really loud and it wont bother ppl... do u think this is possible?

 

[Nigel Goodwin]

... i dont want to by a big hand held global positioning system... i want very small.. ideally match box size...(that i can integrate into my own system) that can triangulate where a signal is coming from and its distance..... why cant i use RF to triangulate if i use a clock speed in like the Mhz or GHz range is it stil impossible....? is there any other methods/types of signal other than ultrasonic that would allow me to triangulate? how bout audio frequency.. but like at 22khz or something (or even possible very low audio frequency that we cant hear)..... like which is just outside of human hearing so that humans cant hear it..... but then u can blast it really loud and it wont bother ppl... do u think this is possible?

Triangulation requires at least two receivers, seperated by a decent angle, and the accuracy depends on the accuracy they can be aligned.

What distance are you looking for?, and what degree of accuracy?.

 

[cyprio7]

the accuracy doesnt have to be spot on.... ideally id like the recievers to be as close together as possible so i can fit it into something handheld and small.. i dont want it to be bulky... recievers probably alligned to a few millimetres accuracy


and as for how accurate i want it to be and dimensions........

in terms of size of the area id say... a field about the length of a average/small sized football pitch, and its width is double the width of a average/small football pitch. given these rough dimensions, id like to know the *approximate* direction.. mayybe accurate to 45degres of where the transmitter is given the approximate size of the football pitch. ..altho of course greater accuracy will be much better but id say 45degres is minimum.

also, do u think what i said about using a high audio frequency that humans cant hear to triangulate from minimum 20m ish away is feasable? obviously greater distance the better.. but id say about 15-20metres is absolutely minimum. and like i said ideally id like the recievers to be as close as possible.. to give me the desired kind of accuracy as i described above. what ya recon? is it feasable?

thanks.

 

[Nigel Goodwin]

the accuracy doesnt have to be spot on.... ideally id like the recievers to be as close together as possible so i can fit it into something handheld and small.. i dont want it to be bulky... recievers probably alligned to a few millimetres accuracy

Do the maths, it's simple enough - for best accuracy the two receivers need to be 90 degrees apart - something handheld isn't anywhere near wide enough.

and as for how accurate i want it to be and dimensions........

in terms of size of the area id say... a field about the length of a average/small sized football pitch, and its width is double the width of a average/small football pitch. given these rough dimensions, id like to know the *approximate* direction.. mayybe accurate to 45degres of where the transmitter is given the approximate size of the football pitch. ..altho of course greater accuracy will be much better but id say 45degres is minimum.

I would have though 45 degrees is easily possible, with just one receiver, but without two widely spaced you've no idea as to the distance.

also, do u think what i said about using a high audio frequency that humans cant hear to triangulate from minimum 20m ish away is feasable? obviously greater distance the better.. but id say about 15-20metres is absolutely minimum. and like i said ideally id like the recievers to be as close as possible.. to give me the desired kind of accuracy as i described above. what ya recon? is it feasable?

Parabolic and rifle microphones are quite directional, you may be able to get a decent angle with those - but again, you need two readings widely spaced to give any idea of distance.

 

[cyprio7]

to be honest i dont mind too much about distance...getting the distance reading doesnt matter to me for now.... my main goal is to soley calculate the approximate direction to about 45degrees resolution..with as minimum spaceing as possible between the recievers... is it possible to do that with small spacing? the signal will cover 1mm in 3uS theoretically... which means 3MHz... but obviously the clock must be faster than 3Mhz for accuracy purposes.

ideally i wouldnt want the recievers to be spaced more than 10cm apart. so given that i only want to calculate approx direction... do u think this is possible with small distance? i think its theoretically possible, however, i dont have as much experience as you when it comes down to actually doing these things in practice and whether it is feasable in practice is a different story..

thanks

 

[Nigel Goodwin]

to be honest i dont mind too much about distance...getting the distance reading doesnt matter to me for now.... my main goal is to soley calculate the approximate direction to about 45degrees resolution..with as minimum spaceing as possible between the recievers... is it possible to do that with small spacing? the signal will cover 1mm in 3uS theoretically... which means 3MHz... but obviously the clock must be faster than 3Mhz for accuracy purposes.

ideally i wouldnt want the recievers to be spaced more than 10cm apart. so given that i only want to calculate approx direction... do u think this is possible with small distance? i think its theoretically possible, however, i dont have as much experience as you when it comes down to actually doing these things in practice and whether it is feasable in practice is a different story..

For triangulation the two receivers need to be much, much further apart - but if you're not concerned about the distance, and only require 45 degree accuracy, you don't need two receivers anyway. Have you considered using a low RF frequency? (ignoring licence complications), you can then use a ferrite rod aerial which gives a nice sharp null either end of the rod, and in a very compact unit.

I'm presuming you're at one side of the target area so know the rough direction?, as a ferrite rod gives nulls both ways - but should do a LOT better than 45 degrees. Assuming you have a portable MW radio, tune a local station and then rotate the radio for minimum signal - the null is quite pronounced.

 

[dknguyen]

RF

Stuff you asked/said:
------------------------
For the 22kHz sound emitter, I don't think it would work at 20m very well...especially not at distances much farther and especially not in the wind. I also doubt you can find a transmitter than can take that amount of power. You could try to find one, but it would still have a hard time overcoming noise introduced by wind and other things.

When I say GPS receiver I dont mean a handheld consumer version...there exist 3cm x 3cm GPS board out there that interface directly with a microcontroller. The accuracy, however, is 10m though, 5m if you have WAAG in the area. If you need higher accuracy and like the GPS method, then it basically becomes the RF triangulation method- the only difference is now instead of using the GPS satellites, you have your own transmitters. If you want to make the mobile unit less bulky, you would turn it into a transmitter and have multiple receiver "satellites". But you still have the timing synch problems.

Triangulation:
-------------------------------
GPS basically works this same way. For simplicity, let's just say that there are multiple transmitters (satellites) synchronized to each other and they send out a unique signal at regular intervals, all exactly in synch with every other satellite. The position of each transmitter is stationary and is therefore always known. The receiver measures the time it takes for each signal from each tranmistter to reach it and can calculate its distance from each transmitter. If you have 2 (or 3) or more transmitters, the distances will only intersect at one point and you can figure out where you are relative to the transmitters.

Alternatively, you can make a mobile transmitter and have multiple stationary receivers measure the time it takes for a signal to travel from the transmitter to each receiver, calculate the distance and figure out the position of the receiver from the intersection of the distances.

The problem is that all the transmitters and receivers have to be in synch with each other, and for a RF signal travelling the speed of light this time-synch has to be very very close. GPS satellites use atomic clocks for this purpose and this purpose alone. Its probably something you won't be able to easily and cheaply achieve and achieve very compactly at that.

Indoors, what is done is that an ultrasonic sound pulse and an IR light pulse are sent at the same time from the mobile transmitter (or multiple stationary transmitters). The receivers figure out the distance by measuring the time difference between the arrival of the much faster light pulse and the much slower sound pulse. This does not require synching everything together and since sound is so slow, it does not require very fast electronics either...but you need line-of-sight for the IR to reach you and calm air for the ultrasonic pulse to reach you. I am also fairly sure sunlight will mess around with IR receivers a lot. This won't work very well outdoors.

If you can get a 1GHz receiver you can measure distance with RF just fine. light travels 30cm in a single GHz clock cycle (1ns). But you need multiple stationary transmitter or receivers if you want direction and they have to be very far apart to be effective. Ideally on 2 (or 3 preferably) corners of the field, but then you need to synch all the transmitter and receiver clocks very precisely.

Triangulation using signal strength:
-----------------------------------------
You would need directional receivers that can rotate mechanically and each lock onto the direction where the signal is strongest (or null depending on your antenna) and then the intersection represents where the mobile transmitter is. Or instead of rotating directional receivers, you could use receiver arrays each facing outward in a different direction (then you don't have to rotate the receiver to "search".) Precision mechanics/motors are needed to rotate the receivers...a bit bulky, but very doable if you aren't overly concerned with distance and don't need absolute direction accuracy. You probably don't even need multiple receivers if you don't want location and only want direction.

Summary:
------------------
RF TRIANGULATION USING TIMING:
1 transmitter and receiver = distance
multiple transmitters or receivers = direction (and distance)

RF TRIANGULATION USING SIGNAL STRENGTH:
1 transmitter and receiver = direction
multiple transmitters or receivers = distance (and direction)

I would say an RF triangulation using signal strength is your best bet if you use a single receiver array with a 180 degree span. This should allow you to fully track direction without need for timing or "rotating" to search for the object. All that is required of the transmitter is to transmit a signal continuously.

 

[cyprio7]

thanks dknguyun for your detailed reply. well i think direction is a lot more important to me. but id like to keep the device small. nigel what kind of sizes are those ferrite rod aerials.. can u get really miniture ones? iv given up my obsession with distance now coz it seems too complicated to do both.. direction is more important... so u think i can do that using ferrite rod aerials and keeping it compact? but the prob with that is that it will give u 2 nulls wouldnt it? therefore, you have to know which way the transmitter is in terms of 180degrees. is that correct? is there a way to get round this? im just concerned about direction, but being able to do this using the smallest components possible, by signal strength or by timing... anything. nigel could you please tell me a bit more about the ferrite rod aerial method of measuring the strength please and how it finds the direction etc i dont know anything about it to be honest, i had a look on google but its just comin up with pages an pages of inductor coils and tuned circuits.. not really the info that u wer tellin me

thank u

 

[dknguyen]

Direction

I was thinking about the two nulls, but is the receiver handheld? Or is it sitting at the edge of the field? If it was at the edge of the field, you could immediately rule out one null...

I don't know much else about these

RF should have no trouble getting through a greenhouse.

50m is not a problem for RF. Just like ultrasonics, higher radio frequencies get weaker more quickly so more power needs to be used to transmit a high frequency signal the same distance as a low frequency signal. But suppose you use 2.4GHz (unlicensed band), it still isn't a problem for 50m.

If you use the radio modules here as an example:
**broken link removed**

Looking at the XBee ZigBee Module, it is able to work 300ft (90m) with just 1mW of power at 2.4GHz.

Example Ferrite Aerial Rod:
**broken link removed**

And this aerial is of a much lower frequency than 2.4GHz(in case you didn't know, antenna length is proportional to signal wavelength it is designed for, so higher frequency = shorter wavelength = shorter antenna) and its already just 5" long!

 

[Nigel Goodwin]

nigel could you please tell me a bit more about the ferrite rod aerial method of measuring the strength please and how it finds the direction

It doesn't find the strength, that's not an accurate enough indication, it's likely to vary with all sorts of external factors. However, in your later posts you've said distance doesn't matter, and you just need a rough direction?.

Ferrite rods are used in medium and long wave radios, and they have the tuning coil wound around them - they receive on the SIDE of the rod, and have sharp nulls either end. So you can easily get a direction, but not know which way it is.

The classic DF technique is to measure the direction from one point, draw the line on a map, move a substantial distance away and take another reading, draw that line on the map - where the lines cross is where the transmitter is located. It's common practice to take at least three readings, and more if possible, as it gives you more chance of been accurate.