A Simple Underwater Radiolocation System

Cave Mapping
The traditional method of mapping an underwater cave is to take compass bearings, and water depth, at every point where the guideline changes direction. The length of each section of line is also measured. This allows the position of points in the cave to be determined with respect to the starting point. Unfortunately, in a complex cave, like Tank Cave, many readings are required to reach the more remote sections of the cave. This can give a significant error in determining the position of the remote sections.

A radiolocation system allows a point in the cave to be found from above ground. Conventional above ground surveying techniques can then be used to locate the point very accurately. This allows remote sections to be correctly positioned on the map.

Location using a Magnetic Field
Normal radio frequencies cannot penetrate through water and rock. However a magnetic field can penetrate water and limestone very easily. The diagram shows a simple idea for locating a point in an underwater cave. A bar magnet is hung vertically from the roof of the cave, and an observer on the surface measures the magnetic field until he finds a point where the field is vertical. This point is directly above the bar magnet in the cave, and is called "ground zero".

Magnetic field produced by a bar magnet in a cave passage

The observer can also measure the thickness of rock below his feet. He looks for a point on the ground where the magnetic field emerges at 45 degrees to horizontal. By measuring how far this point is from ground zero, and multiplying this distance by 1.77, the depth of the bar magnet below the surface can be found.

Unfortunately this method would not work in practice. The magnetic field from the bar magnet will be too weak on the surface and it will be swamped by the Earths magnetic field. The field from a bar magnet is called a "dipole field" and this drops of very rapidly with distance from the magnet. It diminishes in strength as the cube of the distance. In other words twice the distance means one eighth of the field strength.

Principle of Radiolocation
Instead of a bar magnet the radiolocation system uses a coil, which is energised with alternating current at an audio frequency. The magnetic field produced by the coil is the same shape as the field of a bar magnet, but it is now an alternating field. This field can be picked up by a second coil, which is connected to an amplifier and headphones. The operator can now "hear" the magnetic field. The earth's magnetic field has no affect, since it is not changing with time and does not generate any sound.

The magnetic field direction is determined by a processing of "nulling". The receiver coil is turned until the signal can no longer be heard. This is called the "null" position. At this point the axis of the receiver coil is at right angles to the magnetic field. At ground zero the magnetic field is vertical and the receiver will null when the coil is horizontal. To find ground zero exactly the coil must null horizontally no matter in which horizontal direction the coil is pointing.

The "nulling" method can also be used to find the 45-degree point, in order to measure depth.

Design of the Pinger Transmitter
Adrian Richards and I have constructed a simple radiolocation system. We call it "The Pinger" because of the sound it makes.

For underwater use the transmitter needs to be small and easy to handle. Unfortunately the traditional designs of radiolocation systems use quite large coils, typically 600 mm or more in diameter. Winding it on a core of material with a high magnetic permeability can reduce the size of the coil. However a core of laminated iron (like a transformer core) may not be suitable because of excessive losses at the audio frequencies used.

Carlo Virgili & Ken Smith with a Pinger © Andrew Seifried

The pinger uses a core of laminated mu-metal with a length of 310 mm, and a cross section of 12 x 10 mm. This allows a long thin coil to be wound with characteristics similar to a much larger "air cored" coil. This coil, together with driver electronics and batteries, fits into a 600 mm length of 50 mm diameter PVC tube. The tube is permanently sealed at one end and has a threaded O-ring cap at the other. The inner workings can be removed via the cap for battery replacement or maintenance.

The pinger is weighted at one end and has a slight positive buoyancy, so that it floats upright in the water. At the top end is a central nylon spike, which is placed on the roof of the cave. The pinger can be left floating in this position and it will hang exactly vertically to give the correct orientation of the magnetic field.

The pinger is small and simple to use. It is fitted with two plastic rings for clipping to the diver is a similar manner to a stage bottle. Three identical Pingers have been made to this design so far.

The driver electronics for the pinger was designed with simplicity in mind. The coil is connected in parallel with a capacitor to form a tuned circuit, which resonates at 1.16 kHz. A tuned circuit allows a large alternating current to be generated in the coil, which in turn generates a strong magnetic field.

A simple oscillator provides the 1.16 kHz signal and this is amplified by a single chip audio amplifier and fed to a driver winding on the transmitter coil. Eight Alkaline C cells provide the 12-volt power supply required. The current drain when transmitting continuously is about 190 mA. The current consumption was reduced to about 70 mA by adding a circuit to pulse the signal on and off with an on time of about 33%. With this modification the battery life is more than 24 hours.

Ken Smith using the receiver coil to locate ground zero © Tim Payne

The pulse rate is different for the three pingers that have been made. This allows each pinger to be uniquely identified by its signal.

Design of the Receiver

The receiver uses a coil identical to the transmitter coil. It is also tuned to resonate at 1.16 kHz. Use of a tuned receiver minimises interference from other signals, such as power lines and electric fences. The signal from the receiver coil is amplified by a single chip audio amplifier and fed to the headphones. The electronics is housed in a small die cast box, which can be mounted on a waist belt.

The receiver coil is mounted in a PVC tube. This is carried in one hand when searching for the pinger. The tube is fitted with two spirit levels. One indicates when the coil is horizontal, for ground zero determination. The other indicates when the coil axis is at 45 degrees, for depth measurement.

Using the Pingers
Placing of the pingers is usually done by volunteer divers. As the diver is preparing to enter the water the pinger is switched on and the end cap screwed on.

Ken Smith locating the 45 degree point for depth measurement © Tim Payne

The pinger can be clipped to the diver's vest. One or more pingers to be easily carried without any interference to the enjoyment of the dive.

At Tank Cave there is almost no electrical interference, and the distinctive "beep beep" of a pinger can be heard at up to 90 metres from the transmitter. The pulsing of the signal assists in finding the pinger. It seems to be easier to hear a weak pulsing signal than a weak continuous signal. Plenty of signal strength was available for measurements at Tank Cave, since the pingers were rarely more than 20 metres below the ground surface.

The diver is asked to place the pinger on the roof of the cave above a specified survey point. The diver is also asked to measure the roof and floor depths at this point. Metal clips must not be left on the on the pinger. They can cause the pinger to sink, or not hang vertically. Clips are normally left on the guideline nearby for use when the pinger is collected.

Phil Prust locates the pinger in position © Tim Foster

On the surface we can estimate when the pinger will be placed, and we have a reasonable idea of where to find it. The pinger can easily heard within 50 metres of ground zero. With a bit of experience the orientation of the magnetic field can be used to find ground zero quite quickly. Within five minutes the position and depth have been recorded and we can go looking for another pinger.

Later in the day the pingers can either be moved to new survey points or brought out of the cave. Bringing them out at the end of the day is preferred because the batteries can then be turned off and saved for more pinging on the next day.

Future developments
The main deficiency with the pingers at present is the range. They are fine for Tank Cave, but do not have sufficient range for the 100m depths found on the Nullarbor.

Prior to a recent trip to the Nullarbor a 50 gain preamp was added to the receiver coil. This allowed the signal to be heard through 100m of rock, but accurate measurements by "nulling" could not be achieved. The main problem was the large amount of amplifier noise (hiss) that tended to drown out the signal. Future developments planned are:

  • Use of a low noise preamp.
  • Adding a band pass filter to the receiver to filter out broadband noise.
  • Increasing the pinger output power.
  • Auto switch off to save the batteries.

Thanks and Acknowledgments
Thanks are due to:

  • Adrian Richards for the construction of the receiver and other useful help and advice.
  • Tony Carlisle, whose Nullarbor cave radio, first used in 1991, gave inspiration for this project.
  • Brian Pease, whose informative web site gave me much needed technical information. Doing a web search for "Brian Pease" is probably the best way to find out about radiolocation. See http://radiolocation.tripod.com/
  • All of those people who have helped by carrying pingers underwater and making surface measurements.
  • Landowner, Rob Dycer, for allowing us to have so much fun in, and underneath, his paddock.


Ken Smith