Simple signal transmission using long steel pipe

I am a mechanical/hydraulics engineer with an application which may be more suited to you guys' expertise. What I want to do is transmit a signal from a remote (up to 5km) device through a long piece of steel pipe. The remote device incorporates a sensor which produces a simple signal (on/off) which I would like to detect remotely. Is this feasible to transmit a sigal (electrical or sound) without any signal boosters etc.? If this is in no way feasible I have other solutions but this would make a real neat, simple and cheap solution. Your advice/help is much appreciated.

If the pipe is insulated from ground for its entire length then it could be used in conjunction with an earth return to transmit an electrical signal. However, it would not be allowed to be insulated anyway due to the possibility of lightning strike or induced currents, so that one is ruled out.

Sending an audible signal by mechanical means over the required length would only be possible if the pipeline was not buried or mechanically damped by mounting stanchions or concrete supports. Such supports would very quickly attenuate any mechanical vibration set up by a transmitting transducer. Since it would be impossible to construct a

Pipeline construction generally call for a buried cable or fibre or a radio data system to convey signals.

Can you generate soliton waves in a pipe?

martin

This is second hand, so I can't provide much info. There was a posting to this or a related NG way back in response to a similar question. It mentioned that there is a heating system in Europe that is distributed in some way throughout a building. It transmits the heat as a hot liquid medium in pipes, and commands are transmitted acoustically through the liquid. Useful for old buildings with old systems, where the pipes exist but installing signal paths separately would be difficult. From a physical point of view, acoustics should be good at such a task, sound travels well and rapidly in low-compressibility media such as vapour-free liquids. The substantial acoustic impedance mismatch between the liquid and the pipe would minimise loss. Discontinuities such as bends could be another issue....

My intuition tells me it may be possible to induce a shock wave into the medium being transported in the pipe but due to the damping effect of frequent support structures it would quickly die away (similar to a tidal wave or tsunami being impeded by striking a coastline). The wave may travel unimpeded in the wide expanse of the oceans but it quickly stops when it hits land where it expends its energy in cataclysmic fashion.

Acoustic transmission seems like a possibility, and you can do a preliminary test with a hammer, a contact probe stethoscope (or just an ear to the pipe) and a couple of cell phones. Where is the pipe located anyway?

-- Joe Legris

Folks in the oil industry transmit information by modulating the pressure of the drill mud. If your data rate is very low, you could do something like that.

If you have a pipe in the soil, you can still send an electrical signal down it. The soils resistance is enough higher than the pipes. You will be dealing with a very small signal. The real problem is the other contact of the transmitter and receiver. They need to be as far from the pipe as practical. Your signalling frequency needs to be quite low. I'd suggest the local mains frequency divided by 2 as a good first guess.

On Tue, 11 Apr 2006 18:54:05 +0800, bruce varley wrote in Msg.

What a baroque description of a mean central heating system. I like it.

robert

If the pipe can be grounded at both ends, a current transformer could induce a signal at one end, and another could pick it up at the other end. Leakage to earth along the way would cause losses, but I suspect it's workable at several km.

Not hard to try: a couple of regular metering-type CTs, a signal generator and an audio amp at the drive end, another amp and headphones to receive.

John

distributed in

liquid medium

it.

Patent Speak!

What's in the pipe? What's it made of? What diameter? How are sections attached?

I'm thinking of a waveguide, but the losses could be too high for something not optimized for RF use.

sorry for the delay in replying - this is my first chance to check the group. The steel pipe is one continuous length of 2" diameter with a bore of 1-1/2" and will usually contain sea-water. There are no supports required as the pipe is suspended (almost) vertically in a well-bore. Sending pressure pulses is a possiblity but oe Legris's idea of transmitting a sound wave is interesting.

An RF waveguide is definitely out of the question. I'm not sure if hydrostatics follows the same principles as electromagnetics do, but you should figure out what maximum wavelength will propagate along a pipe of these dimensions and with what losses. Then, some information theory wizards can help you select an appropriate modulation scheme.

The problem is relatively simple with two values (on and off). A two tone scheme might suffice, with one for 'on' and the other for 'off'. If you have some ideas about the characteristics of the ambient noise (RMS pressure vs frequency), the signal power can be calculated for a maximum error rate.