Leaky Feeder

From a recent post ...

As an experiment i got one of these   http://www.banggood.com/DSP-PLL-87-108M ... 93164.html
this has the advantage of being a sound card as well as an fm transmitter.

... which received this apt warning by reply ...

Unfortunately no radio station except maybe radio 3 would broadcast like this but it does show how good FM radio can sound.
and this from a transmitter that cost a whole £4-46

...and at a power that could well have Ofcom banging on your door.

So how do you stop RF getting past your front door?

An RF distribution system would be ideal but, whilst TVs all have aerial sockets, how many FM radios, particularly of the portable kind, have them?

The answer could be a leaky feeder. We are used to having FM (and some AM) coverage in places such as the Dartford Tunnel in which leaky feeder has been installed and it is also used in mines and underground railway tunnels for two-way communication, so how does it work?

I first encountered it at The Stock Exchange in 1970 when STC were carrying out experiments on the Market Floor.

We used their Starphone 2-way UHF radios and the idea was that the Members might appreciate this form of instant communication with their staff during trading hours.

To do this would require using a number of channels outside the normal bands and the only way that the system could be licensed was on the strict condition that it did not radiate outside the building - hence the leaky feeder.

It comprised a special type of coax which, as the outer sheath was translucent, like the inner dielectric, its construction could be clearly seen.

It was about ½" in diameter with an outer conductor of copper foil. Had the foil completely covered the dielectric it would have been no different to the trunk cables that carried our Cable TV system round the City but, in this case, only half of the dielectric was covered - see (a) in the figure below.

This enabled it to function as a conventional feeder but also to radiate for a short distance all round the cable and it was installed above the false ceiling of the Market Floor in a pattern which ensured an even distribution.

I've never seen any of this feeder for sale but have been pondering for some time as to whether it might be possible to make some?

One idea I had was to use 15mm copper water pipe, split in two, possibly using a jig saw and suitable jig to guide the pipe. Then find a suitable centre conductor and suspend it somehow in the middle (b) but how?

This pair of videos from You Tube gave me an idea.

My first idea was to create a cross section as shown in (c) and slice it into narrow pieces, then Tie-Wrap them to the pipe as shown. This produces a style of construction akin to Bamboo cable. The only sample I've seen of this was about 2" in diameter with a solid copper centre conductor over ¼" in diameter! (This was not a cheap cable to manufacture!)

The centre conductor had thin discs of polyethylene on it every 6 - 8" which, in turn, were inside a relatively thin walled polyethylene tube, so the inner dielectric was mostly air, resulting in a very low loss.

If it is possible to put sufficiently accurate angled slots in the block of HDPE, it would be possible to hold the centre conductor in place without any other fixings as shown in (d). The HDPE supports are only necessary to keep the centre conductor in place but both inner and outer should be so rigid as to only need a very small number for each length of pipe. (e)

A sample of 15mm copper pipe I have has an internal diameter of 13mm but I think some thinner walled varieties exist these days so it would be as well to measure any that you obtain.

This is just one of many on-line cable calculators available to calculate the size of inner conductor required: http://www1.sphere.ne.jp/i-lab/ilab/tool/cx_line_e.htm

The spec of the FM transmitter does not state the impedance but the 75cm whip suggested points to 50Ω - 75Ω. As all the cable and connectors most folk will have will be 75Ω, I'm using 75Ω here.

So, input D as 13mm, noting that it is the internal measurement which is important, and for the dialectric constant Ɛr enter 1, the value for air. Entering a value of 3.7mm for d, the centre conductor diameter, gives Zo as 75.4Ω - you can experiment with different values for whatever is available.

4mm, for example, give about 71Ω but with a thin walled pipe, 14mm moves this back to 75Ω again, so expect some variation in the final result.

I have some ideas for the 75Ω connectors that will be required at each end but I'll leave them for a later post to allow time for reactions to the idea. As many of these as required can be made and daisy chained with normal 75Ω coax, so you could have one in a shed feeding one or more in the house at the other end of the garden, if you wish!