TV radar

In the last few months my TV reception has been increasingly affected by a particularly strong double image, caused by reflection of the signal from a nearby object. This is cause for concern as I am now liable to pay more for my TV licence as I am receiving the same programme twice, so I set about finding the source of the reflections with a view to cutting it down with an axe.

The reflections are generated because the path, p, taken by the reflected signal to my television set is longer than the direct path, d, from the transmitter resulting in a second, weaker, image on screen.

The difference between the two distances is the path difference, pd, where:


In order to determine pd the time difference between the direct and reflected signal must be determined. This will allow the source of the reflection to be confined to somewhere on an ellipse where the transmitter is one focus and my television is the other.

To determine the time difference for the reflection, knowledge of the TV signal is required. A complete PAL TV line is 64us from start to finish. This includes the line sync pulses and colour burst. The picture portion of the line is 56us from the left hand side to the right hand side of the screen. In practice the actual bit you see on screen is probably slightly less than 56us as the line will inevietably extend over the edge slightly to ensure the whole screen is full of picture.

So to find the time difference for the reflection, divide 56us by the width of your screen and multiply by the distance from the ‘real’ image to the reflection. To get the path difference of the signal multiply the time difference by the speed of light (299792458m/s). I used a PC video capture card to get more accurate measurements than using a ruler. To get maximum accuracy I first calibrated my video capture card to determine the time taken for each pixel to be drawn. The oscilloscope waveform above shows the signal with gives the image below (a rather badly adjusted colour bar). The six coloured bars (i.e. not black or white) can be seen as the fuzzy bits in the signal above. They take 34us to be drawn (measured from the oscilloscope) and are 417 pixels wide. So each pixel represents 81.5ns, giving my TV radar a resolution of <24m.

Next the path difference was found from a TV picture. The reflection can clearly be seen in the image below and is about 151 pixels from the original image. This corresponds to a delay of 12.31us or a signal path difference of 3691m.

Using the maps on, the position of my house in Greenwich and the transmitter at Crystal Palace were determined. An ellipse with the appropriate values of A and B was then drawn.

Next, the ellipse was overlaid on a map of the area and likely candidates for the reflection investigated.

Buildings and hills well within the ellipse surround the receiver from the North to the South-East, shielding it from reflections from this direction. It is unlikely that reflections would be received from the transmitter end of the ellipse, so the most likely direction is from the North/North-West. A quick glance out of the window in this direction (which would have saved me a lot of time doing all this) revealed the likely source of the interference as the Canary Wharf office developments:

The ellipse cuts right through the area occupied by these buildings:

As can be seen from the photo, these buildings provide a huge surface area and are orientated at almost the optimum angle relative to the transmitter for maximum reflection in my direction.

This effect seems to have been confirmed by pointing the antenna at Canary Wharf, giving the image below.

Now the weaker signal is to the left of the main image, showing that it arrived first. This suggests that the antenna is pointing at the source of the reflected signal.

So there it is. My television is now an effective Canary Wharf detecting radar. Probably the most interesting thing I’ve seen on TV for a long time.

Nov 10

Wales A


won 30-9

John Barnard (Eng)

Tweet about this on TwitterShare on FacebookShare on Google+Share on LinkedInPin on PinterestShare on RedditShare on StumbleUponEmail this to someone