Band 1 TV Offset Measurement Using The Icom PCR1000

The following is an original article which appeared on the Skywaves website in the 2000s. I have just been asked if I still had this article. Thankfully, the wonderful archive.org Wayback Machine is still able to time travel back into the past to look at historical information on the web.

Some of the links below may be dead and images are missing, though I may try to find some suitable replacements along the way.

Just bear in mind that most of the band one TV transmitters in Europe are now silent, including many of those referred to in the article below, thanks to the wonderful digital revolution which has killed our enjoyment of many aspects of traditional radio.
_________________________________________________________________________________________

Band 1 TV Offset Measurement Using The Icom PCR1000

The Skywaves Magazine for August & September 2002 carried my article "Band 1 Video Offsets" in which I described the techniques needed to identify many TV transmitters from precise measurement of the carrier frequencies. The TV-DX community has had more experience of the technique in the past two years, and interest in the technique is growing, so I felt it was time to bring this article up to date, and also make it available for newcomers to the technique.

Using Video Carrier Offsets
Some TV DXers have used this method of precision video carrier measurement as an aid to identifying band 1 television signals. 

When there is a DX opening on band 1 television you may find there are several signals fighting to take control of a particular channel. While there may be one individual signal that dominates, there could be half a dozen others that don’t get through and you’ll probably never know they were even there. Many of the signals are very weak and may only be detectable on a scanner. But wouldn’t it be interesting to be able to discover the identity of these signals? How do we go about doing so?

TV video carriers transmit in the AM mode, so by listening in SSB you can hear the carriers, even when they are very weak. Scanners have a distinct advantage here as they have much narrower bandwidth than a TV receiver, so the weaker signals may be received relatively easily, however you usually need quite a strong signal in order for the actual picture to be seen on the television.  Furthermore, there are various software “spectrum analysers” freely available from the internet which you can install on your PC and these can play a very important role in the measurement and consequently the identification of these carriers.

If you take an audio feed from your scanner into your computer sound card, the software can “see” and display carriers that are so weak that they are not audible to the human ear. Now this is getting quite far removed from the more traditional methods of our DXing hobby as we are no longer listening to the audio or watching the video from a TV station.  But just how accurate is this method of carrier detection when using a scanner?  This is one of the questions I have been asked recently.

Personally, I don’t think there can be any replacement for the traditional spoken or visual identification, but do not rule out other methods which can enable you to narrow down the possibilities. The system of measuring TV carriers is simply an aid to discovering the location of some of those weaker and otherwise unidentifiable pictures.  By recognizing the carriers we are usually able to get a good idea of the area or even the actual which is being received.  Whichever way you look at it you have to admit this does provide an interesting slant on the hobby and gives you useful information that would not otherwise be available.  So how much guesswork is involved? 

Fortunately for us, video carriers on band 1 are not always bang on frequency. Given that channel E2 video is 48.250 MHz, and R2 is 49.750, etc. you might expect to find all the stations on that precise frequency, but in practice a system of frequency offsets are used by broadcasters to minimise the subjective effect of co-channel interference.  It has been found that offsets which are integer multiples of 1/12th of the line frequency give less objectionable interference when propagation opens up than when carriers are more closely synchronised. Hence the offsets are agreed among neighbouring transmitting authorities. This disperses the nominal frequencies around each channel frequency, and when the frequency errors of individual transmitters can be distinguished, we have a possible means of identifying the transmitter without ever having to resolve the video.

That's the theory. In reality, some transmitters are inclined to drift around a little but this is not usually a problem due to the wide spacing of the carriers.  In fact the occasional drift can be a positive boon!  Not only can we now 'see' these carriers with computer software, but we can watch them fade in and out, see how clean they are, and analyse any other properties that they might exhibit.

As you may know, there are two standards for line and field frequencies in use globally.  The spacing between video and audio carriers is also different.  So the moment a carrier becomes available it may not be difficult to work out what you are receiving.  The offsets are only one clue. You can also check the sound and vision spacing, assuming of course that you are receiving the sound.

The use of the 6m propagation maps which can be found on the internet can point you in the right direction too, as can actually listening to the Tv audio on your scanner if the signal is strong enough.  Usually with a sporadic E opening on band 1 you are able to work out a “footprint” or general area of reception.  Now you begin to see that there is probably less guesswork in this than you might have thought.

Scanners can give you the possibility to receive some very low power outlets of TV stations.  In the UK we may be able to receive double hop E signals which tend to be weaker.  As an example, countries like Syria, Iran and Jordan and the UAE are often received here in the North Midlands. Transatlantic TV has also been possible on many occasions – from north and south America and the Caribbean!

I will explain the method I use to calculate the offsets and show you some screen grabs of stations received using Spectrum Laboratory software which is available to download, free of charge by clicking here.

Precision
As we are looking at precise frequency measurement here you will find that some band 1 video carriers drift a little, but usually only by a few Hz.  However, a few transmitters like the E2 transmitters at Grunten in Bavaria – 48.260.422.0 and Bantiger in Switzerland – 48.250.000.0 are precisely “locked” on a set frequency which never alters so you can use these transmitters as reference frequencies to help you calibrate your receiver.  Another one is the Austrian transmitter at St Polten on E2a – 49.750.000.0.

Offsets
Offsets are sometimes referred to as being ~p or ~m (i.e. an "8p" offset or a "5m" offset, etc.) which again indicates roughly how far plus or minus the offset is.

Reference Frequency
In order to measure a carrier frequency accurately, it is necessary to calibrate the receiver to the required degree of accuracy. We are fortunate in western Europe in having several standard frequency transmissions easily available, which can be used in various ways to calibrate the receiver.

Spectrum Laboratory
I use a software spectrum analyzer called Spectrum Laboratory, or SpectrumLab for short.  This is packed with useful features and lets you zoom in close on the carrier so you can take very precise measurements to accuracies of less than 1 Hz.

Method
I feed the audio from my Icom PCR1000 receiver into the computer's sound card.  I use the Buxton amateur radio beacons on 50.000.000.0 & 70.000.000.0 MHz as my markers because these have a precise and stable frequency as stated on their website.  However, it may be necessary to find an alternative reference frequency if you cannot receive Buxton.  It seems that Buxton has a coverage area of roughly 100 miles radius, judging by reports from DXers. The Band I TV transmitters at Bantiger, Switzerland and Grunten, Germany are derived from atomic standards and can also be used as references for our purposes.  Using the narrow-band techniques we are discussing, these can normally be 'seen' constantly in the English Midlands, and are probably available throughout much of Western Europe. If received at a distance, it will be necessary to ignore any spectral 'streaking' due to aircraft or meteor scatter. Western Europe also has available a wealth of LF & MF Standard Frequency transmissions, and while these are unsuitable for direct calibration of the PCR1000, they can be effectively multiplied up by the use of an off-air frequency standard. We used an off-air standard locked to Droitwich 198kHz, and observed the harmonics of the 10MHz output; these were easily visible up to 200MHz. This way it would be easy to calibrate within 20Hz at 200MHz , whereas direct measurement of the 198kHz carrier would require resolution better than 0.02Hz, which is not really feasible.

Elsewhere in the world, it may be necessary to look to the few remaining HF Time & Frequency Standard transmissions. WWV is still available on 10,000 and 15,000kHz, and the Moscow station RWM is on 9,996 and 14,996kHz, between them probably covering half the globe. You need to set your rx within 1Hz at 10MHz to be within 5Hz at Band I. GPS-conditioned frequency standards are the current method of choice in the scientific community, and if you can borrow one, use the harmonics to calibrate at VHF. Unfortunately, the use of standard frequencies via TV line synchronising pulses is obsolescent, as digital distribution protocols remove the need to transmit real-time sync pulses. It is no longer useable in the UK.

Calibrating the PCR1000
By tuning the PCR1000 to 49.999.000 in USB I get an audible tone from my 50MHz Buxton reference.  This can be seen clearly on SpectrumLab, and should be at 1.000kHz. If the SpectrumLab frequency offset is set to “0” Hz, this will compensate for the 1kHz offset of the receiver tuning, and the received tone should be at the 1000Hz mark.  A setting of -500 to +500Hz for the SpectrumLab display width should allow you to see the initial error, and you can zoom in closer as you calibrate.

To calibrate the PCR1000 it is necessary to use the engineering mode software “EX2099 for IC-PCR!1000” current revision 1.0, which might still be available here.  This software is not easy to obtain so you may need to Google for this or join one of the PCR1000 forums.  Beware with this software though as improper use may cause damage to your PCR1000.

Firstly, you will need to close any PCR1000 controlling software that may be in use, then run EX2099. Switch the “power on”, and input the required frequency (49.999 MHz USB in this example) making sure you turn up the AF gain as this may be too quiet by default.  In the “Adjustment panel” select “EEPROM” and click “START”. Then select “Xtal”.  Adjust the “Ref Adjustment” up or down while monitoring the (Buxton) carrier.  Making note what happens to the positioning of the carrier in Spectrum Lab.  Once you have got the carrier as close to zero offset as possible click on “Rx Set”. Mission accomplished!

When making adjustments to the “Ref Adjustment” you will see that, although you are adjusting the oscillator frequency in roughly 1Hz steps, the carrier at 50MHz will appear to move by around 4Hz at a time as the receive frequency is controlled entirely by the 10.250MHz reference oscillator, and the steps are scaled pro rata.  But you can compensate for this in Spectrum Lab by adjusting the frequency bar in order to get the carrier spot on to 0Hz.  Don’t worry if you are unable to get the initial carrier adjustment spot on to 0Hz, simply adjust the frequency bar. You can be safe in the knowledge that, at 50MHz the calibration will be accurate to 1Hz right across band 1. This can be demonstrated here with Buxton’s transmitters at 50MHz and 70MHz.

Note that this software is not particularly user friendly and has a tendency to revert back to a default frequency in mid-adjustment, so you may need to reset the frequency, mode and AF levels all over again.

Also make sure your receiver has warmed up sufficiently before making these adjustments. My own PCR1000 takes several hours to stabilise though about three hours should be sufficient to make adjustments with the EX2099 software - the longer the better.

By using the reference oscillator adjustment dialog box, it is possible to alter the receiver frequency in steps of approximately 4Hz at 50MHz, or pro-rata at other receive frequencies, until the tone moves to the 0Hz mark. It appears that one can generally set a PCR1000 freq. calibration to better than 5×10e-8. It should therefore be possible to reduce the calibration error to within ±2.5Hz at 50MHz or ±10Hz at 200MHz (short-term).

In use, add 1000Hz to the indicated receive frequency and add the SpectrumLab frequency scale reading (plus or minus) to get the received carrier frequency.

But don’t just take my method as THE only way to measure offsets.  The alternative is to use a locally-generated reference frequency source, for example a spectrum of harmonics derived from an accurate oscillator, and make corrections to the indicated receive frequency at each frequency.  Read the methods of Ian Roberts and Todd Emslie at their website.

More Information about Offsets
Offsets are nominally arranged in units of 1/12th of the line frequency so in the case of 625 line pictures 1 offset unit is 1.302 kHz thus giving us:

Offsets for E2 transmitters:
8m 48.239.584
7m 48.240.866
6m 48.242.188
5m 48.243.490
4m 48.244.792
3m 48.246.094
2m 48.247.396
1m 48.248.698
zero offset 48.250.000
1p 48.251.302
2p 48.252.604
5p 48.256.510
6p 48.258.812
7p 48.259.114
8p 48.260.416
9p 48.261.718

Offsets for R1 / E2a transmitters
8m 49.739.584
7m 49.740.886
6m 49.742.188
5m 49.743.490
4m 49.744.792
3m 49.746.094
2m 49.747.396
1m 49.748.698
zero offset 49.750.000
1p 49.751.302
2p 49.752.604
3p 49.753.906
4p 49.755.208
5p 49.756.510
6p 49.757.812
7p 49.759.114
8p 49.760.416
9p 49.761.718

Offsets for A2 [nom. 55.250MHz] with 525-line 60Hz field video:
Negative 55.240.000
zero offset 55.250.000
Positive 55.260.000

Although these are the listed offset frequencies you may find there is slight variation (a couple of hundred Hz at the very most from personal experience and usually within 50Hz) but you shouldn’t find anything in between these. There are no m’s and p’s as used in Europe.

Thankfully, in practice transmitters are only "nominally" on these offsets, which makes it possible to separate and identify individual transmitters.

For instance, the EBU list gives Kuldiga in Latvia as R1 with an "8p" offset, which gives us a nominal frequency of 49.760.416. The transmitter is in fact on 49.760.459 (most recent readings) but the "8p" figure gives us a rough idea of where to look.  The "actual" frequency can be lower or higher than the nominal offset. Or, on E3 both 55.273440 and 55.247.552 count as "2m" which is nominally 55.247396.  Note that the quoted frequencies are not immutable, but will be subject to the same sort of drift-with-time as your receiver. If the transmitter is subject to periodic maintenance, it is quite likely that the transmitter frequency may occasionally be altered, so that we are dependent on updates from those who are able to identify and measure the carrier frequencies. All of us may be able to contribute to the updating of the databases, if we can achieve the required measurement accuracy.

In conclusion, it is clear that users of scanning software have a big advantage when DXing during the E season, etc:- the ability to detect signals at lower levels which may result in a greater number of loggings; a possible extension of the E season either side as a result; being able to measure offsets during the F2 season on otherwise smeary and unwatchable pictures; the ability to tune into the audio and thus work out the sound and vision spacing.

Here are some screen grabs of video carriers

Above:  Two wide, smeary auroral carriers on E3.  John Faulkner.

Above:  Jet scatter on Liege, E3, measured at 55.250.034.5 (on the right of the image).  The smeary traces which surround it are doppler effect caused by reflections from aircraft.  The carrier itself is in the centre of the doppler smears.  You can also see other weak carriers at 55.249.984.5 (this appears at -15.5 on the frequency reference bar) and a broad mess of meteor scatter can be seen just below that (-15 to -25).  This is rather difficult to see due to the low resolution of the images.

Above:  Warm-up drift of the PCR1000 nicely captured by SpectrumLab software.  The signal is that of the amateur radio repeater GB3BUX in Buxton, Derbyshire on 50.000MHz.

For the club's complete listing of band 1 offsets, please click here.

John Faulkner, with thanks to Julian Hardstone for checking the techie stuff!