Friday, 14 July 2017

VHF data frequencies plus WSJT-X 1.8.0-rc1

Thanks to David, GM4JJJ, for pointing out me that the long awaited sharpening of the blades of WSJT-X, version v1.8.0-rc1 r7847, is now available at the WSJT Home Page link on the right. Follow the link  to WSJT-X and select the "candidate release" of WSJT-X 1.8.0.

Our attention was immediately drawn to a new "mode" (or protocol as some people put it). FT8 offers many advantages over JT65 on VHF. In particular it should suit weak long distance signals affected by QSB and Doppler.

At first glance there are several features in FT8 which I can certainly see will help. Immediately noticeable is a 15 second segment time which is just what we need for DX. This means that a QSO can be completed in a minute. If you want to know why that is important ask a CW operator. On JT65, a mode otherwise well suited to VHF DX work, you can sit and hear a station for a minute, reply, and at the end of 2 minutes get nothing back, never knowing whether you were heard or not.

I used to work with a guy who said "try to see the bigger picture - what does this mean for me?". Tee hee. The bigger picture here is that I might have worked Japan in the time I sat and watched many JA stations calling CQ. Quite possibly they were also watching me and other Europeans calling CQ, and the path just was not open for two minutes. Yes, overall it was open for a hour or more, but between any two stations it was only open for a minute or two.

I can get a whole QSO done on FT8 in a minute ....
A full QSO on FT8 in one minute, with 15 seconds thrown in for the 73s
At this speed you have barely enough time to respond so there is auto sequencing to reply automatically. Anyone who has used 30 second segments on MSK144 will feel very comfortable with FT8.

Not only are the segments shorter, but the synchronisation is at the start, middle and end. A readable signal in a 15 second segment should be 100% decoded, whereas the same length of strong signal would be less likely to decode in a 1 minute JT65 segment and even less so in 2 minute WSPR one, depending on the fading. If I understand it correctly, the chance of decoding is better as the segment gets shorter, thanks too to more frequent synchronisation. That is another plus for FT8 but not at the level of the "fast" JT modes like MSK144 or fast JT9, which can decode tiny fragments of signal. However, it does represent a huge step forward for the terribly weak and fading VHF stations we now hear.

What FT8 seems to represent is a very clever blend of tweaking the protocol with tailoring to the WSJT-X wrapper. So while a better decoding process will bring benefits (-20dB is promised), cutting the segment time and auto sequencing will bring much more chance of working those fleeting Japanese, Chinese and ... well, who knows where? ... stations.

I tested it by working a few stations and some I could not decode due to their clearly late timing. Every well timed station was copied easily.

Apparently it is possible to set it to automatically reply to the first station which comes back to your CQ, and further developments will see it able to reply, not to the first reply, but the weakest reply out of several. Now that is a DX-ers dream.

I worked a couple of stations using FT8 on what I am pretty sure was iono-scatter. It was neither as steady nor as strong as Es, and it certainly looked like that mode. FT8 looks like a great mode for iono-scatter, and once Es has faded it will be very interesting to see how we get on with the even more esoteric propagation methods. I will be particularly interested to try mixed iono/tropo/meteor scatter - type melanges.

EDIT - by melange I mean mixture - not the drug in the film, which was fictional by the way. Nor the "Melange" two stroke oil/petrol mixture machine I used for my Puch motorbike in the 1970s.

MSK144 seems to have totally replaced JT6 on 4m and 6m meteor scatter. I would really like to see FT8 replace JT65 on VHF Es. What chance of that happening? Not much I suspect. Human nature being what it is (the rest of this has been deleted so as not to offend the dozy old duffers in the hobby who think we should be running a museum).
I am not even going to start on the great 6m "operators versus the band planners" debate. What frequency should we use - the one the Region 1 band planners think we should use, or the one that works?

I did raise this before, and I will explain it at some stage for anyone who is understandably confused. But for now, I just do not want to get involved.

I was involved at some stage in the past with the 4m equivalent. I had some discussions with the then RSGB VHF Manager. It was easy to agree with this gentleman that 70.091 was good frequency for WSPR on 4m. Much later a row developed between me and some others versus a self-appointed rule maker about this. We won the argument as 70.091 remained, but I was hurt, bruised and annoyed by the whole episode - enough to put me off 4m WSPR for years.

Lately, several of us have been putting forward 70.176 for JT65. This seems to have stuck. I have worked a number of DXCC on that frequency and I see others doing it too. I have no particular qualification to decide these things, just 40 years of using the band and an interest in using data. That self-appointed rule maker seemed to be more interested in STOPPING me using data.

Thank you to several VHF supporters for occupying 70.176 and making it unofficially ours.

So with the arrival of FT8 a pattern of allocations emerges which would put, in ascending order, FT8 on 70.174, JT65 on 70.176 and by analogy, JT9 on 70.178. MSK144 is already on 70.280. WSPR is already on 70.091.

None of these frequencies is "right". The contesters will complain about infiltrating their precious bands, but then data modes are in use during openings, whereas contests never happen during openings (erm, or so it seems). I do not want to start a whole debate about 70.176 as it is now in use. The other modes as suggested by me fit around this as they do on other bands - leaving the 6m shambles out of it for the moment. So does anyone object?

Total data users on 4m - about 20 maybe. Total bandwidth used = very little, during periods of vanishing brevity. Surely this is not too much to ask? Just somewhere to go to find similar minded folk. A sort of "data cottage" - no maybe we should not suggest data cottaging on a family-oriented blog. A data village pump. That is a better idea. A centre of activity. And anyone who wants to transmit any other mode over it can do so. After all, it is in the "all modes" region.

Actually, it doesn't matter all that much. I have discovered that 4m operators can actually turn their VFOs and find other people. There you are, on 5MHz they transmit outside the band because they do not have the wit to change frequency, but on VHF we can actually listen in different places at different times. Amazing. Chapeau. Speaking of which, isn't Fabio Aru amazing?

This area around 70.176 is pretty good for the various international frequency allocations we have at the moment. It is not perfect, but it might be OK. There might be an argument for moving WSPR to 70.172 to fit in, as 70.091 is not so widely available, but I do not want to start that row up again.

Well, I have said it now. I bet there is an argument about this idea.




No comments:

Post a Comment