[ADXA] UKSMG Proposal - brings up the real problem shared by CW/SSB guys

Wed May 27 10:04:55 EDT 2020

This discussion by G4IDR brings up what I have said about the demise of CW and SSB on 6m. His comments include  “I think at least 80% of the FT8 QSO’s today could have been made on SSB and 95% on CW”.  This points out the overcrowding of the “low - dB signals” segment of 6m. If guys would go to SSB or CW, then QSOs could be made at the rate of 3-4 a minute anyway (split and listening up) instead of software programs getting confused on how to cope. Leave those upper ‘SUBaudible signals’ frequency ranges near 50.313 JUST for subaudible that can’t be heard by ears… 
Sure gets complicated on 6m starting 2-3 years ago. I prefer not using software digital updates to work countries and it would be nice to use “human modes” to communicate, especially when signals are from -5dB and up into the +dB range of audio. Ultimately, to each his own, but this consideration by DXing hams would allow subaudible signals a better chance of being copied by software without having to undergo continuing changes in FT8/FT4 operating techniques and updates.. 
San

Sent from Mail for Windows 10

From: Crownhaven
Sent: Tuesday, May 26, 2020 11:37 PM
To: adxa
Subject: [ADXA] UKSMG Proposal

This was received from the UK Six Meter Group.  Some of you might be members.  I highly recommend joining.  In any case, here it what is up their sleeve.  Just FYI.......
Steve, QQ
Some interesting thoughts from G4IDR?

73
Chris
G4IFX

www.uksmg.org 

From: WSJTX at groups.io [mailto:WSJTX at groups.io] On Behalf Of David Redman
Sent: 25 May 2020 12:29
To: WSJTX at groups.io
Subject: [WSJTX] FT8 on 6m SpE

Listening to the recent 6m Sporadic E openings ( SpE ) with some possible F2 mixed in there then it is apparent to me that the 2.5 / 3kHz window is somewhat impracticable when we have widespread sporadic E openings particularly in busy Europe.
Following on from the recent comments of others I agree that stations need to spread out

In one opening I was copying FT8 signals down to -24 but the strongest was +42 (  a local station 30 miles away ) BUT SpE European ‘local’ stations from southern Spain and Italy were +38. This is not uncommon given how SpE works and comparing FT8 signals with the strength of some SpE contacts on SSB and CW.

So we have a range of signals of some >66dB WITHIN the audio passband. No matter how good your rigs filter is ( SDR or analogue ) it’s not going to help ,  we are asking a lot to decode every signal in this 2.5kHz passband down to the decode limits of FT8  ( can you switch your agc off ( does your noise floor change on the waterfall as the AGC comes into effect on strong signals )  , do you peak the wanted signal , notch the strong , narrow your passband assuming you know where the real weak DX station is in the passband etc etc )

Paraphrasing from W9KNI et al ,  if the station your working is stronger than -18 your not working the real DX

If on 6m SSB you have a 150kHz segment and 2.5kHz signals then that’s potentially 60 stations without overlaps.
Turning this to FT8 with a 50Hz bandwidth in 2.5KHz that 50 stations , ok similar without overlaps
However I do note that at busy times the FT8 QSO throughput rate is very low AND many stations are calling CQ for 10 minutes or more without a reply.

A Proposal
So how about spreading out into say 5 segments 2.5kHz wide ( In this suggestion I am using 2.5kHz rather than 3Khz or 4kHz as it seems to be the most used ). My suggestion of 5 segments is a compromise of spreading out v time to decode all segments
WSJT-X has rig control, most stations use it , so how about adding a little search automation to the WSJT-X program on FT8 and it searches these segments 50.5080 , 50.3105 , 50.3130 . 50.3155 and 50.3180MHz
It could be user selectable to search 1st ( even ) sequence ( 0 to 15 and 30 to 45 , 2nd or both periods ( 0 to 30 secs ) before moving on automatically to the next segment and doing the same. Decoded signals are then indicated by the segment or grouped by segment.
Once a stations signal you want to contact is decoded and you double click on it then WSJT-x would place your TX in that segment and on that frequency so that we don’t have split QSO’s spread over the entire 12.5kHz.

This way QSO’s are spread out , you can use your receiver filters to better effect and I am convinced that the FT8 QSO throughput would go up despite it taking 5 times longer to scan all segments AND you would contact more weaker stations.

This proposal still keeps regional SpE contacts away from 50.323MHZ the Intercontinental DX calling frequency and rightly so.

Just trying to help 50.313MHz might not be the right frequency area for such a proposal  , others can decide and it you don’t like my suggestion fine , make an alternative or carry on as you are.

Frankly after making some FT8 SpE QSO’s , I went lower in the 6m band and had a few SSB SpE and even more weak signal CW QSO’s , much more fun with further distance QSO's and easier on the ears despite my tinnitus.

I think at least 80% of theFT8 QSO’s today could have been made on SSB and 95% on CW , so by spreading out we might be able to ‘hear’ the weak -24dB FT8 signals ( which we would not be able to work on CW ) far more often. 
Currently I don’t think we are consistently reaching the full potential of the weak signal capability of FT8 especially given the enhanced decode capability of WSJT-X-rc1 ( now rc2 ).
I use -24dB as the lowest signal decoded , maybe it’s lower but you get my point

This concept could be applied to other bands BUT PLEASE keep this FT8 mode away from the existing EME CW and JT65 / QRA64 sub-bands.

Of course if MAP65 had FT8 capability it would serve the same effect but not everybody has that external I/Q SDR into MAP65 capability. MAP65 is a very useful program and I believe that there are some possible enhancements to its JT65 capability already trialled but not implements so it’s due for an upgrade anyway. Again MAP65 could decode a 12.5kz bandwidth each cycle but with up to 5 times the processor power needed ?? this might be a limitation compared to the above segmented approach ) and with an improved rig control interface could then move your rig to the required TX frequency.

Notice I use segments not channels and QSO’s / signals not Traffic……. some of you will understand my point.

Dave
G4IDR
VHF up DX contacts rule ok ( EME , F2 , SpE , Tropo etc )

p.s. I write very little code so don’t ask me to do it , I am not a radio or electrical engineer but a decent mechanical engineer who enjoys the technical aspect of amateur radio including construction and trying for weak signal QSO’s and have done so for some 45 years.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.qth.net/pipermail/adxa/attachments/20200527/9d3cb7b4/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: bieehdoleopjmhfd.png
Type: image/png
Size: 15422 bytes
Desc: not available
URL: <http://mailman.qth.net/pipermail/adxa/attachments/20200527/9d3cb7b4/attachment.png>