FT8 – The latest data mode

I’ve posted a lot about the low power, low signal modes, JT65 and JT9 in the past and now there’s a new kid on the block. This one is called FT8 which stands for ‘Franke and Taylor, 8-FSK modulation’, developed by Steven Franke (K9AN) and Joe Taylor (K1JT).

The biggest change from the slow, plodding JT65 and JT9 modes is the speed.  Rather than each transmission period lasting around 50 seconds, each transmission made using FT8 is a mere 15 seconds which means a QSO can be completed in around a minute and a half.  In fact, things happen so fast that the software can now control the transmission sequence.  You reply to a CQ call, if the person you are calling responds to you, the software takes over and makes the exchanges automatically.  It’s so quick that it’s actually difficult to do it manually.  You can switch the automation off and operate manually but I don’t think many people will do that.

I made my first QSO using FT8 just ten days ago and I have to say that my initial impression wasn’t positive.  I was a little frustrated that I wasn’t able to do it manually, like I’ve done with the other data modes in the past but then after a few QSOs, my opinion has changed a bit.  It reminds me of AMTOR ARQ where two stations sync together and switch between transmit and receive very quickly.  I’m actually quite enjoying operating FT8 now.

My impressions after ten days of using FT8 is that it doesn’t handle signals as weak as JT65 or JT9 (the protocol notes below confirm this is the case) and I’m convinced by the signals that I’m seeing on the bands that many stations are running significantly higher power than was common on JT65/JT9.  FT8 reports signals into the + range as well as – and I’m regularly receiving stations up to +9dB.  That’s very loud.

I think that FT8 will be picked up by big stations and I suspect that amateurs with low powered stations and small aerials are going to suffer.  They’re going to struggle to work FT8 with anywhere near as much success as with JT65 and JT9

Thirty seconds of received stations on 20m FT8

Thirty seconds of received stations on 20m FT8

Five minutes of received FT8 waterfall on 20m

Five minutes of received FT8 waterfall on 20m

In the ten days I’ve been operating FT8, I’ve made nearly 50 QSOs and worked 33 DXCC entities.  This is very different to the early days of JT65 and JT9.  I remember calling and listening to both those modes and struggling to find anyone to work.  Every time I’ve had a session on FT8, the band segments have been very busy so the takeup of this new mode has been massive.

Here’s the list of DXCC entities I’ve worked using FT8.

Countries worked using the new FT8 mode

All these have been worked on either 40m using my OCFD or 20m using my Hexbeam.  I’ve never turned the power above 10 Watts.

This is a map of the stations I’ve received here on 20m using FT8 over the previous 24 hours.

Stations received using FT8 on 20m at G6NHU

Stations received using FT8 on 20m at G6NHU

A nice feature of WSJT-X is the ability to upload ‘spots’ to the PSK Reporter site and strangely this isn’t enabled by default.  I recommend you enable this option through Preferences, Reporting and then tick the Enable PSK Reporter Spotting box.

Enable WSJT-X for PSK Reporter

Enable WSJT-X for PSK Reporter

Here’s the official ‘brief’ description of the FT8 Protocol:

WSJT-X Version 1.8.0 includes a new mode called FT8, developed by K9AN
and K1JT.  FT8 uses 15-second T/R sequences, provides 50% or
better decoding probability down to -20 dB on an AWGN channel, and
maintains good performance on Doppler-spread fading channels. An
auto-sequencing facility includes an option to respond automatically
to the first decoded reply to your CQ. FT8 QSOs are 4 times faster
than those made with JT65 or JT9. FT8 is an excellent mode for HF
DXing and for situations like multi-hop E_s on 6 meters, where deep
QSB may make fast and reliable completion of QSOs desirable.

Some important characteristics of FT8:

– T/R sequence length: 15 s
– Message length: 75 bits + 12-bit CRC
– FEC code: LDPC(174,87)
– Modulation: 8-FSK, tone spacing 6.25 Hz
– Constant-envelope waveform
– Occupied bandwidth: 50 Hz
– Synchronization: 7×7 Costas arrays at start, middle, and end
– Transmission duration: 79*1920/12000 = 12.64 s
– Decoding threshold: -20 dB; several dB lower with AP decoding
– Multi-decoder finds and decodes all FT8 signals in passband
– Optional auto-sequencing and auto-reply to a CQ response
– Operational behavior similar to JT9, JT65

FT8 is included in RC1 of WSJT-X v.1.8.0 and can be downloaded for Windows, Linux and OS X from here.

This entry was posted in Amateur radio, Data, FT8, HF and tagged . Bookmark the permalink.

10 Responses to FT8 – The latest data mode

  1. Bas PE4BAS says:

    Hello Keith, a very fine review of FT8. I admit I had some thoughts about the mode myself but gave it some time. I enjoy it now but since I have a slow computer (compared to the ones you buy now) it’s sometimes difficult to react to a CQ call in time. I didn’t analyse my log yet but worked a lot of stations already, much more as a could with JT65/JT9 and many of the contacts are DX. However, I noticed when the frequency is very busy and signals are overlapping it is difficult to make any QSO at all. On 10m I decoded signals of -22dB with FT8, not bad at all…73, Bas

    • g6nhu says:

      Hi Bas,
      It’s nice to hear from you again, it’s been a while since we spoke.
      I have seen your callsign on my screen recently, I think it was probably on 40m but you were calling other stations rather than calling CQ so I couldn’t initiate a QSO with you. If you see me on and I’m CQing, please give me a shout.
      Yes, I certainly appreciate FT8 more now than when I first tried it, it still has that “ooo, new, shiny” feel to it!
      I agree with you, I’ve found it difficult to work some of the stations I’ve called and in fact over the last two evenings, I’ve responded to some strong DX stations multiple times just for them to keep calling CQ because they’ve not heard me. That can be frustrating.
      Good job with a decode at -22dB, that’s very good.
      73 Keith.

  2. John Nemo says:

    Yawn yawn. Data?,no thanks, I prefer to work the stations myself, not have software do it for me. Personally I have found watching the grass grow more exciting. As one chap put it in RADCOM, data enabled him to do some gardening while the software worked the DX for him. What possible satisfaction,or sense of achievement could anyone get from that?. Please understand I do not mean to be unpleasant, I just do not understand the appeal of Data. Regards, John.

    • g6nhu says:

      Hi John,

      The comment about working DX while gardening is a bit tongue in cheek as there isn’t any software that will allow that to happen and I wouldn’t want to do that anyway. It’s all down to individual preference, I enjoy the data modes and have done for many years. It’s good to see how technology has changed over the 35 years since I’ve been licenced and how it’s now possible to pull such weak signals out of the ether.

      I’m just as happy running data modes as I am on SSB or CW. It’s all part of the hobby which is about self training and learning. Keeping up with modern times.

      It’s been said many times before but it’s true – If we were all the same, life would be very boring.

      73 Keith.

  3. Kit says:

    Hi Keith. Good to find your site! I’ve only just started looking at the data modes that have been developed in the past few decades (AX25, Packet SSTV & RTTY were where I was when my callsign went into hibernation in the late 80’s. HI!

    I attended a Colchester Radio Amateurs club meeting last week for a talk on JT65 – and that’s where I had FT8 demonstrated. I tend to think of these modes as propagation tools as there is little or no ability to add a personal touch to exchanges on either of the above modes. However I do find it very satisfying to see my QRP signals being successfully decoded in countries where I have no chance of working SSB at this time. Thanks for the review – I’ve since had a few contacts on FT8 and heard FT8 stations from all over the globe.

    • g6nhu says:

      Hi Kit,
      Good to know you found the site 🙂
      I’ve just added a tiny bit to my post about enabling PSK Reporter – I really don’t know why that’s not ticked by default as I think it’s a great idea for people to upload what they’re hearing as the reporter site is a handy tool.

      I know what you mean about a personal touch. That’s one thing I do miss about FT8 compared to JT65 and JT9 as I had some custom final 73 messages, generally trying to say what power and aerial I’m using. “RR73 5W HEX” was a good one.

      Enjoy the data modes. I certainly do.

      73 Keith.

  4. Jim Erbe says:

    Hi Keith. Nice to have found your site. Thanks. Anyway, I have been working JT65
    for many years having made over 3700 contacts and worked 637 grids. Now we have
    FT8, only working 12 or so, mainly 6M, 20M and 40M. A question I have is how to tell
    if you have worked someone before..what color to use or how to even represent that
    contact like one does in JT65? I have, under General, the “Show DXCC entity and worked before status’ checked. I do paper logging, period. I reply to eQSL contacts.
    But just wonder if the next update to FT8 will have more to select with possibly the “worked before” being shown when someone calls CQ.

    Thanks, W6NIF.

    • g6nhu says:

      Hi Jim,
      It’s a good question and one I really don’t know the answer to. What I did was to rename my existing log file within WSJT-X and start afresh, that way I can easily tell who I’ve worked before using the mode. If I’m going to operate JT65 or JT9 then it’s a simple matter to rename the current log file and put the old one back again.
      I’ve heard people talking about JT Alert, not sure if that still works with the new version of WSJT-X or not or even how it supports logging and notifications.
      73 Keith.

  5. Does FT8 share the same frequency/band segments of JT65/JT9 or do they use the sub-bands of the wider and higher powered data modes?

    Also, as an aside, there are many handi-cappable Hams who can’t speak, hear, or see, or do many of the things we able-bodied people take for advantage. To watch someone, who can only use a suck/blow tube, work Ham Radio data modes, even it it’s with software assistance, is something to behold. They are no longer trapped inside their room “watching the paint dry or the grass grow”.


    • g6nhu says:

      FT8 seems to sit 2kc lower than the JT65 segments – Or at least it does on the bands I’ve tried.
      Robert, that’s a very good point about the data modes and one I hadn’t considered.

      73 Keith.

Leave a Reply

Your email address will not be published. Required fields are marked *