Dual JT9 and JT65 with WSJT-X, without split!

George KF2T

Now that we have DAX, it looks like dual JT65 and JT9 operation with WSJT-X is very practical without resorting to setting up split. Here's how: 1) Set up WSJT-X as normal for JT65 (that is, do NOT check the +2 kHz button), and select JT9+JT65 from the Mode menu. 2) Set up your audio through DAX as has been described elsewhere. 3) Use the click and drag option on your filters to create a receive bandwidth of 3800 Hz (100 Hz to 3900 Hz). 4) In the SSDR Settings > Radio Setup > Voice menu, set your Transmit Filter Bandwidth to those same values (100 Hz low cut, 3900 Hz high cut). That's it. You can now work both modes without resorting to split. You'll probably want to adjust your "Wide Graph" on WSJT-X to show a full 4000 Hz, and you may need to tweak your levels a bit. It works! While writing this up, had a nice JT9 contact at 3100 Hz, signal reports were -01 and -03 between Washington State and Nevada on 40 meters. Thanks to Ed, K0KC for getting me thinking about this! Life just got a little easier.

Ed, K0KC

George, I must be brain dead for not thinking of this! I use the beta version of WSJT-X and there are still some issues at times with CAT. I will give this a try this morning if I can get over the illness that hit me last night. Ed, K0KC

Michael - N5TGL

Not trying to be a wet blanket here, but I did this with my SignaLink USB through the ACC port, so this is not necessarily a DAX benefit. Good idea posting it though for others who haven't tried it! TX profiles will make this even easier!

Ernest - W4EG

George, I am doing the same as you described and it works very well. Thanks for posting it; so other may benefit.

Ed, K0KC

Ok, guys, I have given this topic a little more thought and wish to share the following information: JT65 and JT9 are based upon the transmission of a set of sinusoidal (sine wave) audio tones utilizing an error-correcting code. A perfect sinusoidal waveform (no distortion) has power at only one frequency, the fundamental. If distortion is somehow introduced by a non-linearity or some other process, harmonics will be generated. For instance, a pure 800 Hz tone would have harmonics at 1600 Hz, 2400 Hz, 3200 Hz, etc. The amplitude of these harmonics decreases as the harmonic number gets higher. Since we are talking about WSJT-X, I know that Joe Taylor and his development team have taken great care in ensuring that the audio tones delivered by his software are extremely "pure". I think that it is a pretty good assumption that DAX and the associated Flex-6000 series hardware and software are extremely linear as well and thus these pure tones are delivered by the transmitter to your antenna (if you set your audio levels in DAX correctly). Problems can arise with radios that introduce non-linearities. That 800 Hz tone that I mentioned above will be transmitted along with a lower amplitude 1600 Hz tone and perhaps a lower yet 2400 Hz tone If the transmit bandwidth is set to the nominal 2800 Hz transmit bandwidth (100 Hz to 2900 Hz) that is usually used for SSB signals. In many legacy radios, the transmit bandwidth is not adjustable. The latter versions of WSJT-X (I believe starting with version 1.1) attempt to minimize this potential problem by limiting the audio tones to a range of 1500 Hz to 2000 Hz so that the harmonics fall outside of the nominal transmit bandwidth. Since the audio tone range is limited, split mode is used to adjust the dial frequency so that tones are transmitted at the correct frequency. The dial frequency is adjusted via CAT control. If we do not utilize split operation with our Flex-6000 series radios by un-checking CAT and use a single slice with at approximately 4 KHz bandwidth on transmit and receive as George suggests above, we are counting on the WSJT-X tone output and the Flex-6000 being sufficiently linear so that we do not transmit harmonics. I have tested this assumption and it appears that all is well as I have observed a very clean transmitted waveform over the JT65 and JT9 frequency range with constant output power. These observations are only qualitative, however, as I have not taken any measurements. Not using split operation greatly simplifies WSJT-X operation with the Flex-6000 radios, especially if you are testing beta versions of both software products...thanks for the idea, George! The above principles would also apply to JT65-HF and Joe Large, the software author, has warned users in his user manual that the lower audio frequencies should be avoided if possible. I have observed that JT65-HF also produces very pure tones, but again, I have not made any quantitative measurements. Most of the time the signals are weak enough so that the harmonics, even if transmitted, are not visible on the waterfall and do not impact the decode process, but when conditions are very good and/or the transmitter is running high power, they do show-up...causing QRM. I have observed a number of harmonic signals on the waterfall in my approximately two years of JT65 and JT9 operation. Ok, this has been kind of long-winded, but I thought it would be good to share my understanding of the issue...thanks for listening and please add your comments to the discussion. Ed, K0KC

Michael - N5TGL

One other thing to think about is checking your sample rate as explained in the DAX video. Most machines default to 44.1 sampling, and that will cause harmonics because the math doesn't divide nicely. Setting the sample rate to 48 will give a much cleaner output.

N4AB

I have been using this method for JT9/JT65 since late August, and it works great. I think that someone mentioned this in a post sometime in June or July but I don't remember who it was. As far as the operation is concerned it very easy and does not seem to produce any problems. I just set the slice to USB, select the 4 kHz filter, adjust the TX band width to 100 to 4 kHz, and set the Xmit power to 20 or less. However, I have seen several instances of Multiple Decodes where one signal is decoded every 180 Hz all the way across the bandwidth (4kHz) (sometimes only a few decodes 2, 3, or 4). This usually happens on very strong signals, as would happen if you forget to reduce the RF power after SSB or CW operation. It has happened to me but I don't know if my signal was distorted as a result. With the addition of DAX as the audio transfer to the WSJT-X program the signal is much cleaner both Transmit and Receive, very nice. The Flex 6700 is great for the digital modes (including RTTY) as well as SSB and CW. This is truely a visual radio mode as you only need to see the decoded output on the WSJT-X program. But I keep my audio monitor on just to avoid any distortion problems (hopefully). This is a great Mode and thanks to the Flex 6700 and WSJT-X it easy to get on and operate. 73 Al N4AB

DrTeeth

I have my WSJT-X setup like above, but cannot decode JT9 signals unless I work in JT9 exclusive mode and have a 2 KHz filter in the JT9 sub-band. One signal got a +1 report when I managed to decode it in JT8 mode, but not in mixed mode.

Any idea what I am doing wrong?