Software WSJT-X question

GB46

Very radioactive
Joined
Feb 4, 2017
Messages
589
Location
British Columbia, Canada
Anyone here use WSJT-X for decoding amateur FT8 or FT4 transmissions? Great app, but I'm wondering if there's any way to prevent it from saving WAV audio files to record the transmissions. New ones get saved on each decoding cycle, without deleting previous files, and they start to accumulate in the data folder. They can be deleted manually, of course, but since I have no use for them, I'd prefer that none are saved at all. There's a Save menu, on which I've checked the option "None", but that doesn't seem to work.
 

GB46

Very radioactive
Joined
Feb 4, 2017
Messages
589
Location
British Columbia, Canada
Time limit of 30 minutes expired to edit the above message, Sooooooo...

The WSJT-X homepage at Groups.io is found at main groups.io Group

Hope that helps.
Thanks, it sure does! I didn't need to pose the question to either Joe Taylor or the group, however, because someone else had already answered it: The latest release of WSJT-X fixed the bug, so I just updated it and it has stopped automatically generating those unwanted WAV files. An earlier Google search brought up many complaints about the issue, but nobody seemed to know what to do about it.

I'm not a groups.io member, but Radioreference.com keeps me busy enough online, and is a great source of useful information, a lot of which I've added to my offline documents for quick reference.

Thanks again!
 

GB46

Very radioactive
Joined
Feb 4, 2017
Messages
589
Location
British Columbia, Canada
On the toolbar on top of the main WSJT-X window go to Save and select None.
I was already aware of that, and the option was checked by default, but as I mentioned earlier, it wasn't working. That's why I had to update to the latest version, 2.2.2, in which the bug had been fixed.
 

RocketNJ

Member
Joined
Feb 1, 2003
Messages
934
Location
Pequannock, NJ
Another reason to always check for updates if you have issues. 2.2.2 also added triple decode to FT8 and double decode per cycle to FT4, meaning better decodes. I normally see in escess of 50 decodes per cycle during the day on 14.074.

A couple of tips - make sure your PC clock is accurate. Use a third party program such as Meinberg. Also in the waterfall set bins per pixel to a value that allows a view of 0 to 3000 Hz. The default 2 bins per pixel is usually too small and you lose the higher frequency decodes.
 

n5ims

Member
Joined
Jul 25, 2004
Messages
3,820
Just be aware of a bug in 2.2.2 that affects the log file (the .adi file). The initial tag isn't written correctly when a new .adi file is created. You may get some invalid tag messages and the log may not import correctly. The fix is easy, just change the incorrect "<eh>" to the correct value of "<eoh>" and save the file.
 

GB46

Very radioactive
Joined
Feb 4, 2017
Messages
589
Location
British Columbia, Canada
Another reason to always check for updates if you have issues. 2.2.2 also added triple decode to FT8 and double decode per cycle to FT4, meaning better decodes. I normally see in escess of 50 decodes per cycle during the day on 14.074.

A couple of tips - make sure your PC clock is accurate. Use a third party program such as Meinberg. Also in the waterfall set bins per pixel to a value that allows a view of 0 to 3000 Hz. The default 2 bins per pixel is usually too small and you lose the higher frequency decodes.
Yes, I always see plenty of FT8 activity all day long. Monitoring that helps fill in the gap when nothing else seems to be happening on HF. FT4 is a little less active, but then it's newer, so it may get more popular later on.

I have an app on my computer that can update the clock from WWV on demand, so I usually run it before starting WSJT-X. As for the waterfall, I never use it, preferring the wide graph's spectrum display. I've resized its window to cover most of the controls beneath the message window, since I only use those controls when I want the program to calculate the azimuth and distance to a grid location. Here's the way I have set it up:

WSJT-X_screenshot.jpg

I wish I could lock that graph in place, so that it doesn't disappear into the background when I click on the main screen, but the graph is in a separate window.

Just be aware of a bug in 2.2.2 that affects the log file (the .adi file). The initial tag isn't written correctly when a new .adi file is created. You may get some invalid tag messages and the log may not import correctly. The fix is easy, just change the incorrect "<eh>" to the correct value of "<eoh>" and save the file.
That's another feature I don't use. ALL.txt is sufficient for my logging purposes, as I'm not a ham and just monitor the transmissions.
 

RocketNJ

Member
Joined
Feb 1, 2003
Messages
934
Location
Pequannock, NJ
What do you have the Bins/Pixel set to in the WSJT-X waterfall? (you need to check the Controls box in upper left of waterfall). I am not seeing you decode higher than 2200 Hz. Increase Bins/Pixel so the waterfall shows 0 -3000 Hz. I run mine at a value of 5. Also adjust you RX bandwidth for 0 - 3000 Hz.
 

GB46

Very radioactive
Joined
Feb 4, 2017
Messages
589
Location
British Columbia, Canada
What do you have the Bins/Pixel set to in the WSJT-X waterfall? (you need to check the Controls box in upper left of waterfall). I am not seeing you decode higher than 2200 Hz. Increase Bins/Pixel so the waterfall shows 0 -3000 Hz. I run mine at a value of 5. Also adjust you RX bandwidth for 0 - 3000 Hz.
I use just the spectrum graph, as I don't like waterfall displays. I've experimented with the Bins/Pixel setting a number of times, and could go way beyond 5000 if I needed to, but judging from the stations appearing in the list, nothing higher than 2200 seems to ever come from the receiver, possibly due to some loss in audio response when connected to my computer. That's why I usually use the value of 3, which takes me up to 2200 Hz.

What I have noticed, however, is a spike at what looks like 60 Hz, which you can see in the screenshot. It's always there, even when there is no activity to decode, so probably due to some AC hum or low frequency RFI getting through. The odd thing is that the spike's amplitude is much higher if the R75's DSP noise reduction feature is on.
 
Top