I have been running Linux WSJT-X on a Ubuntu machine for several months, with excellent results.

Recently, WSJT-X stopped decodes, even though there are very good signal showing in the waterfall.

I am using mode, FT8 and Decode:Deep.  I tried the other Decode type, same result.

I checked the timing on the computer, and it is has a very small time error.  The timing looks good in the waterfall, too.

I removed and re-loaded WSJT-X, no decodes.

The "Activity Monitor", no longer shows a spike up in CPU usage, indicating no decodes

Any suggestions, on what to check/do next?

Alan. WA9WUD

On 02/09/2021 15:25, Gene Horr wrote:

There are ferrites on the USB cable, but that is probably irrelevant because there is no RF being sent.  This is only occurring when changing the TX offset.  During all other activities, including transmission, everything works fine.

The problem is intermittent though there seems to be a correlation with changing the offset quickly multiple times.

As mentioned this only occurs in split mode.  Fake it or no split does not cause this problem.

There are no other libraries loaded.

As mentioned it appears that the software is sending a "change the frequency by 0.000xxx Mhz" rather than "change frequency to X.XXXXXX Mhz".  Is this what is occurring?

It is probably heavily out of date but I do have some experience in tracking down command issues on non-radio RS-232 and RS-485 systems.  I have no idea what is used with USB nowadays.  If anyone has a recommended method I could try to monitor the commands to see whether it is something happening on the computer side.

Hi Gene,

all CAT QSY commands are to an absolute frequency, but WSJT-X does monitor the current frequency and calculates the split offset based on that. Therefore incorrect or corrupt frequency queries could have the effect you are seeing. If you can reproduce the issue easily then the best step forward is to enable CAT diagnostics to track down what is happening. CAT diagnostics will create a very big log if you cannot reproduce the issue in a few operations.

To get a CAT diagnostic log put the attached file into your WSJT-X configuration files directory (on MS Windows that is the same as the log files directory "Menu->File->Open log directory" - if you are on macOS or Linux then let me know and I will tell where to put the log configuration file), restart WSJT-X, carry out the minimum actions to reproduce the issue once, then quit WSJT-X. It will have created a file on your Desktop called "WSJT-X_RigControl.log". Send that file to me (g4wjs <at> classdesign <dot> com) for analysis please?

Once you have sent the log file revert to normal operation by deleting the wsjtx_log_config.ini file.

73
Bill
G4WJS.

There are ferrites on the USB cable, but that is probably irrelevant because there is no RF being sent.  This is only occurring when changing the TX offset.  During all other activities, including transmission, everything works fine.  

The problem is intermittent though there seems to be a correlation with changing the offset quickly multiple times.

As mentioned this only occurs in split mode.  Fake it or no split does not cause this problem.

There are no other libraries loaded.  

As mentioned it appears that the software is sending a "change the frequency by 0.000xxx Mhz" rather than "change frequency to X.XXXXXX Mhz".  Is this what is occurring?

It is probably heavily out of date but I do have some experience in tracking down command issues on non-radio RS-232 and RS-485 systems.  I have no idea what is used with USB nowadays.  If anyone has a recommended method I could try to monitor the commands to see whether it is something happening on the computer side.  

Long explanation, see page 78 of the current manual.

73, and thanks,
Dave (NK7Z)
https://www.nk7z.net
ARRL Volunteer Examiner
ARRL Technical Specialist, RFI
ARRL Asst. Director, NW Division, Technical Resources

On 01/09/2021 22:38, Dick McClain wrote:

In following window, what is the meaning of the "a1" and the "?" which is
shown on some lines on the right side?

Thanks and 73,
Dick,, KD4S

Hi Dick,

the WSJT-X User Guide has the answer to that.

https://physics.princeton.edu/pulsar/k1jt/wsjtx-doc/wsjtx-main-2.4.0.html#_ap_decoding

73
Bill
G4WJS.

In following window, what is the meaning of the "a1" and the "?" which is 
shown on some lines on the right side?

Thanks and 73,
Dick,, KD4S

Thanks Bill.

I will try that.

Kind regards - Terry

> make sure you have "CI-V Transceive Mode" disabled on your IC-7300 Makes no difference. Out of curiosity how would this effect the commands WSJT-X is sending out and how would this make the rig change frequency when there is no other equipment on the chain?

Gene,

it was only a suggestion, CI-V Transceive Mode causes the rig to emit unsolicited CAT traffic which can collide with commands from a PC, that can result in very strange behaviour.

We have not had other reports of incorrect CAT control, as you describe, of the IC-7300 with WSJT-X v2.4.0. Are you using the built in Hamlib library for direct CAT control or one of the other proxy CAT control methods?

73
Bill
G4WJS.

On Tue, Aug 31, 2021 at 10:03 PM, Gene Horr wrote:

7300 when in Rig split mode only.  No problem all other modes.  V2.4.0

When changing the TX offset frequency enough to require a change in the split frequency occasionally the base frequency will change.  It does not happen each time but the effect is cumulative.  For example:

Gene:

You haven't said how you're implementing the CAT control of the 7300 but if it's CI-V via the 7300's USB port then you'll need to ensure that there's no RF getting into the USB connection - that can cause all sorts of random and intermittent errors on the USB bus.  Do you have any ferrites on the USB cable between the PC and the 7300?

Also, have you tried using the 'Fake It' option for split, to see if that makes any difference?

FWIW, I very occasionally saw a similar effect on my 7610 (when using 'Fake It' split mode) - the rig's frequency wouldn't always return to the correct 'base frequency' after the end of a Tx period, it would stay on the frequency selected at the start of the Tx period.  I changed the configuration of the ferrites on the USB connection between PC and rig, to put a more substantial common-mode choke at the PC end of the USB cable, and haven't seen the effect since.

--
Martin G0HDB

> make sure you have "CI-V Transceive Mode" disabled on your IC-7300

Makes no difference. Out of curiosity how would this effect the commands WSJT-X is sending out and how would this make the rig change frequency when there is no other equipment on the chain?

For general HF work I choose to call CQ in the time slot with the strongest signals.
There is a benefit to having all the RX signals near the same level without one or more big signals in your RX.

Bob KM4RL

Thanks

On 31/08/2021 21:35, Gene Horr wrote:

7300 when in Rig split mode only.  No problem all other modes.  V2.4.0

When changing the TX offset frequency enough to require a change in the split frequency occasionally the base frequency will change.  It does not happen each time but the effect is cumulative.  For example:

TX Offset     Base   Split Freq
1500           28074       28074
500             28072.5    28074
2500           28072.5    28072.5
500             28072.5    28071.5
2500           28071       28071
500             28071       28072.5
2500           28071       28071
500             28069.5    28071
2500           28069.5    28069.5
500             28069.5    28068
- Choose 10m from the drop down box -
500             28074       28069.50

As a WAG rather than sending a "set frequency to X, set split to Y" the software is sending "increase/decrease split by Z" but occasionally sends the "decrease by Z" to the base frequency rather than the split.  Choosing the band again in the drop down box appears to send a full "set frequency to X, set split to Y" command and everything goes back to where it should.  Until the TX offset is changed again.

Gene
NG1H

Hi Gene,

make sure you have "CI-V Transceive Mode" disabled on your IC-7300.

73
Bill
G4WJS.

7300 when in Rig split mode only.  No problem all other modes.  V2.4.0

When changing the TX offset frequency enough to require a change in the split frequency occasionally the base frequency will change.  It does not happen each time but the effect is cumulative.  For example:

TX Offset     Base   Split Freq
1500           28074       28074
500             28072.5    28074
2500           28072.5    28072.5
500             28072.5    28071.5
2500           28071       28071
500             28071       28072.5
2500           28071       28071
500             28069.5    28071
2500           28069.5    28069.5
500             28069.5    28068
- Choose 10m from the drop down box - 
500             28074       28069.50

As a WAG rather than sending a "set frequency to X, set split to Y" the software is sending "increase/decrease split by Z" but occasionally sends the "decrease by Z" to the base frequency rather than the split.  Choosing the band again in the drop down box appears to send a full "set frequency to X, set split to Y" command and everything goes back to where it should.  Until the TX offset is changed again.

Gene
NG1H

There will always be a frequency change, as the frequencies of the FT4 'areas' are different from the FT8 areas.   As far as why its missing, have you tied doing a Frequency List reset?

Go into Settings -> Frequencies and RIGHT click in the list, from the menu select RESET.    You should then see a 20m FT4 frequency.   

Neil, KN3ILZ

Neil, KN3ILZ

On 31/08/2021 17:18, Jack Olivieri wrote:

When switching from FT8 to FT4 (under the Mode pulldown), I get a different set of frequencies and there is no choice for 14.080. In fact there is no 20M choice at all.
What am I doing wrong?
Jack WA!FEF

Hi Jack,

FT8 and FT4 use different transmit periods so are not compatible, they are used in different frequency sub-bands. The WSJT-X "Settings->Frequencies->Working Frequencies" table defines the frequencies offered on the main window pop up list, they can be mode and region specific. If you have upgraded from an older version of WSJT-X you probably need to reset the list to the current defaults (right-click the table body and press the "Reset ..." button in the contextual pop up menu that appears.

73
Bill
G4WJS.

When switching from FT8 to FT4 (under the Mode pulldown), I get a different set of frequencies and there is no choice for 14.080. In fact there is no 20M choice at all.
What am I doing wrong?
Jack WA!FEF

Bill -

Thanks for the input. The sound card has analog audio out (of course) but I don't know how much, if any amplification, it has. 

I modified my settings by decreasing the KX3 mic gain to 10, and raising the WSJT-x pwr to -15db (sound card still at -8db). Monitoring both the WSJT output and the sound card output with  headphones I hear a clear warble moderately loud. Operating in these conditions appear to give me more frequent QSOs. The signal reports haven't changed, but maybe my signal is now easier to decode at the receiving end?

Marty
KN6OKN

On 31/08/2021 15:23, groups.io@... wrote:

Apologies if this has been asked before.

I have one copy open for my IC-9700 on VHF/UHF and the other for HF on my IC-7300.

I use the call roster on GridTracker and when I double click on a station it populates the correct copy of WSJT-X and if the station is calling CQ then it toggles Enable-TX to on.

However, particularly if it is an HF link from GT and if I am already in Enable-TX on VHF then the Enable-TX is toggled to off on the VHF copy as the HF is turned on.

I am guessing that this is something that GridtRcaker might be "doing/causing" inadvertantly?

Thanks in advance - Terry

Hi Terry,

you can diagnose which end the problem lies by using the test application message_aggregator that ships with WSJT-X. Run that with two instances of WSJT-X and use it to initiate QSOs to see if there is an cross "interference" between WSJT-X instances.

73
Bill
G4WJS.