locked World-wide Digi DX Contest Problem


Bill N4iQ
 

No - Roger I checked this.  The radio itself is not in split operation when In WSJTX Fake It.    The 590S and 590SG has some differences in the CAT commands and perhaps this is a reason the SG radio operates in Fake it and the 590S does not. Or the problem could be the way OMNIRIG is handling the Fake It from WSJT-X with the 590S.   I am not using OMNIRig with the 590SG.  There are other possibilities but not worth my time trying to figure it out.  Setting up WSJT-X  in Split Mode works for both radios so that is what I am using.   As I said previously "SPLIT works fine now.   During the contest turmoil perhaps I did not have the 590S in split for the WSJT-X split mode?  Totally possible.  Who knows."  I was trying a lot of things to find why Fake It was not working with the S radio and perhaps the 590 radio split was not turned back on for WSJT-X split. 

Think we can move on now.   I truly appreciate everyone's replies and help.  Happy WSJT-X contesting!!!!!

Thanks all - Bill N4iQ



Dick- K9OM
 

For Elecraft K3 users, please advise how the following "Operating Tip" from the WW DIGI website factors into all of this?   

"Select “Fake It” in WSJT-X or equivalent software if your radio has CAT control. This allows the software to dynamically adjust your carrier and audio frequencies such that the audio signal is above 500 Hz where there is no low-pass filtering degradation".


Roger
 

Reino

By split in the rig itself I mean was it set up to transmit on VFO A and receiver of VFO to provide the 500Hz offset?

If the rig itself is in split when he uses Fake it then the offset will be incorrect.

Perhaps Bill can advise? Does WSJTX switch off split on the rig when using Fate it?

Roger
G$HZA

On 03/09/2020 10:26, Reino Talarmo wrote:
Roger
Bill told in his latest mail: "My personal fix is to just use SPLIT."
The problem seems to be in Fake it, but the actual reason seems to be a mystery at least now.
73, Reino OH3mA
-----Original Message-----
From: main@WSJTX.groups.io [mailto:main@WSJTX.groups.io] On Behalf Of groups@...
Sent: 3. syyskuuta 2020 11:21
To: main@WSJTX.groups.io
Subject: Re: [WSJTX] World-wide Digi DX Contest Problem
Bill
Is there any possibility that your rigs themselves are stuck in split so that the rigs themselves are providing the incorrect offset?
Roger
GW4HZA


Reino Talarmo
 

Roger

Bill told in his latest mail: "My personal fix is to just use SPLIT."
The problem seems to be in Fake it, but the actual reason seems to be a mystery at least now.

73, Reino OH3mA

-----Original Message-----
From: main@WSJTX.groups.io [mailto:main@WSJTX.groups.io] On Behalf Of groups@...
Sent: 3. syyskuuta 2020 11:21
To: main@WSJTX.groups.io
Subject: Re: [WSJTX] World-wide Digi DX Contest Problem

Bill

Is there any possibility that your rigs themselves are stuck in split so that the rigs themselves are providing the incorrect offset?

Roger
GW4HZA


Roger
 

Bill

Is there any possibility that your rigs themselves are stuck in split so that the rigs themselves are providing the incorrect offset?

Roger
GW4HZA

On 02/09/2020 20:37, Bill N4iQ wrote:
Bill G4WJS  (lots of Bills here....)
"If I understand correctly, you are saying that while using one of the WSJT-X "Settings->Radio->Split Operating" options you selected a Tx audio offset, say 1000 Hz, on a VFO dial frequency, say 14,074 kHz, and observed the transmitted signal was *not* at 14,075 kHz (lowest tone frequency)"
Yes, this is exactly what I am saying.  And as I pointed out before, this also happened to NU4E during the contest so the problem is not unique to me.  We discussed the problem on the phone during the contest.  We both understand how the "split" operation normally functions.
I was not able to clear this problem by rebooting the computer. Probably I can reproduce this again.  First chance I get I will bring up he SO2R operating conditions and see (using 1000 Hz TX audio selection on the waterfall} exactly what the split VFO is reading while transmitting and then measure the frequency it is actually transmitting.  When the problem is present, both the FT signal and the tune button have the same problem.
"Why do you need to use anything other than "Settings->Radio->Split Operating->None" when you are using an all digital audio chain, and a rig with variable Tx and Rx bandwidth?"
As Frode previously pointed out: The purpose of those two settings (Split - Fake It) is to make sure that the FT8/FT4 audio signal stays between 1500 and 2000 Hz (to avoid the risk of intermodulation in the audio passband). If you select an audio TX frequency (the red "goalpost" in the waterfall window) between 1000 and 1499 Hz, say 1200 Hz, you would expect your signal to be transmitted on e.g. 14,075.2 kHz. To transmit an audio signal in the 1500-2000 Hz sweetspot, the USB "carrier" or dial frequency is therefore moved down from e.g. 14,074  to 14,073.5 kHz. Your audio signal will be 1700 Hz, right in the 1500-2000 Hz sweetspot. Your signal will be transmitted on 14,075.2 kHz, which is the sum of  14,073.5 kHz + 1700 Hz.
WSJT-X User Guide as has this info: " Either method will result in a cleaner transmitted signal, by keeping the Tx audio always in the range 1500 to 2000 Hz so that audio harmonics cannot pass through the Tx sideband filter. "
I will be back later with the info you want, Bill.   I *DO* appreciate you responding.
Bill N4iQ


Bill N4iQ
 

OK - not good, or maybe good, depending on your view. Hi. 

I cannot totally reproduce the problem.  However, I can reproduce the problem with the Fake It but the problem most likely is not WSJT-X.   When using Fake It the VFO should shift to emulate operating SPLIT.  I now see my TS-590S VFO does not shift when using Fake It.   This produces the erroneous shift.  Strangely, Fake It works fine on the other SO2R radio which is also a TS590, however it is a SG model.   The data connection to this radio is direct via the USB virtual COM port.  The 590S is communicating via OMNIRig.  Dunno -  this could be the difference in Fake It VFO operation.  My personal fix is to just use SPLIT.
  
Having said all this, I cannot reproduce the WSJT-X SPLIT problem.   so....  SPLIT works fine now.   During the contest turmoil perhaps I did not have the 590S in split for the WSJT-X split mode?  Totally possible.  Who knows.  I am a senior.......   The one fact that makes me think I am still sane is that my friend had the same problem.  He does not use TS590's.

73 Bill N4iQ


Bill N4iQ
 

Bill G4WJS  (lots of Bills here....)

"If I understand correctly, you are saying that while using one of the WSJT-X "Settings->Radio->Split Operating" options you selected a Tx audio offset, say 1000 Hz, on a VFO dial frequency, say 14,074 kHz, and observed the transmitted signal was *not* at 14,075 kHz (lowest tone frequency)"

Yes, this is exactly what I am saying.  And as I pointed out before, this also happened to NU4E during the contest so the problem is not unique to me.  We discussed the problem on the phone during the contest.  We both understand how the "split" operation normally functions.

I was not able to clear this problem by rebooting the computer. Probably I can reproduce this again.  First chance I get I will bring up he SO2R operating conditions and see (using 1000 Hz TX audio selection on the waterfall} exactly what the split VFO is reading while transmitting and then measure the frequency it is actually transmitting.  When the problem is present, both the FT signal and the tune button have the same problem.

"Why do you need to use anything other than "Settings->Radio->Split Operating->None" when you are using an all digital audio chain, and a rig with variable Tx and Rx bandwidth?"   

As Frode previously pointed out: The purpose of those two settings (Split - Fake It) is to make sure that the FT8/FT4 audio signal stays between 1500 and 2000 Hz (to avoid the risk of intermodulation in the audio passband). If you select an audio TX frequency (the red "goalpost" in the waterfall window) between 1000 and 1499 Hz, say 1200 Hz, you would expect your signal to be transmitted on e.g. 14,075.2 kHz. To transmit an audio signal in the 1500-2000 Hz sweetspot, the USB "carrier" or dial frequency is therefore moved down from e.g. 14,074  to 14,073.5 kHz. Your audio signal will be 1700 Hz, right in the 1500-2000 Hz sweetspot. Your signal will be transmitted on 14,075.2 kHz, which is the sum of  14,073.5 kHz + 1700 Hz.

WSJT-X User Guide as has this info: " Either method will result in a cleaner transmitted signal, by keeping the Tx audio always in the range 1500 to 2000 Hz so that audio harmonics cannot pass through the Tx sideband filter. "

I will be back later with the info you want, Bill.   I DO appreciate you responding.

Bill N4iQ


Bill Somerville
 

On 02/09/2020 12:38, Bill N4iQ wrote:

[Edited Message Follows]

Hello Frode

I appreciate your comments.  The problem I was having was - while split and fake it operated just as you described, the actual desired audio frequency was not correct.  I verified this by monitoring my actual transmitted signal on a second radio running WSJT-X.   The ACTUAL transmitted audio had a constant offset from the desired and expected frequency.  NU4E also experienced this problem at his station.  The offset caused severe problems when operating because the frequency (say 1000 hz) I thought was being sent was not sent.  Instead 1500 Hz was actually being sent.  (I don't remember the actual offset - sorry).  I verified this on the second independent radio and WSJT-X setup.  I was making very few QSOs when this happened, even with 1 KW output.  Once I stopped using split or fake it, I verified the ACTUAL transmitted signal on the independent WSJT-X and the problem was gone.  I started making QSOs at good rates.  I run SO2R FT8 & FT4 so verifying this problem was easy.  My second SO2R radio setup never experienced this problem.

Bill N4iQ
Hi Bill,

your comments do not clarify the issue you are reporting. Audio offsets are irrelevant when checking the behaviour, what matters is the RF frequency the signal is sent on. If I understand correctly, you are saying that while using one of the WSJT-X "Settings->Radio->Split Operating" options you selected a Tx audio offset, say 1000 Hz, on a VFO dial frequency, say 14,074 kHz, and observed the transmitted signal was *not* at 14,075 kHz (lowest tone frequency). If that is what you are observing then please tell us what VFO dial frequency is being selected for your Tx VFO when this happens?

As an aside, why do you need to use anything other than "Settings->Radio->Split Operating->None" when you are using an all digital audio chain, and a rig with variable Tx and Rx bandwidth?

73
Bill
G4WJS.


Bill N4iQ
 
Edited

Hello Frode

I appreciate your comments.  The problem I was having was - while split and fake it operated just as you described, the actual desired audio frequency was not correct.  I verified this by monitoring my actual transmitted signal on a second radio running WSJT-X.   The ACTUAL transmitted audio had a constant offset.  NU4E also experienced this problem at his station.  The offset caused severe problems when operating because the frequency (say 1000 hz) I thought was being sent was not sent.  Instead 1500 Hz was actually being sent.  (I don't remember the actual offset - sorry).  I verified this on the second independent radio and WSJT-X setup.  I was making very few QSOs when this happened, even with 1 KW output.  Once I stopped using split or fake it, I verified the ACTUAL transmitted signal on the independent WSJT-X and the problem was gone.  I started making QSOs at good rates.  I run SO2R FT8 & FT4 so verifying this problem was easy.  My second SO2R radio setup never experienced this problem.

Bill N4iQ


Frode Igland
 

Re the 500 Hz "error" when using "Split" or "Fake it". 
The purpose of those two settings is to make sure that the FT8/FT4 audio signal stays between 1500 and 2000 Hz (to avoid the risk of intermodulation in the audio passband). If you select an audio TX frequency (the red "goalpost" in the waterfall window) between 1000 and 1499 Hz, say 1200 Hz, you would expect your signal to be transmitted on e.g. 14,075.2 kHz. To transmit an audio signal in the 1500-2000 Hz sweetspot, the USB "carrier" or dial frequency is therefore moved down from e.g. 14,074  to 14,073.5 kHz. Your audio signal will be 1700 Hz, right in the 1500-2000 Hz sweetspot. Your signal will be transmitted on 14,075.2 kHz, which is the sum of  14,073.5 kHz + 1700 Hz.

If you use the "None" split setting, you will transmit a 1200 Hz audio signal on the 14,074 kHz dial/"carrier" frequency. 

 Ref. the User Guide section 4.2.

Frode LA6VQ 


Bill N4iQ
 

Hi John - I also experienced the same problem with the decode button staying activated after an initial decode sequence.  Like you I could only clear the problem by rebooting the computer.  WSTJ-X latest version.

Another problem I had was the transmit audio frequencies were off by nearly 500 Hz when using split or fake it.  This started happening after about an hour of normal operation.  All of a sudden I was not making QS0s so I monitored my xmt signals on my other SO2R waterfall and found the symptom.  I had to turn off split or fake it to get correct transmit frequency.  I rebooted my computer several times to reset WSJT-X  but rebooting never cleared this problem.  Strangely my SO2R radio 1 instance of WSJT-X never experience either of these problems.  I was using integrated N1MM.   My friend Matt NU4E also experienced this same audio xmit problem at his station and was not able to clear it as well.  We were both using latest WSJT-X versions with N1MM and Win10.

Bill N4iQ


Jim Shorney
 

I also noticed the decode button staying on during the contest. Investigating, I found that one CPU core was hitting 100% for periods roughly coinciding with the decode button behavior.

I'm using WSJT-X in Kubuntu 18 Linux, compiled from source. I have WSJT-X connected via LAN to DXLabs SpotCollector on a very busy Windows box for the extra highlighting of calls. After I tried a couple of other things I found that clearing the band activity pane with right-CLICK-Erase helped. Periodically erasing the band activity or erasing when the monitor button started to act up seemed to settle things down.

I can't say if I was missing any decodes or not, I was usually getting plenty so how would I know? :)

73

-Jim
NU0C

On Mon, 31 Aug 2020 16:27:19 -0700
"Tony Collett via groups.io" <tony.nbs@...> wrote:

Possibly!

More basic system here v2.2.2 W10 Pro 64bit v2004 on a 3GHz Intel Core2 Duo CPU and just WSJT running alongside N1MM with a K3 over the weekend.
Out of contests I run JTALert and DXLab suite (Commander/Log and Spot) so much more going on on the PC.

I've never noticed the decode light stay on before but it was doing it over the weekend during the Digi DX Contest. Didn't appear to lose me any decodes, other than them appearing after the next period had started from time to time.

I did have WSJT crash out on me twice though which it has never done before. Both times I had just logged a QSO and clicked on a decode of a wanted station - not his CQ but one containing reports. Think one of the decodes included a hashed call but not sure of the other occasion.

73
Tony G4NBS


Tony Collett
 

Possibly!

More basic system here v2.2.2 W10 Pro 64bit v2004 on a 3GHz Intel Core2 Duo CPU and just WSJT running alongside N1MM with a K3 over the weekend.
Out of contests I run JTALert and DXLab suite (Commander/Log and Spot) so much more going on on the PC.

I've never noticed the decode light stay on before but it was doing it over the weekend during the Digi DX Contest. Didn't appear to lose me any decodes, other than them appearing after the next period had started from time to time. 

I did have WSJT crash out on me twice though which it has never done before. Both times I had just logged a QSO and clicked on a decode of a wanted station - not his CQ but one containing reports. Think one of the decodes included a hashed call but not sure of the other occasion.

73
Tony G4NBS


John, K9MM
 

I'm running WSJT-X v2.2.2 on a Windows 10 Pro v1909 system with a FlexRadio 6600M using the slice-master interface to Smart SDR v2.62..

Soon after I enabled the "Special Operating Activity" mode for this contest on Saturday, I encountered a problem where the Decode indicator would remain illuminated at the end of the decode cycle and there would be no more decodes. I believe this is the same problem reported during Field Day. Clearing the decode window and hitting ALT-Z would correct the problem, but only momentarily.

I switched back to normal operating mode, but the problem persisted. I shut down and re-started WSJT-X, but the problem persisted. I shut down WSJT-X, slice-master, and the SDR software, but the problem persisted. Finally, I re-booted the computer, and that fixed the problem as long as I didn't enable Special Operating Activity again.

I didn't see anyone else reporting this problem during the contest. Am I the only one who was seeing it? During Field Day many were reporting it. Is there a work-around? Thanks!

73,

John, K9MM