Can not run the Calibration cycle #FreqCal


Bill - NO4ON
 

I have a new IC-7610 that I want to run WSJT-X on.  I do not have the best antenna, Diamond BB7V, but it will suffice in the short term to get me up and running.  I have 2 operation instances of WSJT-X.  One on a iMac, and one on a Raspberry Pi.  I have made contact on both installs.  I have created a FreqCal configuration on each as per the instruction call for.  I can run the calibration numbers on both by checking the Execute frequency calibration cycle in the tools menu.  On each when I stop measuring, and run the Solve for calibration parameters I get a popup box with Calibration Error. Check fmt.all for possible bad data.
I have run the process muntiple time on both machines.  I have deleted the fmt.all on each and always get the same results.  I look at the fmt.all file on each system and I can not see anything obvious.  I do not know what the data is supposed to be, but there are no unreadable characters that I see.  I seem to be missing something fairly obvious, but I just cannot see it.  Any suggestions would be greatly appreciated.  I will also post this on the Facebook app.

Bill Norman
NO4ON


Michael Black
 

Try deleting the fmt.all file and test again.
Did you check the file for bad data like it said?  It's a pretty simple format.

Mike W9MDB




On Wednesday, August 18, 2021, 09:46:18 AM CDT, Bill Norman <no4on999@...> wrote:


I have a new IC-7610 that I want to run WSJT-X on.  I do not have the best antenna, Diamond BB7V, but it will suffice in the short term to get me up and running.  I have 2 operation instances of WSJT-X.  One on a iMac, and one on a Raspberry Pi.  I have made contact on both installs.  I have created a FreqCal configuration on each as per the instruction call for.  I can run the calibration numbers on both by checking the Execute frequency calibration cycle in the tools menu.  On each when I stop measuring, and run the Solve for calibration parameters I get a popup box with Calibration Error. Check fmt.all for possible bad data.
I have run the process muntiple time on both machines.  I have deleted the fmt.all on each and always get the same results.  I look at the fmt.all file on each system and I can not see anything obvious.  I do not know what the data is supposed to be, but there are no unreadable characters that I see.  I seem to be missing something fairly obvious, but I just cannot see it.  Any suggestions would be greatly appreciated.  I will also post this on the Facebook app.

Bill Norman
NO4ON




Bill - NO4ON
 

As I said in my post, I have deleted the fmt.all file many times and rerun the procedure on both platforms many times.  The file structure is a simple txt file, but I see nothing odd about any of it.

Bill NO4ON


Michael Black
 

How about posting your file?

Mike




On Wednesday, August 18, 2021, 11:15:31 AM CDT, Bill Norman <no4on999@...> wrote:


As I said in my post, I have deleted the fmt.all file many times and rerun the procedure on both platforms many times.  The file structure is a simple txt file, but I see nothing odd about any of it.

Bill NO4ON




Bill Somerville
 

On 18/08/2021 15:04, Bill Norman wrote:
I have a new IC-7610 that I want to run WSJT-X on.  I do not have the best antenna, Diamond BB7V, but it will suffice in the short term to get me up and running.  I have 2 operation instances of WSJT-X.  One on a iMac, and one on a Raspberry Pi.  I have made contact on both installs.  I have created a FreqCal configuration on each as per the instruction call for.  I can run the calibration numbers on both by checking the Execute frequency calibration cycle in the tools menu.  On each when I stop measuring, and run the Solve for calibration parameters I get a popup box with Calibration Error. Check fmt.all for possible bad data.
I have run the process muntiple time on both machines.  I have deleted the fmt.all on each and always get the same results.  I look at the fmt.all file on each system and I can not see anything obvious.  I do not know what the data is supposed to be, but there are no unreadable characters that I see.  I seem to be missing something fairly obvious, but I just cannot see it.  Any suggestions would be greatly appreciated.  I will also post this on the Facebook app.

Bill Norman
NO4ON
Hi Bill,

are there enough good readings for each frequency measured after you remove the lines with '*' characters on the end (the bad ones)?

73
Bill
G4WJS.


Bill - NO4ON
 

I do not know what a good record looks like.  I have posted the file.  Why do some of the records have an "*" at the end.  Do you have a reference on what a good record is supposed to look like?  How many records does it take and what makes them bad?

73
Bill NO4ON


Reino Talarmo
 

I do not know what a good record looks like.  I have posted the file.  Why do some of the records have an "*" at the end.  Do you have a reference on what a good record is supposed to look like?  How many records does it take and what makes them bad?

Hi Bill,

That ’*’ as least in your case indicates that signal to noise ratio is too low for a reliable frequency calibration. Perhaps your antenna system is not suitable for other than 5 and 10 MHz signals.

73, Reino OH3mA


 

Hi Reino,

 

Not everyone can have the best antenna system ever, I’ve only got a small suburban back garden and under the new ICNIRP can’t get big antennas up (also it will offend the neighbours).  We should be able to decode signals no matter how they are received. The poorer the antenna the lower the noise MAY be.

 

73 Phil GM3ZZA

 

Sent from Mail for Windows

 

From: Reino Talarmo
Sent: 18 August 2021 19:45
To: main@WSJTX.groups.io
Subject: Re: [WSJTX] Can not run the Calibration cycle #wsft-x #FreqCal

 

I do not know what a good record looks like.  I have posted the file.  Why do some of the records have an "*" at the end.  Do you have a reference on what a good record is supposed to look like?  How many records does it take and what makes them bad?

Hi Bill,

That ’*’ as least in your case indicates that signal to noise ratio is too low for a reliable frequency calibration. Perhaps your antenna system is not suitable for other than 5 and 10 MHz signals.

73, Reino OH3mA

 


--
73 Phil GM3ZZA


Kenneth Williams
 

I thought I would jump in with my own experience and solution.

When I first attempted frequency calibration, I tried using WWV but I would get so many poor readings across the different WWV frequencies that I could never get a good run. This was certainly impacted by the fact that the only antenna connected to my rig was a 40m inverted V.   So, instead of fighting this, I connected my radio to a signal generator, modified the fcal list to take a reading at 1 MHz steps from 1 to 20 MHz and let it rip.  I ended up with usable readings at all points and I was able to execute the calibration with the information that I had gathered.  I doubt that what I did is new but I thought that I would toss this onto the thread.

Of course, before attempting this, it is crucial to verify that your signal generator is not going to overdrive the antenna input to your rig, or to use an attenuator to bring the signal to within a safe range.  Also, WWV will always give a much more accurate signal so unless one is using a GPSDO as a reference to the signal generator,  your calibration will be subject to the accuracy of your signal generator.

Good luck

Ken
KC6PUQ


William Smith
 

I don't think Reino was insulting the OPs antenna system, just stating the fact that he doesn't have enough signal from that antenna to generate a useful frequency calibration file.

While WSJT-x can receive signals way below the noise floor, it's not magical, if there's no discernable signal (especially when trying to get a frequency standard), then there's no useful information to be gleaned from that (lack of) signal.

Try it at night, or in the early morning, and maybe you'll get better results?

OP: pskreporter.com shows you are getting out, so don't worry too much about exact frequency, FT8 is fairly tolerant of frequency accuracy and stability (and even time, though it's easy enough to be within a couple hundred ms of realtime at https://time.is )

Enjoy!

73, Willie N1JBJ


On Aug 18, 2021, at 3:01 PM, Philip Rose via groups.io <gm3zza@...> wrote:

Not everyone can have the best antenna system ever, I’ve only got a small suburban back garden and under the new ICNIRP can’t get big antennas up (also it will offend the neighbours).  We should be able to decode signals no matter how they are received. The poorer the antenna the lower the noise MAY be.


Bill - NO4ON
 

I really appreciate all of the help.  I modified the all.fmt file and deleted all of the records that had the * at the end.  Then when I processed it, it worked fine.  The people who wrote this software are very smart people. I find it hard to believe that the let a record be recorded that stops the process and produces a vague error message.  It would be very easy to just not record the records that were not usable so the process would not fail with no visible solution.

Thanks again, I would not have been able to solve the process without all of your help.

Bill - NO4ON