Hi Jos

KA1GT and I have been looking into this a little further. I have had a look
at the code, searching for the string   Qt::ControlModifier,  which at
least is one way I think of detecting if the CTRL key has been pressed.

I see instances of this in the code, but did not find anything that to my
understanding related to CAT control.

Bob reported yesterday that he also can tune radio knobs and see the sked
frequency change.He tested two rigs, an IC7100 and an FT897D.

Quoting some of Bob's observations:


"With the IC7100 You can indeed change the sked frequency without using the
control key if you spin the frequency control dial on the radio
continuously.  If you make a small change using the dial and stop, it will
reset after a second or so and the sked frequency will not change. However
if you continuously rotate the dial, it will keep occasionally glitching
and the sked frequency will change.

The ft-897D is worse then the IC-7100. Most of the time changing the Rx
frequency with the dial shifts the sked frequency. "

This is similar to what I have seen with the IC735, but I didn't look at it
as closely as Bob.

One question he had, was what PIC are you using for your CAT/synthesiser
control?  He has seen instances where lack of precision in FP variables led
to some inconsistencies in handling large numbers  (with an Arduino).

73

Charlie

On Sat, 28 Sept 2024 at 12:13, Jos F.M. VanderList via wsjt-devel <
wsjt-devel@lists.sourceforge.net> wrote:

> Good morning Charlie,
> Yes, the logfile that I sent to Mike covered a large part of the day
> including the moment that the sked frequency started to drift away.
> I like your idea to test with another radio. Of course I don't have a
> second homemade transceiver for 144 MHz with CAT-control, so maybe your
> idea of using an SDR is worth investigating.
> Still, if such a test shows that my station setup is unique, then the
> question remains what CAT-commands or sequence of events and CAT-commands
> can lead to WSJT-X changing the sked frequency. I have even considered to
> program a PIC processor to behave as an ideal transceiver to test if the
> same behavior occurs.
> I have a puzzle to solve and will inform you of what I find. Thanks again
> for helping me!
> 73, Jos, PA0JOZ
>
> On Sat, 28 Sept 2024 at 07:07, Charles Suckling via wsjt-devel <
> wsjt-devel@lists.sourceforge.net> wrote:
>
>> Hi Jos
>>
>> It does appear your situation is unique.  As far as I'm aware, the rig
>> control software within WSJT-X itself has not changed for years, and this
>> problem has never been reported before. As I mentioned yesterday, WSJT-X
>> has been used for EME on 24GHz where Doppler shifts often exceed 19kHz.  It
>> has also been used successfully on 47GHz, so I'm confident that the issue
>> is not related to the magnitude of the Doppler shift.
>>
>> I wonder if you have some other radio you could test with? This could be
>> for example an SDR controlled by software such as HDSDR or SDR Console.
>> Both of these programs can be connected to WSJT-X,  since they both have
>> emulators (using a virtual com port connection such as com0com).  Audio can
>> be routed via virtual audio cables,
>>
>> Did the data that Mike studied cover the case during which the sked
>> frequency was changing?
>>
>> 73
>>
>> Charlie
>>
>> On Fri, 27 Sept 2024 at 16:50, Jos F.M. VanderList via wsjt-devel <
>> wsjt-devel@lists.sourceforge.net> wrote:
>>
>>> Good afternoon all,
>>> While monitoring the DL0SHF beacon at 10 GHz I noticed strange behavior
>>> of the Doppler tracking.
>>> I set the sked frequency initially to 10368.024 MHz, with Doppler
>>> tracking "On DX Echo".
>>> DL0SHF appears initially at its normal place in the waterfall.
>>> However, when the (absolute) value of the DX Doppler exceeds
>>> approximately 19 kHz the sked frequency gets changed in the Astro window,
>>> without me touching anything. The change is steadily downward with 1 or 2
>>> Hz at a time. At the same time the VFO frequency of the transceiver is also
>>> ordered to a lower frequency. And the result is that the beacon is drifting
>>> slowly upwards in the waterfall.
>>> I also noticed that in the situation that WSJT-X alters the sked
>>> frequency it is sometimes possible to adjust the sked frequency by turning
>>> the tuning knob of the transceiver, without pressing CTRL.
>>> Mike, W9MDB, checked the communication between WSJT-X and the
>>> transceiver and he didn't find anything strange there. So to me it seems
>>> that the problem is in WSJT-X itself.
>>> Before anyone asks: all my oscillators are locked to my 10 MHz GPSDO.
>>> And the transceiver frequency follows the frequency ordered by WSJT-X.
>>> I'll give some examples from this afternoon. Sked frequency is the sked
>>> frequency in the Astro window. Transceiver frequency is the frequency as
>>> shown below the Monitor button in the main WSJT-X window. And DXDop is as
>>> shown in the Astro window.
>>> DXDop:    TRX freq.:           Sked freq.:
>>> -17000    10368.007000    10368.024000
>>> -18000    10368.006000    10368.024000
>>> -19116    10368.004877    10368.023993
>>> -19500    10368.004232    10368.023932
>>> -19999    10368.003480    10368.023479
>>> -20497    10368.002335    10368.022832
>>> Can anyone shed some light on this? Why is WSJT-X altering the sked
>>> frequency if the Doppler is more than 19 kHz? And why isn't the Doppler
>>> tracking keeping the beacon stable in the waterfall? Is it a bug, am I
>>> doing something wrong, is my understanding of the Doppler tracking wrong?
>>> 73, Jos, PA0JOZ
>>> _______________________________________________
>>> wsjt-devel mailing list
>>> wsjt-devel@lists.sourceforge.net
>>> https://lists.sourceforge.net/lists/listinfo/wsjt-devel
>>>
>> _______________________________________________
>> wsjt-devel mailing list
>> wsjt-devel@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/wsjt-devel
>>
> _______________________________________________
> wsjt-devel mailing list
> wsjt-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/wsjt-devel
>
_______________________________________________
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Reply via email to