Hello Rik,
Thank you for your efforts in slowJT9. The latest improvements seem to
be the breakthrough. Very well.
Am 28.12.2018 14:35, schrieb Rik Strobbe:
Hello Stefan,
I switched to 136.8kHz dial:
1215 -9 -0.6 580 5 CQ DK7FC JN49
1225 -4 -0.4 580 5 CQ DK7FC JN49
Thanks for the reprots. I seem to be audible on your side :-)
I will continue to TX a bit longer, maybe Spiros will pick it up,
although this is difficult over that distance in daytime.
About the audio frequency limit for JT9-5:
this is injherent on the decoding technique I am using. The
JT9-5 signal is resampled (speeded up) to a JT9-1 signal and then
decoded by the JT9 decoder of the WSJT suite.
The resampling rate between JT9-5 and JT9-1 is 40960/6912 =
+/- 5.926. In the resampling the frequencies are shifted up by this
factor (so a 500Hz audio tone at JT9-5 becomes 2963Hz at JT9-1.
The upper frequency limit of the JT9 decoder is 5000Hz, that
means 5000/5.926 = 843.75 Hz for the JT9-5 signal.
RR, understood. This is of course sub-optimal because TX stations would
need to re-tune the antenna each time when they want to try another
mode, because they are forced to do QSY. This can be problematic on LF!
So maybe it is possible to integrate an internal upconverter that
solves the problem? I assume many would like to stay at 136 kHz 'dial'
because they are running other instances on WSPR-2, WSPR-15 and so on.
By the way: I have done some testing and have the impression
that the sensitivity of the JT9-5 decoder decreases above 3800kHz. So
for optimal result (at RX end) I would recommend to stay below 640Hz
audio.
OK, so something like
However, it would be possible to allow transmission above 843
Hz. If that would be of any help to you I will incorporate it in the
next version.
Yes please. It won't hurt to have the option.
73, Stefan
Slowing down the spectogram: I will add it to the wish list.
73, Rik ON7YD - OR7T
LF,
I am now TXing JT9-5 on 137.38 kHz. Is there someone who yould
give me a report?
It is good to see that the program did not crash yet on my PC. This is
an improvement. Even the transmisions seem to work now, so far.
Unfortunately the program does still not support TX frequencies above
843 Hz in the JT9-5 mode. So i need to change the (not existing 'dial')
frequency by adding another 600 Hz to SpecLabs upconverter. Actually i
need 12380 Hz to steer my LF hardware to transmit on the desired
frequency. It would be nice if a next version could support all
frequencies up to 24 kHz, so the software becomes more flexible for OMs
using their homemade upconverters, rather than a HF TRX with their
microphone inputs and SSB frequency ranges.
One more feedback: It would be nice if the spectrogram would scroll
slower for the slower modes, and just displays time markers in one
TX/RX interval, i.e. here: 5 minutes.
73, Stefan
Am 28.12.2018 12:15, schrieb DK7FC:
Hello Rik, LF,
Congrats to the latest version! I'm going to install it now and hope it
will work well here too.
What about the JT9-10 mode? It could be very helpful on 2200m these
days.
My VLF TX was running for 3 days without a single interruption. Now i
stopped it and do QSY to 137 kHz. It means that i can receive again
there and hope to see some messages into the monitor field. Also i plan
to transmit, if all works well.
Looking forward to a QSO.
73, Stefan
Am 26.12.2018 21:11, schrieb Rik Strobbe:
Dear all,
I finally found the time to implement JT9-5 and just uploaded v0.9.12.5.
Be ware that this is a ZIP file
that only contains the SlowJT9.exe. In order to use it you need to
install version 0.9.12.4 (if not already done) and replace the
v0.9.14.4 SlowJT9.exe by the one in the ZIP file.
The JT9-5 mode was just briefly
tested but seems to decode fine at -33dB.
All reports on JT9-5 tests are
welcome, as well as any bug report, remark or suggestion.
73, Rik ON7YD - OR7T
Dear all,
Beta-release v0.9.12.4 of SlowJT9 has been uploaded, see http://www.472khz.org/SlowJT9/.
Changes:
- A bug fixed that sometimes caused the application to crash
at the end of a transmission.
- Ability to select double or quadruple tone tone spacing
(Settings->Mode)
- Ability to display the own transmission in the RX frequency
window (Settings->Mode).
I got very few bug reports from the previous version
(0.9.12.3), so either SlowJT9 is running quit stable or people are
getting tired of sending me bug reports ;-)
Bug reports, remarks and suggestions are still welcome.
73, Rik ON7YD - OR7T
Dear all,
I just uploaded a new beta-release of SlowJT9 (v0.9.12.3).
The reporting to debug.txt is been extended (in particular
during TX).
I also fixed a array bound violation that caused crashes once
in a while.
JT9-5 and JT9-10 modes are temporarily disabled, my first goal
is to have a stable version before tackling new features.
Be aware that due to the baudrate changes there is NO
BACKWARD COMPATIBILITY for JT9-2 and JT9-5 to version prior to 0.9.10.
In other words: JT9-2 and JT9-5 send
with v0.9.11 cannot be decoded with versions v0.9.00 - v0.9.03 and vice
versa.
The JT9-1 mode however is still the same and can be sent /
received with all versions (and is WSJT-X compatible).
The purpose of this version is to determine the
stability of the application.
Therefore all test reports, remarks, suggestions and
opinions are most welcome.
I want to thank everybody who gave feedback, suggestions and
bug reports over the past weeks, this was very much appreciated! In
particular Markus, DF6NM, for his patience in explaining me how to
implement windowed sinc interpollation for the resampling.
Whoever is interested can download the SlowJT9 version
0.9.12.3 installation file at http://472khz.org/SlowJT9/.
For now only a Windows version of SlowJT9 is available. But as
the used IDE has cross platform facilities it should be possible to
support other operating systems as well in the future.
|
|
