Re: FT8 time differential - how to resolve?
I commented out the TZ line in .profile and I have WLAN connection, but DT is not better than 1.5!
?
Juergen - DF7II?
|
Re: FT8 time differential - how to resolve?
There is a setting for "delay to transmit" in WSJT-x.? ?Is it possible this has been misconfigured in the software you are using?? Just a guess Gordon KX4Z
They are answering my CQ call. Someone else's CQ call has not been answered yet.
Vince d'Eon VE6LK/AI7LK via <vincedeon= [email protected]> ezt ¨ªrta (id?pont: 2025. m¨¢j. 1., Cs 13:24): Laci are you able to make FT8 contacts when you answer someone else's call?
|
Re: sBitx v3, just installed new MOSFETS and no output
#sBITX_v3
John,
?
Thanks for sharing this wise moment.
?
Reminds me of my first experience with radio that was not successful..
?
I must have been 6 years old, when I thought the music came from the wall outlet..?
Trying to surprise my parents on a sunday morning, I managed to plug in an external speaker with banana plugs in the wall socket with 220 volts..
I've never seen my parents getting fully awake in such short time..???
?
73,
Dick
|
Re: FT8 time differential - how to resolve?
Why not set up a cron to do a time resync every x minutes.
Leo, AI5II
According to my memories (which I regularly record on paper), I remember that a TZ problem appeared in sbitx 3.02, which was solved by the developers.
At this time, the developers entered and the 32-bit and 64-bit systems were divided. The fix appeared as a TZ fix in 3.02.
I remember that at that time I was already using an external application, so the passage of time was visible in three different windows.
That's when the difference appeared, that the correct time was not displayed on sbitx.
The TZ fix covered this. Since the TZ fix was in the sbitx_gtk.c file, I found the difference, comparing it to the sbix_gtk.c file used by the current zbitx v3.052.
My attention was focused on the time setting, and this is missing from the sbitx_gtk.c used by zbitx 3.052.
The modification requires programming work, I just compared the contents of the two files.
Of course there are more differences, but they are necessary due to hardware differences.
If someone requests it, I will upload the corrected TZ file. (not interchangeable!)
--
Gyula HA3HZ
|
Thanks, I was looking on the Buy page.
|
Re: FT8 time differential - how to resolve?
According to my memories (which I regularly record on paper), I remember that a TZ problem appeared in sbitx 3.02, which was solved by the developers.
At this time, the developers entered and the 32-bit and 64-bit systems were divided. The fix appeared as a TZ fix in 3.02.
I remember that at that time I was already using an external application, so the passage of time was visible in three different windows.
That's when the difference appeared, that the correct time was not displayed on sbitx.
The TZ fix covered this. Since the TZ fix was in the sbitx_gtk.c file, I found the difference, comparing it to the sbix_gtk.c file used by the current zbitx v3.052.
My attention was focused on the time setting, and this is missing from the sbitx_gtk.c used by zbitx 3.052.
The modification requires programming work, I just compared the contents of the two files.
Of course there are more differences, but they are necessary due to hardware differences.
If someone requests it, I will upload the corrected TZ file. (not interchangeable!)
--
Gyula HA3HZ
|
ZBitx frequency deviation!
zBitx frequency deviation! Can the frequency deviation be corrected?
|
Re: FT8 time differential - how to resolve?
It appears the timing , besides the initial boot up time sync is not being kept updated.
Initially for ft8,? timing is .very good , around .2 to .4 secs.? ?However, over time it deteriorates to more than 2 secs onward. Using WSJT-X? results in the same issues.
If I issue the ntp time command in a terminal, timing immediately drops to around .2 secs or less. So it appears there is not regular timing adjustments occurring
Help
Joe VE1BWV? On Thu, May 1, 2025, 9:34?a.m. Vince d'Eon VE6LK/AI7LK via <vincedeon= [email protected]> wrote:
I will open an issue in the Github Repo
|
Re: FT8 time differential - how to resolve?
I will open an issue in the Github Repo
|
Re: zBitX not reporting on pskreporter.info
Nope.? I'm missing that too!?
|
Re: Build Instructions for ZBitx Front Panel Software?
Thank you Bhatnagar!
?
This is actually the same thing I tried: Look at the schematic and figure out what the configuration in User_Setup.h has to be, but I always ended up with the gray screen. The Pico was running, I could see that debug messages to the serial port were printed correctly, it was just the display that did not work. When I compared your configuration to mine, this one line made all the difference:
?
#define TFT_SPI_PORT 1
?
Once I added that to my config file, things showed up on the display.?
?
Thank you again!
?
--
Karl Heinz - K5KHK
|
Re: FT8 time differential - how to resolve?
They are answering my CQ call. Someone else's CQ call has not been answered yet.
Vince d'Eon VE6LK/AI7LK via <vincedeon= [email protected]> ezt ¨ªrta (id?pont: 2025. m¨¢j. 1., Cs 13:24):
Laci are you able to make FT8 contacts when you answer someone else's call?
|
Re: zBITX CW Mode - Proof of Concept
Thanks for working on this!!!!? ?ON my sBitx I studied the outputs with a scope and verified that I could get the delays down to 1 ms or so.? ?I also computed the time constants of the stages involved and concluded it should work in that ballpark.? ?33/49 mSec are very very long time periods.? ?For this to be a quality CW rig we have to get those reduced.? ?Thanks again for your work on this!! 73 Gordon KX4Z
at line 655 in sbitx_gtk looks to be where the limits and step rate for the delay is defined
?
{ "#cwdelay", NULL, 1000, -1000, 50, 50, "CW_DELAY", 40, "300", FIELD_NUMBER, FONT_FIELD_VALUE,
? ? "", 50, 1000, 50, CW_CONTROL},
?
If the delay is too long you will miss the start of the message from the operator being contacted.?
?
Near the end of the file sbitx.c is the TX/RX turn around code.? Assuming the bitx is using V4, there are delays totaling 33 ms when changing to transmit and 16 ms changing to receive for a total of 49 ms not responding to the paddles or key.?? My understanding is the zbitx uses diode filter selection instead of relays so there should not be a delay needed for relays to settle.? So I think the only reason for delay is to allow the bidirectional amplifiers to switch.? ( the bitx v6 had an issue where the IF section would oscillate during change over as both rx and tx amplifiers where powered for a very short time causing a burst of unwanted energy transmitted? ).?
?
For a proof of concept you could change all the delays to 1ms.? The transmit output should be checked with a scope to see if any off frequency energy is transmitted.
?
|
Re: zBITX CW Mode - Proof of Concept
at line 655 in sbitx_gtk looks to be where the limits and step rate for the delay is defined
?
{ "#cwdelay", NULL, 1000, -1000, 50, 50, "CW_DELAY", 40, "300", FIELD_NUMBER, FONT_FIELD_VALUE,
? ? "", 50, 1000, 50, CW_CONTROL},
?
If the delay is too long you will miss the start of the message from the operator being contacted.?
?
Near the end of the file sbitx.c is the TX/RX turn around code.? Assuming the bitx is using V4, there are delays totaling 33 ms when changing to transmit and 16 ms changing to receive for a total of 49 ms not responding to the paddles or key.?? My understanding is the zbitx uses diode filter selection instead of relays so there should not be a delay needed for relays to settle.? So I think the only reason for delay is to allow the bidirectional amplifiers to switch.? ( the bitx v6 had an issue where the IF section would oscillate during change over as both rx and tx amplifiers where powered for a very short time causing a burst of unwanted energy transmitted? ).?
?
For a proof of concept you could change all the delays to 1ms.? The transmit output should be checked with a scope to see if any off frequency energy is transmitted.
?
|
Re: FT8 time differential - how to resolve?
Laci are you able to make FT8 contacts when you answer someone else's call?
|
Re: FT8 time differential - how to resolve?
Hello! Unfortunately I also have a time problem. The FT8 transmission is delayed by about 2 seconds. Does anyone have a solution? Thank you, HA8LKT, Laci
|
Re: Build Instructions for ZBitx Front Panel Software?
Karl, The setup file for pico needs information about connections and a few other details. The pin connections are described in the program definitions. I have made it from there and am able to compile it and upload it to pico. I still don't have the zbitx so can not test out fully with sbitx base but I do see the screen and the wait message.? I have uploaded my version of setup for TFT-eSPI in files section under VU2SPF folder. Either you can replace the original (after saving it)? User_Setup_Select.h or put an #include for this file (preferred) there.
The other file platform_local was not required in my build.
Hope it helps.
73s
VU2SPF / Bhatnagar?
|
Re: zBITX CW Mode - Proof of Concept
Jody, I took a closer look at your version of sbitx_gtk.c, in ui_tick(). I see you are trying to call modem_poll() on every tick. But you are asking a lot from the processor because while you are only interested in CW, modem_poll() tries to service all the modes every tick now. I think you need to only call modem_poll every tick for CW mode. And then, you don't have to go read the key again ...
Have you looked at the sbit_gtk.c code for ui_tick()in the 64 bit repository? I really don't know if it will carry over to the zbitx, but it seems very similar to me ... I would love to hear if anyone has actually tried it!
--
Mike KB2ML
|
Re: sBitx v3, just installed new MOSFETS and no output
#sBITX_v3
I did the exact same thing just 2 days ago with a friend, working on his radio. Gordon KX4Z
Please disregard.? Although no "wizz" at troubleshooting, it occurred to me that I'd been using a bench supply that is only 2A at 12V!? Connected up a 4A one I had on a shelf behind me and RF out!? Not the first time I've had a "duh" moment, nor will it be the last.
73 All,
John
N2LCX
|
Re: zBITX CW Mode - Proof of Concept
Jody,
Look down at the bottom of cw_tx_get_sample(), there is this:
if (keyup_count > 0 || keydown_count > 0){
cw_tx_until = millis_now + get_cw_delay();
}
The way to read that is if the key is down, or it is up but we are still counting out the space between symbols, then keep extending our time in transmit mode. Yes the max for cw_delay is 1000 ms, but as long as you keep coming through this code you shouldn't be running up against any time out.
On Wed, Apr 30, 2025 at 07:34 PM, Jody - K3JZD wrote
toggle quoted message
Show quoted text
The problem that I and others are having is the cw_delay, which is capped out
at 1000ms.? Once that times out, and it drops out of Transmit Mode, the
device goes off and gets busy with 'other stuff' and does not always respond
to the next Key Down hardware input in a timely fashion? While my published
experiment, which bypassed a lot of that 'other stuff' was successful, it
wasn't enough.? ?Finding the parameter that caps cw_delay at 1000ms has been
like looking for a needle in a haystack.
--
Mike KB2ML
|
HA3HZ
SP Bhatnagar