QDX on FST4W


Chris Mackerell
 

I reckon that -41 is a glitch for some reason. Doesn't seem
reasonable compared to what else I'm decoding.

Happy to keep on spotting from the Down South 😁

Cheers, Chris

On 2023-01-23 11:17, Rob Robinett wrote:

Hi Chris,

In addition to spotting Paul, you spotted Glenn N6GN at -41 SNR.  I didn't think FST4W-120 decoded at that low a SNR!  Thanks for operating our one Southern Hemisphere WD station.

Rob


Screenshot 2023-01-22 at 2.10.49 PM.png

On Sun, Jan 22, 2023 at 1:36 PM Chris Mackerell <chris@...> wrote:
Yep - I noted a couple of FST4W-120 here in ZL-land recently.

73 Chris ZL2005SWL

On 2023-01-23 08:26, Paul WB6CXC (tech-blog: wb6cxc.com) wrote:
I have just put my modified QDX on 20-meters, running FST4W (-120 for now).  I installed a little reference-clock multiplier (takes 5 or 10 MHz in, delivers 25 MHz to the QDX), and the transmit frequency is rock-solid.  I posted details to my blog: STABILIZING THE QDX
Apparently it's working well, even with my crappy antenna: a 30-meter dipole, one leg horizontal, the other vertical, with an MFJ manual Pi-network antenna tuner bringing the SWR to a reasonable level. 

-- 
Chris Mackerell, Sandy Bay-Marahau Road, Marahau, RD2
Motueka 7197, New Zealand         chris@...


--
Rob Robinett
AI6VN
mobile: +1 650 218 8896


Paul WB6CXC (tech-blog: wb6cxc.com)
 

On Sun, Jan 22, 2023 at 11:31 AM, Paul WB6CXC (tech-blog: wb6cxc.com) wrote:
Apparently it's working well, even with my crappy antenna: a 30-meter dipole, one leg horizontal, the other vertical, with an MFJ manual Pi-network antenna tuner bringing the SWR to a reasonable level. 
I have shifted over to a better dipole:  An off-center-feed multiband that covers 80/40/20/etc, I am still tuning it to lower the SWR to something that won't stress the QDX output transistors.  Signal reports are much better.  I started with the Bodnar GPSDO as the reference, then for a while used a little OCXO-in-a-box I built-- this worked great too.  Now I'm back using the Bodnar, to avoid any long-term frequency drift.


Gwyn Griffiths
 
Edited

Odd about the single -41 dB SNR for Paul WB6CXC at Chris, ZL2005SWL - I have no explanation.
Looking at the spectral spread it was 824 mHz for that spot, but there were others in the 800s with SNR in the -20s.

Paul - your spectral spread to Bryan, KF6ZEO who has a GPSDO Kiwi at 1000 km, has a number of instances of below 20 mHz.
So for tx + ionosphere + rx you clearly have a low spread solution.

Gwyn G3ZIL


Rob Robinett
 

Hi Chris,

In addition to spotting Paul, you spotted Glenn N6GN at -41 SNR.  I didn't think FST4W-120 decoded at that low a SNR!  Thanks for operating our one Southern Hemisphere WD station.

Rob


Screenshot 2023-01-22 at 2.10.49 PM.png


On Sun, Jan 22, 2023 at 1:36 PM Chris Mackerell <chris@...> wrote:
Yep - I noted a couple of FST4W-120 here in ZL-land recently.

73 Chris ZL2005SWL

On 2023-01-23 08:26, Paul WB6CXC (tech-blog: wb6cxc.com) wrote:
I have just put my modified QDX on 20-meters, running FST4W (-120 for now).  I installed a little reference-clock multiplier (takes 5 or 10 MHz in, delivers 25 MHz to the QDX), and the transmit frequency is rock-solid.  I posted details to my blog: STABILIZING THE QDX
Apparently it's working well, even with my crappy antenna: a 30-meter dipole, one leg horizontal, the other vertical, with an MFJ manual Pi-network antenna tuner bringing the SWR to a reasonable level. 

-- 
Chris Mackerell, Sandy Bay-Marahau Road, Marahau, RD2
Motueka 7197, New Zealand         chris@...



--
Rob Robinett
AI6VN
mobile: +1 650 218 8896


Chris Mackerell
 

Yep - I noted a couple of FST4W-120 here in ZL-land recently.

73 Chris ZL2005SWL

On 2023-01-23 08:26, Paul WB6CXC (tech-blog: wb6cxc.com) wrote:
I have just put my modified QDX on 20-meters, running FST4W (-120 for now).  I installed a little reference-clock multiplier (takes 5 or 10 MHz in, delivers 25 MHz to the QDX), and the transmit frequency is rock-solid.  I posted details to my blog: STABILIZING THE QDX
Apparently it's working well, even with my crappy antenna: a 30-meter dipole, one leg horizontal, the other vertical, with an MFJ manual Pi-network antenna tuner bringing the SWR to a reasonable level. 

-- 
Chris Mackerell, Sandy Bay-Marahau Road, Marahau, RD2
Motueka 7197, New Zealand         chris@...


Paul WB6CXC (tech-blog: wb6cxc.com)
 

I have just put my modified QDX on 20-meters, running FST4W (-120 for now).  I installed a little reference-clock multiplier (takes 5 or 10 MHz in, delivers 25 MHz to the QDX), and the transmit frequency is rock-solid.  I posted details to my blog: STABILIZING THE QDX
Apparently it's working well, even with my crappy antenna: a 30-meter dipole, one leg horizontal, the other vertical, with an MFJ manual Pi-network antenna tuner bringing the SWR to a reasonable level. 


Glenn Elmore
 

Be patient.  The present storming is wreaking havoc with spectral spreading.  K6PZB, WB7ABP and N6GN are all transmitting GPSDO referenced FST4W but  the ionosphere is particularly messy. There may be others transmitting well disciplined signals as well.

You can use a Grafana WSPR and FST4W comparisons dashboard to see how/if these spot each other. For the last 24 hours, at least, 40m one-hop spreading  has been worse than corresponding 20m spreading a few days ago.


On 8/18/22 12:30, Bruce KX4AZ wrote:

I've manually hopped between 40/30/20m today using the Airpsy HF+ and WSJT-x for FST4W 120 second mode.  So far the only spots have been for 40m (mainly Glenn's N6GN setup)...perhaps the Airspy oscillator has too much drift for the higher frequencies.


Bruce KX4AZ
 

I've manually hopped between 40/30/20m today using the Airpsy HF+ and WSJT-x for FST4W 120 second mode.  So far the only spots have been for 40m (mainly Glenn's N6GN setup)...perhaps the Airspy oscillator has too much drift for the higher frequencies.


Glenn Elmore
 

Andrew,

Several spots with some interesting spreading reports already. It will be interesting to see what happens when the storming subsides. Thanks for spotting and uploading so this can be collected.

Glenn

On 8/18/22 07:38, Andrew Cowan wrote:

Glenn
its set here with a 4 square to USA through our night-time.
Andrew GM0UDL


Andrew Cowan
 

Glenn
its set here with a 4 square to USA through our night-time.
Andrew GM0UDL


Bruce KX4AZ
 

On Wed, Aug 17, 2022 at 07:14 PM, Elmer G Musser wrote:
Set WSPR mode select band then select Mode again FST4W 

 
To make it simpler to switch bands for FST4W in WSJT-x, it is simple to add the mode/frequency to the chart in the settings menu.  But there is still no ability to enable a hopping mode as with WSPR. 


wayne roth
 
Edited

Done (W2, F2, F5)
WA2N


Elmer G Musser
 

Set WSPR mode select band then select Mode again FST4W 


On Aug 17, 2022, at 5:01 PM, Bruce KX4AZ <bruce@...> wrote:

I have WSJT-x FST4W decoding as 'KX4AZ/T1' on 40 meters now and I see there are several spots of N6GN so far.  Tuned to the same frequency that WSJT-x uses for WSPR on 40 meters.  I'll leave it running overnight to see if there are any other stations I can decode.  Same antenna connected as for the 'KX4AZ/T' spotter.


Glenn Elmore
 

In this vein, As we are just now doing preliminary testing on FST4W on 40m, I would be pleased to have anyone running wsprdaemon to spot WSPR  on 40m modify their schedule to include "<receiver>,40,W2:F2:F5" in order to secode FST4W modes "F2" -120 seconds, and "F5" -300 seconds. 

According to Rob this is not taxing even for an RPI and adds very little overhead.  Having more spotters and spotters worldwide and many hops away would be useful.

Thanks in advance to any so inclined  !

Glenn n6gn

On 8/17/22 15:01, Bruce KX4AZ wrote:

I have WSJT-x FST4W decoding as 'KX4AZ/T1' on 40 meters now and I see there are several spots of N6GN so far.  Tuned to the same frequency that WSJT-x uses for WSPR on 40 meters.  I'll leave it running overnight to see if there are any other stations I can decode.  Same antenna connected as for the 'KX4AZ/T' spotter.


Bruce KX4AZ
 

I have WSJT-x FST4W decoding as 'KX4AZ/T1' on 40 meters now and I see there are several spots of N6GN so far.  Tuned to the same frequency that WSJT-x uses for WSPR on 40 meters.  I'll leave it running overnight to see if there are any other stations I can decode.  Same antenna connected as for the 'KX4AZ/T' spotter.


Glenn Elmore
 

WSJT-X really doesn't care what the source/sink of it's audio encoding is, there's even evidence that FST4W may be beneficial at UHF and microwave. 

Here's the last 24 hours of 40m FST4W from N6GN


and the previous day or so on 20m FST4W


There's still a lot to be learned about when, where and how the benefits can be obtained, especially the longer modes are a bigger challenge, but it is definitely not just useful for LF/MF

Glenn


On 8/17/22 14:20, Jim Lill wrote:

I may be wrong, but doesn't wsjt-x only do FST4W 2200-160m, nothing higher?

-Jim

On 8/17/22 16:10, Bruce KX4AZ wrote:
I don't believe my Raspi4 currently has enough horsepower to add FST4W decoding, but I do have a spare Airspy HF+/WSJT-x combination running on a Win10 PC.  So time for a stupid question....to get the WSJT-x to decode FST4 on say, 40 meters, is it just a matter of first starting it up in WSPR receive mode on 40m, then switching to FST4W mode without changing the frequency?  Or do I need to change to a different frequency for that mode.  Sorry for asking, I haven't been following the FST4W discussions in detail.


Jim Lill
 

I may be wrong, but doesn't wsjt-x only do FST4W 2200-160m, nothing higher?

-Jim

On 8/17/22 16:10, Bruce KX4AZ wrote:

I don't believe my Raspi4 currently has enough horsepower to add FST4W decoding, but I do have a spare Airspy HF+/WSJT-x combination running on a Win10 PC.  So time for a stupid question....to get the WSJT-x to decode FST4 on say, 40 meters, is it just a matter of first starting it up in WSPR receive mode on 40m, then switching to FST4W mode without changing the frequency?  Or do I need to change to a different frequency for that mode.  Sorry for asking, I haven't been following the FST4W discussions in detail.


Bruce KX4AZ
 

I don't believe my Raspi4 currently has enough horsepower to add FST4W decoding, but I do have a spare Airspy HF+/WSJT-x combination running on a Win10 PC.  So time for a stupid question....to get the WSJT-x to decode FST4 on say, 40 meters, is it just a matter of first starting it up in WSPR receive mode on 40m, then switching to FST4W mode without changing the frequency?  Or do I need to change to a different frequency for that mode.  Sorry for asking, I haven't been following the FST4W discussions in detail.


Rolf Ekstrand
 

Thanks Rob,

I suspected that something was not right with the QDX  as I saw multiple decodes on the same station and cycle on Gwyn's Grafana chart.    OK back to the drawing board. Then again the QDX was designed for the fast FT 8 cycles and not FST4W. Let's see if it can be fixed or we have to look for something else to get FSTW4 going

On another note when "fiddling" with my set up here I crashed the Kiwi/wd yesterday as it suddenly a time after the wd update did not connect to my network.  Well to make a long story short I re-installed the WD and the puter operating system  and then I re installed the Kiwi operating system.  It took me a while but I got it up and running again, and it runs fine now.  Just another learning lesson for this old fart  He he.   

Rolf K9DZT


Rob Robinett
 

Hi Rolf,

It is good to have another FST4W beacon on air.  

However the spectral spreading of your signal in reports from WD listeners suggests that you will receive relatively few reports.  The 'metric' column in the table below documents the spreading in millihertz and those receive stations see other remote (i.e. propagated via the ionosphere) FST4W beacons with much lower numbers.

Glenn N6GN and Gwyn G3ZIL have been studying this spreading issue and problems with the QDX as FST4W transmitter in the hope that problem can be addressed.

Rob

tutorial=> select time,mode, "SNR",metric, tx_call, rx_id,receiver,"tx_dBm",freq from wsprdaemon_spots_s where tx_call='K9DZT' and mode!=2 and freq>1 and freq<21 order by time  desc limit 30;
        time         | mode | SNR | metric | tx_call |   rx_id   |  receiver   | tx_dBm |   freq
---------------------+------+-----+--------+---------+-----------+-------------+--------+----------
 2022-08-16 12:24:00 |    3 | -13 |    315 | K9DZT   | W3ENR     | BURLEYMAN   |     33 | 7.040015
 2022-08-16 12:24:00 |    3 | -14 |    382 | K9DZT   | W3ENR     | BIGEARS     |     33 | 7.040015
 2022-08-16 12:24:00 |    3 | -19 |    441 | K9DZT   | KJ6MKI    | KIWI_0      |     33 | 7.040015
 2022-08-16 12:24:00 |    3 | -21 |    460 | K9DZT   | KFS       | KIWI_Omni_A |     33 | 7.040015
 2022-08-16 12:24:00 |    3 | -20 |    439 | K9DZT   | KFS       | KIWI_SE     |     33 | 7.040015
 2022-08-16 12:24:00 |    3 | -15 |    410 | K9DZT   | W3PM      | KIWI_0      |     33 | 7.040015
 2022-08-16 12:24:00 |    3 | -15 |    449 | K9DZT   | W3PM      | KIWI_1      |     33 |  7.04002
 2022-08-16 12:24:00 |    3 |  -5 |    377 | K9DZT   | KA7OEI-1  | NUT_KIWI4   |     33 | 7.040014
 2022-08-16 12:24:00 |    3 | -11 |    414 | K9DZT   | KA7OEI-1  | NUT_KIWI2   |     33 | 7.040015
 2022-08-16 12:24:00 |    3 | -21 |    552 | K9DZT   | KP4MD     | KIWI_0      |     33 | 7.040014
 2022-08-16 12:24:00 |    3 | -17 |    463 | K9DZT   | AI6VN/KH6 | maui75      |     33 | 7.040015
 2022-08-16 12:24:00 |    3 | -18 |    504 | K9DZT   | KPH       | KPH_HF_76   |     33 | 7.040016
 2022-08-16 12:24:00 |    3 | -23 |    627 | K9DZT   | ND7M      | KIWI_1      |     33 | 7.040015
 2022-08-16 12:10:00 |    3 | -17 |    431 | K9DZT   | W3PM      | KIWI_0      |     33 | 7.040015
 2022-08-16 12:10:00 |    3 | -15 |    448 | K9DZT   | W3PM      | KIWI_1      |     33 | 7.040008
 2022-08-16 12:10:00 |    3 | -12 |    368 | K9DZT   | W3ENR     | BURLEYMAN   |     33 | 7.040015
 2022-08-16 12:10:00 |    3 | -13 |    379 | K9DZT   | W3ENR     | BIGEARS     |     33 | 7.040015
 2022-08-16 12:10:00 |    3 | -17 |    515 | K9DZT   | KJ6MKI    | KIWI_0      |     33 | 7.040015
 2022-08-16 12:10:00 |    3 | -19 |    517 | K9DZT   | KFS       | KIWI_Omni_A |     33 | 7.040015
 2022-08-16 12:10:00 |    3 | -19 |    606 | K9DZT   | KFS       | KIWI_SE     |     33 | 7.040015
 2022-08-16 12:10:00 |    3 |  -3 |    342 | K9DZT   | KA7OEI-1  | NUT_KIWI4   |     33 | 7.040014
 2022-08-16 12:10:00 |    3 | -10 |    318 | K9DZT   | KA7OEI-1  | NUT_KIWI2   |     33 | 7.040015
 2022-08-16 12:10:00 |    3 | -22 |    532 | K9DZT   | KP4MD     | KIWI_0      |     33 | 7.040014
 2022-08-16 12:10:00 |    3 | -15 |    534 | K9DZT   | AI6VN/KH6 | maui75      |     33 | 7.040015
 2022-08-16 12:10:00 |    3 | -16 |    677 | K9DZT   | KPH       | KPH_HF_76   |     33 | 7.040015
 2022-08-16 11:56:00 |    3 |  -3 |    343 | K9DZT   | KA7OEI-1  | NUT_KIWI4   |     33 | 7.040013
 2022-08-16 11:56:00 |    3 | -10 |    381 | K9DZT   | KA7OEI-1  | NUT_KIWI2   |     33 | 7.040015
 2022-08-16 11:56:00 |    3 | -15 |    412 | K9DZT   | W3ENR     | BURLEYMAN   |     33 | 7.040015
 2022-08-16 11:56:00 |    3 | -16 |    409 | K9DZT   | W3ENR     | BIGEARS     |     33 | 7.040015
 2022-08-16 11:56:00 |    3 | -12 |    738 | K9DZT   | KJ6MKI    | KIWI_0      |     33 | 7.040015
(30 rows)

tutorial=>




On Tue, Aug 16, 2022 at 6:15 AM Rolf Ekstrand <rekstrand@...> wrote:
Greetings y'all,

I am now running my QDX on 20 and 40  usually early evening and after sunrise.  I am using wsjt-x  and a "surplus" raspi 3B  as the easy way to get x-mitting  FST4W.  I did reduce the power ( lower the operating voltage) so the QDX is now putting out about 1.5 W  on 20 and 2 W on 40.  This as the QDX cooling is designed for FT 8 and is getting a tad hot at 5 W WSPR and FST4W. For that reason I have mostly run F:2    The antenna is a a ground mounted R7 for now.   The only drawback is the reporting as WSJT-x decodes FST4W but WSPR.net reports the decodes as WSPR.   

Rolf K9DZT





--
Rob Robinett
AI6VN
mobile: +1 650 218 8896