Date   

locked Re: What should Radio TX Power Setting really be? #wsjt-x #WSJTX_config

Alan G4ZFQ
 

without adding excessive quantisation noise.
Phil

The only time I have read about quantisation noise is in respect of detecting really weak signals.
Can you point me to something that explains the problem you describe?
How much noise will it add, will it really be important?

73 Alan G45ZFQ


locked Re: Time Synch. #Timesync

William Smith
 

OK, and what is the error message you get when trying to install it on Win7?

I know you know that Win7 support ended January 14, 2020, so the chances that your problem has a root cause of some kind of malware are non-zero. https://www.malwarebytes.com is your friend.

73, Willie N1JBJ

On Jul 24, 2021, at 2:53 PM, Dave Tucker Nu4N <dwtucker19@...> wrote:

Correction. It will load on Windows 10 but not Windows 7.
--
Thanks 73's de NU4N/Dave


locked Re: What should Radio TX Power Setting really be? #wsjt-x #WSJTX_config

Don Roden
 

Bob,

I think the PWR slider *IS* the computer soundcard level.  I set the PWR slide to reliably trip the Signalink PTT and just a little more.  This puts the TX level on the Signalink between 9 and 12 oclock and allows a smooth control from a max of 100 watts down to zero on my TS2000.

Don W4DNR

  

On 2021-07-24 11:59 am, Bob McGraw - K4TAX wrote:

That being the case, then why is the PWR slider in the WSJT-X application available for use in both the TUNE and TRANSMIT mode and why does the application allow for saving those settings?

Thus if the PWR slider is set to a maximum of 0 dB, then the computer sound level must be reduced to a much lower level.  And/or the Line gain of the transceiver be reduced to a much lower level.   In these instances, the granularity of the adjustments becomes significant steps in terms of level adjustments. 

Guess I don't understand your point and thinking. 

73
Bob, K4TAX

On Sat, Jul 24, 2021 at 07:47 AM, David Fanelli wrote:
First, set the "Pwr" slider on the Human/Machine Interface (HMI) to 0dB and never touch it. There are two reasons to keep the audio output of WSJT-X at full power. The first is that you need sufficient drive to activate the VOX of either the RIGblaster or Signalink. The second is quantization noise. Running the audio at full output reduces quantization noise.





locked Re: What should Radio TX Power Setting really be? #wsjt-x #WSJTX_config

Don Roden
 

David,

We are in agreement.

I set the power slider to get a good solid PTT on the Signalink. then set the rig's output power with the Signalink TX control ,,, usually around 9-12 oclock.

Works well.

W4DNR

On 2021-07-24 7:06 am, David Fanelli wrote:

Willie,

I use an TS-2000X as my primary rig for FT8 with a RIGblaster Advantage, but� I have used the same procedure on an Icom IC-781 with a Signalink with excellent results.


First, set the "Pwr" slider on the Human/Machine Interface (HMI) to 0dB and never touch it. There are two reasons to keep the audio output of WSJT-X at full power. The first is that you need sufficient drive to activate the VOX of either the RIGblaster or Signalink. The second is quantization noise. Running the audio at full output reduces quantization noise.

Second, set the output power of the rig to maximum power, and have the output meter read ALC.

Third, set transmit audio level of the soundcard interface to minimum by turning the "XMT LEVEL" (for the RIGblaster) or "TX" (for the Signalink) to minimum and push the "Tune" button on the WSJT-X HMI. Increase the transmit audio level of the soundcard until you start seeing ALC action, and then reduce the the transmit audio level to reduce ALC action to zero.

To set transmit power, always use the soundcard interface transmit audio level to reduce the quantization noise.


73 de kb5pgy

On 7/23/2021 8:22 PM, William Smith wrote:
Hey Bob,
 
I've been looking for a way to do that since I started playing with FT8, but I can't seem to find an inexpensive way to do this properly, can you point me to some equipment and best practices that'll allow me to do that?
 
I seem to be doing OK with 'turn the (audio) Pwr up to full power and back off a bit', but I don't know for sure...
 
Thanks!
 
73, Willie N1JBJ
 

On Jul 23, 2021, at 12:30 PM, Bob McGraw - K4TAX <rmcgraw@...> wrote:

(c) a means to monitor their own signal







locked Re: Time Synch. #Timesync

Alan <n5pa@...>
 

I just downloaded and installed the Meinberg NTP Service on Windows 10 and it downloaded, installed, and runs fine. I like that it runs as a service at startup. I still have my USB GPS Dongle so that I can have a time source when I am operating portable or without an Internet connection. 

73,
Alan Clark, N5PA
Ellisville, MS
Email:  n5pa@...
URL:  http://www.n5pa.com

On Jul 24, 2021, at 1:54 PM, Dave Tucker Nu4N <dwtucker19@...> wrote:

Correction. It will load on Windows 10 but not Windows 7.
--
Thanks 73's de NU4N/Dave



locked Re: Operating portable #WSJTX_config #FT8

 

UK regulations require 6 character location to be sent (or more loquacious data like 15 km E Edinburgh, EH54 9JE or Livingston) when operating away from home. It just means I shouldn’t use FT8 otherwise than at my home location. However now  that conditions are improving can we use other modes, please.

 

I see nothing on the bands other than the 3 kHz band for FT8 weekday afternoons.

 

73 Phil GM3ZZA

 

Sent from Mail for Windows 10

 

From: Reino Talarmo
Sent: 24 July 2021 19:15
To: main@WSJTX.groups.io
Subject: Re: [WSJTX] Operating portable #FT8 #WSJTX_config

 

>If I read the online WSJT help files correctly, the wizards of WSJT indicate that in FT8 Contest Modes, appending the /P suffix will suppress transmission of the 4 character grid locator in the "CQ" message.  If a country's regulatory agency requires that location data be sent when an amateur station transmits using FT8 mode at other than the operator's permitted location, the operator can't also append the /P suffix. 

Hi Karl,

It seems that you got it wrong, it is other way round. Actually it is very easy to test what happens. Just type in Settings – General into My Call: whatever you need to study. E.g. to test whether OH3MA/P is supported I type it in and I will see immediately on the main screen
“CQ TEST OH3MA/P KP20”, when I have selected NA VHF test. Without any special activity I’ll see “CQ OH3MA/P KP20”. If I put into DX Call field any call sign and hit Generate Std Msgs, I’ll see all possible messages. Also /R is supported, but not /1 etc. or /M or anything else.

Note that none of those is supported with non-standard call signs such as OJ0/OH3MA/P.

73, Reino OH3mA

 


--
73 Phil GM3ZZA


locked Re: Time Synch. #Timesync

 

Correction. It will load on Windows 10 but not Windows 7.
--
Thanks 73's de NU4N/Dave


locked Re: Clock Sync #Timesync

Alan B
 

I too thought minpoll of less than 4 was internally rounded up to 4, but I find that minpoll of 3 works (polls every 8 seconds) on my local NTP servers with the version of Meinberg that I'm using. It's a year or so old and works fine so I haven't bothered updating the several machines running it. It is of course not proper to use polling that often with NTP servers other than your own.

The laptop that I run Meinberg on does sleep when it chooses to, and Meinberg recovers pretty quickly, far faster than a reboot.

There are many configurations for NTP servers that will work, but they often have secondary problems, the main one being they don't work offline, they want to see other servers before they will start serving. Comment out all the other servers from your NTP server config file and see if it will start serving anyway once the PPS and NMEA are stable. That's what we need for offline use, whether it be in the field or at home when the network is down.

I also found some Raspbian things that were interfering with the performance like Wifi power saving mode that need to be turned off, at least for nodes using Wifi, otherwise the jitter rises to crazy values.

I included the NTPsec in the document since Frank N6OI chose and used that, and he wrote that section. I didn't follow that path on my install. Lots of choices.

I just checked time.is on the desktop here. It shows 242 milliseconds off, while 2 PPS servers here and one internet servers show 1-3 mS. Seems like time.is doesn't have good accuracy.

73 de w6akb




On Sat, Jul 24, 2021 at 9:47 AM William Smith <w_smith@...> wrote:
In my experience, the Raspian releases tend to lag software packages like gpsd signficantly.

In order to get numbers for "Seen" and "Used" satellites, I had to install gpsd 3.22

When I did apt-get update/ugrade, suddenly I was "upgraded" to 3.17, which doesn't report Seen/Used.  And that took a while to untangle.

apt-cache policy gpsd still shows gpsd 3.17 as the one it'll install.  And 3.17 has a release date of 2017-09-07

For basic functionality it probably works fine, but I like to add monitoring for stuff like sync and number of satellites and SNR and such, so it can send me an email if something goes wrong.

73, Willie N1JBJ



On Jul 24, 2021, at 12:32 PM, Jim Pennino via groups.io <penninojim@...> wrote:

If you find a write up that says you have to compile something or install something outside of the distro, it is guaranteed to be well out of date.

For a Pi and a USB GPS and either Buster or Linux, it is just point gpsd to the correct device.





locked Re: Clock Sync #Timesync

 

With a lot of help from this thread, I have established a GPSD/Raspberry Pi server.  So far it has much less jitter and delay than the GPS pucks and any time source on the internet.  But still if I use ‘time.is’ as a reference, I see a +36 millisecond time offset reported there.  Is that due to internet data lag?  NTP Time Server Monitor reports a 0.285 ms offset.

 

__________

Dan – K4SHQ

CFI/II

 

From: main@WSJTX.groups.io <main@WSJTX.groups.io> On Behalf Of William Smith
Sent: Saturday, July 24, 2021 11:48 AM
To: main@wsjtx.groups.io
Subject: Re: [WSJTX] Clock Sync #Timesync

 

In my experience, the Raspian releases tend to lag software packages like gpsd signficantly.

 

In order to get numbers for "Seen" and "Used" satellites, I had to install gpsd 3.22

 

When I did apt-get update/ugrade, suddenly I was "upgraded" to 3.17, which doesn't report Seen/Used.  And that took a while to untangle.

 

apt-cache policy gpsd still shows gpsd 3.17 as the one it'll install.  And 3.17 has a release date of 2017-09-07

 

For basic functionality it probably works fine, but I like to add monitoring for stuff like sync and number of satellites and SNR and such, so it can send me an email if something goes wrong.

 

73, Willie N1JBJ

 

 



On Jul 24, 2021, at 12:32 PM, Jim Pennino via groups.io <penninojim@...> wrote:

 

If you find a write up that says you have to compile something or install something outside of the distro, it is guaranteed to be well out of date.

For a Pi and a USB GPS and either Buster or Linux, it is just point gpsd to the correct device.

 


locked Re: What should Radio TX Power Setting really be? #wsjt-x #WSJTX_config

 

Because people will complain if they can’t adjust the output of WSJT-X. But seriously, you need to balance the audio levels, get as high an audio level as you can get to the radio without adding excessive quantisation noise. To me, this means generating the digital version using as many bits as it can, and then converting to analogue as high as the radio accepts. In the case of a USB Audio Codec radio like the IC-7300, this means keeping the digital signal as unprocessed as possible between WSJT-X and the radio – I am presuming having 0 dB attenuation in the Windows mixer means it’s not processed. I hope that my PC does not perform a DAC/ADC conversion when using a USB Audio Codec.

 

Then you need to rely on the volume control (again digital) in  the radio not to add too much noise. I have mine set to 25% which I hope just lops the two LS bits off the digital value of the audio.

 

As I said before I then have the Power drive set to 50% and get 36 W out of the radio. Any more than that and I will probably be non-compliant with Ofcom guidelines on public RF exposure.

 

73 Phil GM3ZZA

 

Sent from Mail for Windows 10

 

From: Bob McGraw - K4TAX
Sent: 24 July 2021 18:11
To: main@WSJTX.groups.io
Subject: Re: [WSJTX] What should Radio TX Power Setting really be? #wsjt-x #WSJTX_config #wsjt-x #radio

 

That being the case, then why is the PWR slider in the WSJT-X application available for use in both the TUNE and TRANSMIT mode and why does the application allow for saving those settings?

Thus if the PWR slider is set to a maximum of 0 dB, then the computer sound level must be reduced to a much lower level.  And/or the Line gain of the transceiver be reduced to a much lower level.   In these instances, the granularity of the adjustments becomes significant steps in terms of level adjustments. 

Guess I don't understand your point and thinking. 

73
Bob, K4TAX

On Sat, Jul 24, 2021 at 07:47 AM, David Fanelli wrote:

First, set the "Pwr" slider on the Human/Machine Interface (HMI) to 0dB and never touch it. There are two reasons to keep the audio output of WSJT-X at full power. The first is that you need sufficient drive to activate the VOX of either the RIGblaster or Signalink. The second is quantization noise. Running the audio at full output reduces quantization noise.

 


--
73 Phil GM3ZZA


locked Re: Operating portable #WSJTX_config #FT8

Reino Talarmo
 

>If I read the online WSJT help files correctly, the wizards of WSJT indicate that in FT8 Contest Modes, appending the /P suffix will suppress transmission of the 4 character grid locator in the "CQ" message.  If a country's regulatory agency requires that location data be sent when an amateur station transmits using FT8 mode at other than the operator's permitted location, the operator can't also append the /P suffix. 

Hi Karl,

It seems that you got it wrong, it is other way round. Actually it is very easy to test what happens. Just type in Settings – General into My Call: whatever you need to study. E.g. to test whether OH3MA/P is supported I type it in and I will see immediately on the main screen
“CQ TEST OH3MA/P KP20”, when I have selected NA VHF test. Without any special activity I’ll see “CQ OH3MA/P KP20”. If I put into DX Call field any call sign and hit Generate Std Msgs, I’ll see all possible messages. Also /R is supported, but not /1 etc. or /M or anything else.

Note that none of those is supported with non-standard call signs such as OJ0/OH3MA/P.

73, Reino OH3mA


locked Re: Operating portable #WSJTX_config #FT8

Karl Beckman WA8NVW - NNV5BH
 

Roger -
Regarding use of the callsign suffix /R, the FCC in the USA previously required amateur repeater stations to use the /R suffix if they used a Morse code identifier rather than voice.  There was also a period back in the 1970's when new repeater station authorizations were issued calls beginning with WR.  Both of these USA/FCC policies have since been abandoned, but the use of '/R' in any context is not currently prohibited by the FCC in the USA.
 
The other common use of "/R" in amateur contesting is for 'roaming' stations in some grid square events.  I'm not an avid contester, therefore those details should be explained by those who have the knowledge and experience. In all cases, amateur operators should be aware of limitations on the use of call signs imposed by licensing regulators and the amateur best practices.
 
If I read the online WSJT help files correctly, the wizards of WSJT indicate that in FT8 Contest Modes, appending the /P suffix will suppress transmission of the 4 character grid locator in the "CQ" message.  If a country's regulatory agency requires that location data be sent when an amateur station transmits using FT8 mode at other than the operator's permitted location, the operator can't also append the /P suffix. 
 
Last, thank you, Roger, for serving as a moderator of this group, usually a thankless job at best. Know that your work is noticed and appreciated by many.
--
Karl  WA8NVW  OH
WA8NVW@...
in WSJTX@groups.io


locked Re: What should Radio TX Power Setting really be? #wsjt-x #WSJTX_config

Bob McGraw - K4TAX
 

That being the case, then why is the PWR slider in the WSJT-X application available for use in both the TUNE and TRANSMIT mode and why does the application allow for saving those settings?

Thus if the PWR slider is set to a maximum of 0 dB, then the computer sound level must be reduced to a much lower level.  And/or the Line gain of the transceiver be reduced to a much lower level.   In these instances, the granularity of the adjustments becomes significant steps in terms of level adjustments. 

Guess I don't understand your point and thinking. 

73
Bob, K4TAX


On Sat, Jul 24, 2021 at 07:47 AM, David Fanelli wrote:
First, set the "Pwr" slider on the Human/Machine Interface (HMI) to 0dB and never touch it. There are two reasons to keep the audio output of WSJT-X at full power. The first is that you need sufficient drive to activate the VOX of either the RIGblaster or Signalink. The second is quantization noise. Running the audio at full output reduces quantization noise.


locked Re: What should Radio TX Power Setting really be? #wsjt-x #WSJTX_config

Bob McGraw - K4TAX
 

The RSP-1 or RSP-1a  SDR receiver connected to your computer with HDSDR or other free software will do the job.  You can either terminate the input to the SDR receiver with a 50-ohm resistor or just leave it unconnected as there is likely enough RF around your desk to see your transmit spectrum. 

73
Bob, K4TAX


locked IC7300 Ham Lib error #Cat_RigControl

John , NA2NY
 

Lost split control of IC7300 with 2.5.0-rc3, keep getting hamlib error.
Hamlib error: Command rejected by the rig

icom.c(1791):icom_set_mode_with_data return(-9)

rig.c(4018):rig_set_split_mode return(-9)

rig.c(4303):rig_set_split_freq_mode return(-9) while setting split TX frequency and mode

Fake it works and if I set split on rig it reverts back to no split.

Any one else having problem?

John   NA2NY


locked Re: Clock Sync #Timesync

William Smith
 

In my experience, the Raspian releases tend to lag software packages like gpsd signficantly.

In order to get numbers for "Seen" and "Used" satellites, I had to install gpsd 3.22

When I did apt-get update/ugrade, suddenly I was "upgraded" to 3.17, which doesn't report Seen/Used.  And that took a while to untangle.

apt-cache policy gpsd still shows gpsd 3.17 as the one it'll install.  And 3.17 has a release date of 2017-09-07

For basic functionality it probably works fine, but I like to add monitoring for stuff like sync and number of satellites and SNR and such, so it can send me an email if something goes wrong.

73, Willie N1JBJ



On Jul 24, 2021, at 12:32 PM, Jim Pennino via groups.io <penninojim@...> wrote:

If you find a write up that says you have to compile something or install something outside of the distro, it is guaranteed to be well out of date.

For a Pi and a USB GPS and either Buster or Linux, it is just point gpsd to the correct device.


locked Re: Time Synch. #Timesync

William Smith
 

Dave,

I tried the program on my windows 10 but will load on windows 7 but keep getting the same error.
Sorry, I don't understand, can you try again?

73, Willie N1JBJ


locked Re: Clock Sync #Timesync

Jim Pennino
 

Re Alan W6AKB write up and random thoughts:

Very good job.

I have never tried NTPsec as the primary advantage of that version is security enhancements as you say which are really only relevant if you are running a public server, which I do not. The installation procedure for NTPsec on the Pi seems to me to be overly complex for a newbie user.

Using legacy NTP with gpsd, Pi4 with Buster and the Ultimate HAT, installation is pretty trivial and there are a lot of write ups out there that are obsolete. Basically it is just turn off the serial console in the configuration editor, edit config.txt to turn off bluetooth, assign the correct GPIO pin for PPS, set the baud rate to 9600, and then point gpsd to the correct device. If you find a write up that says you have to compile something or install something outside of the distro, it is guaranteed to be well out of date.

For a Pi and a USB GPS and either Buster or Linux, it is just point gpsd to the correct device.

All my USB GPS machines are getting a long term (several hours) time accuracy of a bit over 2 milliseconds and the Ultimate HAT a bit over 2 microseconds. I have had several USB GPS devices that were much much worse than 2 milliseconds. All of them wound up in the trash.

I would never let a Windows machine go into sleep mode as I have found various strange things happen when it wakes up. If power is an issue, shut down the computer. It will only take NTP a couple of minutes to correct the time well within WSJT requirements on a power up. Perhaps this has been fixed in Windows but I find I can get along just fine without sleep mode.

AFAIK, no version of ntp will accept a minpoll value of less than 4 and will silently change anything less to 4. This can be verified by setting it to something less and running ntpq and looking at the poll value. The minpoll value is a power of 2, thus minpoll 4 says the poll time will be 2^4 or 16 seconds. In my opinion setting the poll time to less than 16 seconds will do little to nothing to time accuracy and one would be better off with minpoll 4 iburst. The iburst command says on startup to make several queries to more quickly get the system time corrected, but in practice with a GPS it seems to make less than a minutes worth of difference in the time required.

In my opinion setting the stratum of a GPS based system is also pointless as any client machine will select or reject the server machine based on the client's evaluation of all the servers configured on the client, selecting which ever server it thinks best, and essentially ignore any stratum number less than 16. A stratum of 16 basically says I have crap for time and use me at your risk.

If you are setting up a network, e.g. as in a field day event, and if you are providing DNS services for it, then on your DNS server configuration for your domain you can do something like:

time                    IN  A           192.168.0.100          ;pi1
time                    IN  A           192.168.0.101          ;pi2
time                    IN  A           192.168.0.102          ;pi3
time                    IN  A           192.168.0.103          ;Fred's windows machine
time                    IN  A           192.168.0.104          ;linux

Then in a client's ntp.conf have a line like:

pool time.mydomain.foo
 
Having your own pool has the advantage of requiring all the clients to only have the pool configured on their machines and changes to the pool become transparent to all the users as long as whoever maintains the DNS server keeps it up to date if things change making it one less thing that has to be coordinated among all the participants.

Jim WB6DKH


locked Re: Time Synch. #Timesync

 

I tried the program on my windows 10 but will load on windows 7 but keep getting the same error.
--
Thanks 73's de NU4N/Dave


locked Re: What should Radio TX Power Setting really be? #wsjt-x #WSJTX_config

Ray
 

ALC indication is not an indicate of ALC threshold or action. In the Elecraft K3, K4 ALC action atarts  beyond 4 bars. In most Icoms, powers setting  sets ALC threshold, so at 100% there is no ALC action below 100 watts out. 

FT8 unlike PSK is a constant carrier signal, ALC is not really an issue.
Read the manual, this will come as a surprise but the person designing the radio does know a bit about it.
Ray
W8LYJ

11221 - 11240 of 38068