Re: [wsjt-devel] Possible feature request, time skew notification?

[Jim Pennino via wsjt-devel]

Get a VK-162 G-Mouse available on Amazon from various vendors for about $15.
It is recognized by Windows and creates a virtual com port, making it usable by 
ntp, which the WSJT installation documentation recommends being installed.
No RFI problems.

Add the following two lines to ntp.conf:

server 127.127.20.n minpoll 4
fudge 127.127.20.n time1 0.030

Where n is the COM port number that Device Manager says is the GPS puck.
There is a long thread about this on the main group titled Clock Sync.
m...@wsjtx.groups.io | Clock Sync

| 
| 
|  | 
m...@wsjtx.groups.io | Clock Sync


 |

 |

 |


Jim WB6DKH


___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] Possible feature request, time skew notification?

[John Nogatch via wsjt-devel]

> 2. In outdoor setting (sounds like it is more applicable) you can get a 
> trusted time from a GPS receiver.

I had some experience with this recently, using this USB GPS
https://www.ebay.com/itm/284227196983
with GPS2time 
https://vk4adc.com/web/index.php/software-projects/55-vk4adc-utils/181-gps2time
on a Win10 laptop.

We got this to work well, but there were some problems:
1. An older USB GPS works on linux, but because it has the Prolific
chip, it does not work on Win10.
2. When the GPS was unplugged, the RF noise level dropped several S-units.
 This was mentioned on the VK4ADC page, and the GPS does not need
to remain connected after the time has been set.
3. Several turns around a ferrite bead on the GPS cable reduced the
noise, but the GPS cable is only 2m long.
4. A USB extender cable was needed so that the GPS could be outside,
on the other side of the wall.
5. GPS2time has a checkbox for Daylight Savings Time, which must be
checked appropriately.
6. GPS2time can set the grid square before starting wsjtx, which would
be useful for a rover station.

-John AC6SL


___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] Possible feature request, time skew notification?

[Gary McDuffie via wsjt-devel]

> On Jul 18, 2021, at 09:26, Carey Fisher via wsjt-devel 
>  wrote:
> 
> If you simply look at the waterfall, it's clearly obvious whether your clock 
> is off or not.

You and George are correct.  Simply looking at the waterfall will get you 
within the two seconds required.  After that, the df will give you all you need 
over a couple minutes time of monitoring.

If you run your waterfall so narrow that you can’t tell, you merely need to 
expand it for a few minutes.  I’ve seen some postings that show some people run 
their waterfall so narrow (vertically) that they can’t tell a thing about the 
timing or quality of incoming signals.  That’s the cost of setting up that way.

Use the tools within the program itself.

Gary - AG0N

___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] Possible feature request, time skew notification?

[Carey Fisher via wsjt-devel]

If you simply look at the waterfall, it's clearly obvious whether your
clock is off or not.
73, Carey, WB4HXE

On Sat, Jul 17, 2021 at 10:17 PM Paul Bramscher via wsjt-devel <
wsjt-devel@lists.sourceforge.net> wrote:

> I was at a monthly outdoor ham event in the greater St. Paul/Minneapolis
> area that's drawn loyal attendance the past few years.  An operator
> setup FT8 outdoors, but had some issues getting it to decode.  We could
> tell it was getting sound from his port, but there was no decoding.
>
> He was using Windows, whereas I'm Linux only so wasn't much help.
> Another ham suggested he compare his system clock with his cellphone,
> and that fixed it.  I believe he said he was off by 5+ seconds.
>
> I'm wondering whether it might be possible, algorithmically to somehow
> detect that a person's clock may be out of sync and present a
> notification message.  Of course there's the DT/Time Delta column when
> you are able to decode.  But, presumably, he had reached the point of no
> return.
>
> I don't know how this might be possible without an accurate reference
> clock -- but possibly based on the average timing of heard signals, or
> some sort of signature within them, to cause a message to the user that
> s/he may want to check their clock's accuracy?
>
> Just wondering what might be possible in this regard, and I suspect it
> would help portable users, especially those who've been offline for a
> number of days/etc. and their system clocks had degraded.
>
> 73, KD0KZE / Paul
>
>
> ___
> wsjt-devel mailing list
> wsjt-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/wsjt-devel
>


-- 
Carey Fisher
careyfis...@gmail.com
___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] Possible feature request, time skew notification?

[George J Molnar via wsjt-devel]

Wouldn’t a simple glance at the Wide Graph provide clear evidence of timing 
errors? Even without WWV, GPS, or software changes, it seems trivial to 
identify the error and walk in a correction.

George J Molnar
College Park, Maryland 
KF2T | FM19ma



___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] Possible feature request, time skew notification?

[Alex Deligiannis via wsjt-devel]

I’m using a different approach, with a USB mouse GPS and the NMEATime2

from http://visualgps.net/index.html#nmeatime2-content 



I’m always on time.

73s

Alex

On 18/07/2021 07:58, Wolfgang K. Meister via wsjt-devel wrote:



Yes, there is a utility http://www.dxshell.com/jtsync.html that will
do exactly what you want. It syncs either via ntp or if there is no
network it is using a very clever methode do average received FT8/FT4
signals to set your clock. Extreme late or early transmissions in
time are ignored.

73s
Wolfgang OE1MWW


 Ursprüngliche Nachricht 
Von: Paul Bramscher via wsjt-devel 
Datum: So., 18. Juli 2021, 04:14
An: wsjt-devel@lists.sourceforge.net
Cc: Paul Bramscher 
Betreff: [wsjt-devel] Possible feature request, time skew notification?

I was at a monthly outdoor ham event in the greater St.
Paul/Minneapolis
area that's drawn loyal attendance the past few years.  An operator
setup FT8 outdoors, but had some issues getting it to decode. We could
tell it was getting sound from his port, but there was no decoding.

He was using Windows, whereas I'm Linux only so wasn't much help.
Another ham suggested he compare his system clock with his cellphone,
and that fixed it.  I believe he said he was off by 5+ seconds.

I'm wondering whether it might be possible, algorithmically to somehow
detect that a person's clock may be out of sync and present a
notification message.  Of course there's the DT/Time Delta column when
you are able to decode.  But, presumably, he had reached the point
of no
return.

I don't know how this might be possible without an accurate reference
clock -- but possibly based on the average timing of heard signals, or
some sort of signature within them, to cause a message to the user
that
s/he may want to check their clock's accuracy?

Just wondering what might be possible in this regard, and I suspect it
would help portable users, especially those who've been offline for a
number of days/etc. and their system clocks had degraded.

73, KD0KZE / Paul


___
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

Re: [wsjt-devel] Possible feature request, time skew notification?

[Jim Shorney via wsjt-devel]

Every ham with a modern HF rig has a simple, accurate, and reliable way of 
getting the correct time. It's called WWV. If you can't hear WWV on at least 
one frequency you either need to do some more antenna work or the bands are in 
such bad shape you won't make many contacts anyway.

Other parts of the world have their own WWV equivalents as well.

73

-Jim
NU0C

On Sat, 17 Jul 2021 21:26:35 -0700
Pearce Kieser via wsjt-devel  wrote:

> Like you said the issue here is getting an accurate reference time.
> 
> Two options you might consider:
> 1. If you’re internet connected you could get time a trusted over network 
> time protocol (NTP) 
> 2. In outdoor setting (sounds like it is more applicable) you can get a 
> trusted time from a GPS receiver.  I remember seeing YouTube videos showing 
> how to use gps as the time source for wsjtx on Linux.  Not sure how easy it 
> would be to do on windows or upstream into this app.  But a usb gps might 
> give you the data you need to detect and alarm on that time drift. (Also note 
> there are other radio time mechanisms besides gps in different modes and 
> bands)
> 
> 73, 
> Pearce - N2LW
> 
> > On Jul 17, 2021, at 7:19 PM, Paul Bramscher via wsjt-devel 
> >  wrote:
> > 
> > I was at a monthly outdoor ham event in the greater St. Paul/Minneapolis
> > area that's drawn loyal attendance the past few years.  An operator
> > setup FT8 outdoors, but had some issues getting it to decode.  We could
> > tell it was getting sound from his port, but there was no decoding.
> > 
> > He was using Windows, whereas I'm Linux only so wasn't much help.
> > Another ham suggested he compare his system clock with his cellphone,
> > and that fixed it.  I believe he said he was off by 5+ seconds.
> > 
> > I'm wondering whether it might be possible, algorithmically to somehow
> > detect that a person's clock may be out of sync and present a
> > notification message.  Of course there's the DT/Time Delta column when
> > you are able to decode.  But, presumably, he had reached the point of no
> > return.
> > 
> > I don't know how this might be possible without an accurate reference
> > clock -- but possibly based on the average timing of heard signals, or
> > some sort of signature within them, to cause a message to the user that
> > s/he may want to check their clock's accuracy?
> > 
> > Just wondering what might be possible in this regard, and I suspect it
> > would help portable users, especially those who've been offline for a
> > number of days/etc. and their system clocks had degraded.
> > 
> > 73, KD0KZE / Paul
> > 
> > 
> > ___
> > 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

Re: [wsjt-devel] Possible feature request, time skew notification?

[Wolfgang K. Meister via wsjt-devel]

Yes, there is a utility http://www.dxshell.com/jtsync.html that willdo exactly what you want. It syncs either via ntp or if there is nonetwork it is using a very clever methode do average received FT8/FT4signals to set your clock. Extreme late or early transmissions intime are ignored.73sWolfgang OE1MWW Ursprüngliche Nachricht Von: Paul Bramscher via wsjt-devel Datum: So., 18. Juli 2021, 04:14An: wsjt-devel@lists.sourceforge.netCc: Paul Bramscher Betreff: [wsjt-devel] Possible feature request, time skew notification?I was at a monthly outdoor ham event in the greater St. Paul/Minneapolisarea that's drawn loyal attendance the past few years.  An operatorsetup FT8 outdoors, but had some issues getting it to decode.  We couldtell it was getting sound from his port, but there was no decoding.He was using Windows, whereas I'm Linux only so wasn't much help.Another ham suggested he compare his system clock with his cellphone,and that fixed it.  I believe he said he was off by 5+ seconds.I'm wondering whether it might be possible, algorithmically to somehowdetect that a person's clock may be out of sync and present anotification message.  Of course there's the DT/Time Delta column whenyou are able to decode.  But, presumably, he had reached the point of noreturn.I don't know how this might be possible without an accurate referenceclock -- but possibly based on the average timing of heard signals, orsome sort of signature within them, to cause a message to the user thats/he may want to check their clock's accuracy?Just wondering what might be possible in this regard, and I suspect itwould help portable users, especially those who've been offline for anumber of days/etc. and their system clocks had degraded.73, KD0KZE / Paul___wsjt-devel mailing listwsjt-devel@lists.sourceforge.nethttps://lists.sourceforge.net/lists/listinfo/wsjt-devel___
wsjt-devel mailing list
wsjt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/wsjt-devel

Re: [wsjt-devel] Possible feature request, time skew notification?

[Pearce Kieser via wsjt-devel]

Like you said the issue here is getting an accurate reference time.

Two options you might consider:
1. If you’re internet connected you could get time a trusted over network time 
protocol (NTP) 
2. In outdoor setting (sounds like it is more applicable) you can get a trusted 
time from a GPS receiver.  I remember seeing YouTube videos showing how to use 
gps as the time source for wsjtx on Linux.  Not sure how easy it would be to do 
on windows or upstream into this app.  But a usb gps might give you the data 
you need to detect and alarm on that time drift. (Also note there are other 
radio time mechanisms besides gps in different modes and bands)

73, 
Pearce - N2LW

> On Jul 17, 2021, at 7:19 PM, Paul Bramscher via wsjt-devel 
>  wrote:
> 
> I was at a monthly outdoor ham event in the greater St. Paul/Minneapolis
> area that's drawn loyal attendance the past few years.  An operator
> setup FT8 outdoors, but had some issues getting it to decode.  We could
> tell it was getting sound from his port, but there was no decoding.
> 
> He was using Windows, whereas I'm Linux only so wasn't much help.
> Another ham suggested he compare his system clock with his cellphone,
> and that fixed it.  I believe he said he was off by 5+ seconds.
> 
> I'm wondering whether it might be possible, algorithmically to somehow
> detect that a person's clock may be out of sync and present a
> notification message.  Of course there's the DT/Time Delta column when
> you are able to decode.  But, presumably, he had reached the point of no
> return.
> 
> I don't know how this might be possible without an accurate reference
> clock -- but possibly based on the average timing of heard signals, or
> some sort of signature within them, to cause a message to the user that
> s/he may want to check their clock's accuracy?
> 
> Just wondering what might be possible in this regard, and I suspect it
> would help portable users, especially those who've been offline for a
> number of days/etc. and their system clocks had degraded.
> 
> 73, KD0KZE / Paul
> 
> 
> ___
> 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