Topics

QCX, QLG1, v1.05 ... WSPR?

 

Can anybody confirm that QCX, QLG1 and firmware v1.05 work together properly with WSPR?

I enjoyed good returns on WSPR with both my QCX 40 and my QCX 20 prior to  v1.05 and prior to QLG1.
Yesterday with v1.05 only I got typical returns on the 20m unit while I was building my newly arrived QLG1.
Since I connected the QLG1 which, by all indications, is working correctly, and I used it to calibrate the 20MHz and 27MHz clocks, I haven't heard a peep from wsprnet.

Diag on QLG1 is "A -- f-- t12 s26" (s value drops to 5 occasionally, but no lower). At some point there was something after "A" and something after "f", as it appeared and disappeared immediately after I pressed "select" to update.

The QLG1 lamps flash, as described, the QCX's identify it and respond as anticipated. The calibration values on the ref and system clocks look reasonable, settling at 27,003,887 and 20,006,197 respectively after the heartbeats.

I was about to hit "Send" saying "nothing seems wrong, behaviorally, yet I haven't heard a squeak out of WSPRnet since connecting the QLG1." but on further gazing I did notice that with both QCX's, the "t" and "s" fields blank out occasionally too.

Julian N4JO.

Mike Besemer - WM4B
 

Works fine for me since 1.05 came out.

 

Mike

WM4B

 

From: QRPLabs@groups.io [mailto:QRPLabs@groups.io] On Behalf Of Julian Opificius
Sent: Wednesday, May 27, 2020 2:14 PM
To: QRPLabs@groups.io
Subject: [QRPLabs] QCX, QLG1, v1.05 ... WSPR?

 

Can anybody confirm that QCX, QLG1 and firmware v1.05 work together properly with WSPR?

I enjoyed good returns on WSPR with both my QCX 40 and my QCX 20 prior to  v1.05 and prior to QLG1.
Yesterday with v1.05 only I got typical returns on the 20m unit while I was building my newly arrived QLG1.
Since I connected the QLG1 which, by all indications, is working correctly, and I used it to calibrate the 20MHz and 27MHz clocks, I haven't heard a peep from wsprnet.

Diag on QLG1 is "A -- f-- t12 s26" (s value drops to 5 occasionally, but no lower). At some point there was something after "A" and something after "f", as it appeared and disappeared immediately after I pressed "select" to update.

The QLG1 lamps flash, as described, the QCX's identify it and respond as anticipated. The calibration values on the ref and system clocks look reasonable, settling at 27,003,887 and 20,006,197 respectively after the heartbeats.

I was about to hit "Send" saying "nothing seems wrong, behaviorally, yet I haven't heard a squeak out of WSPRnet since connecting the QLG1." but on further gazing I did notice that with both QCX's, the "t" and "s" fields blank out occasionally too.

Julian N4JO.

 

Thanks. Are you using the GPS unit?

N3MNT
 

W

N3MNT
 

Works fine for me as well.  Not sure what you mean by hitting send.  When you enter beacon mode with parameters sent correctly, the transmission is automated.Are you leaving GPS connected while using WSPR?

Arv Evans
 

Julian

Not using a GPS here although I do have a couple old Garmin-35 units left 
over from when I ran a tracking system company.

My time clock for digital modes is a 10 MHz temperature stabilized crystal 
oscillator with a looooong divider chain down to 1 PPS.  It is zero-beat with 
WWV via a direct-conversion receiver (SA-612 driven by 10.000 MHz oscillator). 
With a DC oscilloscope or DVM on the mixer output it is possible to see errors 
of less than  1/4 Hz as the beat-note caused DC output slowly moves up and 
down.

For accurate date and time I Use NTP on a Linux PC. 
For non-moving locations NTP synced output from the Linux "date" 
command can be used to provide a GPS like output string.  Hans and 
myself have discussed this in the long distant past and have even used 
it to test systems that require GPS-RMC input strings.

Also have strong DGPS signals from a 325 KHz station in eastern Oregon, 
but have never been able to find if DGPS LF transmitters are synched to 
satellite GPS frequency.  It they are synched that might be a good source of 
time, location, and frequency signals.  Thinking in this direction is possibility 
of building a TRF receiver that phase-locks to the 325 KHz DGPS carrier, 
followed by a divide by 325,000 to get 1 PPS.

Bottom line, I am interested in digital modes but not yet interested enough 
to actively participate.  At 78 years old it sometimes takes lots of motivation 
to get me started on big projects.    I spend more time working in my shop 
with a Chinese made mini-Lathe Lathe and on a CNC controlled DIY milling
machine.I have built and use a disk sander, belt sander, surface grinder, 
and have a 225A welder for really big fabrication projects.

Arv
_._

On Wed, May 27, 2020 at 4:05 PM Julian Opificius <n4jo@...> wrote:
Thanks. Are you using the GPS unit?

 

Hi Bob,
"Send" as in "send my reply": what I meant was I was about to send my reply when I noticed that in the GPS data page, with the unit connected, not only are the first two fields dashed, but the last two were also intermittent.

I had been trying to get the 20m unit working all day with no success. So... um... it turns out WSPR doesn't work if you leave the QCX in practice mode...which would explain why I got nothing all day :-(

But my GPS data still isn't showing up properly on either unit.... Anybody else experience that?

Julian N4JO.

Hans Summers
 

Hi Julian
 
But my GPS data still isn't showing up properly on either unit.... Anybody else experience that?

This is a known issue which has been mentioned several times in this group. There is some kind of timing race condition where the display update occurs BEFORE all the data has been received from the GPS. It must be somehow location dependent since I have hardly ever seen it here. 

This applies to all the firmware revisions of the QCX. It has not been tackled by me yet. But it is not a serious issue, just a cosmetic thing in the GPS Info display screen. 

So... don't worry about it

73 Hans G0UPL
http://qrp-labs.com

 

Thanks for that, Hans,
I won't worry, then; everything else is working perfectly.

The battery backup is a really nice touch: I can swap the GPS unit between my two QCXs and it picks right up again in seconds.

Julian N4JO.

Arv Evans
 

Hans & Julian

Several years ago I was playing with a GPS receiver and Arduino and experienced 
something similar.  It would come and go depending on time-of-day.  Research 
indicated that it was dependent on how many satellites were in view.  I had tested 
for valid versus invalid RMC strings in software and found that when too many 
satellites were in view there were so many RMC strings that my software was not 
completing analysis of each string before the next stanza of strings was received.

Does this sound like it might be the problem?  My quick-and-dirty fix was to limit the 
number of RMC strings processed to just the first 4 of each stanza.  No, not an 
elegant fix, but it worked.

Arv
_._



On Thu, May 28, 2020 at 12:48 AM Hans Summers <hans.summers@...> wrote:
Hi Julian
 
But my GPS data still isn't showing up properly on either unit.... Anybody else experience that?

This is a known issue which has been mentioned several times in this group. There is some kind of timing race condition where the display update occurs BEFORE all the data has been received from the GPS. It must be somehow location dependent since I have hardly ever seen it here. 

This applies to all the firmware revisions of the QCX. It has not been tackled by me yet. But it is not a serious issue, just a cosmetic thing in the GPS Info display screen. 

So... don't worry about it

73 Hans G0UPL
http://qrp-labs.com

Hans Summers
 

Hi Arv

There's only one RMC sentence but I know what you mean - there are some sentences of which there are a variable number, because they contain information about each satellite. 

Yes it's something like that. I know what the problem is and where it is in the code. Since the question does keep arising intermittently I will probably one of these days, invest a few hours to put it to had finally!

73 Hans G0UPL 

On Thu, May 28, 2020, 23:13 Arv Evans <arvid.evans@...> wrote:
Hans & Julian

Several years ago I was playing with a GPS receiver and Arduino and experienced 
something similar.  It would come and go depending on time-of-day.  Research 
indicated that it was dependent on how many satellites were in view.  I had tested 
for valid versus invalid RMC strings in software and found that when too many 
satellites were in view there were so many RMC strings that my software was not 
completing analysis of each string before the next stanza of strings was received.

Does this sound like it might be the problem?  My quick-and-dirty fix was to limit the 
number of RMC strings processed to just the first 4 of each stanza.  No, not an 
elegant fix, but it worked.

Arv
_._



On Thu, May 28, 2020 at 12:48 AM Hans Summers <hans.summers@...> wrote:
Hi Julian
 
But my GPS data still isn't showing up properly on either unit.... Anybody else experience that?

This is a known issue which has been mentioned several times in this group. There is some kind of timing race condition where the display update occurs BEFORE all the data has been received from the GPS. It must be somehow location dependent since I have hardly ever seen it here. 

This applies to all the firmware revisions of the QCX. It has not been tackled by me yet. But it is not a serious issue, just a cosmetic thing in the GPS Info display screen. 

So... don't worry about it

73 Hans G0UPL
http://qrp-labs.com