Application in for Volunteer Monitor Program

Posted on May 18, 2019 by Bob Beatty

The ARRL and FCC have been wrangling over the Official Observer (OO) program for a bit, and the OO program was shut down.  In its place, a Volunteer Monitor program is in the process of being established.  Details here.

After a bit of poking around, I found the application, filled it out, and submitted it.  Given my interest in metrology, it seemed a natural fit.  We shall see.

(Edit 5/30/19: I received a confirmation that my application had been received by the program manager, who said “We are working though all this in the next several months”.  Sounds good)

Posted in Volunteer Monitor | Leave a comment

First FT4 QSO

Posted on May 16, 2019 by Bob Beatty

I finally bit the bullet and installed WSJT-X version 2.1.0-rc5  I figured why let everyone else have all the fun!  (I can admit that debugging new software can be a huge time investment).

Anyway, installation went fine, but you must read the user’s manual in order to understand how to “Clone” your original settings into a new one that can be modified to add the frequencies required for FT4.  Just Read The FT4 Protocol Manual and you will be golden!

Within less than 2 minutes of starting I had my first FT4 QSO with W5I.  And the QSO, which included one retry on the “CQ”, took 42 seconds — 30 seconds if you take out that retry.

The waterfall does look different because the FT4 signals are significantly wider and the TX/RX cycle is MUCH shorter.  The image below illustrates this well.  The top 2/3 is FT4, while the lower 1/3 is FT8.

Since FT4 isn’t part of the official ADIF standard yet (a few more weeks to go), LotW isn’t accepting FT4 as a valid mode.  You need to have it force “DATA” in tQSL.  Also if you use HRD like I do you need to add the new mode into HRD.

In HRD:

In HRD Log, click on Tools/Configure/Modes then use the ADD button to add in FT4.

(Ed: Feb 4, 2020 — The comments below, concerning TQSL, were based on the ARRL recommendations prior to their adding support for FT4 in a version of TQSL that released a week after this post.  Things have been fine since then, and assuming you have a current version of TQSL, you shouldn’t have to do anything (so no need to add a new ADIF Mode and change it to DATA))

In TQSL:

File>display or modify preferences>ADIF Modes.  Then click on “Add” and enter FT4 in the “ADIF Mode” box.  Then highlight “DATA” in the list of “Resulting TQSL mode” entries, then click “OK”.  You are all set!

Posted in Digital Mode, FT4, FT8, LotW, WSJT-X | 4 Comments

… then one, and done!

Posted on May 3, 2019 by robert

With the help of Manuel EA5TT and a CW QSO on FO-29, plus a QSO with Frank WA2NDV on the next pass, I have received LotW confirmations for 100 different grid squares, which qualifies me for VUCC Satellite.

Many thanks — this has been a fun chase that took 45 years to complete — Satellite QSO #1 was via AO-7 back in 1974 (Using my New York call sign WB2ZIB).

Posted in Satellite, VUCC | Leave a comment

And then there were two…

Posted on May 1, 2019 by robert

A nice surprise this May Day, for my VUCC Satellite effort.  I am now up to 98 confirmed grids.  So five new grids were added in April.  Perhaps I will get two more in May!

Posted in Satellite, VUCC | Leave a comment

Joe Taylor’s upcoming FT-4 and the value of bit shaping

Posted on April 24, 2019 by robert

I was watching a video of Joe Taylor giving a presentation at the Fair Lawn Amateur Radio Club (Thanks, Bob W1YRC, for that reference).  Joe started giving some details of his upcoming FT-4 protocol (starts at 15:35 mark).  FT-4 might become a replacement for digital contest work.

What really caught my eye was the importance of bit-shaping to reduce the bandwidth.  My club had been pondering a “group build” project that would do something radio related but allow members to get their feet wet with an arduino-like project.  One of the ideas was a Morse Code training tool.  As these things tend to do, it quickly spiraled out of control with requirements for Farnsworth timing, variable side tone frequencies, printouts for checking, etc.

While it was pretty easy to get a variable frequency side tone generated (essentially a square wave or PWM output), they sounded somewhere between horrible and dying cat bad.  The source of the problem is our old friend — sharp edge transitions, which is rich in harmonics.  It became clear that some waveform shaping was needed, and the old “raised cosine” technique immediately came to mind.  That requirement complicated things considerably.  Gaussian Filters were also considered.

Anyway, back to Joe and FT-4.  Joe was pointing out that while the standard RTTY signal had two tones spaced 170 Hz apart, and one might assume it was a narrow signal, in fact, due to keying issues, it is actually quite broadband.  Strong stations may well wipe out 1 KHz of spectrum space simply because of their bandwidth.  He then showed a graph comparing RTTY to FT-4:  RTTY taking up 3 KHz to the -60 dB point, while FT-4 taking up less than 200 Hz.

Critical to that reduced bandwidth is the use of a Gaussian Filter to shape the 4-tone FSK signal.  In the diagram below, the unfiltered signal is shown on the top in red. The shaped signal is shown in blue.  Those are then plotted against a standard RTTY signal (in purple).

The above diagram is from the FT-4 Protocol Document.

 

Posted in Digital Mode, FT-4, FT8 | Leave a comment

April 2019 FMT Results: Green Bar!

Posted on April 15, 2019 by Bob Beatty

Well the results are in, with 112 people submitting measurements, 74 of 112 (including me) made the Green Band (< 1 Hz error).  Five of those had errors less than 0.1 Hz error.

My results were 0.26 Hz high on 80-meters (72 PPB), and 0.43 Hz low on 40-meters (61 PPB); for an average error of 66 Parts Per Billion (PPB).   Matt, KA1BQP (another Newport County Radio Club member here in RI) beat me with an average error of 64 PPB.  Our scores put us exactly in the middle of the pack.

Comparing it to April of 2018, I did MUCH worse.  Last year I had an average error of 13.4 PPB (being off 0.01 Hz on 80 and 0.17 Hz on 40).

In an earlier post, I had stated three different values for each band; math based, visually based (visual average), and hunch (based on my best guess of what Doppler was doing).  I submitted the math based approach.  Looks like I would have done better by using a visual average of the waterfall — but that is just this time.

Technique Audio Freq Freq Guess Actual Freq Delta Result
80 Math 998.14 3,599,278.14 3,599,277.88 -0.26
80 Visual 998.10 3,599,278.10 3,599,277.88 -0.22 Better
80 Hunch 998.20 3,599,278.20 3,599,277.88 -0.32 Worse
40 Math 1,011.99 7065531.990 7,065,532.42 0.43
40 Visual 1,012.03 7065532.030 7,065,532.42 0.39 Better
40 Hunch 1,012.20 7065532.200 7,065,532.42 0.22 Better

 

Posted in FMT | Leave a comment

Siglent SDG 2042X Stability Test

Posted on April 13, 2019 by Bob Beatty

I want to replace my dead HP 3335A with a new Siglent SDG 2042X.  After all, the latter is 40 years newer, 20 pounds lighter, and maybe 1/8 the volume.  It is also silent as a church mouse (the HP’s fans produced an obnoxious roar while keeping its ECL logic from melting).  The HP cost about $14,000 in 1980, and the Siglent sells for under $500 new today.  We have come a long way in the last 40 years!

The executive summary is that it is more than capable of replacing the 3335A, from a stability point of view.  However, it lacks HP’s wonderful built-in step attenuator that allowed one to set an output signal level like -84.1 dBm and expect it to be spot on.  The Siglent has no dBm feature, only volts or millivolts. (Correction, it can be set to dBm from -50 to +23 dBm).  I set mine to an output level of 632.4 mVpp, which is the equivalent of 0 dBm, then adjust it using an external attenuator box like a JFW Industries 50BR-001.

The Siglent allows sub-hertz frequency entry.  Using the keypad you can enter 1.23456789 MHz, you will get an output signal that is 1.234 567 890 MHz and will see that final digit bouncing around by no more than 1 count up or down.  If you use the knob, you can expect to change the output in steps of 1 hertz (or 10/100/1000/10000, etc).  Watch out if you enter a frequency using the keypad that has a sub-hertz resolution, then use the knob to adjust it later.  You will no longer see those sub-hertz digits on the display, but they are still there.  Best to enter a frequency to the nearest hertz using the keypad before using the knob. (Edit – I verified that you can enter a frequency, using the keyboard, with 1 millihertz resolution).

The spec is a tad misleading as it claims to be a 40 MHz signal source.  This is true for sine wave only.  All other functions have significantly lower upper limits (25 MHz for square wave, 1 MHz for a ramp).

Posted in FMT, Gear | Leave a comment

Elecraft K3s DDS/PLL Induced Drift

Posted on April 13, 2019 by Bob Beatty

I wasn’t very clear when I was posting about my pre-FMT calibration.  But I did describe that the K3s was not a particularly effective rig to use for a FMT.  This is because of how the digital synthesizer was implemented.  Even with the optional high-stability oscillator, which my K3s has installed, the issue remains.

The behavior is very evident in the Spectrum Lab screen capture below.  The K3s was fed a Rubidium based 10 MHz signal that is stable in the microhertz range.  The K3s is in CW mode with a 500 Hz Pitch selected , so we would expect it to produce a 500.000 Hz tone.

On the bottom of the illustration, in the waterfall area, the horizontal dashed lines are 5 minute interval time stamps, so we are seeing the variation in tone over a 15 minutes interval.  During that time, the DDS/PLL is performing a stair-step adjustment that is causing the tone to vary from about 501.2 Hz to 501.8 Hz.  One might describe this as 501.5 Hz, plus or minus 0.3 Hertz.

By the way, the older K3 would do exactly the same thing.  This is not a defect, as the K3s is working exactly as designed.  It was done to maximize dynamic range and minimize unwanted noise from the synthesizer.  The designer saw no reason to have its short term stability be better than +/- 0.3 Hertz.  For ham use, that is more than acceptable.  It is highly stable a longer term.  It will remain at 501.5 Hz +/- 0.3 Hertz for many days.

By the way, this is also pretty much what you can expect for Doppler on a 20-meter band signal anyway.  So the rig is just as stable as the ionosphere is in the short term.

Posted in FMT, Gear | Leave a comment

So Long OLD Friend!

Posted on April 13, 2019 by Bob Beatty

And just like that, my 1980 vintage HP 3335A Synthesizer died this afternoon when I turned it on to get a better graph of my K3s PLL frequency hunting.  Dreaded “Unlocked” indicator and crazy display.  I have another that I’ve already cannibalized for parts, so there is some chance I can repair the one that was working.  But this also gives me an excuse to check out my Siglent SDG-2042X, which sports a 10 MHz external reference.  It all depends on how they implemented their PLL if it will be useful at the sub millihertz level like the HP was.  If the Siglent works, then I will pull the HP Oven controlled Reference out of both and eBay them.

Good timing, however, as I was able to use the HP to verify that everything was all set correctly prior to the FMT.  I will miss the old fella.

Posted in FMT | Leave a comment

Preliminary Results for April 2019 FMT

Posted on April 12, 2019 by Bob Beatty

I was able to participate in the FMT last night (around 10 PM local time).  This was a pretty big deal for me, because I haven’t been in the basement that late at night since I was in the hospital last October (my equilibrium has been quite bad, especially at night, making the journey from basement to bedroom a challenge).

I was fumbling wildly with the Spectrum Lab settings during the initial call-up on 40-meters.  I thought I’d be able to locate my target tone (about 1000 Hz) quickly.  I wasn’t.  I finally got things locked in right as Connie K5CM went carrier down.

The results below are simply an audio analysis and don’t reveal my receiver dial frequency.  The audio frequency is either added or subtracted from the RX dial frequency depending on which side of the carrier the tone is received.

The image above is a screen capture from DL4YHF’s Spectrum Lab (Audio Signal Analyzer) taken during the 40-meter run (somewhere near 7.065 MHz).  You can see how the signal spreads out over the two minute key-down interval.  I saw a minimum of 1011.6762 Hz, and a maximum of 1012.240 Hz.  Visually I called the tone at 1012.03.  After mathematical analysis, it seems more like 1011.988 Hz.  The signal shifted over a range of 0.360 Hz (the actual frequency is somewhere in that noisy band).  I might “hunch” a frequency of 1012.20 if I believed I knew what Doppler was at that moment.  It will be interesting to see which of those three frequencies are closer to the mark (math=1011.99, eye=1012.03, hunch 1012.20).

The image above was taken during the 80-meter run (somewhere near 3.999 MHz).  I saw a minimum of 997.8961 Hz, and a maximum of 998.3294 Hz.  Visually I called the tone at 998.10.  After mathematical analysis, it seems more like 998.138 Hz.  The signal shifted over a range of 0.234 Hz.  I might “hunch” a frequency of 998.20 if I believed I knew what Doppler was doing then.  It will be interesting to see which of those three frequencies are closer to the mark (math 998.14, eye=998.10, hunch=998.20).

Results will be posted Sunday April 14 at 10 PM — Connie K5CM is that quick!

Posted in FMT | Leave a comment