Just Testing

Path of 500 foot Beverage On Ground.  Feedpoint is at upper right.

 

In the early part of 2018 I installed a 500 foot long Beverage antenna on the ground (BOG), and I have been pleased with its performance.  With this antenna, I have been able to receive WSPR spots from Europe, Hawaii, Australia, and throughout the Continental USA and the Caribbean.

During the late Spring of 2018, I installed an Inverted L for use as a transmit antenna.  Its vertical element is roughly 53 feet tall and the horizontal tophat extends for 210 feet.

The BOG runs along an axis of 230-250 degrees from its feedpoint to its termination.  The Inverted-L runs along an axis of 12-25 degrees from its feedpoint.

I was curious how reception from the Inverted-L would compare with that of the BOG, so on the night of October 17, 2018 I set up two identical receive stations, with one station being fed by the BOG and the other being fed by the Inverted-L, and then I monitored both WSPR and JT9 on both stations from 0000 to 1200 UTC on October 18, 2018 and compared the results using the WSPRNet reports for each station (one station used the callsign W3SZ and the other station used the callsign W3SZ/IL).

I used identical openHPSDR Hermes transceivers for this test.  Each Hermes was running OpenHPSDR-PowerSDR mRx PS v 3.4.9 with two instances of WSJTX 1.9, one instance of WSJTX on each Hermes receiving WSPR transmissions and the other instance on each radio receiving JT9 transmissions.

I presented the results of that overnight comparison of these two antennas in mid October of last year. That comparison is here.

Those results were a bit surprising to me, which may just reflect my inexperience.  The BOG DID do better than the Inverted-L, as I expected, but the performance difference between the BOG and the Inverted-L, while highly statistically significant, was not as large as I would have expected. Those results were performed before the 630M band had really “opened” for the year, and so I wanted to run another overnight test now that the band is open. This test was performed overnight on Janary 11 – January 12, 2019 which was a better-than-average night here at W3SZ, as you can see from my report which was included in John Landridge KB5NJD’s 630M Daily Report for today, here.

On WSPR on this January night, the BOG spotted a total of 1041 receptions.  The Inverted-L spotted a total of 1046 receptions during the same time frame. The Inverted-L had more receptions because it copied my WSPR transmissions whereas the BOG did not.   There were 948 simultaneous receptions by the two stations that allowed for direct comparison of simultaneously received signal strengths for the two antennas.  Receptions were obtained of a total of 31 unique callsigns. 14MDA, DL6TY, EI0CF, F1AFJ, KR6LA, LA3EQ, PA0A, and PA3ABK were received only by the BOG. W3SZ (as noted above) was received only by the Inverted-L.  The remaining 23 unique callsigns were detected by both antennas.  There were 93 Receptions that were made by the BOG that were not detected by the Inverted-L, and 48 receptions (other than W3SZ) that were made by the inverted-L but not by the BOG. 

The graphs below will show signal strength (or difference in signal strength) in WSJT units (dB) on the Y axis, and either callsign or azimuth on the X axis.  For each X value all data points for that X value will be shown, along with the mean and standard deviation error bars for those values.  In addition, to make interpretation easier, the Azimuth, Callsign, and Distance to the received station from FN20ag (W3SZ location) will be shown in red near the upper error bar.  Where that text data is not given on the graphs of signal strength vs azimuth, the omission is because two or more stations shared a single azimuth value.  A horizontal green line will mark the 0 dB signal level on each graph.

The first graph below compares the signal strengths of simultaneously received signals on the BOG and the Inverted-L by displaying the value (BOG SNR minus Inverted-L SNR) vs Callsign.  A positive value (above the green line) indicates that the BOG heard better than the Inverted-L, and a negative value (below the green line) indicates that the Inverted-L heard better than the BOG.  As indicated above, the mean Y value for each X value is shown in red, as are the error bars above and below the mean value. 20 of the 23 stations received by both antennas were on average heard better on the BOG than on the Inverted-L. 

The graph below shows the same parameter displayed for the Y values, (BOG SNR minus Inverted-L SNR), but the X axis parameter is now azimuth of the received station from W3SZ. You can see that the greatest advantage of the BOG was between 334 and 64 degrees (and especially between 0 and 64 degrees), over a range of approximately 90 degrees, centered at approximately 20 degrees azimuth. The stations between 0-64 degrees had 8-10 dB better SNR on the BOG than on the Inverted L. The 3 stations between 194 and 204 degrees azimuth were the only stations that had worse SNR on the BOG than on the Inverted L. This difference was slight, only 1-1.5 dB.

As you can see in the graph immediately above, 16 of the 23 stations had the error bars completely above the green 0 dB line. Only one station, K2BLA, had the error bars completely below the green 0 dB line.  A t test of all of the simultaneously received BOG and Inverted-L signal strength results gave a p value of 2 x 10^-16 for the comparison. In general, a p value of less than 10^-2 is considered to indicate a significant difference, so this results indicates a highly significant difference between the received signal strengths of the BOG and Inverted-L antennas, in favor of the BOG. Although the mean difference between the two antennas was 4.2 dB SNR, there was great directional variation as noted, from 10 dB advantage for the BOG over 0-64 degrees azimuth to 3-6 dB advantage for the BOG for 334-360 degrees, to 1-3 dB advantage for the BOG for 213-293 degrees, to 1-1.5 dB deficit for the BOG from 194-204 degrees.

Of course, these graphs don’t show the azimuths of the stations that only the BOG received, namely DL6TY, EI0CF, F1AFJ, KR6LA, LA3EQ, PA0A, and PA3ABK. These stations have azimuths (listed in ascending order) of 41, 44, 47, 47, 49, 56, and 286 degrees. This provides further evidence of the substantial superiority of the BOG in the azimuth range of 0-64 degrees.

Interestingly, the BOG does not have an obvious peak in its performance relative to the Inverted-L in its “preferred” direction of 230-250 degrees pointing along its axis and directed away from its feedpoint.  Of course, a 500 ft BOG would not be expected to have much directivity at 630M, for its wavelength assuming a velocity factor of 55% would be only (500/0.55 * 12/39.37)/630 = 277/630 = 0.44 wavelengths.  So it would be even less directional than the elevated 89m-long 160M (0.56 wavelength) Beverage in this illustration taken from ON4UN’s text where the -3 dB points are at approximately +/- 70 degrees:

And my Inverted-L, if I am interpreting L.B. Cebik’s work correctly, would have minimal if any directivity on 630M; see the plot below for a 50 x 50 Inverted-L on 3.55 MHz.  With my Inverted-L dimensions, at 630M my Inverted-L would be expected to be even less directional than the plot for 3.55 MHz in the illustration below, and the azimuth plot of the 3.55 MHz antenna is nearly a perfect circle:

What is really surprising to me is that azimuth range where the biggest advantage for the BOG is seen is at azimuths centered around 20 degrees, which is close to the “180 degree” expected null of the BOG, which would be expected to be between 50 and 70 degrees.  And we can’t explain this anomaly by invoking directivity of the Inverted-L, because that should have almost no directivity.

I guess other explanations must be in play here; perhaps including the signals angles of arrival at each antenna, distortions of one or both antenna’s expected patterns by local terrain and adjacent structures, the unavoidably asymmetric ground system of the Inverted-L (due to adjacent buildings), and other factors.

To sort this out, I thought that it might be helpful to look at the difference in signal strengths between the two antennas for each callsign/azimuth received as a function of time, so I produced the graph below.  It shows a small graph of (BOG SNR minus Inverted-L SNR) vs time for each simultaneous reception for each callsign received. In other words, it plots the difference in dB of the SNR for signals simultaneously received by both antennas vs time for each individual station received.  I set this up so that the X axis will autoscale for each individual graph to give the best display of datapoints for that particular station.  As a result, the X axes differ from graph to graph, so stay alert as you peruse this image.  If the image is too small for you to view it comfortably, you may find it helpful to right-click on the image and then open it in a separate tab so that you can enlarge it to better see the details.  A few of the graphs (e.g. KR7O, VE1YY) where the station was only received over a very short interval show minutes instead of hours on the X axis:

Unfortunately, this analysis provided me with no additional insights regarding the mechanisms causing the substantial relative performance differences between the two antennas at different receive azimuths. There was a suggestion that the difference between the two antennas tended to diminish after approximately 1000 UTC in some cases, as the band “closed” for the day.

Below are, without comment, some graphs of the raw data for each antenna with means and error bars shown, with x axis factors as above being either Callsign or Azimuth from W3SZ.

The graph below lacks a title. It shows SNR vs Time for the BOG.

 

Compared with the prior results that were obtained in October, 2018, the differences between the two antennas are much more apparent on the current study, likely because the “openness” of the band provided a much greater variety of signals for the current analysis than was present in October.

73,

Roger Rehr

W3SZ

Path of 500 foot Beverage On Ground.  Feedpoint is at upper right.

 

In the early part of this year I installed a 500 foot long Beverage antenna on the ground (BOG), and I have been pleased with its performance.  With this antenna, I have been able to receive WSPR spots from Europe, Hawaii, Australia, and throughout the Continental USA and the Caribbean.

During the late Spring, I installed an Inverted L for use as a transmit antenna.  Its vertical element is roughly 53 feet tall and the horizontal tophat extends for 210 feet.

The BOG runs along an axis of 230-250 degrees from its feedpoint to its termination.  The Inverted-L runs along an axis of 12-25 degrees from its feedpoint.

I was curious how reception from the Inverted-L would compare with that of the BOG, so on the night of October 17, 2018 I set up two identical receive stations, with one station being fed by the BOG and the other being fed by the Inverted-L, and then I monitored both WSPR and JT9 on both stations from 0000 to 1200 UTC on October 18, 2018 and compared the results using the WSPRNet reports for each station (one station used the callsign W3SZ and the other station used the callsign W3SZ/IL).

I used identical openHPSDR Hermes transceivers for this test.  Each Hermes was running OpenHPSDR-PowerSDR mRx PS v 3.4.9 with two instances of WSJTX 1.9, one instance of WSJTX on each Hermes receiving WSPR transmissions and the other instance on each radio receiving JT9 transmissions.

The results were a bit surprising to me, which may just reflect my inexperience.  The BOG DID do better than the Inverted-L, as I expected, but the performance difference between the BOG and the Inverted-L, while highly statistically significant, was not as large as I would have expected.  I will give a summary of results first, and then show some graphs.  This report will be confined to the WSPR results.

On WSPR, the BOG spotted a total of 875 receptions.  The Inverted-L spotted a total of 788 receptions during the same time frame.  There were 780 simultaneous receptions by the two stations that allowed for direct comparison of received signal strengths between the two antennas.  Receptions were obtained of a total of 23 unique callsigns, with one of those 23 callsigns being a one-time-only reception of WB0DBQ at -30 dB by the Inverted-L that was not detected by the BOG.  The other 22 unique callsigns were detected by both antennas.  There were 90 Receptions that were made by the BOG that were not detected by the Inverted-L, and 12 receptions (including the one of WB0DBQ just mentioned) that were made by the inverted-L but not by the BOG.  Most (73/90) of the receptions by the BOG that were missed by the Inverted-L were at distance of greater than 1100 km, whereas most (7/12) of the receptions by the Inverted-L that were missed by the BOG were at distances of less than 600 km.  The successful decodes by the BOG that were missed by the Inverted-L were all at signal levels of -24 dB or weaker, while half of the decodes by the Inverted-L that were missed by the BOG were -20 dB or stronger.  So it seems, qualitatively, that the BOG did a better job on the weaker, easier to miss signals than did the Inverted-L.

The graphs below will show signal strength (or difference in signal strength) in WSJT units (dB) on the Y axis, and either callsign or azimuth on the X axis.  For each X value all data points for that X value will be shown, along with the mean and standard deviation error bars for those values.  In addition, to make interpretation easier, the Azimuth, Callsign, and Distance to the received station from FN20ag (W3SZ location) will be shown in red near the upper error bar.  Where that text data is not given on the graphs of signal strength vs azimuth, the omission is because two or more stations shared a single azimuth value.  A horizontal green line will mark the 0 dB signal level on each graph.

The first graph below compares the signal strengths of simultaneously received signals on the BOG and the Inverted-L by displaying the value (BOG SNR minus Inverted-L SNR) vs Callsign.  A positive value (above the green line) indicates that the BOG heard better than the Inverted-L, and a negative value (below the green line) indicates that the Inverted-L heard better than the BOG.  As indicated above, the mean Y value for each X value is shown in red, as are the error bars above and below the mean value.

Received Callsigns are listed alphabetically, left-to-right.  Nineteen of the 22 stations were on average heard better on the BOG than on the Inverted-L, and 3 stations were heard better on the Inverted-L.  You can see that the greatest difference between the two antennas was for AA1A, which was on average received 12 dB better on the BOG than on the Inverted-L.  The next-greatest spread between the antennas was for AA3GZ, with the BOG being better by 7 dB for this station, but there were much wider error bars for AA3GZ’s results due to fewer simultaneous data points being acquired.  Next were AA8HS and K1BZ, both stations being received about 3 dB better on the BOG.  Of the remaining stations, all except for VA3TX, VE7BDQ, and K2BLA were on average better heard on the BOG than on the Inverted-L, although in some cases the difference in received signal strength was very small.  13 of the 22 stations had the error bars completely above the green 0 dB line, and only one, VA3TX, had the error bars completely below the green 0 dB line.  A T test of all of the simultaneously received BOG and Inverted-L signal strength results gave a p value of 2 x 10^-11 for the comparison. In general, a p value of less than 10^-2 is considered to indicate a significant difference, so this results indicates a highly significant difference between the received signal strengths of the BOG and Inverted-L antennas, in favor of the BOG.

The graph below shows the same parameter displayed for the Y values, (BOG SNR minus Inverted-L SNR), and the X axis parameter is azimuth of the received station from W3SZ.

In the graph above, it looks like the BOG is most superior to the Inverted-L at azimuths of less than 90 degrees (specifically, based on the available data, at azimuths of 67-87 degrees), and it does not have an obvious peak in its performance relative to the Inverted-L in its preferred direction of 230-250 degrees.  Of course a 500 ft BOG would not be expected to have much directivity at 630M, for its wavelength assuming a velocity factor of 55% would be only (500/0.55 * 12/39.37)/630 = 277/630 = 0.44 wavelengths.  So it would be even less directional than the elevated 89m-long 160M (0.56 wavelength) Beverage in this illustration taken from ON4UN’s text where the -3 dB points are at approximately +/- 70 degrees:

And my Inverted-L, if I am interpreting L.B. Cebik’s work correctly, would have minimal if any directivity on 630M; see the plot below for a 50 x 50 Inverted-L on 3.55 MHz.  With my Inverted-L dimensions, at 630M my Inverted-L would be expected to be even less directional than the plot for 3.55 MHz in the illustration below, and the azimuth plot of the 3.55 MHz antenna is nearly a perfect circle:

What is really surprising to me is that the station where the biggest advantage is seen, AA1A, is at azimuth 67 degrees, which is almost exactly in the 180 degree null of the BOG, which would be expected to be between 50 and 70 degrees.  And we can’t explain this anomaly by invoking directivity of the Inverted-L, because that should have almost no directivity.

I guess other explanations must be in play here; perhaps including the signal’s angle of arrival at each antenna, distortions of one or both antenna’s expected patterns by local terrain and adjacent structures, the unavoidably asymmetric ground system of the Inverted-L (due to adjacent buildings), and other factors.

To sort this out, I thought that it might be helpful to look at the difference in signal strengths between the two antennas for each callsign/azimuth received as a function of time, so I produced the graph below.  It shows a small graph of (BOG SNR minus Inverted-L SNR) vs time for each simultaneous reception for each callsign received.  I set this up so that the X axis will autoscale for each individual graph to give the best display of datapoints for that graph.  As a result, the X axes differ from graph to graph, so stay alert as you peruse this image.  You may find it helpful to right-click on the image and then open it in a separate tab so that you can enlarge it to better see the details.  A few of the graphs (e.g. VK4YB, K9FD, AA3GZ) where the station was only received over a very short interval, show minutes instead of hours on the X axis:

Unfortunately, this analysis provided me with no additional insights, although there was a suggestion that the difference between the two antennas tended to diminish after approximately 1000 UTC in some cases (e.g N1HO, WD8DAS), as the band “closed” for the day.

Below are, without comment, some graphs of the raw data for each antenna with means and error bars shown, with x axis factors as above being either Callsign or Azimuth from W3SZ.

 

I will take a look at the JT9 data next, and if it offers any new insights, will report it in similar fashion to the above.

73,

Roger Rehr

W3SZ

 

 

 

 

 

 

This is a followup to our October 2018 discussion at the MWL just past regarding measuring sun noise / moon noise.  This was originally written on October 7, 2018 and was updated December 15, 2023 with some additional details pertinent to using this method with an ExtIO device other than the RTL ( such as the SDRPlay RSPduo ), and for using virtual audio cables from VB Cables rather than the VAC virtual audio cables.

The key to getting an accurate measurement is in having sufficient bandwidth of the sampled signal, whether the technique used is an old analog method or a more technologically advanced SDR-based method.

Paul Wade W1GHZ wrote an excellent although now dated article on this subject in the 1990s:
https://www.qsl.net/n1bwt/chap10.pdf 

Paul wrote this article before SDR’s had come onto the scene for Amateur Radio use, and so he used a wide-band (several MHz bandwidth) analog system using a 10 GHz transverter, some interdigital filters, and an amplifier feeding an HP 432 power meter.

I was going to duplicate Paul’s system more a decade or so ago when I ran into Al Ward W5LUA at Dayton, and he indicated that I should advance to the current century and use an SDR-IQ ( or maybe it was an SDR-14, I don’t remember ) taking the IF signal from the 10 GHz transverter along with SpectraVue running in the Continuum mode to do this. I already had an SDR-IQ, and so that is what I have done ever since.

VK3UM also wrote a detailed paper on sun noise in 2008. He had not yet embraced the digital age:
http://www.vk3um.com/SunNoise_Measurements.pdf 

Here is a brief paper on using SpectraVue for measuring sun noise. There are no startling revelations in it:
http://www.pa0ehg.com/spectravue.htm 

So, basically, my opinion is that using anything other than an SDR to measure sun noise (or moon noise) these days is dinosaur-like behavior.  Below is my cookbook for setting up a system to use a cheap $20 RTL-SDR dongle for measuring sun and moon noise.

What You Need to Do This:

RTL SDR Dongle such as are available on Amazon:

https://www.amazon.com/RTL-SDR-Blog-RTL2832U-Software-Defined/dp/B0129EBDS2

or another device that has an ExtIO driver, such as the SDRPlay RSPduo:

https://www.sdrplay.com/rspduo/

HDSDR Free SDR Software:

http://hdsdr.de/

ExtIO_RTL2832U.dll dll file:

http://hdsdr.de/download/ExtIO/ExtIO_RTL2832.dll

or if you are using the RSPduo, an ExtIO file from here:

https://github.com/SDRplay/ExtIO_SDRplay/releases/tag/2.1

Zadig Driver Utility (not needed if you are using the RSPduo or similar device):

http://zadig.akeo.ie/

Virtual Audio Cable or equivalent (this is NOT freeware):

https://software.muzychenko.net/eng/vac.htm

or my current preference, VB Cable, available from VB-Audio Software:

https://vb-audio.com/Cable/

SpectraVue Free Software:

http://rfspace.com/RFSPACE/SpectraVue.html

In case you get stuck, the RTL-SDR Quick Start Guide at:

http://rtl-sdr.com/QSG

or if you are using the RSPduo, the guide for their ExtIO:

https://www.sdrplay.com/docs/sdrplay_extio_user_guide.pdf

1.  Steps to Do This if you are using an RTL-SDR dongle:

1.  Plug your RTL-SDR dongle into a USB port.  DO NOT install any of the software that came with the dongle, but let Plug and Play try to install it.  If you had already installed any software drivers that came with the dongle, uninstall them.
2. Go to http://zadig.akeo.ie/ and download Zadig, if you haven’t already done so.
3. Run Zadig and go to Options->List All Devices and make sure this option is checked.
4. Select “Bulk-In, Interface (Interface 0)” from the drop down list. Ensure that WinUSB is selected in the box next to where it says Driver. (Note on some PCs you may see something like RTL2832UHIDIR or RTL2832U instead of the bulk in interface. This is also a valid selection). (Do not select “USB Receiver (Interface 0)” however).
5. Click “Replace Driver” (or “Install Driver” if the button says that). You might get a warning that the publisher cannot be verified, but just accept it by clicking on Install this driver software anyway. This will install the drivers necessary to run the dongle as a software defined radio. Note that you may need to run zadig.exe again if you move the dongle to another USB port, or want to use two or more dongles together.
6. Download HDSDR from http://hdsdr.de/, using the download button at the bottom of the page, if you haven’t done so already.
7. Use the installer you just downloaded to install HDSDR, if you haven’t done so already.
8. If you haven’t done so already, go to http://hdsdr.de/hardware.html and download the ExtIO_RTL2832U.dll dll file from the table entry “RTLSDR (DVB-T/DAB with RTL2832) USB” (Direct Link).
9. Copy the ExtIO_RTL2832U.dll file into the HDSDR install folder which is by default set to C:\Program Files (x86)\HDSDR.

1a.  Steps to Do This if you are using an SDRPlay RSPduo:

1.   Download the SDRPlay ExtIO files from the link given above: https://github.com/SDRplay/ExtIO_SDRplay/releases/tag/2.1
2. Run the file SDRplay_ExtIO_Installer_2.1.exe to install the ExtIO files, following the instructions in the guide that you got from https://www.sdrplay.com/docs/sdrplay_extio_user_guide.pdf , installing the files into your HDSDR install folder which is by default set to C:\Program Files (x86)\HDSDR.

2.  Next Steps common to all devices:

1. Open HDSDR. You might be asked to select a .dll file. If you are using the RTL-SDR, then choose the ExtIO_RTL2832U.dll file you just copied over and then click Open. If you are using the SDRPlay RSPduo, then click on the file ExtIO_SDRplay_RSPduo,dll. It is okay if you do not see this screen as long as you have copied the ExtIO dll file over properly in the last step. 
2. Whether or not you saw the screen above upon opening HDSDR for the first time, next click on the “SDR Device [F8]” button in HDSDR.  This will present you with a window where you can set up the hardware parameters particular to your device.  The image below is the window that pops up when you are using an RSPduo:
3. I originally recommended, in 2018, that you download and install Virtual Audio Cable from https://software.muzychenko.net/eng/vac.htm  This is not freeware, and the trial version is not suitable because it produces a periodic audio message that will interfere with the test measurements.  If your system is 32 bits, select the 32 bit download.  If it is 64 bits, then select the 64 bit download.  However, in 2023 there is a freeware alternative (actually donationware, but you don’t need to donate) VB-Cable the use of which is described in item #5, below.
4. If you are using the Virtual Audio Cable (not the VB Cable), next open the Virtual Audio Cable Control Panel and set the sampling rate of Cable 1 to 192000 in both “SR” boxes, as is illustrated below.  Then click the “Set” button in the “Cable Parameters” box, then click “Restart Audio Engine” and finally, click “Exit”.    Note that when you start it, Virtual Audio Cable should always start in administrator mode.  If it does not, then you may get an error when you try to “Restart Audio Engine”.  If that happens, exit the program and start it by right clicking on its icon and then left-clicking “Run as administrator” instead of double-left-clicking on the icon.  You can also set the program to always run as administrator by right-clicking on it, then clicking “Properties”, then clicking “Compatibility”, then clicking “Run this program as an administrator”, and then clicking “OK”.
5. If you are using the VB Cable, download it from the following URL:  https://www.vb-audio.com/Cable/VirtualCables.htm.  Unzip the file and run the install file named “VBCABLE_Setup_x64.exe” if you have a 64 bit system or the file “VBCABLE_Setup.exe” if you have a 32 bit system.  You will need to set the sampling rate for the cable to 192000 and the bit depth to 16, and you want the cable to be stereo.  To set all of this, go to the Windows Control Panel and click on “Sound” and then find the VB cable that you are going to use on the “Playback” tab and right-click it and select “Properties”.  Go to the “Advanced” tab and select “16 bit, 192000 Hz (Studio Quality)”.  Then go to the “Spatial Sound” tab and make sure that is set to “Off”.  Then click “OK”.  Now select the “Recording” tab and on this tab right-click the same VB cable and select “Properties”.  Go to the “Advanced” tab and select “2 channel, 16 bit, 192000 Hz (Studio Quality)”.  Click “OK”  and then click “OK” again to save your settings.
6. In HDSDR, click on “Soundcard [F5]” and select “Line 1(Virtual Audio Cable)” if you are using VAC, or the proper VB Cable if you are using VB Cable.  Then click “OK”. 

1. Click on “Bandwidth [F6]” and under “Output” click on 192000.  Then click the “X” in the upper right corner of this window.  Do NOT change the Input Bandwidth!  
2. Click on “Options [F7]” and then on “Output Channel Mode for Rx” and on “IF as I (Left) / Q (Right)”. 
3. Set the HDSDR LO frequency to a clear segment in the 2M band (or whichever band you are receiving on, or possibly an IF frequency) and make sure that “AGC Off” is showing in the buttons located beneath the recording controls.  You can fine tune the frequency either by clicking in the RF spectrum, or using the Tune numbers.  Make sure that you are in “FM” mode by clicking on the “FM” button if it is not already illuminated.  After you click the “FM” button, you can use the slider to the right of the buttons below the recording controls to bring the bandwidth to 192000 if it is not already set to that value, by pulling the slider to the very top of the vertical slider bar:
4. Next set the output filter width to 200 kHz using the pull-down box at the right edge of the “Tune” frequency adjust line.  Set the filter width to 200 kHz as shown below:      
5. Next pull the red bandpass filter line in the bottom right corner to the right until it disappears off the screen.  You will see the frequency of the bandpass edge increase as you do this.  I pull it all the way to 192000, which is well outside the HDSDR window.
6. Download SpectraVue from http://rfspace.com/RFSPACE/SpectraVue.html and install it.
7. Run SpectraVue and set InputDevice to Soundcard. 
8. Click on “Soundcard IN Setup” and set SoundCard to “Line 1 (Virtual Audio Cable)” if you are using VAC, or to the VB Cable if you are using it.  Also set Sample Rate and BW Limit to 192000 and set Center Freq to 96000 and make sure “Stereo (Complex I=R, Q=L)” is checked.  Then click “OK”. 
9. In the main SpectraVue window, click on “Continuum” to set the receiver to that mode.  Then make sure that Span is set to 0.192000 MHz and then click “Start-F12”.  Then click “Auto Scale (A)” and that should position the signal trace near the bottom of the screen.  You can click below the arrow at the extreme right of the display to move the signal trace higher on the display.  I use settings of 32 FFT Ave, 0 Smoothing, FFT/BLK 16384, 0.1 dB/Div V Scale.  The “Demod On” box at the right edge of the window should be unchecked.
10. To measure sun or moon noise, just feed the 144 MHz receive output of your tranverter (or whatever your input is) into the input SMA jack of the RTL-SDR dongle or the 50-ohm Tuner 1 input of the RSPduo, or use a splitter to put the device in parallel with the receive input of your 144 MHz-28MHz transverter.  The signal level of the RTL-SDR dongle (once it has warmed up) or the RSPduo is quite stable, as you can see below with the dongle’s input seeing a 50 ohm termination. 
11. Good Luck!

Please let me know if you have questions or comments!

–W3SZ  10-7-2018

I have been having good success with my 630m station performance for both transmit and receive, although as previously mentioned I want to improve my transmit antenna and have plans to do so in short order.

I had originally planned to right-off-the-bat optimize the matching of my transmit antenna, but when I installed it I quickly saw that although the impedance was mismatched at the antenna end of the coaxial transmission line, the apparent SWR was not bad (less than 2:1) at the amplifier position in the shack, due to the effects of common mode current and the impedance transformation produced by the particular length of coaxial cable that runs from the antenna to the transmitter. So I lazily decided to not worry about getting better antenna tuning until I put up a more permanent transmit antenna. And then when I started having good success, any thought of messing with additional matching at this point vanished.

So I have been running with the antenna tuned to resonance at my transmit frequency of 0.47457 MHz but with an impedance mismatch, all of this being measured at the input to the low pass filter attached to the input to the variometer at the base of the inverted L. The image below shows the VNA results with the VNA attached at this point, at the base of the inverted L:

You can see that at the resonant frequency of 0.475 MHz at the base of the antenna SWR = 2.96, Z = 18.4 ohms, Rs = 17.2 ohms, Xs = -6.6 ohms, and phase = -163 degrees.

With the inverted L tuned in this manner, in the shack all looks well, with a nice low SWR, and with the Wattmeter on the ENI 1040L amplifier telling me that all of the power put out by the amplifier (120 Watts) is going to the load, with no reflected power being measured back at the amplifier.

Not being much of an RF guy, I figured that this tuning was the best I could do until I finished matching the antenna by correcting the impedance mismatch. So I wasn’t worried about pursuing things further before putting up the planned new transmit antenna, which should go up “any day now”.

But last week there was an interesting posting on the 600 Meter Group email list by Ike Blevins, KW7T which I’ve quoted here:

Signed on 630m cw back in January. Made a couple of contacts with my trusty homebrew 6AG7/1625 low powered transmitter. Of course I’ve read all the comments about measuring antenna current, but kept right on saying to myself…..too much fuss and trouble and probably won’t make any difference anyway. Yeah, right!

Long story short, thoughts about measuring the antenna current kept nagging at me until this morning I decided only one way to find out. Dug around in the junk box for about an hour, found the parts and after months of putting it off, made a junk box RF ammeter in about 20 minutes. Didn’t have to buy a single item.

Hooked the meter between the coax out and the loading coil tap at the base of my old ground mounted Hustler 6BTV, which I have modified for 630m (about 30 ft. tall). Key down and everything looked normal, but no reading at all on the RF current meter. Mmmmmmmm…..must have wired something wrong or found bad parts. But, on a hunch, before starting over again from scratch, I decided to move the input to a different tap on the loading coil. When I first made the loading coil I added several other taps, but ended up using the one that seemed to work the best (wrong!!). Moved the tap two turns from where it was and the RF meter came alive, from zero to PEGGED!

Spent two hours goofing around with different shunt values so as to get the meter in mid range. Otherwise, no way to know which way to go with the inductance for max RF transfer.

Right now I’m just in the ballpark, with more trimming to do, but thanks to the RF current meter I’m now getting at least TEN TIMES more power into the antenna! When you begin with 20 watts max, that’s a real big jump.

Now my plate dip, forward power out, max RF antenna current, and max field strength readings are all occurring at the same time in tuning. Moreover, I’m rockbound, and though I have several 473.5 khz crystals, none of them ever worked until now (old FT-241’s). So I was pretty much stuck on either 474.5 or 472.7 khz. So even though the transmitter uses a separate tube for the crystal oscillator, somehow my less than accurate loading of the final, due to antenna mismatch, was reflecting all the way back to the 6AG7 and preventing it from working with some crystals.

Whole new world down below the broadcast band. The old fly by the seat of your pants, clip and trim, close is good enough, etc…..a sure recipe for failure. Another lesson learned the hard way. Spend an extra 1% of your time doing it right and save yourself months of 99% frustration and mediocre results. Looking back, nothing short of a miracle that I made any cw contacts at all 🙂

Because LFers in the know seem to really hang their hats on putting an RF ammeter at the input to the antenna as the best way to maximize the power you are really sending into the ether,  I had already procured several RF ammeters prior to reading Ike’s email.  I had obtained the RF ammeters on eBay at bargain-basement prices, as they were being sold as “Steampunk” decorations rather than as RF devices.  They looked like they were in great shape from their pictures, so I took a chance and purchased them.  For those of you who aren’t familiar with Steampunk, it is a style of design and fashion that combines historical elements with anachronistic technological features inspired by science fiction.  Here is a brief article describing it.

After reading Ike’s note, I decided that it was time for me stop being lazy, and to put one of my Steampunk devices to work.  So I put the RF ammeter between the antenna terminal on the variometer and the bottom end of the inverted L, as is shown below:

There is less than a foot of 50 ohm coaxial cable between the variometer and the low pass filter, and then there is about 150 feet of coax between the low pass filter and the amplifier.

When I hooked up the ammeter as shown above with the variometer tuned to resonance at the operating frequency and then retuned the variometer for maximum antenna current, I got a huge increase in antenna current. Then I took a quick look to see what was going on in the shack. The image of the wattmeter on the amplifier’s front panel that I saw in my cellphone showed that instead of reading 120 watts as it had been before, it was reading more than 300 watts forward output. So I quickly turned down the drive to the amplifier so that I would get a similar reading on the wattmeter on the amplifier to what I was getting before, roughly 120 watts.  According to the meter on the amplifier, only 60 of these 120 watts were reaching the load, with the remainder being reflected.  Reducing the indicated power from 300+ watts to 120 watts resulted in my reducing the drive level in my PowerSDR software from 30 to 2! So if I was operating in the linear portion of the amplifier’s curve, I had reduced my power output from the amplifier from about 120 watts down to 8 watts. Interestingly, with the drive level at only 2 and with the amplifier putting out 8 watts with the new variometer tuning, I was still getting 2 amps of antenna current, which is what I got with 120 watts of transmitter power output with the old variometer tuning (antenna tuned to resonance at the operating frequency).

So why was the wattmeter on the amplifier indicating more than 300 watts out? Because the wattmeter depends on seeing a 50 ohm load for accurate results. When the impedance is significantly removed from 50 ohms the wattmeter will not give accurate results. To get a better idea of what power I was actually putting out with a PowerSDR drive level of 2, I ran the amplifier into a 50 ohm dummy load using this amount of drive, and the wattmeter showed a bit less than 10 watts, about 8 watts, just as the calculation predicted.

With the variometer now tuned to give maximal antenna current, the VNA results with the VNA hooked up to the LPF at the base of the inverted L now looked like this:

You can see that the resonant frequency is now 0.464 MHz, about 11 kHz below the operating frequency of 0.4757 MHz. At the resonant frequency, Z is 20.3 ohms, Rs is 17.6 ohms, Xs is -10.1 ohm, and phase is -154 degrees. But look at the results at the operating frequency of 0.475 MHz!  The impedance is 50.9 ohms, very close to the 50 ohm impedance of the coaxial transmission line.  So although the SWR is now lousy in the shack, with the “antenna” impedance matched to the transmission line there is maximal transfer of power from the transmission line to the antenna.  Holy complex conjugate!  For fun, I detuned the variometer and then adjusted it using the VNA to achieve as close to 50 ohms as I could for the impedance measured by the VNA at the input to the low pass filter at the base of the antenna at my operating frequency of 0.4757 MHz.  I then disconnected the VNA and checked the antenna current using the RF ammeter (which I now leave in the circuit at all times, as removing it changes the match slightly).  Adjusting the VNA to give a 50 ohm impedance at the operating frequency always resulted in peak antenna current.  Nice!

So what does the match look like at the amplifier in the shack with the new settings that maximize power transfer to the antenna?  It is not pretty!:

You can see that the SWR has jumped from less than 2 to greater than 6 and that the impedance is 240 ohms, with a substantial reactive component (Xs = 161.4). But nevertheless, FAR MORE power is now being delivered to the antenna.

Since changing the tuning I have been running WSPR with just 8 watts and I am getting WSPR reports that are similar than what I was getting running 120 watts before changing the tuning as outlined above. This result makes sense, since the antenna current under the two conditions is the same. The amplifier that I am using is specified to be protected from damage when it is running at 400 watts output into anything from a short circuit to an open circuit at all phase angles. So I do not think that running 8 watts even at an SWR of more than 6 is likely to damage it. It has an “overload” circuit that trips if it receives excessive reflected power, and that circuit has not been activated.

What would it take to match the load properly? Well, it would actually be very simple to achieve a good match, and if I were not planning to retire this antenna in a few days I would order the necessary parts and do it.

I put the values obtained by the VNA placed at the input to the LPF at the base of the antenna at the operating frequency of 0.4757 MHz with the variometer adjusted to give resonance at that frequency (Rs = 17.2, Xs = -6.6, phase = -163 degrees) into ON4UN’s L Network Calculator, and it showed that an L network with a series inductance of 10.18 uH and a shunt capacitance of 9262 pf will match the impedance presented when the variometer is so tuned:

The desired result is solution #2 in the image. There are always at least 2 solutions, and sometimes 4. You DO NOT want the first solution because a series capacitor or a parallel inductor would result in very high RF voltages. And a series capacitor would need to tolerate the entire RF current going to the antenna. These problems are not present with the second solution.

To check the solution, I put the calculated values into Wes Hayward’s Smith Chart program, and you can see that the calculated values produce an excellent result, with an SDR of 1.0078. The blue dot at the origin (r=1 on the horizontal axis) indicates a “perfect” SWR obtained by this matching network. The white cross slightly below the horizontal axis between r=0 and r=0.5 is the “starting point” before matching with the L network:

The schematic used to create the Smith Chart is here:

The ON4UN software was included in my copy of the 4th edition of his book, ON4UN’s Low-Band DXing. Unfortunately, the software won’t run with modern Windows versions. So I run it on my 64 bit Windows 10 machine by using vDos, a DOS emulator. vDOS is recommended by PCWorld, and it is free, so it wasn’t hard for me to decide to use it.

The Smith Chart software by Wes W8ZOI was included in his book Introduction to RF Design. It must also be run on an old Windows machine or using vDOS on a modern Windows machine.

You can read more about RF Ammeters for LF use here.

I hope that you found the above interesting!

73,

Roger Rehr
W3SZ

Above is a map of the WSPR stations that I have received over the past 2 weeks. Because the WSPRnet site map will only display data for up to 24 hours, I used the aprsinfo site to generate this map.

You can see in the map above that the aprsinfo site map code has a “bug” that doesn’t properly map paths that cross the international date line, so my path to VK4YB goes off the left side of the map at the International Date Line and then continues from the International Date Line to VK4YB’s QTH in Moorina, Queensland on the right side of the map. The map engine that I use in my Aircraft Scatter Sharp program had the same bug which I was just alerted to recently by ZL2DX, and it took 12 pages of corrections to fix the problem, which was not in my code, but in the GreatMaps engine .dll file which I had to recode and recompile. So if you are using an old version of Aircraft Scatter Sharp and are planning to do some aircraft scatter or troposcatter across the IDL, as does ZL2DX, then you should download the installer for the latest version of my program from here.

In any event, you can see that my WSPR receive coverage extends over the continental USA and on to Hawaii and to Australia, heading west. I did not copy Europe during the last 2 weeks, even though there were a number of trans-Atlantic WSPR receptions during that period. You can see however from the map of all stations reporting such contacts below, that there were many more WSPR NA–>EU reports (green dots) than EU–>NA reports (red dots), at least over the past 24 hours. I only show the past 24 hours because when one tries to display all WSPR receptions over the past 1-2 weeks, the hard limit placed by the aprsinfo website on the number of reports displayed on the map actually reduces the number of trans-Atlantic reports that get displayed compared with the number displayed for a 24 hour period. The reason for this is obvious if you think for a second or two. It appears that most US stations are not copying Europe consistently, and that only what I would consider to be the “Super Stations” in the USA on 630m are hearing the European spots consistently. I had copied Europe last fall on a few occasions, but have not copied Europe for at least a couple of months. This may be because I had a receive antenna issue that I just discovered and then corrected late on 3-31 UTC…time will tell, but 630m receive conditions will continue to get worse as we move into Spring, so I may have missed my chance for more Eu receptions during this season. Here is the map of WSPR 630m trans-Atlantic receptions by all stations over the past 24 hours:

As I previously noted, my WSPR transmit coverage is NOT as good as my receive coverage; see the map below, which shows my WSPR transmitter coverage, and which is again taken from aprsinfo for the same reason as the maps above.

You can see that my transmitter is well heard over the Eastern half of the USA, to somewhat west of the Mississippi (Dallas, Oklahoma City, and Salem, SD being the 3 Western-most stations copying my 630m WSPR signal).

I would like to extend this coverage at least to the west coast.

My current transmit antenna is an inverted L which was described in previous blogs. It is roughly 15 feet high and 100 feet long. My next iteration transmit antenna will run between trees. I estimate that it will be 60 feet tall and have a 200 foot top hat. If we model this using the 472khz.org calculator to which I have referred many times before in these blogs, and which has worked so well for me, we get the following results for my proposed transmit antenna:

You can see that with ground loss estimated to be 50 ohms, 99.5 Watts Transmitter Power Output is required to put out 5 Watts EIRP, with an antenna current of 1.372 Amps (and an antenna voltage of 2.04kV).

Contrast this with my initial calculation result, shown below, for the 15 ft / 100 foot inverted L I am currently using with same 50 ohm estimated ground loss, where 1492.7W TPO is required to put out 5 Watts EIRP, with an antenna current of 5.158A and an antenna voltage of 17.37 kV!:

I did work on the ground system for my current 15 ft / 100 ft inverted L, as I described in a previous blog, and with that I got the ground loss down to 25 ohms. This did improve my transmit calculations, as you can see in the textbox below, but even so I need 511 W TPO to get 5W EIRP, and my antenna current for 5 W EIRP is 4.054 A and my antenna voltage is 13.65 kV. So the new transmit antenna should be a significant improvement over the old transmit antenna.

And if I reduce the ground loss to 25 ohms for the new transmit antenna, as I did for the old transmit antenna, then the numbers really look good, as you can see in the yellow textbox below: 32.7 Watts TPO to achieve 5W EIRP, with antenna current 1.069A at 5W EIRP, with antenna voltage 1.60 kV.

So I am hoping to reach the West Coast with the new transmit antenna. We shall see. In any event, the above should make quite clear the necessity to optimize the transmit antenna, and particularly the transmit antenna ground, in order to minimize ground losses.

73,

Roger Rehr
W3SZ

I still have a very basic antenna system here both for Rx and Tx, and I have substantial improvements planned, but even so I have made some improvements that are interesting from a technical standpoint and have also significantly improved transmit capability here. I am still using the 500 foot BOG for receive, and my receive capability signficantly outstrips my transmit capability, so I have no plans to upgrade receive capability until it is more nearly matched by my transmit capability.

My Inverted L with 15 foot height and 60 ft horizontal element when matched with the BC-306A variometer showed an Rs (series resistance) of 200 ohms at 475 kHz. Last week I improved the ground for the transmit antenna by connecting its ground to the star ground of the tower near its vertical element. This star ground is also connected to the star ground of the second tower, and also connected to a ground ring that encircles the building housing my station. Every 8-10 feet along the entire length of this extensive ground system is an 8 foot ground rod. After I connected the transmit antenna to this extensive ground system, Rs fell to 25 ohms at 475 kHz.  On the left below is the VNA graph before the improved grounding was added, and on the right is the VNA graph after the improved grounding was added.  Rs is colored burnt-orange in both graphs:

---

You can see that the SWR before improving the ground was approximately 4:1, and after improving the ground the SWR was about 2:1. I improved the SWR further, to 1.35:1 using a Nye-Johnson MB-V-A antenna tuner.

After doing all of the above, I also added 10 70-foot radials centered around the vertical portion of the Inverted L.

The next day I went up to Hilltop to do some work, and I was surprised to find my antenna lying on the ground. I decided to remake the inverted L with #12 stranded wire, and to extend it from 60 feet to 100 feet for the horizontal segment, by angling it a bit at the end farthest from the vertical element, so that I could extend it beyond the second tower without coming close to that tower. I was hoping to get some improvement with this longer antenna, not only because it was 66% longer, but also because I had moved the one end away from the tower to which it had been attached.

At first I was very excited, because this change seemed to give me a 12-17 dB boost in my signal strength at K3MF, who is the closest-in station to me at 74 km, compared with prior measurements. Prior to my putting up the new antenna, my WSPR signal at Wayde’s was on the order of -10 dB, and immediately after I put up the new antenna the first six spots I got from Wayde ranged from +2 to +7.

But here is where it gets interesting! After I got these first very encouraging spots from Wayde, the situation changed and the next spots that I got from Wayde were on the order of -13 to -20 dB; see the graph below, where NEW Tx antenna spots are in pink and OLD Tx antenna spots are in blue. You can also see NEW vs OLD Tx antenna spots for many other stations on this graph. K3MF at 74 km, WA3U at 77 km, WB3AVN at 128 km are the closest stations to me.

You can see on the graph that WA3U also shows a dichotomy of signal levels, but in his case the NEW Tx antenna spots by him are roughly 8 dB WORSE than the OLD Tx antenna spots by him. K3MF is the station near the middle of the graph with 6 pink dots above the green zero-dB baseline, and WA3U is 6 groups (4 labeled callsigns) to the left of K3MF. If you want to enlarge the graph for easier viewing, then right click on it and click “Open image in new tab” and then move to that tab to view the enlarged image.

It turns out that the apparent 12-17 dB boost in my Tx signal that I initially saw after putting up the new antenna is a “time of day” effect. My initial spots from K3MF after I put the the new antenna were all before the band opened for the night, and all of my spots by K3MF before I put up the new antenna were after the band had opened for the night. My spots with the new antenna by WA3U were all after the band had opened for the night, whereas my spots by WA3U with the old antenna had all been before the band opened for the night.

I believe the reason for the initial “good” spots of my new Tx antenna by K3MF is that these measurements were made before the band “opened” for the night when K3MF was getting my groundwave signal but not yet getting distant skywave noise to increase the noise floor at his site. Once the band “opened” and markedly increased the noise baseline at his QTH, the signal-to-noise ratio of my signal there decreased even though my absolute signal strength likely didn’t change and this produced the subsequent “bad” spots of my Tx signal from his QTH.

Because all of my prior spots by K3MF had been at times that were after the band had opened, I didn’t notice this behavior of my signal worsening after the band opened before, because I had no point of comparison.

I believe that this behavior was more prominent at K3MF and WA3U than at the other stations that received me because they are the closest stations to me and so my groundwave signal predominates over skywave more at their stations than at the other stations.

WB3AVN is next closest at 128 km, and his spots of my Tx signal were also much better before the band opened, at -14 to -3 dB, whereas after the band opened and the noise floor rose my signal strength there was -17 to -28.

I made a plot of nighttime vs daytime signal strength and you can see the results in the graph below. The close-in stations K3MF, N3FL, W3LPL, and WB3AVN with both nighttime and daytime measurements are clustered in the center-right portion of the graph. All of them show the dichotomy in signal strengths, with the daytime signal strengths being significantly stronger than nighttime signal strengths :

You can see on the two graphs below what the noise floor looks like at my location before the band opens (Top) and after the band opens (Bottom). There is at least an 8 dB difference, with the noise floor of course higher after the band opens:

The effect is reversed for stations farther away than these close in stations, for the more distant stations cannot receive my signal at all until the band opens.

---

---

I did the calculation of antenna parameters for my new Inverted L with the 100 foot horizontal top wire using the calculator at 472kHz.org as I had done for the old antenna. You will recall from the prior blog that covered this that I was very impressed with the accuracy of the online 472 kHz calculator in terms of estimating the antenna capacitance, as the number calculated was very close to the measured value. For the new antenna with height 15 feet (2.615M) and length 100 feet (30.48M), the calculated antenna capacitance is 223 pF, and my BK Precision 875B RLC meter measures a capacitance value of 236 pF…quite impressive!:

With the new antenna and estimated ground loss of 50 ohms, 100 watts Transmit Power Output (TPO) gives me 340 milliwatts EIRP. 150 watts TPO would give me 510 milliwatts EIRP. So I will need to increase my Tx antenna efficiency by 9.8 times if I am to achieve the maximum-permitted 5 watts EIRP at these TPO levels. If I push the amplifier to its maximum of 400 watts, then I only need to increase antenna efficiency by 3.7 times.

If the ground loss could be reduced to 25 ohms, then the antenna performance would improve even more, with 511 Watts TPO required to get 5 Watts EIRP and with 100 Watts TPO giving just under 1W EIRP (978 mW) and 150 Watts TPO giving 1.47 Watts EIRP.

With the current Tx and Rx antenna setup I am currently collecting more than 1000 WSPR spots per night on other stations. The most distant station out of the 39 unique callsigns that I have received in the past 2 weeks was W7IUV in DN07dg at a distance of 3549 km.

Since I started transmitting 12 days ago I have had my transmitter spotted on WSPRnet by 57 different amateur stations and 1 SWL. The furthest spot of my transmitted signal was by W0SD in EN13gp at a distance of 1812 km. 15 stations have copied me at distances of over 1000 km, and an additional 17 stations have copied me at distances of more than 500 but less than 1000 km.

In the past 2 weeks I have completed 16 QSOS, all but one of them using digital (all JT9 except for one FT8), with the remaining QSO being CW. Most of this has been done operating remotely, as described in the last blog. I am now using my Hermes transceiver exclusively, having retired the K3 from MF and LF duty.

The WSPRNet map at the top of this page shows both stations that I have received and also the stations that have received me using WSPR. Below are maps of the JT9 stations that I have received on 630m in the past 24 hours (top map), and of the stations that have received my JT9 signals in the past 2-4 hours (bottom map):

73,

Roger
W3SZ

This was a busy week, but I managed to work a few more stations after my first contact with Wayde, K3MF, which I described in the prior blog. Since I worked Wayde, all of my subsequent contacts have been digital, using JT9.

On Wednesday I worked John, W3HMS twice, first using FT8 and then using JT9. I then worked Eric, NO3M about an hour and a half after I worked John. That was done while I was at Hilltop. Later that evening I returned home and transmitted WSPR with a 20% duty cycle for a while. I then switched to JT9, and after making that change I no longer had transmit output. I tried rebooting the remote computer but that made no difference. I could not reboot the K3 that was was using for both Rx and Tx because I had not set up pin 25 on the accessory socket on the back to act as a remote “ON” button. So that was the end of my transmitting on Wednesday.

My long term plan was to use one of my openHPSDR SDRs for MF and LF work anyway, so on Thursday rather than wring up the accessory connector for the K3, I substituted one of my openHPSDR Hermes transceivers for the K3. I added 25 dB of attenuation between the Hermes RF output and the input to the ENI 1040L amplifier, as that amount of attenuation keeps the input power going to the amplifier under 1 milliwatt. With the 25 dB attenuation in place, a drive level of 21 on the Hermes gives me 100 watts Transmitter Output Power.

I wanted to use the SDR rather than the K3 because all of the SDR controls are available on the remote VNC screen that I see at home when operating remote, whereas the K3 is “invisible” when I operate remote. Plus with the SDR I get two nice spectral displays and waterfalls that extend beyond what WSJTX shows me. The SDR gives me two receive slices, and I use one to run an instance of WSJTX in WSPR mode, and the other to run another instance of WSJTX, this one in JT9 mode. That way I can monitor activity for both modes simultaneously. In fact, I can use one slice to monitor 630m and the other to monitor 2200m simultaneously! You can see on the image at the top of the screen that both WSPR activity and JT9 activity are robust on a Friday night.

Being “fresh meat” on the band, I quickly made four first-time contacts in less than 30 minutes as soon as I fired up the radio this evening, working one after the other WA3ETD, K9SLQ, N3FL, and K9KFR. Below is a table showing details of my 630m contacts thus far.

Call	Mode	Grid	Azimuth	Distance km	Report Sent	Report Rcvd
K3MF	CW	FM19sr	28	80	599	449
W3HMS	FT8	FN10mf	266	88	-21	-15
W3HMS	JT9	FN10mf	266	88	-24	-21
NO3M	JT9	EN91wr	240	385	-07	-25
WA3ETD	JT9	FN33lq	32	450	-04	-27
K9SLQ	JT9	EN70	277	781	-10	-24
N3FL	JT9	FM19ka	215	172	-08	-17
K9KFR	JT9	EN71	281	804	-10	-24

In WSPR mode, my transmitter has been heard a bit further out than my longest JT9 contact, as you can see from this list of stations who have heard me on WSPR:

Timestamp	Call	MHz	SNR	Drift	Grid	Pwr	Reporter	RGrid	km	az
2018-03-17 05:12	W3SZ	0.475722	-16	0	FN20ag	0.001	VE1YY	FN85ga	1149	59
2018-03-15 02:32	W3SZ	0.475735	-26	0	FN20ag	0.001	N1DAY	EM85	818	232
2018-03-15 03:22	W3SZ	0.475729	-22	0	FN20ag	0.001	VE2PEP	FN46hc	747	28
2018-03-12 00:38	W3SZ	0.475731	-23	0	FN20ag	0.001	W4KZK	EM97	538	236
2018-03-15 03:44	W3SZ	0.475732	-24	0	FN20ag	0.001	N1JEZ	FN34im	521	24
2018-03-15 03:44	W3SZ	0.475731	-20	0	FN20ag	0.001	AA1A	FN42pb	482	64
2018-03-17 04:08	W3SZ	0.475763	-8	0	FN20ag	0.001	W1XP	FN42fo	450	53
2018-03-17 05:10	W3SZ	0.475728	-15	0	FN20ag	0.001	WA3ETD	FN33lq	450	31
2018-03-15 03:14	W3SZ	0.475731	-20	0	FN20ag	0.001	NO3M/3	EN91wr	385	296
2018-03-16 22:52	W3SZ	0.475728	-25	0	FN20ag	0.001	WA3TTS	EN90xn	347	277
2018-03-16 21:18	W3SZ	0.475730	-19	0	FN20ag	0.001	K3RWR	FM18qi	221	195
2018-03-16 22:22	W3SZ	0.475730	-18	0	FN20ag	0.001	K3RWR/3	FM18qi	221	195
2018-03-14 22:52	W3SZ	0.475732	-14	0	FN20ag	0.001	KJ4YBS	FM28bh	218	178
2018-03-12 00:30	W3SZ	0.475731	-23	0	FN20ag	0.001	W3LPL	FM19lg	145	220
2018-03-16 22:52	W3SZ	0.475731	-23	0	FN20ag	0.001	K1BZ	FM19ne	144	213
2018-03-16 21:26	W3SZ	0.475730	-15	0	FN20ag	0.001	N3FL	FM19ua	142	192
2018-03-15 02:32	W3SZ	0.475731	-24	0	FN20ag	0.001	K3FOX	FM19qf	129	206
2018-03-17 04:08	W3SZ	0.475730	-19	0	FN20ag	0.001	AA2UK	FM29pv	114	111
2018-03-14 21:28	W3SZ	0.475731	-21	0	FN20ag	0.001	W3HMS	FN10mf	85	267

That I would be heard at greater distances with WSPR than I was able to achieve with JT9 QSOs is of course expected for several reasons: WSPR is inherently more sensitive (-31 dB WSJT units) than JT9 (-27 dB WSJT units), the WSPR reception only requires one-way communication, and there are many more chances over time for a station to copy me using WSPR than JT9 just by virtue of total minutes spent transmitting in each mode to name a few.

I have been able to receive much further than I have transmitted. Those of you who have followed my blogs know that earlier in the winter I was able to receive WSPR from Hawaii on a number of occasions. In the past 2 weeks I have received 41 unique calls; but two of these, the longest two “receptions” are either false decodes or pirates. Here is a list of stations received by me on 630m in the past 2 weeks, in descending order of path length. Starting with and including EA5DOM they are “real” decodes. Note that the azimuth values reported by WSPRnet here are the reverse azimuth from W3SZ’s perspective:

Timestamp	Call	MHz	SNR	Drift	Grid	Pwr	Reporter	RGrid	km	az
2018-03-15 02:50	P57SCV	0.475703	-23	0	ED90	0.2	W3SZ	FN20ag	11106	4
2018-03-13 08:26	P60PCD	0.475595	-26	-1	FF43	0.2	W3SZ	FN20ag	8554	356
2018-03-09 01:40	EA5DOM	0.475607	-26	0	IM98xn	1	W3SZ	FN20ag	6312	298
2018-03-13 10:40	W7IUV	0.475692	-28	0	DN07dg	1	W3SZ	FN20ag	3549	86
2018-03-12 04:28	ZF1EJ	0.475695	-20	0	EK99ig	2	W3SZ	FN20ag	2390	11
2018-03-14 01:58	WA4YHC	0.475640	-22	0	EL44	0.2	W3SZ	FN20ag	2249	35
2018-03-16 09:06	KE7A	0.475779	-27	-1	EM12kx	1	W3SZ	FN20ag	2049	61
2018-03-16 10:12	W5OXC	0.475650	-28	4	EM13mm	5	W3SZ	FN20ag	2006	62
2018-03-11 06:08	K5DNL	0.475709	-25	0	EM15lj	5	W3SZ	FN20ag	1924	67
2018-03-13 10:40	K1UTI	0.475795	-25	3	EL86wx	0.5	W3SZ	FN20ag	1583	19
2018-03-17 02:44	N4WLO	0.475676	-25	0	EM50uo	0.1	W3SZ	FN20ag	1547	43
2018-03-16 10:32	W0DJK	0.475700	-29	0	EN34gh	5	W3SZ	FN20ag	1505	101
2018-03-16 10:12	W4BCX	0.475625	-9	0	EL98pd	5	W3SZ	FN20ag	1417	17
2018-03-16 09:50	K2BLA	0.475692	-14	0	EL99hb	2	W3SZ	FN20ag	1340	20
2018-03-16 10:12	WD8DAS	0.475719	-27	0	EN52hx	1	W3SZ	FN20ag	1154	101
2018-03-16 10:12	VE9GJ	0.475701	-18	1	FN77ha	2	W3SZ	FN20ag	1133	232
2018-03-14 01:14	WA4SZE	0.475771	-29	0	EM65	0.2	W3SZ	FN20ag	1108	58
2018-03-12 00:28	W9XA	0.475795	-20	0	EN51uu	0.2	W3SZ	FN20ag	1048	96
2018-03-09 01:40	KA9OKH	0.475648	-26	0	EM67fx	10	W3SZ	FN20ag	1030	72
2018-03-14 23:56	WA9CGZ	0.475619	-26	0	EN61ch	1	W3SZ	FN20ag	1002	93
2018-03-13 10:40	KU4XR	0.475769	-13	0	EM75xr	2	W3SZ	FN20ag	869	52
2018-03-11 08:02	N1DAY	0.475743	-14	-1	EM85sg	5	W3SZ	FN20ag	797	44
2018-03-11 08:02	KC4SIT	0.475609	-7	0	EM85tg	5	W3SZ	FN20ag	792	43
2018-03-13 05:46	K9KFR	0.475681	-18	0	EN71	0.5	W3SZ	FN20ag	775	97
2018-03-11 23:30	KN8DMK	0.475651	-28	0	EM89oo	2	W3SZ	FN20ag	587	81
2018-03-13 09:22	VE3EFF	0.475742	-10	0	FN15rj	1	W3SZ	FN20ag	572	175
2018-03-11 06:08	VE3CIQ	0.475767	-18	0	FN15wd	1	W3SZ	FN20ag	542	179
2018-03-11 04:12	W1IR	0.475788	-6	0	FN34lp	5	W3SZ	FN20ag	542	207
2018-03-11 03:24	W4KZK	0.475690	-16	0	EM97xe	0.2	W3SZ	FN20ag	493	45
2018-03-11 03:24	AA1A	0.475684	-18	0	FN42pb	1	W3SZ	FN20ag	482	247
2018-03-16 21:56	WA3ETD	0.475721	-27	0	FN33lq	2	W3SZ	FN20ag	450	213
2018-03-11 03:24	W1XP	0.475666	-19	0	FN42fo	1	W3SZ	FN20ag	450	236
2018-03-10 23:02	WA1OJN	0.475715	-1	0	FN32pi	5	W3SZ	FN20ag	357	231
2018-03-11 03:44	AE2EA	0.475764	-17	0	FN12fr	2	W3SZ	FN20ag	303	154
2018-03-17 02:34	KA1AL	0.475704	-22	0	FN31gm	0.02	W3SZ	FN20ag	252	237
2018-03-12 06:10	N2EIK	0.475699	-24	0	FN21	1	W3SZ	FN20ag	155	210
2018-03-16 13:12	W3LPL	0.475742	+1	0	FM19lg	5	W3SZ	FN20ag	145	39
2018-03-12 02:36	N3FL	0.475755	-11	0	FM19ua	0.02	W3SZ	FN20ag	142	11
2018-03-16 13:12	WB3AVN	0.475670	-11	-2	FM19og	5	W3SZ	FN20ag	132	32
2018-03-11 21:26	W3HMS	0.475628	-27	-2	FN10mf	0.05	W3SZ	FN20ag	85	87
2018-03-09 16:50	K3MF	0.475797	+14	0	FM19sr	2	W3SZ	FN20ag	74	35

We will end this blog with two interesting tidbits regarding my Inverted L transmit antenna. You will recall that based on the variometer settings I estimated that the inductance needed to match the antenna was approximately 700 uH. And in the previous blog I used the 472kHz.org website antenna simulator to estimate the inductance required to match the antenna. That calculator estimated that the antenna’s capacitance was 151 pF and that 745 uH would be required to match the antenna. On Wednesday I measured the capacitance of the antenna with no matching attached using my BK Precision 875B RLC meter and obtained a value of 157 pF. At resonance (475 kHz) the impedance of the antenna is purely resistive, with no reactance. Thus, at resonance Xc – Xl = 0.

So,

2 x Pi x f x L – 1 / (2 x Pi x f x C) = 0

Solving for L,

L = 1 / (4 x Pi^2 x f^2 x C)

where

f = frequency in Hz
L = inductance in Henries
C = capacitance in Farads

Plugging in
f = 475000
C = 157 x 10^(-12) Farads

gives a value of 715 uH as the inductance value required to match the antenna. So the prior estimates of 700 and 745 uH were pretty darn close to the value of 715 uH obtained by measurement of the capacitance of the antenna and the calculations shown above!

The second interesting tidbit is that my inverted L, before I moved the vertical element further away from my tower, was extremely sensitive to small changes in antenna position caused by shifting breezes. This is not surprising; with the vertical element of the Inverted L within 2 feet or so of the vertical tower there is significant capacitance added to the inverted L’s Xc, and as the vertical section moves closer to and farther away from the tower this capacitance value will change.

My intuition was that this variation in capacitance was likely greater than any variation caused by motion of the much longer horizontal section of the antenna, so I felt that if I could move the vertical section of the inverted L further away from the tower and “tighten up” the vertical section of the Inverted L so that it would swing less in the breeze, that this would reduce the variation in capacitance and thus reduce the changes in antenna tuning during windy periods. This proved to be the case.

Watch the SWR and Power changing with time on the video shown below that was taken before I moved the vertical section of the Inverted L further from the tower and tightened up this vertical section so that, being more taut, its excursion would be reduced and thus the variation in antenna capacitance would be reduced, thus reducing the changes in Xc and antenna matching.

Although the SWR and power readings shown in this video are both incorrect, as the Nye tuner was not designed for these frequencies, they do give relative readings. When the Nye antenna tuner’s SWR meter shows 2:1, the actual SWR as indicated by my miniVNA Pro is 1.05:1. And the RF power reading has proven to be double the actual value over a wide range of RF power values.

The video stops suddenly because I was concerned about the safety of the transmitter with these wide excursions in SWR and power. Even though the amplifier is supposed to be safe and protected from damage even with open and shorted loads, I did not want to risk the small possibility of damaging the amplifier due to sub-optimal loading conditions and excessive reflected power. Although it is nice to have protection circuits, it is even better (and definitely wiser) not to create circumstances where you have to rely on them.

The next thing that I need to do is to improve my transmit antenna efficiency, which is currently about 0.00291. With this efficiency, my 100 watts TPO gives me only 291 milliwatts EIRP or about 5.8% of the maximum permitted EIRP. All other factors being equal, just raising my EIRP from its current value to 5 watts would increase my signal strength by 10 x log(5/0.291) = 12.4 dB.

In the process of improving my transmit antenna system, I will also upgrade my variometer (actually, replace it with a homebrew variometer that can handle higher powers and be remotely tuned) so that I can increase my power from 100 watts to 400 watts and remotely adjust the antenna matching if need be as the antenna resonant frequency changes due to changes in environmental conditions.

Of course, if I achieve a high-enough antenna efficiency, a TPO of less than 400 watts will give me 5 watts EIRP. With my current antenna efficiency, I would need 1719 watts TPO to achieve 5 watts EIRP. So there is a lot to be gained by improving my transmit antenna efficiency.

Once that is done and I can be heard by everyone I hear, then it will be time to improve the performance of my receive antenna, which is still the 500 foot BOG (Beverage On the Ground). I have been looking at adding additional receive antennas that would be complementary to my 500 foot BOG. W8JI has an excellent website filled with information on antennas for the low bands. The index page is here. W8JI’s index page for low-noise receiving antennas is here. There is a list of antenna types at the top of this page and clicking on one of the links in that page will take you to the appropriate antenna type. I found Tom’s page on Magnetic Receiving Loops interesting. He basically says that what many people think about the way in which magnetic loops operate is incorrect, and that the construction method used by many folks is wrong.

73,

Roger W3SZ

In a recent blog post I discussed tuning my short 75 foot (15 ft vertical and 60 ft horizontal) Inverted L with my BC306-A variometer. So yesterday I set up a sked and worked K3MF on 473.6 kHz, using CW. Wayde was 599 at my location, and I was 449 at his. I was running somewhere between 25 and 50 watts to the inverted L. I am not sure which, because the ENI 1040L amplifier said I was putting out 25 watts, and my Nye Viking MB-V-A Antenna Tuner said that I was putting out 50 watts. So we don’t know if my EIRP was 72 milliwatts or 145 milliwatts (calculation done using http://www.472khz.org/pages/tools/antenna-simulator.php):

The ENI 1040L amplifier has 55 dB gain, so 1 mw in gives 500 watts out. I am not yet ready to run 500W, because I don’t think that the BC306A variometer would handle that. John, W3HMS and I both guessed that it would handle about 150 watts, but I want to play it safe and so I have not run more than 100 watts transmitter output at this point, which would give me an EIRP of about 291 mw. That “100 watts” is however the Nye Antenna tuner reading, because I want to be conservative, so it might be that my maximum power out has been 50 watts, giving an EIRP of 146 mw or so. The ENI 1040L amplifier is rated for 500 watts out at any SWR with any phase angle. It has a very simple front panel: AC On/Off switch, Power and Overload lamps, BNC RF input and output, a power meter, and a meter switch for either forward power or power delivered to load:

I started off using CW because the first test of digital modes showed that with this high-gain amplifier even when running just 50 watts TPO (transmitter power output) there was enough RF getting into the USB cable running between the computer and the radio so that the TPO jumped up to 200 watts when my Elecraft K3 was put into transmit without any audio input. This had never happened with any of my VHF/UHF/microwave work, even when running 1500 watts JT65 on 144 MHz. This occurred even though I had fed the USB cable through some cylindrical toroids.

I got a good recommendation on a particular shielded USB Cable with built-in ferrite chokes at each end Tripp Lite USB 2.0 Hi-Speed A/B Cable with Ferrite Chokes, so I ordered some of those from Amazon and they arrived today. I installed one of them between the computer and the Elecraft K3, and I also improved my station grounding by running thick braid between the radio and the computer, the radio and its power supply, the radio and the amplifier, the amplifier and the Nye Antenna tuner, and the Nye Antenna Tuner and the copper bus bar that is connected to the building safety ground and to the extensive ground rod system that surrounds my building. Slide # 103 of this presentation shows you what you need to do. A fairly common practice of running each piece of gear individually to a common point is NOT the proper way to do this. See this pdf file by Jim, K9YC: http://k9yc.com/GroundingAndAudio.pdf

After installing the new USB cable and improving my station grounding, I have no problems with RF getting into things, at least up to 100 watts TPO or so.

I only had time to run 4 transmit WSPR cycles today before needing to return home for dinner, but I was able to get decodes on each of the 4 periods from W4KZK who is in EM97 at a distance of 538 km, 4 decodes from W3LPL in FM19lg at 145 km, 3 decodes from N3FL in FM19ua at 142 kM, and 1 decode from KJ4BYS in FM28bh at 218 km:

Next steps will be to improve transmit antenna and build a higher power variometer so that I can run more power.

73,

Roger Rehr
W3SZ

 

You can barely see the inverted L running between the two towers in the photo above.  In a prior blog, I characterized its receive performance, which was worse than a lower and shorter inverted L, and worse than both my 250′ and my 500′ Beverages.  Nevertheless, because this inverted L is already in the air and because it is convenient to the shed where I would house the variometer that I would use to tune it, I decided to make it my initial choice for a transmitting antenna for 630m, realizing that I will want something better in the future.

The antenna is 60 feet (18.29 meters) long, so it is a very small fraction (3%) of a wavelength at 630m.  Thus it will need a lot of inductance to bring it to resonance at my desired frequency of 475.2 – 475.4 kHz.  This is the RF frequency for signals in the range of 1000-1200Hz on the WSJT-X waterfall when the radio dial frequency is 474.2 kHz.

The BC306A variometer that am using for this project was used on B17 and B24 bombers during World War 2.  It has an inductance range of 85 to 1210 uH.  A photo of the innards of the BC306A is below:

Changing the inductance of the BC306A is accomplished by first choosing the range of inductance using the switch near the top of the variometer, and then fine tuning the inductance using the rotary control near the bottom of the variometer that rotates the direction of a small coil positioned within the larger main variometer coil. The inductance ranges provided by each of the switch postiions on the BC306A are:

Switch Position	Minimum Inductance (uH)	Maximum Inductance (uH)
1	shorted between terminals	shorted between terminals
2	85	303
3	265	595
4	550	915
5	835	1210

Below is a picture showing the BC306A attached to the bottom end of the inverted L and the miniVNA Pro hooked up to the variometer.  The variometer and the miniVNA Pro combination was used to tune the antenna to resonance by adjusting the variometer range switch and movable coil and looking at the results on a tablet wirelessly connected to the miniVNA Pro by Bluetooth.  To the left of the variometer in the photo is the Lenovo Tab 4 Plus tablet that I used to run the BlueVNA (Bluetooth VNA) software that controls the miniVNA and displays its results.

Using the miniVNA is very handy, because you can place it at the antenna and be looking at the results and controlling it from the tablet in another location thanks to Bluetooth communications between the tablet and the miniVNA Pro. I forgot to take a shapshot of the miniVNA Pro’s BlueVNA software display on the tablet’s screen showing the antenna’s characteristics before I started tuning it with the variometer. So the first snapshot I have was taken after I had gotten the resonant frequency, which before I added the variometer had been well over 1 MHz of course, tuned down to 530 kHz. The screenshot showing how things look after the initial tuning to get the resonant frequency down to 530 kHz is below. You can see the yellow SWR curve dipping down from more than 9:1 to about 4:1 at 0.53 MHz. At that same frequency the impedance, Z, dips down to about 185 ohms, and the resistance of the antenna rises to about 185 ohms. The reactance goes from capacitive to inductive at the resonant frequency. The return loss is poor at about 5 dB, consistent with the poor SWR:

As you can see in the screen grab below, some tweaking of the movable coil knob on the BC306A brought the resonant frequency a bit too low, to 430 kHz, with the impdedance and resistance just over 200 ohms and the SWR slightly greater than 4:1 at resonance:

A VERY SLIGHT tweak of the variometer knob back in the other direction to give less inductance brought the resonant frequency to 474 kHz, with the SWR about 4:1, the impedance and resistance both just under 200 ohms, and the return loss about 5 dB. You can see all of this below:

Having declared success with this result, I put the variometer into the shed and rechecked the resonant frequency with the miniVNA Pro and all was still OK, so I hooked the 50 ohm coax from the variometer to the 50 ohm hardline for the run back to the shack and took the miniVNA inside and attached it to the 50 ohm coax right where the radio would ordinarily go during normal use. The length of the coax run affected the match, and so the miniVNA Pro now showed that the resonant frequency was now 464 kHz. The miniVNA Pro also showed that at the resonant frequency the SWR was 1.6:1 and the impedance and resistance were a bit less than 50 ohms, as is seen below:

At this point the battery for the miniVNA Pro died, so I had to power the miniVNA Pro from the USB port of the computer in the shack. Because of the way the charging circuit for the miniVNA Pro is designed, that meant that I could no longer use the Bluetooth connection to the tablet, so in order to continue to use the miniVNA Pro while it was charging I had to switch over from using the BlueVNA software running on the Lenovo tablet to using the jVNA software for the miniVNA running on the computer. I left the miniVNA hooked up to the coaxial cable at the radio position and connected to the computer running the jVNA software. I then went out to the variometer in the shed about 150 feet away and established a VNC connection between my iPhone and the computer running the jVNA software. Therefore, I was able to tune the variometer while standing at the shed at the antenna to again achieve best match at the desired frequency while watching the results produced by the miniVNA in the shack displayed on my iPhone. The image below shows the result. The display produced by the jVNA software is different than the display produced by BlueVNA, but both give much the same information. You can see that resonance is at 475.28 kHz, the return loss RL is -21.06 dB, the impedance Z is 53 ohms, the resistance Rs is 52.2 ohms, and the SWR is 1.19:1:

Clearly there is more work to be done, but this is a good start and a real demonstration of how useful the miniVNA Pro is. Being able to tune the antenna for best-match-at-the-radio-position while standing at the variometer in the shed at the base of the antenna without needing to run inside to view the result of each change in tuning was a great improvement from having run back and forth to complete the tuning process. Without the remote view capability provided by the miniVNA Pro it would have been necessary to tune the variometer hooked to the antenna in the shed at the base of the tower, then run inside to look at the results displayed by the miniVNA Pro at the radio position in the shack, then run back out to the shed to retune the variometer, then run back inside to view the result, continuing this back and forth task until the antenna was tuned properly. As it was, with the miniVNA Pro, the tuning was accomplished in a couple of minutes all while standing at the variometer position in the shed at the base of the antenna. It took much longer to write this short blog than it did to complete the tuning process!

You can read about the miniVNA Pro at http://miniradiosolutions.com/minivna-pro/

Here are 3 web pages on variometers:
http://www.qsl.net/in3otd/variodes.html
http://w5jgv.com/11.7uHy_Delta_Variometer/
https://wg2xka.wordpress.com/the-variometer/

73,

Roger
W3SZ

I. INTRODUCTION

There are 10 Amateur Bands between 50 MHz and 10 GHz inclusive, with widely differing propagation characteristics. There are currently (as of 2-20-18) 5 slow JT modes and 3 fast JT modes, the latter allowing 15 second or shorter receive cycles.  Although the program developers categorize FT8 as a slow mode and I have therefore included it in that category, it also uses 15 second receive cycles. Many of the modes have multiple sub-modes and if you add up all of those, there are currently 32 modes and sub-modes to choose from.

Furthermore, on a given band propagation might be via tropospheric scatter, via meteor scatter, via rainscatter, via aircraft scatter, via Aurora, or via EME, for example (the possibilities depend on the band). Each of these different propagation modes will have its own path attenuation and frequency dispersion/frequency shift characteristics.

Also, station equipment characteristics need to be considered when deciding on the best mode/sub-mode to use.  The characteristics that are important to consider include the frequency stability of the sending and receiving stations, ERP, and the noise figure (sensitivity) of each station’s receive system.

Although the choice of mode/sub-mode for a given QSO attempt at first glance seems complicated, in fact in order to determine the optimal digital mode to use for a given QSO attempt one just needs to compare a few characteristics of each digital mode/sub-mode to the appropriate corresponding characteristics for the path/band/propagation mode to be used, and the best match will thereby be easily selected.

The mode/sub-mode characteristics that need to be considered in this process are:

1. Sensitivity of the mode (i.e., how weak a signal can it copy.  Conventionally expressed as dB above the noise, or in the case of the WSJTX modes, in “JT dB”)
2. Tone spacing (separation between two adjacent signal tones, expressed in Hz)
3. Receive cycle time (total time used for signal reception, sandwiched between two transmit sequences)
4. Time required for a single complete message to be received. This is NOT necessarily the same as #3 above, because several modes include multiple sequential transmissions of the complete message in a single receive cycle. For example, for JT9H the time required to transmit the complete message is 0.425 seconds, even though the receive cycle time for this mode can be set to approximately 5, 10, 15, or 30 seconds.

The band/propagation mode characteristics that need to be considered are:

1. Signal-to-noise-ratio at the receiver
2 Frequency dispersion and/or Doppler shift caused by the scattering objects in the case of rainscatter, aircraft scatter, Aurora, EME
3. Time available to complete the QSO. This might be an hour or more for a non-contest experimental session, or as short as a couple of minutes for an aircraft scatter contact with an airplane flying perpendicular to the inter-station path
4. The time available to complete reception of a single complete message element. This ranges from a tiny fraction of a second for a meteor scatter path to the full receive cycle time for a stable tropo path.

The main station characteristics affecting mode/sub-mode selection are:

1. Effective radiated power
2. Receive system sensitivity
3. Frequency stability.

How does all of this come together?

1. The poorer the signal-to-noise ratio of the received signal at the receiving station, the more sensitive must be the chosen mode/sub-mode.
2. The greater the path frequency dispersion/Doppler shift/frequency instability of the transmit and receive stations, the greater the tone spacing will need to be.
3. Short receive cycle times permit more rapid QSOs, but mode sensitivity is reduced for shorter cycle times, all other things being equal.
4. With meteor scatter under usual non-shower conditions with only very short meteor pings available, only modes such as MSK144 with very short (0.020-0.072) message completion times are useful. Aircraft scatter has less stringent requirements in this regard, but relatively short message completion times are often needed for successful aircraft scatter contacts on the microwaves, where ISCAT-A, ISCAT-B, and JT9F, JT9G, and JT9H may therefore be needed.

The slow QSO modes as defined by the WSJT/WSJTX developers are FT8 (which has 15 second receive cycles), JT4, JT9A-H Slow, JT65, and QRA64. Except for FT8, each of these slow modes have one-minute-long (actually ~47-49 second) receive cycle times.

The fast QSO modes as defined by the WSJT/WSJTX developers are ISCAT-A, ISCAT-B, JT9E-Fast, JT9F-Fast, JT9G-Fast, JT9H-Fast, MSK144, and MSK144-sh. The JT9 fast sub-modes suffer significantly in terms of sensitivity when compared with the corresponding JT9 slow sub-modes, as you will see below. MSK-144 and MSK144-sh have a niche role for meteor scatter and will not be discussed further here.

II. Specific Mode/Sub-mode Characteristics

Now we are going to review the specific characteristics of the modes, limiting the discussion to what is important for operating in the frequency range of 50 MHz through 10 GHz, inclusive, and dropping MSK144 and MSK144-sh from the discussion. If you want more extensive information on the modes/sub-modes, you can start here:

WSJTX Manual: Protocols

First we will consider mode/sub-mode sensitivity. Note that within each of the modes mentioned above, as sub-mode tone spacing is increased the sub-mode sensitivity will decrease, all other things being equal. Also, reducing receive cycle time will reduce sensitivity by about 1.5 dB each time the receive cycle time is halved (10*log10(sqrt(2))).

To obtain the data in column one of the table below, I performed a MATLAB/Simulink simulation by adding incremental amounts of Gaussian white noise to WSJTX signals taken from the audio transmit output of WSJTX 1.9.0-rc1 for each of the modes/submodes analyzed here. Signal-to-noise ratios of ~5 to 50 dB were specified in the MATLAB code, with the added noise being gradually increased until the signal-to-noise ratio was so low that the signal could no longer be decoded. The signal-to-noise ratio specified in MATLAB below which decoding was no longer possible was recorded for each mode/submode, and the degradation of this value relative to the value of this parameter relative to JT9D for each mode/submode is given in column one of the table below.  Positive numbers in column one indicate performance that is poorer than JT9D, and the larger the number is, the poorer the performance of that mode/sub-mode.

The WSJTX Manual given at the link shown above gives the sensitivity of JT9D in “JT dB” units as -24 dB. Using that value and adding the result given in column 1 for each mode/sub-mode, we can calculate a sensitivity in “JT dB” units for each mode and sub-mode. These values are listed in column 2 of the table below. The third column in the table below gives the tone spacing for each mode/sub-mode listed.  The tone spacing values for each mode were taken from the Protocols section of the WSJTX Manual.

Mode	Degradation from JT9D dB	Sensitivity JT dB	Tone Spacing Hz
JT9D	0	-24	13.89
JT9E Slow	0	-24	27.78
JT4C	0	-24	17.5
FT8	3	-21	6.25
ISCAT-A 15	4.5	-19.5	21.50
JT9E-Fast 30	6.3	-17.8	27.78
ISCAT-B 15	6.5	-17.5	43.10
JT9E-Fast 15	8.5	-15.5	27.70

Most of the time you will have only a qualitative sense of what the signal-to-noise ratio will likely be for a given path, unless you have worked that path before.  If you have worked the path before, you will have a QUANTATIVE idea of what the signal-to-noise is likely to be, and what digital modes have worked for you on that path before.  Acquiring this information is a GREAT reason to test things out with other stations “BEFORE the contest”, so that you can go right to the mode/sub-mode most likely to succeed for a given path and band when the contest comes.  My program AircraftScatterSharp will also give estimates of the expected quantitative signal-to-noise values for both troposcatter and aircraft scatter propagation, expressed as dB above the noise.  This is shown in the image below;  the signal strength in dB above the noise is shown in the row labeled “Marg” ( short for “Margin”).  You can read more about AircraftScatterSharp here:

http://w3sz.com/NEW_W3SZ_AircraftScatterSharp2017.pdf

I didn’t list the other slow JT modes/sub-modes in the table above, because (unlike the fast modes) their sensitivities are listed in the WSJTX manual and don’t need to be determined experimentally. The WSJTX manual specifies that the slow modes JT65A-C all have sensitivity of -25 dB. JT4A is specified to have sensitivity -23, JT4B -22, JT9A -27, JT9B -26, and JT9C -25 dB (JT units). I tested the JT4 submodes C, D, E, F, and G even though those sensitivites are given in the manual as a check on the correctness of my experimental methods. My experimental results for the JT4 submodes’ sensitivites are within 1 dB of the results listed in the WSJTX manual.

It is also important to choose the submode with the optimal tone spacing for the path. Picking a mode/sub-mode with tone spacing less than the frequency excursion/dispersion of the path will result in degraded decoding. So choosing a mode with tone spacing at least as great as the frequency dispersion/shift that will be seen during the receive period is important. On the other hand, choosing a mode/sub-mode with greater tone spacing than is necessary will result in reduced sensitivity, thus needlessly throwing away dBs of signal to noise ratio.

Tone spacings for JT4 vary from 4.375 Hz for JT4A to 315 Hz for JT4G. The tone spacing doubles with each increment, except that JT4C has 17.5 Hz spacing and JT4D has 39.375 Hz spacing.

Tone spacings for JT9 vary from 1.736 Hz for JT9A through 222.222 Hz for JT9H. The tone spacing doubles with each increment. For the FAST JT9 sub-modes (but NOT the Slow sub-modes) the data rate increases so that the time required for a single complete message to be received halves with each increment in tone spacing, going from a time of 3.4 seconds for JT9E-Fast to 0.425 seconds for JT9H-Fast. There are no differences in the tone spacing between the corresponding JT9 fast and slow sub modes.

So what you need to do to pick the best mode and sub-mode for a particular QSO is:

1. Decide whether you need the extra sensitivity of the slow modes like JT4, JT9, or even JT65. Pick (a) either one of those slow modes or (b) FT8 or one of the fast modes based on how much sensitivity you need. Refer to the table above for a hierarchy of mode/sub-mode sensitivity, and make your selection accordingly. For a path with stronger signals, you should be able to get by with any of the modes listed (considering only signal strength), and can use the modes nearer the bottom of the table to speed up the time to completion of the QSO. However, for weak signals, you will need to stick with the modes near the top of the table.

2. Decide how wide your tone spacing needs to be based on system frequency instability, path frequency dispersion due to Aurora, rainscatter, etc., and Doppler shift if doing EME or Aircraft Sctatter. Match your expected frequency dispersion/Doppler shift/system frequency instability value with a mode/sub-mode that has compatible tone spacing. The tone spacing of the optimal mode/sub-mode needs to be at least as large as the total of your expected frequency shift/dispersion parameters.

That is all there is to it! Pick the mode and sub-mode based on the sensitivity and frequency tolerance that you think you need based on the band, propagation path/mode, and equipment characteristics as outlined above. If you have sufficient signal strengths to potentially use FT8 or one of the fast modes, you can reduce your QSO time to 25% of what it would be with the slow modes. You can always try a fast mode (or FT8) first and then drop back to a slow mode if you don’t have enough oomph to complete the QSO with one of the fast modes.

For example,if you are in a hurried contest situation and you want to do QSOs with 15 second receive cycle times, then you are limited to FT8, ISCAT-A, ISCAT-B, and JT9E-Fast through FJ9H-Fast. Of these choices, FT8 has the best sensitivity, so if the given path/mode frequency dispersion and transmit-receive instability of the system (including both the transmitting and receiving stations’ instabilities) is less than 6.25 Hz over the 12.6 second receive period, then FT8 is likely your best 15-second-T/R-cycle choice. If your system frequency instability is worse than that, then you should pick another mode instead of FT8, using the tables above as your guide. For the greatest chance of success, always choose the mode that has the greatest sensitivity that will also satisfy the receive cycle duration and frequency spread characteristics of your particular situation.

The best mode for a quiet-condition 10 GHz terrestrial QSO using troposcatter is NOT necessarily the best mode for a rainscatter or Aurora or EME 10 GHz QSO. The optimal terrestrial mode/sub-mode might also be the “right” mode/submode for aircraft scatter if the airplane is flying right down the direct path between two stations so that Doppler shift is negligible, but it will NOT be the right mode/sub-mode if the aircraft’s path is at a significant angle to the direct path so that there is significant Doppler shift of the received signal. You can play with various aircraft/path geometries for various frequencies using my program AircraftScatterSharp and see what Doppler shifts result from various aircraft scatter geometries at various frequencies, or review the table on page 23 of my recent paper on Aircraft Scatter, given at the NEWS Conference in 2017. These resources can be found at:
http://w3sz.com/AircraftScatter.htm
A pdf of the paper is here:
http://w3sz.com.com/w3sz/NEW_W3SZ_AircraftScatterNEWS_2017_Paper.pdf

III. A Few Examples

There is of course much more frequency dispersion when rainscatter is the propagation mode than there is with a “quiet” tropo path. In January, 2018 while helping the NN3Q/R team get setup prior to that month’s contest, we happened to be out in the rover van during a rainstorm. Dave, K1RZ, was kind enough to get on the air so that we could try a rainscatter digital contact. I expected that there would be a lot of dispersion as there was no direct path to Dave, and the entire signal would be scattered. So I chose JT4G, as it has the largest available tone spacing, 315 Hz. It has a specified sensitivity of -17 dB, so by choosing the extremely wide tone spacing, we were giving up 4 dB relative to JT4C and 6 dB relative to JT4A. We were able to complete the contact, with Dave’s signal levels -16 in the NN3Q/R rover van and our signal report from Dave at -20 dB. The figure below shows the WSJT-X main window for this contact.

As you can see from the figure below, the frequency dispersion during this contact was on the order of 50 Hz or slightly less, and there was no appreciable drift or Doppler, so we could have used JT4E (78.5 Hz tone spacing, -19 dB sensitivity) or JT9F (55.56 Hz tone spacing, -22 dB sensitivity) if signals had been too weak for us to complete the QSO using JT4G (315 Hz tone spacing, -17 dB sensitivity), and we likely would have been successful due to the greater sensitivity of JT4E and JT4F compared with JT4G. Fortunately, we had enough signal strength even with JT4G and so this was not an issue.

Based on the fact that, as noted above, the fast JT9 modes have a 6-9 dB disadvantage relative to the slow JT4 and JT9 modes, we likely would not have been able to complete the QSO using one of the JT9 fast modes. Although FT8 and ISCAT-A have better sensitivity than the JT9 fast modes, the frequency dispersion on the 10 GHz rainscatter path was too wide for FT8 (6.25 Hz tone spacing) and ISCAT-A (21.5 Hz tone spacing). ISCAT-B had borderline tone spacing for this path (43.1 Hz) and in addition ISCAT-B has similar sensitivity problems as the JT9 fast modes, with a sensitivity of -17.5 dB.