Saturday, March 11, 2017

Propagation Sanity Check

This morning we received an email from Finland!  It's always nice to hear from our "neighbors" across the north pole.  I know we share a common bond when it comes to tough propagation.

Hello, Sirs.
It is quite foolish to ignore JT65 spotter on.
The whole of Europe called on you.

None of the answer does not come from there.

OH4SS, Matti

I was certainly sorry to miss the opening.  

So it would seem Matti is implying one shouldn't run JT65 spots to PSK Reporter if not in the shack. Now that's an interesting point of view, and one I haven't encountered before.  I think we have a little disagreement here with Matti, but maybe there are others out there who feel the same way. 

"Foolish"?  Let me explain my rationale.

Followers of the blog -- are there any? -- will know there are special challenges the ionosphere lays on us in the high latitudes.  What they teach you in radio school about propagation being basically reciprocal is total nonsense above 60 degrees or so.  It's actually hard to figure out when openings may occur.  The second element to this puzzle besides geomagnetic instabilities is the rapidly changing daylight in the Far North.  In Fairbanks, daylight changes 6-7 minutes per day throughout much of the year.  That has huge implications for HF, as one week is literally completely different from the next. at KL2R we make frequent use of spotting software and other tools to monitor the bands. That includes CW Skimmer reporting to RBN and WSJT-X reporting to PSK Reporter.  The data is helpful to other hams who are anxious to see when the bands are open to Alaska.  What's not to like?

The other reason we frequently run receivers and spotting software when the shack is unattended is to gather data for serious technical analyses.  The data sets are useful for comparing station performance, for instance.  JT65 actually yields quantitative results on received signal-to-noise ratios from various locations.  We can compare, say, AL7ID just a few miles away.  Even better, years of study have clearly demonstrated the myth of reciprocal propagation.  Approximately 80% of the time, the KL2R signal SNR will be about 6-10 dB worse at the remote station than received here.  What that means in practical terms is that when we receive JT65 at around -15 dB (good by most standards), the other station is likely to receive us at very marginal levels, if at all.  So there you have it.

Finally, let me just say presence on PSK Reporter is not the same as a call for CQ.  The spotting network does not substitute for actual RF.  So who's "foolish"?  I suggest those who would call a station without hearing their signal.  Whatever the mode, many Europeans have a bad habit of throwing out their callsign into the ether in the hopes the DX will log them, like a message in a bottle. It's remarkable that within hours of working a pileup I will get requests by email, like "Please tell me if I worked you."  Seriously?  If I worked you, you would know it.

Friday, March 3, 2017

Experience with SDRs

A couple of years ago, N1TX was sufficiently impressed with the QST review of the FiFi SDR from Germany to order a kit and put it together.  After experimenting with it as a standalone receiver, an RF Space IF-2000 board was added to one of the FT950s.  The FiFi then became an integral part of operations, providing bandscope functions and feeding VE3NEA's CW Skimmer.  HDSDR was used to drive the FiFi, and HDSDR was kept in sync with the logging and CAT software through Omnirig.  It proved very effective, and a second FiFi was added to the FTdx-5000MP.  No interface board is required, since the FT5k radio has an IF output port.  The guidance provided by G4ZFQ was invaluable to setting everything up.

The FiFi is a "sound card" SDR and a good entry-level package.

Of course, while the FiFi, SDR-Play,  Soft Rock, and others like it are somewhat plug-n-play, soundcard-type SDRs have inherent limitations due to the way frequency conversion and filtering are implemented.  Note I said "somewhat plug-n-play".  The need to tweak multiple parameters to derive optimal performance and calibration can be a pain.  

Enter Brad Forker KC7JLU, who contacted us in 2016 to discuss installing a remote SDR in Fairbanks.  I was intrigued by a system set up so anyone in the world could tune in to the Alaska receiver via the internet and Soon the CloudIQ was delivered from RF Space and connected to the KL2R antennas.   

The stand-alone “Cloud” mode includes a built-in internet server. In this mode, the radio performs the tuning and demodulation of signals and transmits the demodulated information back to a PC, OS-X, Linux or Android client anywhere in the world.  Control, display, and audio information can be supported over a modest internet connection.  There were few issues with streaming and control, which was used to great success during KL7FWX Skywarn on-air activities in December.  Lower noise levels and better antennas at KL2R yielded positive results. Download the RemoteSDR client to listen. Note: The unit is now on an amplified loop antenna at WL7CW's hilltop location.

A few brief experiments with the SDR in IQ mode were very convincing.  In this mode, the CloudIQ streams raw 24-bit IQ data at 1.8 MHz to software like SpectraVue running on a machine on your LAN.  Wired and wireless rates are sufficient, but streaming this amount of data over the internet may not succeed. (Hence, the Cloud mode).  However, run on the local network as either a stand-alone receiver or IF receiver, CloudIQ adds some powerful capabilities.  Having a completely independent, wideband receiver in the shack is advantageous.  Driving a radio these days without a bandscope is like flying blind.  Spectravue also allows for synchronization to a local PC-controlled transceiver for true point-and-click ability to select frequency, bandwidth, and mode settings.  We added a second CloudIQ to the FTdx5000 as an IF receiver.  It works great with the FT950 + IF-2000 as well.

One thing not tested yet with the CloudIQ is CW Skimmer.  The FiFi feeds IQ data via HDSDR into a VB-Audio virtual audio cable "connected" to the skimmer for monitoring 24 kHz bandwidth.  It is not clear from brief internet searches how to feed IQ data from CloudIQ.  Another concern is whether or not it is possible to maintain the SpectraVue bandscope function with point-and-click tuning while operating in skimmer mode due to port limitations and conflicts.  Even without CW Skimmer, it's a nice asset during the contests.