Does your two meter all-mode stay tuned to 144.200 most of the time generating the all-too-familiar white noise background of an idle band? Do you occasionally call CQ only to be rewarded by the same quiet white noise response? Tropo openings are all too rare, and E-skip conditions are all but non-existent.

How would you like to put your all-mode radio to work, making DX contacts when and where you want to? All you need is a good dose of persistence and an introduction to High Speed CW (HSCW) via meteor-scatter (MS). I started using this mode in November '97 and it has put true excitement back into VHF.

Since I am merely a neophyte, I will pass along references where you can find out more. By the way, the equipment here is modest. A Yaesu FT-290 mobile/portable all-mode pushes a standard 100 watt brick amp into a Cushcraft 17B2.

The quest for "anytime DX" begins with a brief understanding of meteor physics. On any given day over 12 billion meteors impact the atmosphere! Meteor velocities range from 10 to 100 km (6 to 60 mi.) per second depending on whether the earth's velocity in space is adding to or subtracting from the velocity of the meteor around the sun. Meteors ionize in the atmosphere about 100 km (60 mi.) high which is right in line with what we know as the E-layer of the ionosphere. Meteors come in two types, "sporadic" and "shower". We are primarily interested in the sporadic type since these occur every minute of every day. Meteors come in two flavors, those with "underdense" trails and those with "overdense" trails. The underdense meteor is in plentiful supply daily whereas the overdense is more the exception, except during shower periods. The underdense variety provides a very short reflection from the meteor head as it burns with the trail re-radiating RF rather than reflecting it. The overdense type provide good reflectivity sometimes for a long enough period to complete a contact via SSB. There are many more technical items relating to meteor physics(1) but this is enough for now.

Many of you have worked schedules during the major meteor showers. You know that the QSO is pieced together from structured transmit/receive periods. Sometimes the short "bursts" of meteor reflections resemble a "ping" to the ears. Occasionally a strong long burst occurs allowing significant information to be transferred.

Now let's concentrate on the shorter bursts. In my limited experience these appear to run from 10 to 650 ms, which translated is from "no time" to better than half a second. These short bursts are out there every hour of every day. There are more during the early morning hours around sunrise simply because the speed of the earth in space adds to the incoming meteor speed. By the same token there are fewer in the afternoon but they are still there in quantity. Now the challenge becomes one of how much information can be transferred during these short reflections? These problems have been faced before. Even the first U.S. hams to make 2 meter contact via MS used high CW keying speeds in their testing to the point that tape recordings had to be reduced by half for decoding (2).

Step forward the computer age and real High Speed CW. We are talking about speeds of up to 6000 letters per minute and even faster. This would translate to 1200 words per minute! At this speed a 400 ms burst will carry at least 40 letters which is more than enough for communications.

I first found references to HSCW while browsing the Internet. I had been absent for over 20 years from ham radio and I was curious how "weak signal" VHF work had changed. I was puzzled by the discovery that the Europeans have been using HSCW since the early 80's. They are working 3 or 4 different stations per hour using this mode. Dxpeditions on MS are done by a team of ops in order to maintain this QSO rate around the clock. Grid hunting with HSCW is very popular giving as one operator put it "hundreds of new grids (now at #651...) over the years." (3)

This mode of communication via MS was more successful than anything I had ever heard of. One operator commented, “...you don’t need major showers for a MS QSO !!! If I make a sked for tonight or tomorrow morning... in the 650-1000 mile range, we are 95% sure to complete the QSO.”4 The more I read, the more I became anxious to try this new mode.

It was apparent that there were several ways to transmit and receive HSCW. There are programs which handle the transmit job only that will run on a computer without a sound card. One such program is MSSOFT by Ilkka Yrjola, OH5IY5. This shareware program also features a MS scheduler and a fantastic MS path efficiency analysis section.
Still another way to go is to use a memory keyer and modified tape recorder for transmit and receive respectively. However, the majority of U.S. operators are using a setup which keeps both transmit and receive duties on one computer within one type of computer program.
One such program like this is MSDSP (Fig. 1). This program works in conjunction with the computer soundcard to record the received signal, save operator marked bursts or pings, and allow playback of saved signals at reduced speed. It features an "up-converter" which raises the audio pitch of the slowed-down CW note. While the saved signals are being reviewed, this program is also transmitting the operator’s data required at the time. This program, written by Tihomir Heidelberg, 9A4GL, is available as shareware in a beta version(6). There are a few bugs with the program but it is working well and making QSO's on a daily basis.
Another program in use is Cooledit. This commercial product is available in a shareware version direct from the Syntrellium homepage (7). I decided to use Cooledit simply because it was the easiest and fastest to get going for me. This program allows you to manipulate wave files. First I downloaded the individual files needed to build up the data I would need to transmit at the proper times(8). Next I created a group of files such as my callsign by opening and appending each individual letter together (Fig 2).
I then made up a “switch-box” as suggested by Andy Flowers, K0SM (Fig. 3). This is a basic computer-to-radio connection with a switch on the front which pulls the PTT line to ground. Finally it was time to put everything together and see if it worked. After floundering for a few hours trying to find the controls over the soundcard in the computer, I was able to test the receive portion. For transmit testing I asked a friend across town to record my signal and play it back to me on FM simplex where I could check it out. I really did not know what I was listening for other than key clicks because the HSCW sound was strange to my ears. After making some adjustments, I felt like I was ready for the big test., a real sked via MS! One e-mail letter was all it took. I would meet with Shelby Ennis, W8WN, on monday and thursday mornings.
As the hour approached for the first schedule I got nervous. I opened up multiple copies of Cooledit and loaded all the possible files I might need to transmit. I made sure the computer' clock was calibrated to WWV. I even made a test transmission to make sure all cables and settings were correct. Then the time came. Our sked dictated that I transmit first. I flipped the switch and began transmitting. The next one minute period would be Shelby's turn to transmit and my turn to receive. When I flipped the switch to receive I instantly heard a long loud MS burst with the unmistakable sound of Shelby's HSCW signal. I was so shocked to hear his signal on the first receive period that I just about forgot everything I had been practicing to do! After several more minutes of confusion, I discovered that I was transmitting the wrong data back to Shelby. By the time I figured out what I should do I was just about out of schedule time. Ultimately, it took about 3 skeds before I got all the procedures down and the coordination to work through a complete schedule. Then I enjoyed working three HSCW stations in exactly three days. All new grids.

Since that time I have learned a great deal. Although dry runs are helpful in preparing, nothing compares to running actual skeds. The more the better. Proper procedures in regard to frequency settings and the required data for a complete QSO must be understood (9). Transmit and receive periods are only one minute long, therefore it is important to stay alert. The radio must be calibrated precisely.

An excellent way to get acquainted with the sound and "feel" of HSCW is to download "ping practice" wave files from the several Internet sites offering them (10). These files can be played by any wave file player but to really work into the burst recorded you will need one of the free programs mentioned above. See figure 4 for the Cooledit display of the burst recorded Dec. 21st during a sked between AA6HA and KD5BUR. Can you see the characters?
By now I hope that you are ready to know more and eventually become active on HSCW. If you live within 1000 miles or so of EM23 I hope you will schedule a contact with me. I also work SSB MS skeds and would be glad to confirm this grid for you. E-mail me at:
Jim, KD5BUR for skeds or more information on anything I’ve described here.

High Speed CW is fun. It requires a dose of patience, a little bit of study and practice, and sharp operating skills. All the things I think hamming is about. Best of luck and may you have Good Reflections!

         Notes:

      1  Walter F. Bain, W4LTU, "VHF Propagation by Meteor-Trail Ionization,"
         Beyond Line of Sight, ARRL Publication
      2  W1HDQ, "The Earliest Efforts on 144 MHz Meteor Scatter,"
         QST, October 1954
      3  Joe Mutter, PA0JMV, Letter to Shelby Ennis, Sept. 4, 1997,
         WEB Version 
      4  John, PE1OGF, Letter to Shelby Ennis, Sept. 4, 1997,
         WEB Version 
      5  MSSOFT, OH5IY's Meteor WEB Page 
      6  MS-DSP, 9A4GL homepage 
      7  Syntrillium homepage, (Shareware: Cooledit96)
      8  K0SM Home Page 
      9  Meteors, MS, HSMS, Computers and Amateur Radio 
     10  KD5BUR HSCW page 
     11  K0SM HSCW page 
     12  Make More Miles on VHF 

      MSDSP, written by Tihomir Heidelberg, 9A4GL, is popular among HSCW operators.
      Above the digital clock is the main receive buffer. When a burst is
      heard, the operator marks the buffer spot and saves the burst to one
      of eight buffers above. During the next transmit period the saved
      buffer areas are replayed while varying the speed and tone with controls
      to the left. Transmit data is changed and selected as necessary at lower
      right.

      Cooledit, a wave file editor, allows the creation of files to be used
      during transmit periods. Here, my call has been assembled by pasting
      together the six individual character .wav files. A complete library
      of character files can be obtained as freeware from K0SM (see text).

      The HSCW interface box by K0SM. Allows the computer sound card to be
      connected to a transceiver. An optional isolation transformer (Radio
      Shack 273-1374) may be included.
      The switch is used to control transmit and receive functions by pulling
      the PTT line to ground. Shielded cable is preferred.

      Cooledit displays a received burst from a sked between KD5BUR and AA6HA.
      This portion of the burst is 170 ms long and shows the signal increasing
      in strength. The first action with a Cooledit recording is to "paint" the
      approximate area of the window where the ping or burst is located. Then by
      using the Zoom and In/Out buttons, the time scale is expanded to find the
      code characters. Pedro Angel, EA4EOZ has a no-code license and says that
      Cooledit solved his CW problemsx.