Monthly Archives: August 2008

Poor Man’s Guide to WSPR Spots

Okay, so I’ve been playing around with generating various kinds of visual signals on the MEPT subband on 30m. That’s the 100hz wide band just above 10.140Mhz, and it’s fun, because there are a number of “grabbers” which have the spectrum displays from their radio receivers available on the web. But there is a minor problem: there aren’t that many of them here in the U.S., and propagation often makes it difficult to find them.

But there is an alternative: WSPR. WSPR is a beaconing mode designed by K1JT, Joe Taylor, a Nobel Prize winning Princeton physicist. He is also the author of the program WSJT, which is used by many EME and weak signal operators to allow them to exchange messages at very low signal levels. WSPR encodes a callsign, grid and power level as a 162 symbol long sequence of FSK tones 1.46 hz apart, lasting for about 110 seconds. Joe’s code is open source, which is awfully cool, but a lot of it is written in FORTRAN and I’ve had difficulty getting it to run on my Macintosh. It’s probably that I’ve got a slightly odd development environment, but it’s been frustrating.

And here’s the other thing: I really don’t need all the heavy lifting to process receive level stuff. Ultimately, I’d like to be able to send this stuff with just a little microcontroller, with a fixed message. That isn’t hard, but it’s hard to get the exact details of the protocol description without digging into the source code.

But I did finally figure out how to generate a compatible message. It turns out that you can get a Windows command-line program from Joe’s website, called WSPR.EXE. If you read really carefully, you will realize that it can be tricked into telling you which tone it sends in each of the 162 slots for a messag that you specify. I specified a message containing my call (KF6KYI) my grid (CM87) and my power (27dBm or 0.5 watt), and it dumped the tones. From then on, all the work I’ve been doing with writing CW and Feld Hell encoders helped, and I generated a little WAV file that had the right tones in it. I imported that into iTunes, and then played it into my FT-817 at the beginning of an even numbered minute. I then went to the automated WSPR logging site, and…

Voila: I started showing up.

I think another hour’s worth of work, and I could have the entire thing automated, and it would totally work. Very neat.

Thunderbirds are go!

Yesterday, I took the wife and the future daughter-in-law to Travis AFB for their airshow. It was a blast. Lots of cool planes, culminating in a nice show by the Air Force Thunderbirds. I snapped a lot of photos, this one probably being the best.

MV beacon on 30m

Well, inspired by Bruce’s “Flying W” beacon, and realizing that it was a pretty nice and efficient way to send out an easily identified beacon signal, I coded up a really simple little program to generate the “MV” beacon signal (my initials). I then powered on the FT-817, while I didn’t see anything registered on Bruce’s grabber, Vernon’s VE1VDM logger managed to deliver the goods:

I started with the nominal 5 watts, of which I probably radiate a bit more than 1 watt because of antenna losses (I’m guessing). I then tried cranking it down to 500mw, which would have been about 100mw ERP, and we seemed to have lost the signal. I raised it a bit, and we’ll see how we do in a few more minutes.

Amateur Radio Station W1BW in the Morning

I’ve been trying to monitor (intermittently) some of the beacon activity on 30m. This morning, I managed to capture Amateur Radio Station W1BW, who is transmitting a “W” shaped carrier every few minutes on 10.140050. Here’s my spectrogram from this morning’s recording. The first one seems to be the strongest, but if you look carefully you can see the second one and if you imagine hard, a third:

I wonder how much power he’s running.

Addendum: Accidently left my recorder running all day, and while I didn’t see much of W1BW, I did see WB3ANQ again: his familiar sawtooth was booming in for several hours.

YouTube – Cute-1.7 + APD II Earth Movie

YouTube – Cute-1.7 + APD II Earth Movie
From the amsat-bb:

We release the movie taken by Cute-1.7 +APD II.
It was taken above Japan.

The movie is very short. But, it is the 1st movie taken by 3 kg
satellite in the world. From this movie, we could confirm that
the satellite rotates at about 0.2 rad/s.

We are planning to take pictures and movies continuously.
So we’re very happy if you cooperate in our operation.

This movie was taken at 2008/07/17 0:02:14(UTC)


Speed of Light, measured tonight…

Well, I was bored, and it was getting close to midnight. There was nothing really going on on any of the amateur radio bands, so I got to thinking. Light travels about 186,000 miles per second, or about 186 miles per ms. On 10Mhz, NIST broadcasts a standard time signal from stations in Colorado, and in Hawaii. A little Wikipedia work told me that they are located:

CM87ux -> BL01cx: bearing 254.4°, distance 2479.3 miles (to Boulder)
CM87ux -> DN80xq: bearing  73.8°, distance 1095.7 miles (to Hawaii)

The difference is about 7.44ms. So, I recorded some of the time signal tonight, where I could hear both signals. The start of the minute is indicated by a 1000 hz signal from Boulder and a 1200 Hz signal from Hawaii. Here’s the spectrogram I came up with:

Click on it to see the enlarged picture. Each horizontal pixel is 1ms. Sure enough, you can see that the lower frequency path is about 7 pixels to the left of the upper trace. Physics confirmed!

(Yes, there are better ways to detect the start of each pulse, this is just what I had on hand. Glad to see it worked though.)

Morning K & M beacons, testing my new 500hz CW filter

Ordered a new Collins 500hz filter for my FT-817 from W4RT, and installed it (really simple, remove some screws, disconnect speaker, plug in the module in the obvious place, reconnect speaker and screws and you are done). It really does help with cw and digital. As an example, this morning I tuned down around 7.039 to listen to the K and M letter beacons, and recorded 15 seconds first without the filter, and second with it. While the signals are clearly legible in both, it’s certainly less tiring than listening without it.

Here\’s the MP3.

And the resulting spectrogram. Note, you can see a QSO going on above the signals here, which is totally removed when the filter kicks in.

Ether busting in the AM…

Well, I woke up this morning, kind of bleary eyed, and decided to see if I could reach any of the other QRSS grabbers during the morning hours. VE1VDM’s grabber is apparently only receiving his transmit signal, but W1BW was up and running…

There I am, drifting up in frequency across the bottom. Copy is pretty difficult. I think I was starting my callsign about 39 minutes after the hour, but… frankly, I have a hard time reading it out here. But, the signal is at least getting out.

If I do this for real, I need to solve the frequency drift issue.

Addendum: Okay, it’s the evening now, and I’m doing another test. I started out with 5w, sending a faster version of my callsign in DFCW. Then, I cut the power to only 1w, and shifted to DFCW in a slow mode. You can see it starting below. Getting through to W1BW pretty well. I’m gonna let the signal drift up a bit for now, and see if it stabilizes.

Addendum: Eric, WY7USA (nice call) copied my attempt at sending out a signal. I didn’t mention that I was trying to send out a slow version of SMT Feld Hell, because I didn’t have any luck with receiving it, but apparently Eric did. Witness his screen grab:

QRSS Grabbed from CM87ux -> FN85ki

Well, I was experimenting a bit. I dusted off my QRSS3 iTunes beacon idea, and set it going. VE1VDM had his “big ears” grabber going, so I started watching it. He’s in grid FN85ki, which is roughly a 3000 mile path. To help assure success, I was sending 5 watts (horrendous, I know, but I am still testing this stuff and it isn’t yet dark here, so propagation which was terrible all day hasn’t really started up yet). Anyhoo, here’s the screen grab from his grabber:



I’m the signal around 10.140070. If you stare at it closely, you might just see the Morse for KF6KYI repeating (-.- ..-. -…. -.- -.– .. for those who don’t know Morse). I tried sending my slow hell signal as well, but that didn’t cut though very well. I think that’s because I’m splitting the signal into 14 different carriers, and that’s enough to reduce the signal below the detection threshold. Oh well. I think you can also see a slight upward trend in frequency as the transmission continues. As the radio heats up, it probably drifts up slightly.

Addendum: A few minutes later, we got this screen capture. You can see my signal drifting up into some WSPR traffic (which shouldn’t be below 10.140100, naughty!). I discontinued transmission at this point.

Here’s a link to the Greenfield QRSS Grabber

Addendum2: Here’s my attempt to send DFCW (like regular Morse, but “off” times are sent as well, at a frequency about 6 hz lower).

You can see that the temperature stability isn’t that great.

As the transmission continues, the drift does as well:

Addendum3: I dropped the power to 500mw, the lowest setting that the FT-817 can pull off. Here you see that the drift appears to be leveling out, and you can just barely make out the signal. Still, pretty impressive…

Slow Feld Hell Examples…

Hellschreiber is a method of sending text over radio. It basically is a kind of primitive fax machine: it sends each character as a 7×14 matrix of dots. If the receiver hears a signal, it plots a dot, otherwise it plots a space. Each character is scanned left to right, bottom to top, and are sent at 2.5 chars per second.

But all this playing around with QRSS makes me want to try for something different. I wanted to send a character every 20 seconds or so, so I represent each of the 14 rows of the character by a oscillator. Each oscillator is spaced 1Hz apart. All rows for a column are thus sent simultaneously, and we send each column for three seconds.

If you don’t have any noise, it’s really easy to decode:

Same message, with lots of noise…

Here is what the audio sounds like with the noise in place. I can barely hear the signal most of the time, but it does sort of warble in every once in a while.

I’ll have to try an on-the-air test sometime soon.

Addendum: You can see little spurs at the beginning and end of each column. That’s because each of the oscillators snaps on to full volume. That generates a little click, which contains lots of energy. Slowing on that pop would help keep the signal narrow. But the reality is that in practice, it probably doesn’t matter at all.

@ SIGGRAPH 2008

Well, I’m sitting at The Standard (a frankly far too chic hotel for a forty something computer geek like myself), it’s not quite 7 A.M. and I spent my first day at SIGGRAPH. I’m hear mostly for the benefit of recruiting: sitting in the booth, answering questions and showing up at our Pixar User’s Group Meeting. We are handing out 20th anniversary Renderman walking teapots: very nice, I managed to get one for Josh, but haven’t picked up one of my own: I’ll try to later and get a picture here.

I’m not really attending papers (you can get links here) but there seems to be quite a bit of buzz about Intel’s Larrabee architecture. Broadly speaking, the trend of GPUs has been to slowly work to expand both the number and capability of the different functional units: more shader units, that can execute more arbitrary code, and more texture units, which can present results which are available to more units. Larrabee leapfrogs this: we are back to having X86 cores (not my favorite architecture, but ubiquitous) which are fully general, linked together by a fast shared cache with scheduling done in software. To me, this represents the obvious end game to the evolution of GPUs. Companies like nVidia have been trying to tell us that we can use GPUs to do more general computation: Intel has delivered an architecture where that claim is much more obviously true.

Oh well, I’m gonna get some pancakes at IHOP, then walk the show floor a bit. I want to try to see what the state of the art in stereo monitors is, and maybe see who I can bump into. I’ve got booth duty again at 1:30 (more teapots handed out at 2:00) and then the User’s Group meeting at 6:30. If any readers are at SIGGRAPH, feel free to come by the booth and say hi.

SWL on 30m…

I was monitoring the 30m PSK31 subband this evening, and it seemed to be fairly quiet and also picking up some pretty good DX. I heard WP3UX and WP4U chatting with ZL4AS. The first two stations are in Puerto Rico, about 3600 miles away, and the ZL is about 7000 miles away. I think that might be the furthest signal I’ve picked up via PSK31. I guess my pitiful antenna isn’t entirely pitiful. Still, neither of them could apparently hear me, so I’m not getting out all that well. Still, it’s fun. One of the Puerto Ricans managed to get F5TA as well, but I couldn’t hear any of him.

Oh well, that’s about it for the night. Back to some late night Olympic coverage, then bed.