Thursday, July 16, 2009

Blog Post 8 – JT65A—HF Weak Signal Mode by K1JT, Joe Taylor

Blog Post 8 – JT65A—HF Weak Signal Mode by K1JT, Joe Taylor

Wednesday July 29, 2009

This is Curt Black, WR5J, with the Educational Radio Net –
Please standby for a JT65A Digital QST
.
Before the net. Please download tonight’s software – WSJT, Weak Signal Modes by K1JT and set it up according to the blog and WA-DIGITAL Yahoo Group instructions.
.
The software at 5.5Mbytes is just a little too big to be downloaded to the Yahoo site, so folks need to go to the source:
.
http://physics.princeton.edu/pulsar/K1JT/wsjt.html
and grab the current WSJT version 7.03 – the download will include the users guide in English.
.
WSJT facilitates basic digital communication using protocols explicitly optimized for a number of different propagation modes.
.
Specifically:
• FSK441 for meteor scatter
• JT6M for ionospheric scatter
• JT65 for EME at VHF/UHF, and for HF skywave propagation
• JT2, JT4, WSPR: experimental modes
.
Once you have the HF Terrestrial Mode, JT65A under your belt, you will be poised to take on the others on your own. There are some quirky things about the software, and it helps a lot to have someone show the way, but they tend to be consistent from protocol to protocol.
.
Andy, K3UK, a british sleeper cell agent in North America has produced a great guide to getting started with JT65A. It is what I used when I first tried the software and it worked great. For the net, we’ll just work our way through the setup and try sending a few transmissions back and forth.
.
The Complete Bozo's Guide to HF JT65A
( a work in progress)
by Andy K3UK

Version 2.03 Last updated March 7 2009
Images are not availabe on the BLOG - they just aren't working -
please see
.
http://groups.yahoo.com/group/wa-digital/files/
.
This article may be freely re-distributed if credit to the author is given

Introduction:
This article was originally composed in 2007. At that time JT65A was a new mode for HF, and the WSJT software was new and puzzling to most hams. Since that time, the mode and software have become quite popular and less puzzling. Another application, Multipsk, has also added JT65A. WSJT also added another mode, WSPR. A brief K3UK Guide to WSPR mode has been added to the end of this article.
.
I have left the article mostly in its original form, since it was aimed at the total beginner . I have made a few small updates to reflect some new aspects of WSJT.
----------------------------------------------------------------------
.
As a complete bozo, I am quite qualified to write this guide. I have been using the mode for a whole month now and have had "QSOs" with dozens of stations in several continents. For example, South Africa on 40 meters. All DX worked with low power and a simple piece of wire in a tree. I have no idea why the software works but if I, K3UK, can figure out how to get it to work.....anyone can!
.
If you are smart, and know more than me about this software, please feel free to email me via k3ukandy@gmail.com I will be happy to add your help to this document. Please let me know if I have added anything that is just plain stupid, or wrong!
.
History:
.
The JT65A discussed in this guide is a submode within a suite of applications collectively referred to as WSJT. The WSJT software has been available for several years (2001) and was initially designed by Joe, K1JT. This amazing piece of software has evolved to the point that there is now a team, rather than one individual, that are developing the software. That team consists of Joe K1JT, DL3LST, KK7KA, ON/G4KLX., N5HY, VA3DB, and James Courtier-Dutton . This team is to be commended for their contributions to the advancement of amateur radio communications.
.
These non-bozos have already written a document that serves as a useful guide. I have read it several times; it takes several readings for this bozo to understand just a little. If you would like to read the smart people's guide, you can find it at http://physics.princeton.edu/pulsar/K1JT/WSJT_User_600.pdf
.
The first HF terrestrial use of JT65 appears to have been a JT65B QSO on August 14 2006 between Victor UA0LGY and Tetsu JE5FLM. The recent explosion of HF activity began after VE3FGU, and several of his friends, started experimenting with the JT65B on March 23 , 2007 and JT65A on March 24. 2007.
.
On March 28th 2007, several members of the digitalradio reflector (http://groups.yahoo.com/group/digitalradio/ ) noted some unknown digital signals on 20M and on March 30th concluded that they were JT65A signals. The experiments of Mike VE3FGU, Tetsu JE5FLM, David WD4KPD, Marshall W9RVG, Seli TF3AO, and others, quickly exploded in to the hottest HF mode for weak signal DXing and experimentation.
.
What Skills Do I Need ?
.
1. The ability to set your computer's clock accurately and keep it synchronized 2. The ability to connected your radio to a sound card for receive and transmit (receive only, if an SWL). 4. Able to use a computer mouse or keyboard (fingers, feet , or nose will work just fine).
.
What can I do with this software ?
.
This is not software that is designed for a "rag-chew" ( a British-American colloquialism for having a long conversations) . If you want to find out how Bert's operation went, try PSK 31 , DominoEx, Olivia , or some other digital mode well suited for keyboard chats. This software is for brief exchanges of information. Mostly callsign, location, signal report, and confirmation of the signal report. Since it is very good for weak signal detection, it is a great QRP HF mode. You will find the software useful for monitoring propagation and, if you are a DX hunter interested it a quick 5NN and moving on , this is a mode for you! Why not a "rag-chew”? Well, you could, but the software is designed for brief exchanges of no more than 5-6 words and each exchange takes one minute. So, it would take you about 6 minutes to establish contact with Bert and find out that the surgeon removed the wrong organ.
.
The software was designed for EME and Meteor Scatter communications but I will leave that for another bozo to write about. I have to use a spell-checker to spell meteor.
.

Where do I get the software ?
.
They don't make it easy to obtain. A quick trip to Google will give you several dead links and even a visit to WSJT "Smooth Jazz”, an FM station in Florida. We want "jazz" of a different kind, so go to http://www.physics.princeton.edu/pulsar/K1JT/ for the most recent version. If you have a REAL slow computer, there are older versions that will work well, check http://physics.princeton.edu/pulsar/K1JT/Download.htm for these.

.
Installing the software:
.
Assuming you remember which folder you downloaded the software to, go to that folder and click on the WSJT7.EXE file. The install will default to your program files folder and the usual path will be c:\program files\wsjt7\wsjt7.exe. Once you have done this, boot up the program. Take a deep breath; prepare to be confused and amazed.
.

Setting Up The Software:
.
This should take less than a minute . When the program boots up, you will see a brief dos-type window pop up. Ignore this for a while, it s a useful window but something you may never need to pay much attention too. A few moments later you will see the WSJT main screen. Don't panic if the "waterfall" is dark and devoid of signals, this is normal at this stage. Click on MODE and select JT65A.
.
Here is what you will see when the main window opens up .

.

Ignore the bit about the moon unless you are thinking of a moonlight walk with that new sweet-thing, or plan to bounce radio signals off the moon. Yeah right...like that's really possible. In fact, yeah .. like BOTH are really possible !
.
Now, we will set it up for your callsign and your location. To do this, you need to click on SETUP and then select OPTIONS.
.



In the box named MY CALL, enter in your call sign. Next you need to figure out where you are and then you place your grid locator in , you guessed it... the box called GRID LOCATOR. Since you are a complete bozo like me, you may not know what your grid square is. Grid locators can get very specific, to within a few metres. For JT65A HF work, you really don't need to be that precise. The grid square for your post office or town centre will do. Americans can go to http://www.arrl.org/locate/locate.html . Britons can go to http://www.aber.ac.uk/cgi-bin/user/auj/locate.pl . It may be hard for some American and Britons to realize , but there are other countries in the world. That is good , because we need other countries to make DXing more exciting., and provide places to go on vacation. If you live in one of these other places , try http://www.qsl.net/ad7db/yourgrid.html A new alternative is F6FVY's Google Maps server that allows you to find your location on a Google map and your grid square will pop up on the map. Check http://f6fvy.free.fr/qthLocator/fullScreen.php . The WSJT software does not require grid locators but the grid locators are used in the default mode instead of an RST report.
WR5J likes http://www.qrz.com - put in your call sign and look yourself up. Then click on the green “Click for more detail” bar. Then click on the link to the “Explore on Grid Map” and check out your location – you can scroll around and actually update your location if it is off in the FCC database. You should so definitely do that…use the EDIT RECORD link.
.
Setting the Radio for Transmit
.
OK, so you have entered in your callsign and location. If you want to make the software key your radio when transmitting, you need to tell WSJT which comm port you are using for "PTT". My station uses a virtual serial port , port 5. You will likely use 1, or 3 if you are using a regular serial port. Enter your port where it says PTT PORT.
.
Just one more thing to do. The software can calculate the distance between your grid locator and the received station's grid locator. Most sensible folks use metres but a few people clutch to a old system referred to as "miles" (mainly those Brits and Americans, again). Remember, I am a complete bozo, so I use miles. Click on the ML or KM in the DISTANCE UNIT area.
.
That's it. Leave all the other stuff alone.
.

Receiving signals :
.
Assuming you already have the audio from your radio receiver connected to the input of your sound card, you are close to being ready to receive. Remember though, the software was written by somebody completely intelligent. There are many useful things that can be done with the software, the smart author thus decided that the receiving of signals needs to be activated. Your WSJT main screen should now look like this.
.



Note that I have highlighted a button called MONITOR. Note also that the other area I highlighted has your callsign and grid square inserted. Now, click on the MONITOR button with your radio tuned to any HF frequency and you will eventually see signals on what looks like a waterfall. I say eventually, because this waterfall is one thoughtful waterfall. It needs to think a while about what it is receiving and then, precisely when your PC's clock signifies the start of a new minute, the display will come to life with some interesting squiggly lines. Assuming you see some squiggly lines , you are in good shape and ready to move on. To be honest, this "waterfall" is not a waterfall in the traditional sense. The waterfalls that you see in Digipan, Multispsk, fldigi, MMTTY, Winwarbler, DM780, MixW, etc etc, display the signal on a continuous basis and indicate the intensity of the received signal. This display in WSJT actually takes a snap shot of what you have received in the past 48 seconds and displays it all at once. The moon-bouncing meteor dodging hams that use the WSJT software, use this aspect to analyze minute details that enable them to determine if the Martians have actually entered our ionosphere. It is an important "main graphical area" ( as the official guide calls it) , even for HF applications. More on how to interpret this later.
.
Frequencies :
.
We might as well get the radio tuned to frequencies known to be used for JT65A on HF. The most commonly used frequencies are 14.076 and 7.076 in North America 3576 (North America ) , 3796 (Europe) , also 18102, 18106, 10147, 21076, 24910 . 1805 to 1808 . 14076 or 7076 are the best places to start. These are DIAL or VFO frequencies, most use USB but there are some that use LSB. By far the simplest method of figuring out where the action can be found is to use your web browser and go to http://www.chris.org/cgi-bin/jt65talk . This web site is the place to go for announcements of who is active and what frequency they are using. The website is maintained by Chris, N0UK. Yes, he is N0UK and I am K3UK. We are both part of a British plot to take control of the JT65 world and lay the foundation for re-acquisition of colonial America in 2076. The Queen visiting the Kentucky Derby was also part of this plot. IF you do not trust us Britons, there is an American you might try. Check in at http://www.w6cqz.org/ . This American is non-bozo, he is clever enough to develop a system of "reverse beacons" , you can transmit your JT65A signal and see if other stations automatically receive your signal.
.
So tune to one of the above frequencies , get a cup of coffee and relax for a minute or to.
.
Tuning Accuracy :
.
Now, some important stuff. The smart JT65A users will eventually point this out to us bozos, so I figured I will tell you now. The software decodes signals over a wide range, up to 600 Hz. So you can decode several signals within this 600 Hz range . You can be decoding all of them but if you transmit you could be several hundred Hz from the other station's transmit signal. This , to the JT65A afficinados, is a crime and war crime tribunals may get formed. Even a bozo like me can understand this after a while. It's like working "split" and taking up more spectrum than is needed. It is also like not being zero-beat in CW and trying to work a person who has a 200 Hz filter active. If you are 300 Hz from his frequency, he is not going to hear you. Don't fear, this is easy to address, more on this a little later in the document..
.
Let's not worry about accuracy at the moment, lets just tune a signal. After a few minutes on an active frequency your main screen may look like this
.


The blue highlighted text is my work, your screen will not automatically highlight text..
.
The highlighted text is...
.
163500 11 -7 -0.7 -73 3 # DL5SWB DK7OM JO53 OOO 1 0
.
What the heck does this mean? Smart guys already know, but bozos like me are still learning. It tells you that at 1635 (and zero seconds) the software decoded DK70M responding to DL5SWB. Just like a regular phone contact "P5DX de K3UK". . It also tells you that DK7OM is in grid square JO53 . This string of information also tells me that the dB was averaging -7 during the minute or so the software was decoding DK7OM's signal . This is a signal to noise type of reference. 7 dB in to the noise. This is a good signal, WSJT will decode signals well below -20 dB. The minus 73 under the DF area tells me that I am close in frequency , about 73 Hz off, good enough for now. I'll save the other info in that string for a later discussion.
.
What About The Rest In The Window. ?
.
The "RO" RRR" and "73" are just snippets of a conversation used to confirm the QSO. Its like tuning in your radio and hearing someone say "Roger" or "73" . More on this later. Version 5.9.7 has an option that suspends use of the RO , RRR, and 73 messages. This option was specifically added for the new HF application of JT65A. The RO , RRR, and 73 messages in the default format were designed with EME and Meteor Scatter QSO in mind and are as brief as possible. For HF application of JT65A, the need for very brief bursts of information is not paramount., so the new version has added an option to include callsigns and signal reports . The RO , RRR, and 73 messages are messages in a special format and can be decoded and even weaker signal levels that the other messages. So when signals are very weak on HF you may still see these messages.
.
That Weird Waterfall Thingy :
.
As mentioned earlier , that weird waterfall thingy (the "main graphical display" ) can tell you a lot. The rookie user needs to know a few basics that will enable you to be closer to the received station's frequency. In the picture below you will see a red spike on the spectrum display.
.

You can take your mouse and click on the red spike (see center of picture above) and that will get you close to the right frequency. The peaked blue spike tells you something useful too, I will add that at another date.
.
Those RRR and 73 reports that are decoded will show up as orange or purple lines on the waterfall.
.

Reportedly, clicking on the purple peak will also serve as an tuning aide, I have not tested that yet. For meteor scatter and EME work, the orange, purple, and green lines can be valuable in digging out a signal. For JT65A on HF, they are not as important.
.
Using a "waterfall" that has to think for a minute before it updates information can be frustrating. Don't worry, the author has another trick up his sleeve, a second waterfall! This is a REAL waterfall just like the ones Peter Martinez and Skip Teller invented for use old PSK31 users. Click on VIEW at the top of the main window and then select SPECJT. A real waterfall will open up (if you are using an older version you may need to manually boot up "spectran.exe" for this ). This waterfall can detect RF generated by a flatulent butterfly 1000 kilometers away. Eventually you will stare at this waterfall and see the tiniest hint of a signal and be amazed as the software decodes information from it. Note on the picture below, you will see this second waterfall, a received signal and the mouse cursor indicating tuning the signal to the sync tone. For HF work, this second graphical display will be the most important display for identifying the presence of a signal and in accurately tuning in the station.
.


Clicking with the mouse, on the longer sync tone will get you very precisely tuned. The sync tones are actually 1270 Hz, you and your dog should have no difficulty learning to recognize the sync tone both in the SpecJT, and via your ears (actually, your dog may have difficulty understanding the waterfall display, sorry to point this out ) . The authors of this software think of bozos like me and actually send the sync tone quite frequently, so you get lots of chances to spot it. If you have monitored the active HF JT65A frequencies you will have heard a few JT65A signals. These tones sound a little like the old Throb mode, and a little like very slow DominoEX. This throbbing type signal actually sends the important sync tone about 65 times in a typical sequence, the other tones in a sequence are sending information. The tones actually change a little during a "QSO", and some of the messages actually don't use the sync tone in the same way. That's too complicated for a bozo's guide, but the aforementioned smart people's guide explains it well. . The general rule of thumb, detect a signal within the SpectJT that looks like JT65A and click on the LEFT edge of the signal. Be aware that the SpectJT will display all kind of other signals within the range of the waterfall, so you may see PACTOR bandits, PSK Rangers, RTTY rompers, Hell raisers, and a plethora of other digital modes. JT65 is fairly easy to spot after a couple of attempts, the only other digital mode that looks a little like it in the SpectJT is MFSK16, but this sounds very different.
.
The signal in the picture above is a very weak signal. Sometimes just a very very slight line has shown in the SPECJT waterfall and I have successfully decoded a signal. Remember, you don't need to be that precise to just decode signals; you need to get precise to be transmitting as close as possible to their transmitted signal. The reason you want to be close is because the smart JT65A operator my not want to be receiving signals more than a certain distance from his transmit signal. He may have, for example, set the software to ignore signals from more than 200 Hz away from his signal. So, if you are 300 Hz away, he will never decode your signal. I often listen with a 600 Hz range so that I can get a sense of who is active on the band, but switch to 200 Hz when trying to have QSOs. To set the range, left click on the number in the TOL box to raise the setting, right-click on the number to lower the setting. Click the "freeze" button to maintain this.
.

Remember to raise TOL to 600 and Freeze "off", if you are just looking for QSOs.
.

While the SpecJT window is open, click on OPTIONS and select JT65 DF Axis. This will help further in precision tuning. While there, you can also play with the palette settings and vary the appearance of the waterfall. There will be a few occasions where you cannot even see a signal even in this highly sensitive waterfall. The different palettes will make it possible to see extremely weak signals, try "hot" for example.
.
Note at the bottom of SpectJT there is a "gain" area, just to the right of the digital clock. You can adjust the gain of the waterfall with this control. You can also adjust the brightness and contrast of the SpecJT waterfall. Try that, if your ears detect a signal but you don't see it on the waterfall. The gain control can be an important factor if you experience a situation where you are unable to decode signals, please see the "trouble" section of this document.
.


A good Bozo's guide should be short and sweet, but I need to explain a few more useful tips about receiving a signal. Take a look at this
.

The stronger signal at the bottom right-hand corner is VE5MU. The weaker signal is K4CML. They are working each other but not exactly on the same frequency. Their QSO copied at my station shows....
.
022200 3 -16 -1.5 -143 4 * VE5MU K4CML FM17 1 0
022300 19 -4 1.4 -22 3 * CQ VE5MU DO70 1 0
.
Pretty close though and will work unless they have their TOL set to less than 100.
.
Now take a look at this
.

VE5MU and VE7TIL copied at my station on 40M, both almost exactly tuned with each other. My station recorded their QSO
024800 9 -7 1.5 -121 3 # VE7TIL VE5MU DO70 OOO 1 0
024900 3 -9 0.1 -116 3 # VE5MU VE7TIL CN89 OOO 1 0
.
Although -121 and -116 DF from where I was tuned, they were within 5 Hz of each other.
.
Please note, the callsign in the above pictures do NOT appear in the SpecJT, I added them for illustrative purposes.
.
I am sure you wonder what all the other numbers mean, this will be added at a later date but it is worth mentioning that the last number in the line (the zero as in " 1 0 " ) is an indicator of how confident the software is that it has decoded correctly. The software is quite sophisticated and sometimes acts like a human ear/brain, it guesses what it heard. The confidence scale goes up to 10 , supposedly anything under a three is questionable. I'm still trying to figure out what most are given a zero but even when obviously accurately decoded. The smart author of software, Joe W1JT, took pity on this bozo and sent me the following...
.
"You mention the two numbers at the end of each line of decoded text. If the first of these is 1, it means that "Deep Search" was not required and decoding will nearly always be 100% correct. If the first number is 0, the second (some number between 1 and 10) gives an indication of relative confidence of a Deep Search decoding. Since the default list of "plausible" callsigns is VHF/UHF and EME oriented, Deep Search will not be very useful on HF unless the operator becomes fairly sophisticated in using WSJT".
.
Deep Search has no relation to Linda Lovelace and has apparently no connection to the Deep Throat that did Nixon in. It is a software function designed to help the moon-bounce and meteor scatter hams find fragments of RF and match the fragments to known users of the software. It is recommended that you turn this feature off in the main graphical display area , by going to OPTIONS/DECODE/JT65 and then checking "no deep search" . Leaving it on does not cause global warming but does, on occasion, cause a "phantom signal" to call you. If you see a call to you when you have not been transmitting, it is possible that Deep Search caused it.
.
One more item on receiving...
.
In the SpecJT waterfall you will notice a receive noise level indicator in the lower right-hand corner. It displays the noise level in dB. I have been told that things do not decode very well if you have something like a 4-5 dB level (above zero, not below) . I have also noticed that it does not decode very well when you have a very low figure, like -20 dB. A number of factors can influence this and the noise level will change from band to band. The biggest single factor will be whether you are getting enough, or too much, audio from your radio to the input of your soundcard. Many digital mode interfaces like Microham or Rigblaster products have knobs that control the audio levels. If your RX noise level is too high or low, adjust the audio so that your noise level is around 0 db without a signal present. Smart people have also suggested it works better with AGC off, play around and see. You can experiment with your rig's filters, noise reduction circuitry, AGC settings , RF gain settings , etc, etc.
.
Transmitting Signals/Having A QSO.
.
I've said this three times already but... to transmit you first need to understand that the smart JT65A guys all expect you to transmit at a certain time. That time needs to be synchronized, so synchronize your PC's clock with Internet time or WWV. I use an application called Dimension 4 that keeps my clock synchronized. Why? Because the software needs time to decode your transmitted signal and also switches to transmit at certain times. If your timing is off, the other station may have already switched to transmit and thus not decode your signal, or their decode period would be so short that it did not get a chance to decode all your information. You can get the freeware Dimension 4 at http://www.thinkman.com/dimension4/download.htm
.
Transmitting Frequency.
.
Popular digital mode software like MixW, Multipsk, MMTTY, have functions that maintain a relationship between the received and transmit frequency. These functions, often via a "net" on/off feature, cause the radio and/or soundcard to adjust when you click on a signal in a waterfall. The WSJT software has a similar function and is labeled in the main graphical display as TxDF= . In the older versions one had to always match your transmit and receive frequency manually. It may be useful to first understand how to do this, and then learn the new TxDF= feature. So, lets begin an manual reception and transmit.
.


Here we see a CQ from P5DX , my radio was tuned to 14076 USB but my eagle eyes (and bat-like ears) spotted a signal further down the waterfall. I clicked on the left-side of the signal on the SpectJT waterfall and the green scale markers shifted to show that I am tuned about 500 Hz below where I started. Although working North Korea is becoming routine for me, I take pity of the poor guy and decide to work him once again. If I just transmit, my rig will transmit at the zero mark, 14076 USB. He is 500 Hz below, at 14075.500. So all I really have to do is manually turn my dial 500 Hz lower, to 14075.5 and then...
.


Here you can see the old received signal shifted 500 Hz to the left and more importantly the new receive frequency is close to my transmit frequency at 14076 USB. Simple , right?
.
To get more adept at precise matching of the received frequency and transmit frequency, use your mouse to move left and right on the SpecJT waterfall. Note how the frequency and DF changes.
.


Mouse pointer at 881 Hz and 388 Hz below transmit frequency.
.

What About This New Fangled "TXDF= feature?
.
Not only am I a complete Bozo, I can be a little mean. I threw the word "fangled" in here just to puzzle the folks that bring you the Russian and Spanish translations of this document. I think we should offer a prize to someone who can tell us the origin of the word "fangled". Anyway, I deviate from the mission at hand. Actually "deviate" is a key term for this section of the guide. When you tune a signal and successfully decode it, the decoded text will tell you how far your tuned frequency is from what will be your transmit frequency. For example , this decoded text
.
035100 10 -15 3.4 -160 3 * VK7CAV N5BO -16 1 0
.
Shows that the DF was 160 hz below where my transmit is set. The very smart author of the software has now enabled us to match the receive and transmit frequency instantly via a simple double left-click with your mouse.
.

TxDF set to default zero TxDF adjusted 98 Hz and matched
.
To do this, locate the signal in the SpecJT, aim your mouse on the left edge of the signal and double click on it. Now single left click on the TXDF and the deviated frequency will be indicated You , despite being a bozo, will have achieved something akin to mastering rocket science...your transmit frequency and receive frequency will match! Now, actually the smart author is doing this via having the software fiddle around with the capabilities of your soundcard. Your radio is not adjusted, don't worry about that "split" button being activated or, God forbid, that RIT thingy being turned on !
.
A word to the wise: Users of some RTTY or PSK31 software might be used to the fact that once you have instructed your software to match receive and transmit frequencies, they stay matched even as you re-tune and work stations on other frequencies/bands. This WSJT software requires that you make the adjustment each time you are going to transmit, assuming the station you want to work is on a different frequency than the one you just worked. For example, suppose you worked a station that had a DF of 233 Hz. You left-clicked on TxDF and it matched your settings so you were exactly on the same receive and transmit frequency. Now that you have worked the station, you see on your screen that KT2Q has been decoded with DF of -321 Hz. You will need to again click on the TXDF button for it to match the DF , before you transmit.
.
Audio levels:
.
As with any digital mode that relies on your PC's sound card, you need to pay a little attention to the quality of your transmitted signal. Typically, when using modes like Olivia or the old PSK31, people suggest you keep your ALC level to the very minimum. One of the non-bozo types emailed be recently and suggested that the nature of the JT65A signals is such that you don't have to worry about this as much , and can give a good amount of ALC. I still keep mine fairly low, old habits...
.
Another friendly ham reminded me that this mode is full duty cycle and he suggested that people should not use full power for long periods. This mode does not need a whole lot of power. A ham, Tony KT2Q, popped up one day and said he had a "Bird meter in line". Being not only a bozo, but a British bozo, I wondered whether a bird meter was something similar to "a chic magnet", perhaps? Maybe it can detect the fairer sex at a thousand paces? Well, it turns out that this KT2Q guy is one of them non-bozo hams. The bird meter had a "plug" in it . So does my bath tub, but the bath tub does not help with most of my radio projects,. Well the "plug" enables him to calibrate his radio output and accurately determine how low his power is . I was able to detect and decode his JT65A signal when he was using 250 milliwatts on 40 meters. Another ham, some 600 miles away from KT2Q, was able to detect him below 50 milliwatts. So, you don't need to have your transmitter set for high power. I often run just 5-10 watts on 20M.
.
CQing.
.
If you want to CQ, you have to decide whether you want to transmit "1st minute" or "second minute". This helps to maintain world peace and keeps QSOs orderly. "First minute" means 00 and any even numbered minute 02 , 04, 06, and so on. Second minute means 01 and any odd numbered minute , 03, 05, etc.
.
So, if you were transmitting a CQ on the first minute, you should set the main window to TX 1st and select your CQ message. You should also use the auto transmit method. As illustrated below.
.


At the precise minute, your rig will transmit. It will send the CQ for about 48 seconds . You then will wait until the beginning of the next odd numbered minute and , if lucky, you may hear or see a signal responding to your CQ. Please be aware that you may NOT see a signal in the SPECJT or hear a signal with your ears but may actually be decoding a signal. After about 50 seconds in to the minute, a blue decode button on your screen should be active and around 55 seconds ,or so , any decoded signal will pop up on the screen. On my slow computer, the decoded signal sometimes does not pop up until several seconds in to the next minute . This can be annoying and confusing, so if this happens to you turn on QUICK DECODE and the decoding process will start about 5 seconds earlier. You will then get the decoded text in time to read the information before your next transmission. To turn on QUICK DECODE go to go to DECODE in the main graphical area, select JT65, and then Quick Decode as illustrated below

.



If, after CQing, you receive something like this
. 171500 0 -7 -0.1 471 3
It probably means no one responded, or your rig did not hear anyone respond. This is essentially random noise.
.
If you received something like this ...
165200 10 -11 -0.2 -27 5 * YOURCALL WQ5S EM13
.
It means WQ5S heard you and is responding with your callsign , his callsign, and his grid locator which serves as a report (they may also respond sometimes with and R and a number instead of a grid square, this is a signal report) Here is an actual example..
.

165100 12 -7 -0.1 97 3 * CQ VE5MU DO70 1 0
165200 10 -11 -0.2 -27 5 * VE5MU WQ5S EM13 1 0
.

Lets pretend you are VE5MU (haven't we all wanted to be VE5MU at some point in our lives ?) You called CQ, WQ5S answered. Now what the heck do you do ? The new version now gives you a choice of what responding message format you can use. If you left-double click on the callsign of WQ5Q you will see
.

Some standard messages have incorporated your callsign and his callsign automatically. However, the new recommended format for HF use of JT65A is that you access a different message set. Instead of left-double clicking on a received callsign , RIGHT-double click. At my location this causes a 20% increase in carpel tunnel syndrome symptoms, but hopefully all it causes at your house is the following.

.

Note, I have now switched from using VE5MU as an example to using UA0AET. In the above box, you will see that it not only copied the callsign of the other station when I right-double clicked, it also copied the signal to noise ratio which is illustrated in the TX2 and TX 3 boxes . In this case it is -7dB, a good signal! If you used the right-double click maneuver, you should note that the Auto-ON button has turned red. This means that your radio will begin transmitting automatically at the top of the next minute. Since you are the station calling "CQ", you most likely would have already selected auto-on.
.

Before each transmission, you need to determine which of the possible six messages you will be sending. Obviously, if CQing, you will select TX6. So, if P5DX in North Korea has heard your CQ and has responded with:
.
VE5MU P5DX PM27
.
Now what the heck do you do? Well, just like in most other amateur radio modes, you respond with his/her callsign to let him know your heard him/her. You also respond with a signal report. Under the new "HF enhanced" version of WSJT-JT65A, you would respond with TX3 which is... "P5DX K3UK R-7"
.
P5DX would then send his report as ...VE5MU P5DX RRR (TX4)
.
You would then respond with either message TX4 also or some operators just skip to message TX5 and sent the 73.
.
P5DX may also send 73 but it is not required.
.
All of this assumes you are hearing each other well. Suppose you got the first couple of messages, but when P5DX took his/her minute to confirm your report with RO, you never received it. The convention is that you just send your previous message again. P5DX will receive it again (hopefully) and realize that this means you did not receive his RO. He will send it again and , if you copy it this time, you will send..... This format can be repeated for any of the message stages .

Writing this down in a Bozo guide can make it more complicated than it really is. Remember that smart bloke that wrote the software? A dude called Joe. Joe is so smart that he stuck simple way of helping us bozos right in to the software. If you hold your SHIFT key down and press the F5 key at the same time, you will see something like this ...

.

Actually, I broke it down for us bozos. The chart above explains the newer method intended for HF use under fair to good conditions.
.
Alternative Message Format
.
As mentioned earlier, the new HF enhanced version should work well for most HF conditions. However, you may run in to some "old timers" who use the older method, or find someone using the older "short hand" messages because they can be decoded better when a signal is very weak. So, it may be useful for this Bozo's Guide to explain this message format too.
.
The author's chart best illustrates this:
.




Don't ask me what the OOO actually stand for, but this exchange means that you have received WQ5S’ information. If WQ5S received your 000's, he will respond with simply with RO, no call sign, just RO. This means he has received your 000's. After you receive his RO, you send message number TX4 by selecting it before your turn to transmit arrives. TX4 will transmit at the top of the minute and send RRR. Again, no calls signs, just RRR. This means you have received everything for a valid QSO, : A valid QSO consists of his callsign , the fact that he has your call sign correct, the acknowledgment that he received your OOO's . He will send you RRR to let you know that all is correct his end. Often, you then send the message number TX5 with a 73 in it. Some people type in extra info in this message like
.
73 TU good night ur best -7 db. This is optional and usually is typed in to TX1, message number 1.

.
This will sounds confusing but after a couple of contacts you will have it easily figured out. If you did not receive the acknowledgment of the OOO's or an RR, just resend and hopefully you will get it the next try.
.
Remember, the teams designing the software are not bozos, they are smart blokes and their inventiveness is extremely helpful to us bozos. In their cleverness, they designed a mode that actually changes the nature of the digital tones depending on what is going on in the QSO. This is almost like switching from SSB phone to CW in the middle of a QSO because you want to make sure your weak signal got through. In JT65A (and other modes in WSJT) a CQ message , TX1, will appear differently on the SpecJT waterfall that other messages. The RO or 73 messages will appear different too. Let's take a look at the picture below
.

Short Hand Message in older format and for WEAK signal decode
(I have RRR confused with RO, I will re-do this graphic later when I get a better capture, but it does basically illustrate what you need to know)
.
I have illustrated , in the above picture, the typical patterns you will see in the SpecJT. With a little experience you will be able to visually SEE what is going on in a QSO, and know which part of the exchange is taking place. You will also be able to HEAR the difference because the tones sound different to the human ear. Knowing what the different stages (messages) of a QSO sound like, can be useful when working weak signals in a pile-up.
.
You will note that the CQ signal is quite distinctive, much wider than the short-hand messages.. Some people get confused by the fact that the RO, RRR messages appear without a call sign. Was he really coming back to me? It is confusing, that is why K1JT revised the software and added the new message formats that can be accessed via double right-clicking on a callsign. These messages have callsigns throughout and avoid the confusion.
However, the nature of these "shorthand" messages is such that they can be decoded a further -5dB in to the mud ! This can make the difference between working that rare DX station or not.
.
Some people will send you a message that says something “Best -18 TU 73". These smart people are manually typing the message in to the TX1 message text box and sending it instead of the optional TX5 message. Since it is sent after a QSO has been validated by RRR, it is perhaps not that essential that the message get through, so they use the wider , less robust, TX1 message.
.
Responding to a CQ
.
Take a look at this
.
182800 9 -13 0.4 -124 4 * CQ N4WO EL88 1 0
182900 10 -15 0.0 -73 3 * CQ DK7OM JO53 1 0
183000 7 -10 0.0 -124 3 * CQ N4WO EL88 1 0
183100 7 -19 -0.1 -73 3 * CQ DK7OM JO53 1 0
183200 7 -15 0.0 -124 21 * CQ N4WO EL88 1 0

.
N4WO is CQ on the first minute ( 1828, 1830, 1832, ) and DK7OM is transmitting the second minute (1829 and 1831) .
.
Both are within 61 Hz or so of each other but are not hearing each other. If you want to work them, you have a choice. Who do you want to work? The new HF enhanced version makes the choice a little easier. If you want to work N4WO there are three things you need to do..
.
1. Determine which minute N4WO is transmitting.
2. Set the messages group with callsigns and signal report.
3 Match the received frequency with your transmit frequency
.


Setting 1st or Second Minute : In the above example, N4WO was transmitting at 183200, 32 minutes past the hour, and even numbered minute. This is the first minute, so you need to unchecked the TX1st box so that you transmit the second minute. You will do this just once at the beginning of the QSO. The second minute sequence will be maintained throughout.
.
Setting messages group with callsigns and signal report.: If you wanted to work N4WO you would have seen something like...
.
183200 9 -13 0.4 -124 4 * CQ N4WO EL88 1 0 Right-Double click will populate the message groups with the callsigns and signal reports..
.

Matching Received and TX Frequency: After you have selected N4WO, above, you can then double click on the signal in the SpectJT (or on the red spike in the main graphical display ) and then click on the TXDF button. It will turn red and the frequency deviation will be shown. Using the above example, it would read TXDF= -124. That's it, you are matched!
.

So, at the top of the minute, your rig sends TX1, which is N4WO VE5MU DO70
.
N4WO receives this information and now knows that you are calling him and his signal report is -13 db.
.
He sends TX2.... VE5MU N4WO -8
.
He is letting you know that he is receiving you with a signal to noise ratio of -8 dB.
.
You then send... N4WO VE5MU RRR or the 73 message.
.

This tells N4WO that your received his R-8 report and that you are "rogering" the report. N4WO will like this because when he receives "RRR" or 73 it makes his contact official. He may reply with another RRR, rogering your information or he may skip straight to a 73 message. "VE5MU N4WO RRR" or "VE5MU N4WO 73"
.
So, the QSO is "good" now that you have received his "RRR" or "73" .
.
Instead of TX5's basic 73. Some folks get a little fancy and send a brief message , perhaps like "Tnx John 73". This is usually done by typing the "Tnx John 73" in to the message box TX1.
.
You will notice that when transmitting, the message box you have selected changes to a certain color. More on this later, it just signifies what type of message you are sending. I am mentioning it here because some new users panic when they see the message box turn RED. Red does not mean your rig is about to explode, it does not indicate anything bad at all!
.
Ok, I just worked North Korea on 160M, where do I log it. ?
.
Clicking on the "log QSO" button will log the QSO but you will need to first tell WSJT which band you are on, the software does NOT read your radio's frequency. To do this, click BAND in the main graphical display and select the band. Note, the bands are labeled via Frequency not wavelength. e.g "14" is 14 mHz or 20 metres.
.
You can view the log by opening the log.txt file with a text editor , like Wordpad or Note pad. log.txt can be found in your main WSJT folder.
.
In the folder there is also another interesting text file, called all.txt. This tells you knot only who you log but who you received and when you transmitted.
e.g.
.
010100 3 -17 -0.4 -13 4 # K3UK ZS6BUN KG43 OOO 1 0
010300 3 -13 -0.1 -16 3 # K3UK ZS6BUN KG43 OOO 1 0
010414 Transmitting: JT65A RO (Shorthand)
010500 10 -19 -15 1 RRR ?
010612 Transmitting: JT65A RRR (Shorthand)
010700 10 -16 -15 4 73 ?
010815 Transmitting: JT65A 73 (Shorthand)
010900 6 -13 0.3 -16 1 * TNX ANDY -11 1 0


.
I'm hooked, where can a get more of this J65A drug?
.
Go to http://www.chris.org/cgi-bin/jt65talk this is a great resource for meeting other who are looking for QSOs. There is a large, very active email discussion group at http://groups.yahoo.com/group/digitalradio/
This group has a lot of discussion about J65A HF and other digital modes. A WSJT specific group can be found at http://groups.yahoo.com/group/wsjtgroup/
.

I've worked a lot of JT65A and am not getting the same high anymore, is there a more potent drug?
The bigger "high" can be gotten from EME or Meteor Scatter operations. Once addicted to this, you become known as a Ping Jockey. See http://www.pingjockey.net/cgi-bin/pingtalk for more on this addiction.
.

Trouble :
.
Remember that "DOS-type" window that I told you to ignore at the beginning of this document. This can actually tell you if there is a problem with the sound card device or if your PTT circuit is not open. If you suspect a problem of this nature, log on to http://groups.yahoo.com/group/wsjtgroup/ and ask for help their.
.
Soundcard Clock
.
If you have an obvious JT65A signal that you can hear and the waterfall sees but it is NOT decoding, you may want to make sure your soundcard clock is accurate. Most digital software like MultiPSK, MixW, and Winbarbler, have features that will enable you to adjust your soundcard and keep it accurate. The WSJT software displays your clock rate as highlighted below
.

You want your settings to be around a 1.0 .
.
It appears that WSJT also can adjust the soundcard rate, check the SETUP/OPTIONS /RATE IN /RATE OUT area.

.
Audio Levels in to the Soundcard
.
As mentioned earlier in this document, you may experience problems in decode if the audio level is too strong or too low. You can lower or raise these values first with the Gain Control slider in SpecJT and additionally with your soundcard line-in control mixer. With no signal present, -2dB seems to be a good level. These are not always critical but if you cannot decode an obviously clear signal you might want to make sure you are not in the -20 dB or + 20 db range. See below for key areas in control of the audio level.

.

If you cannot get enough audio from your rig to the PC soundcard, consider using the MIC-IN jack on you soundcard. Be careful not to put TOO much audio in to your soundcard.
.
Acknowledgment:
.
As a complete bozo, I needed help understanding things, Thanks to all that responded to my questions in the first few weeks of learning. Thanks to David WD4KPD, Tony KT2Q. Chris, N2YYZ, Steinar LA5VNA, Mike VE3FGU, Colin K4CML, John VE5MU, Bill N9DSJ, George G4PCI, Bruce, N5SIX, John W5UBW, Walt K5YFW, Scott VE7TIL, John N2JH, Al WB7SWW. Marshall W9RVG. Joe K1JT. More to be added soon.

.
K3UK WSPR Guide
I am not sure if it is really needed nowadays, but I just updated my
quick guide to WSPR by K1JT. This reflects the latest significant
advancements that Joe has programmed.
.
K3UK's Quick Guide to WSPR Revised July 26 2008.
.
There are now two options for using WSPR. WSPR.exe is the manned
beacon only software . The latest beta release of WSJT now includes a
version of WSPR that allows for two-way "QSOs". WSJT also includes
modes FSK440, JT6M, JT65 variants, JT4 variants, JT2, and may also
add JT6a soon,
.
This quick guide has been updated to reflect the WSPR changes. The
quick guide to the QSO mode appears towards the end of this guide.
.
General Information:
.
Download software from
.
http://physics.princeton.edu/pulsar/K1JT/WSPR100.EXE
http://physics.princeton.edu/pulsar/K1JT/WSJT700r973.EXE
.
Get detailed help from
.
http://physics.princeton.edu/pulsar/K1JT/WSPR_Quick_Start.TXT
http://physics.princeton.edu/pulsar/K1JT/WSJT_User_600.pdf
http://www.obriensweb.com/bozoguidejt65a.htm (a little dated )
.


*** WSPR : Very LOW Power Attended Beacon Mode ***
.
1. First , remember it is NOT a QSO mode. It is "manned experimental
propagation transmitter". Stations transmit the audio tones at varying
times (as determined by the software) for TWO minutes. It is a "slow"
mode, so does well in very weak conditions. MANNED is important it is
not intended for unattended operations. Get the software from
http://physics.princeton.edu/pulsar/K1JT/WSPR100.EXE
.
2. Make sure your PC clock is sync'd with an Internet time server. If
your clock is off by more than 5 seconds or so, other stations my have
trouble decoding you.
.
3. Keep you power low, most are one watt or less, some 5 watts. More
than 5 watts can cause too strong a signal and the software sometimes
has difficulty decoding very strong signals.
.
4. In the SETUP/Options area of the software. Enter your callsign,
grid square , PTT port, and power in dBMm (example 30 dBM = one watt).
Go to http://www.radioing.com/eengineer/rfcalc.html for some dB to
watt conversion help. (tnx K2MO).
.
5. That is all you need to do for set-up , if you have more than one
soundcard in your PC, you may need to additionally change the device
settings if you want to use the NON-default soundcard.
.
6. Tune to a known WSPR/MEPT frequency. 30M is the most popular with
10.138,6 the standard listening frequency.
.
7. In the main WSPR window, enter 10.1386 in the RX box and then
choose a TX frequency
.
8. The TX frequency is the frequency the soundcard will generate the
tones. Your RIG should NOT be in split mode. Just keep your VFO on
10.1386 USB. Your TX audio tones should be around 1500 Hz above the RX
frequency. The software has a default 30M TX frequency, I suggest you
change it so that everyone is NOT transmitting on the same frequency.
I use 10.140119 In the following example, you will see what other
frequencies people are using ..
.
080409 0120 4 -16 0.7 10.140163 VE7TIL CN89 30
080409 0120 5 -14 -0.6 10.140198 WA7ADK DN31 18
.
VE7TIL was received by my rig on 10.149163
WA7ADK was received by my rig on 10.140198
.
Other slow mode weak signal operators use near by frequencies, WSPR
users are asked to make sure that you do not set you transmit settings
BELOW 10.140100 since the "visual" MEPT folks use that area.
.
9. In the main WSPR window you can choose to select the "RX" button
just to receive ONLY or you can select 20%, 25%, or 33% T/R option.
The T/R option is a ratio of how often your rig transmits in one hour.
I usually choose 20%/. The software then listens most of the time but
20% of the hour it will transmit. Sometimes , when you boot up the
software it may wait several minutes before it decides to transmit. If
you want to force a transmit, choose the TX option. Make sure you turn
this option OFF eventually because your rig will transmit for 2
minutes , pause for a few seconds , and then transmit for another two
minutes and repeat UNTIL you turn it off! Please note that the
software , when it is going to transmit, waits until the top of an
even minute. It does NOT transmit every even minute if you choose T/R
20, 25, 33%, but when it does transmit it will wait until the top of
an even minute.
.
10. When you run the software at first, you will see an inactive
waterfall. The current version has a waterfall that updates ONLY every
two minutes. After your software have run for a few minutes you may
see WSPR signals as narrow lines that are HORIZONTAL on the waterfall.
The waterfall does not display the signal "live" but a later version
may add this capability
.
11. After the software has run for a few minutes you may receive
decoded signals, they will be displayed as... 080409 0200 2 -14 1.0
10.140128 KL7QR DM33 30 The above line is date, time decoded, sync,
signal to noise ration in dB, DT, received frequency, callsign, grid
square, noise level. Please note that this mode is very NARROW, about
6HZ wide. You can find a lot of signals in the 200 Hz wide frequency
range that the software monitors. The software has the ability to
decode several signals at the same time, I recently decoded FIVE
signals at the signals. I started a web page for people to spot each
other, it became so popular that is was moved to another server and a
database of all spots was created. You can use a small batch file to
automate the uploading of your received signals to the database. Go to
http://www.wsprnet.org/ for details. This site is now run by W1BW and
the coding was done by N8FQ. As I type this, the following stations
are connected
Users seen in last 10 minutes:
9V1LF - F4FQM - F6IRF - K1JT - K3UK - K7ZTM - KS7S - LA6TPA - N3TFM -
TF3HZ - VE7KPB - W1BW - WD4KPD
.
and some of the latest in the database are...
Power Reported Distance Date Call Freq SNR Grid dBm Rec by loc km mi
.
2008-07-26 16:36 PA2ABC 10.140208 -10 0 JO22jq
+37
5.012 F1TE IN94qs 966 600
2008-07-26 16:34 N2BJW 10.140159 -17 0 FN34im
+27 0.501
W2RXG EL88xu 1907 1185
2008-07-26 16:34 NN6RF 10.140113 -25 0 CM87uw
+23 0.200
W2RXG EL88xu 3840 2386
2008-07-26 16:34 PA2ABC 10.140204 -23 0 JO22jq
+37
5.012 9V1LF OJ11vh 10499 6524
2008-07-26 16:34 PA2ABC 10.140206 -19 0 JO22jq
+37
5.012 TF3TTY HP94ac 1984 1233
2008-07-26 16:34 PA2ABC 10.140184 -16 0 JO22jq
+37
5.012 UNLIS JN89bo 861 535
.
There are many DX stations to be spotted via WSPR.
.
-------------------------------------------------------------------------


.
*** TW0-WAY QSO WSPR MODE ****
.
The latest beta version of WSJT , available from
http://physics.princeton.edu/pulsar/K1JT/WSJT700r973.EXE, now allows
for two-way communication. The transmitted messages differ in format
from the one-way WSPR mode , therefore this version of WSPR is NOT the
version you want to use to automatically upload your reception reports
to WSPR.org. Here is how to do two-way WSPR:
.
Download and install WSJT . Boot up the software. Click on SETUP
then OPTIONS. Enter in your grid square, call sign , PTT port , and
setting for your sound card. My system has one sound card and I use 0
for both Audio IN and Audio Out. Exit the options area.
.
Now click on MODE and select WSPR . Next click on the MONITOR button
towards the middle of the screen. It should turn green and you are
now set up to begin receiving audio from your radio (sorry, I forgot
that part...you DO need a radio capable of being in SSB mode , one
that is very stable. The radio needs to have audio in and out cables
connected to the souncard of your computer).)
.

At the bottom of the screen you will now see a little box entitled RX
Noise. It will give you a reading. Vary your audio level IN to the
computer so that it is around 0 dB when there are no signals present.
The Rx noise level will increase when a strong signal is detected.
The main graphic display in WSJT is a powerful tool but a tad
overwhelming for the rookie. Read the K3UK Complete Bozo's Guide to
HF JT65A for more about this display method, or even read the original
WSJT manual! I suggest ignoring the main graphical display and
instead advise that you use a "waterfall" display with WSJT. Click on
VIEW and select SPECT JT.
.
Now you will see a waterfall-type display. You can click on OPTIONS
and set volume levels if you wish. Select FREQUENCY AXIS. Select a
speed suitable for you PC hardware. I simply use 3.
.
Almost there now.
.
Turn your dial to 10139, and wait a while. If WSPR is present you
will see a very narrow vertically displayed signal . If you see one,
you need to tune it in using the software. Take your mouse and click
on the signal that you see in SPECJT. After you have clicked, you
will see a green line at the top of the SPECJT screen. It will shift
to where you clicked. Consider the green line to be a guide as to the
frequency range , or range on the waterfall, that you will be capable
of decoding signals . If you suddenly see a signal outside of the
range of the green line you need to again click your mouse and the
green line will now move to where the new signal was spotted. The
green line range can be changed. Click on the TOL button in WSJT and
the TOLerance number should change from 50 to 100 Hz. I say "should"
because I swear I could do this a few days ago, today when I tried
nothing changed.
.
If you came across a signal that is in the middle of a transmission,
you may not be able to decode it. The signals start at the TOP of the
minute and last TWO MINUTES. You will need to wait for the station to
start again if you missed it initially. Again, make sure your
computers clock is very accurate.
.
That's pretty much it for receiving Two-way WSPR. With your dial set
at 10.139 USB you may see some signals at around 1500 Hz in the
SPECJT waterfall. These are usually the ones calling CQ in WSPR mode.
You will probably see several manned beaconing stations at around the
1200 Hz level. They will not be interested in a two-way QSO. Do not
call CQ or initiate a two-way WSPR QSO in the beacon section of the
band. You can , since there are no rules, but it is requested that
you voluntarily conduct WSPR QSO a little further up (in the same area
JT65A HF signals can be found).
.

TRANSMITTING.

To answer a CQ you should first double click on a received callsign.
That callsign will then appear in the message area in the lower
right-hand section of WSJT. You need to select a message that you
intend to reply with. They are self-explanatory, but just in case,
check http://physics.princeton.edu/pulsar/K1JT/WSPR_mode.pdf for
details from the author Joe K1JT. Click the AUTO IS OFF and it will
change to AUTO IS ON, if you so desire. The will allow your
transmissions to start automatically at the top of the minute. It
usually requires you to have the message you intend to send already
selected (click on the little circle to the right of the message0 .
You can also send 'on the fly” by clicking on TX1 through 6. If you
start transmitting late, too far past the top of the minute, the other
station may not get enough data to decode you. You can do lots of
other fancy stuff to the message format but I suggest you read the
WSJT Guide or K3UK Bozo Guide for more on that. You can change the
number in the S report to indicate the level of reception.
.
Enjoy the latest from K1JT
.
73 de Andy K3UK
..

Blog Post 7 – EasyPAL—Digital Slow Scan Television software using DRM – Digital Radio Mondiale

Blog Post 7 – EasyPAL—Digital Slow Scan Television software using DRM – Digital Radio Mondiale **

** MAHN-DEE-AL (like DEE rhymes with BEE and AL as in Senator Al Franken)

Wednesday July 22, 2009

This is Curt Black, WR5J with the Educational Radio Net Please standby for a Digital SSTV QST

Before the net. Please download tonight’s software – EASYPAL and set it up according to the blog and WA-DIGITAL instructions.

As we discussed last week with MMSSTV, Slow Scan TV (SSTV) is a way to get the high-information density of visual images shoe-horned into the audio-bandwidth typically used by the human voice of about 2.5kHz. For full motion NTSC video requires a bandwidth of about 4.5MHz or 9.0MHz if you use both sidebands as you transmit– so minimally our shoehorn is working with a ratio of about 2000/1. We accomplish this by using more than 1/30th of a second to transmit our images. In this case 30 to 300 seconds is more typical.

Digital Radio Mondiale or DRM means "Digital World Radio." DRM is a fairly new digital radio standard for use by HF broadcasters. The DRM standard uses a bandwidth from 4.5 KHz and up to and beyond 20 KHz using OFDM modulation. It provides FM quality stereo audio over HF as well as the ability to send data. Dream is a software implementation of a DRM receiver.

The HamDream software is a modified form of Dream by Cesco HB9TLK. HamDream uses only 2.5 KHz bandwidth and is the basis for several 2.5 khz DRM programs. WinDRM replaced HamDream software by Cesco and it uses either 2.3 KHz or 2.5 KHz bandwidth. It also has a digital voice mode. HamDRM is a Windows DLL program by Cesco based on his WinDRM program.

EasyPAL is a piece of software by Erik, VK4AES, that uses DRM encoding and allows us to send images. PAL is just the European standard that corresponds to our NTSC video standard. EasyPAL, just like MMSSTV allows us to fit our image into a standard sideband voice bandwidth. However, instead of audio frequencies corresponding to brightness and colors, images are disassembled pixel by pixel and sent as a digital stream that is decoded and reassembled into our image.

This digital format allows us to send considerable additional information including data on our station, the filename, image data, and ultimately, much more than just images. If you look at the menu bar you will see “LOAD ANY” which is the clue that we can send any type of file on your computer. Just like with FLDIGI, with great power comes great responsibility. We have that same training issue we did with FLDIGI of not sending a megabyte when 2 kilobytes will do.

Another amazing thing about this software is that we can request “fills” or retransmission of any blocks not received perfectly (hit that BSR button – BAD SEGMENT REQUEST and a message is formatted that reports to the sending station the specific segments numbers that require retransmission. The sending station can collect these reports (Visible with the FIX button) and then select them one at a time to retransmit the needed data for a perfectly received transmission at each station.

OR, if you prefer, you may avoid all this re-transmission stuff. You can Reed-Solomon encode everything you send to increase the probability your information will make it through the first time. Just select the encoding appropriate to the band and mode – for tonight’s demo through an FM repeater, I’ve got “very light” encode on. RS Encoding is controlled by the check box next to the RS1 to RS4. Left click to toggle encoding. Right click to change from light encoding to heavy.

If you choose RS4 – your recipient can loose up to 50% of the data and still put your file back together perfectly without any resending.. That is why you may see the picture (or file) appear before the transmission is done. You may have enough data to assemble a perfect picture even though the progress bar is only ½ of the way through the transmission. Tonight I’m using between 16 and 64 simultaneous data streams. Click the QAM button to set the number of data carriers used 4-16-or-64.

Getting Started

Pull the software off the Yahoo Group Files Section at http://groups.yahoo.com/group/wa-digital/files/

Or go to the source at http://vk4aes.com/

or

http://kc1cs.com/ or http://www.g4rob.co.uk/easypal.htm

If you need help, someone on the DIGSSTV newsgroup can probably help you. Be sure to include the version number in any report of a problem. Also keep up to date and use the latest version. http://groups.yahoo.com/group/digsstv/

When looking for folks to QSO with, there is our own WA-DIGITAL

http://groups.yahoo.com/group/wa-digital/post

Also try http://www.obriensweb.com/sked/ for skeds a little farther afield.

http://groups.yahoo.com/group/digsstv is a good resource on frequencies and folks to QSO with.


Here are some suggested SSTV frequencies :

-10 Meters : 28.673 28.677 28.680=calling frequency 28.683 28.686 28.690=K3ASI repeater 28.700=ON4VRB repeater
-15 Meters : 21.334 21.337 21.340=calling frequency 21.343 21.346 Avoid SSTV around 21.350 because there is a Phone DX Net running
-20 Meters : 14.230=calling frequency 14.233 14.236 14.239

Avoid SSTV on 14.227 because there is a Phone DX Net running

80% of all SSTV traffic is done on those frequencies so please don't transmit SSTV pictures BETWEEN those suggested frequencies, it will QRM a lot .

3Kc spacing is really a must for not interfering on nearby stations

The Ten Commandments
of Slowscan

by Dave Jones - KB4YZ

1. Use voice before sending SSTV.

2. Wait for voice and SSTV traffic to finish before sending SSTV.

3. Choose an SSTV mode that is proper for the image to be sent, band conditions, and the receive capability of the receiving stations.

4. Announce the SSTV mode used prior to sending.

5. Transmit on frequency as confirmed by calibration of the VFO with WWV.

6. Send straight pictures as confirmed by calibration of the clock timing with WWV.

7. Send quality pictures with call sign on image.

8. Send full frame.

9. Avoid sending a CW ID unless required by regulations.

10. Describe the picture only after it is confirmed that it was properly received.


=============================== Setup

Setup: Initial Setup - Enter/change Callsign, Soundcard, and CommPort.
Click the "OK" or "Apply" button to update/exit.

The setup for CommPort includes options for
"FAC Green". Using an external circuit this
option could be used to mute the speaker while
receiving the digital signal.

It is recommended that you close the program
and re-start after setting up or making changes
to the soundcard settings.

RX Input -
Allows quick access to Recording Control.
Select the input where the received audio comes into the soundcard.

TX Volume -
Allows quick access to Master Volume for soundcard.
All inputs should be Muted except for the Master Volume and Wave.

RX Input and TX Volume are not shown when running under Vista.

BSR Mode Automatic (recommended) -
Will automatically switch the DRM transmit mode required by the Bad Segment Request.

BSR Mode User Select (advanced)
Allows the user to force the DRM transmit mode to a mode other than
that requested by the Bad Segment Request.

Expanded GUI -
EasyPal Full runs in a window that is 925x570.
The user may optionally select the "Expanded GUI" which will
resize the window to fill out the available space on screen.
Generally, this will be a 4x3 aspect ratio but the actual
number of pixels would be determined by the screen resolution.

W/Fall Color -
Allows the user to select the color of the waterfall.
Using "negative" will display black on white line drawings more
effectively provided that the image is made negative prior to
being sent. When the received is viewed in the "negative"
waterfall, it will appear with the proper contrast.

Very Light Encode -
Selects the Encode1 level of RS encoding - ".rs1" file extension.
Transmission time will be increased by 13%.
At least 89% of the segments transmitted must be received.

Light Encode -
Selects the Encode2 level of RS encoding - ".rs2" file extension.
Transmission time will be increased by 31%.
At least 76% of the segments transmitted must be received.

Medium Encode -
Selects the Encode3 level of RS encoding - ".rs3" file extension.
Transmission time will be increased by 54%.
At least 64% of the segments transmitted must be received.

Heavy Encode -
Selects the Encode4 level of RS encoding - ".rs4" file extension.
Transmission time will be increased by 86%.
At least 51% of the segments transmitted must be received.

Encoded files will have interleaved redundancy using Reed-Solomon
error correction. The encoded file with redundant data is
sent so that the original file may be created even though not all
segments were received.

There is no progressive receive in this mode. The received file
will decode after a sufficient number of segments are copied and
the image will appear as soon as the decoding is finished. This
may happen even before the transmission is complete. Receiving
RS encoded files is automatic - no need to select for receive.

Use of Encode may reduce the need for BSR's and FIX'es.
Compatable only with others running EasyPal.
A stand alone decode program is available that will allow those
not using EasyPal to decode the .rs files manually. See:
http://www.tima.com/~djones/rsdecoder.htm

Default (resolution) -
Will resize down a large image to fit within 640 x 480 pixels.
Small images that are PASTED will be resized larger to fit
within 640 x 480 pixels.

HiRES (Downsize if > 1280*1024) -
Will resize down a large image to fit within 1280 x 1024 pixels.
Small images that are PASTED will be resized larger to fit
within 1280 x 1024 pixels.

HiRES (Best)
Will not resize image.
This is the best setting to use when an external program such as
"sstvPics" is used to resize the image prior to pasting into
EasyPal.

The HiRES setting will not affect the image when "LoadAny" is used.

Show RX Screen -
Selects the RX tab to view the RX Screen and RX information line.
Gamma -
Allows for the adjustment of the brightness of the received image.
RX Filename -
The filename of the received image. Note this when selecting
the proper file to select for the "Send Selected Request" BSR.

Click on image to display full screen.
Animated GIF files will not be displayed full screen.

Show TX Screen -
Selects the TX tab to view the TX Screen and TX information line.
Secs -
This is the number of seconds it would take to transmit this image.
K -
This is the file size for the image to transmit. Does not apply
to images loaded using LoadAny.
Slider -
The setting of this slider changes the level of the JPEG-2000
compression that determines the file size.
SET -
If the slider has changed position after the image has been loaded,
it will be necessary to click the SET button so that the JPEG-2000
compression can be reapplied.
TX Filename -
The name of the file to be transmitted. The filename will change
when the transmission starts.

TX mode = RX mode -
This provides a way to quickly change the TX mode to match the
DRM mode just received.

Define Additional Save Directory -
Received files will be saved into a user selected folder in addition
to being saved in the "Autosave" folder.
(Not working)

=============================== Action

Action:

Send Text -
Turns the TX window into a text editor for the sending of text.
Text may be typed, pasted, and edited prior to transmitting.

Session Log -
Logs the calls and time for current session only.

QuickTxMode -
This screen serves as a quick reminder for what DRM modes are
suitable for different conditions.

USE Repeater -
Show Repeater Commands - Check to enable "Repeater" menu.



============================== LoadPic

LoadPic:
Brings up Window for selecting an image file to load, resize,
and convert into JPEG-2000 (JP2) format. When transmitted,
the filename is changed to "############-originalfilename.JP2".
The "############" is a 12 digit number representing the
current year/month/day/hour/minute/seconds.

=============================== LoadAny

LoadAny:
Allows for the loading of most any file type to be transmitted.
The file will be renamed but will keep the original extension.
This is useful when sending animated GIF files where
conversion to JPEG-2000 is not desired.

=============================== Copy

Copy:
Standard image copy to clipboard from the RX or TX screen.

=============================== Paste

Paste:
Standard image paste from clipboard to the RX or TX screen.
Paste to the TX screen will compress and convert the image
to JPEG-2000 format. The image may be resized based on the
setting for HiRES.

=============================== FIX

FIX:
Brings up Last FIX Request. Use when the receiving station
requires the same FIX sent again.

After receiving a BSR, you will have a message on the title bar:

it may be a message like this:
"-W9VMT-Requests-63/92 segments for-070707122834-cat1.jp2"
Just click "FIX NOW" to send the FIX.

or it may be a message like this:
"You can repair 63/92 segments in 070707122834-cat1.jp2 for W9VMT"
Just click "FIX NOW" to send the FIX.

or it may be a message like this:
"This BSR is not for you"
You do not have this file.
This BSR request is for another station.
In this case, you will have to get the file repaired
before you will be able to send any fixes.

or it may be a message like this:
'You do not have this file for "K9NP ERIC" to Repair'
You do not have this file.
This BSR request is for another station.
In this case, you will have to get the file repaired
before you will be able to send any fixes.

If there is no message in the title bar and you think that
you received the BSR request OK, and you should be able to
send the FIX, go ahead and send the FIX anyway. It may work!

The FIX sends "FIX" in the waterfall before the data.

Any station running EasyPal that successfully receives an image,
can send the fix for any other station that sends an BSR for that
same image from EasyPal. (The original image may be sent from any
program using a compatable DRM mode.) If the original image was
transmitted with a filename that had spaces within the filename,
it may require the original sender to do the FIX.

There is no need to reload an image to fix someone.

The "Last FIX Request" includes the Call from the station that
sent the BSR, mode, segments, and filename. If the mode details
are in any way different from the mode the file was sent, then the
FIX will not work. Ask for an "Old Type" BSR instead.

=============================== WAV

WAV:
Brings up the "Play Wave File" list. Selecting a WAV file
and clicking "TX Now" will immediatelly play the audio from this
WAV file. Create the wav files using "WFPic",
"WFTxt" or an external program such as HamPAL or DIGTRX.

These special wave files must be in the main EasyPal folder:
BEACON.wav - repeater sends at preset intervals when "Beacon" is checked.
BEGIN.wav - sent before each DRM file transmission.
BSR.wav - sent before each user initiated BSR transmission.
DELETED.wav - repeater sends this when selected file is deleted.
DIR.wav - repeater sends before the RS Encoded file (repeaterdir.txt.rs2)
in responce to "View Repeater Directory".
FILEFAIL.wav - repeater sends before the BSR.
FILEOK.wav - repeater sends when it successfully receives a file.
FIX.wav - sent before each FIX transmission.
NOTDEL.wav - repeater sends this when selected file is not allowed to be deleted.
NOTEXIST.wav - repeater sends this when selected file is not available for deletion.
PAUSED.wav - repeater sends in responce to "Shutdown ALL/Specific Repeater".
REPFIX.wav - repeater sends before the FIX transmission.
REPLAY.wav - sent before repeater sends out the last received file.
REPLY.wav - repeater sends in responce to "Interrogate Repeater".
RESTART.wav - repeater sends in responce to "Restart ALL Repeaters".
SELECTED.wav - repeater sends in responce to "Send Back Specified File".
TIMEOUT.wav - repeater sends this when selected file is too large to send.
TOOBIG.wav - repeater sends this when selected file is too large to send.
WEBCAM.wav - repeater sends before the current webcam picture.

=============================== WFPic

WFPic: (Waterfall picture)
Transforms the image on screen into a grey scale image that
becomes analog encoded audio to produce a like image in the
waterfall.

=============================== WFTxt

WFTxt: (Waterfall text)
Transforms a user provided text message with the choice of
various sizes and fonts into analog encoded audio to produce
the text as an image in the waterfall.

To make a waterfall ID click "WFTxt". Type your call into the
Waterfall Text window using all caps. You may want to add a space
between each character. Select a plain bold font to make it
easier to read when displayed in the waterfall. Click Save and
use your call as the filename. To send this, click WAV and
select the file, then click "Tx Now".

=============================== Clear

Clear:
Clears the selected (RX/TX) screen.


=============================== Info Area

Callsign -
The callsign of a transmitting station will appear in the upper
left corner while receiving an DRM transmission.

RX audio level bar graph.
(Higher incoming audio produces more green.)

% -
RX signal quality level bar graph - equivalent to
SNR (Signal to Noise Ratio). There will be more green area with
increased signal quality.

TX -
DRM submode selected for transmit.
Right click mode parameter to toggle through settings.
Any changes to the mode are saved for future sessions
when the program is closed.

Default DRM Mode should be Mode B, Width
2.5 khz, Error HI, QAM 16, and LeadIn 24.
Interleave is always Long and cannot be changed.
These settings are the best choices for general use
on the HF bands.

LeadIn 12 will reduce transmit time by reducing
the number of redundant segments. For very short
transmissions such as Text messages, LeadIn 12
may be too short a time for all stations to
achieve MSC sync before the transmission is
finished. In this case, select LeadIn 24.

RX -
DRM submode received.
FAC must be green before the mode can be detected and the
callsign decoded. In addition to those mode settings that
may be changed on transmit, the RX will also reveal the
interleave setting. Lng for long and Sht for short. The
receive mode settings are automatic and do not require the
operator to make any changes for receiving any DRM mode.

RX mode display: (red/green bars)
MSC - Green light indicates MSC (Main Service Channel) lock.
FAC - Green light indicates FAC (Fast Access Channel) lock.
Frame - Green light indicates Frame Sync.
Time - Green light indicates Time Sync.
IO - Green light indicates Soundcard is active.

Data cannot be decoded until the MSC has a lock. So the MSC
must be green before you will be able to copy anything.

Total -
"Total" is the number of segments in the image file being received.

OK Segs -
"OK Segs" is the number of segments successfully decoded so far.

Position -
"Position" is the segment number that was last decoded.

not processed -
This is the number of segments that still need to be received.

Transmit progress bar indicates the % transmitted.
(Visable only during transmit).

=============================== Buttons

TRANSMIT -
Starts the transmission of the file.

Tune -
Immediatelly sends the three tone tuning tone for 7 seconds.
Green markers in waterfall should match the "Tune" signal
or pilot carriers.

Replay RX -
Replays last received image file. It may be renamed but it
will have the exact same content.

ABORT -
Allows for stopping a picture, WAV, BSR, or FIX transmission.

Encode -
Left click to toggle RS encoding for transmitted files.
Right click to toggle through the four levels of encoding.

ProgRX - Progressive RX
Check to enable viewing image as it is coming in.
Using Progressive RX may slow other processes
such as the waterfall display.

BSR Requests -
Brings up the BSR Requests window.
For those transmissions that have a partial file stored,
there will be a file name listed. The sender's call is
listed first. Click a file name from the list to display
a partial image (if available) and other information.
There may be more than one file for the same image.
If this is the case, you should select the one that has
the fewest number of missing segments. Selecting the
file first is only necessary when using the
"Send Selected Request".

"Send Selected Request" -
CAUTION - TRANSMITTING THIS TYPE OF BSR MAY LOCKUP
NON-COMPATABLE SYSTEMS. (DIGTRX and HamPal)
This type of BSR is available even after EasyPal has been
closed and restarted. Use when you require a third party
FIX from another station running EasyPal. You must select
the file from the list to be fixed before clicking the
"Send Selected Request" button.

"Old Type BSR (compatable all)"
Use this type BSR when requesting segments from non-compatable
systems. Only the original sender or one that replayed the
file will be able to fix you with this type of BSR.

"Old Type Fast BSR (not compatable)"
This is the normal type of BSR to use when requesting segments
from the original sender that is also running EasyPal.
This type of BSR is a bit different from the BSR format used
in the past. It is called a "Fast BSR" because it uses a form of
shorthand to indicate contiguous segments. This new "Fast BSR"
will send faster because the text file listing the missing
segments is smaller. One drawback to the use of the new
"Fast BSR" is that older programs such as HamPal may not
recognize all the segments required and if there are many,
HamPal may only be able to send fixes for two or three
segments at a time. This is only a problem when pictures are
sent from HamPal to EasyPal.

"Old Type" BSR's do NOT require that you select the filename.

"Old Type" BSR may not be available. You will get the message,
"nothing to fix". This is most likely the case where you have
received a BSR from another station but not copied any of the FIX.
In this case, have the original station send part of the file
and abort. The "Old Type" BSR should now be available.

"Bad Segment Report"
Each type of BSR starts the transmission of a short text file
that lists the segment numbers that were missing on the last
received file (or selected file).
During this transmission, many instances will be sent to ensure
that the information gets through. When many segments are
required, the transmission time for the BSR may be much longer.

You may be better off to simply resend the original file – possibly with

a higher Reed-Solomon error correction setting.

=============================== TABS

RX tab -
Click to view RX screen and RX information line.

TX tab -
Click to view TX screen and TX information line.

View tab -
Click to view thumbnails of Last RX Pictures.
Click to view thumbnails of Last TX Pictures.

Send to TX -
This allows for a replay of the thumbnail image.
Transfers selected thumbnail to TX window. JPEG-2000
compression is applied and filename is changed when
transmitted. Quality is reduced each time this type
of replay is used.

Full Screen -
Displays selected thumbnail to full screen.
Image displayed will always be the original size.
The image is displayed against a black background.
Animated GIF files might not be displayed.
To return to the program screen, click the mouse
or hit the "Esc" key.

Delete to Recycle Bin -
Selected thumbnail will be deleted.
Since not all selected thumbnails will actually be
deleted, it may be necessary to open the "RxFiles"
or the "TxFiles" folder to manually delete these
files. Selecting the very last thumbnail to
delete for the second time may actually delete the
first.

Send to Repeater Directory - ("USE Repeater" checked)
Transfers selected thumbnail to "Repeater" folder.
Additional JPEG-2000 compression is applied based on
setting of slider on the TX window. If "Encode" is
checked, then this level of RS Encoding will be
applied and the filename extension will be changed
to the one for that level of RS Encoding. Quality
may be reduced for this type of conversion.

Thumbnails are standard JPEG (.jpg) images that are
re-compressed copies of the files received and sent.
Image files in the "LastRX" folder may not have the
quality and characteristics as the original files sent.

The original files received are stored in the
"Autosave" folder. These files should be identical to
the quality and characteristics of the original files sent.

=============================== NOTES

The waterfall may be paused - just left click to pause/restart.

When EasyPal is closed, the contents of the "Transient" folder is deleted.
Temp files in other folders are also deleted.

Partially received files (ones with missing segments) will be saved
into the "Corrupt" folder. They are stored here until they are repaired
or manually deleted by the user.

Hints will appear after mouse pointer is placed over selected buttons
and areas.