HF Digital Error correcting? Also, what's up with PSK31?

Discussion in 'General Technical Questions and Answers' started by N0NS, Oct 9, 2008.

Thread Status:
Not open for further replies.
ad: L-HROutlet
ad: l-rl
ad: Subscribe
ad: Left-2
ad: Left-3
ad: L-MFJ
ad: abrind-2
  1. AG3Y

    AG3Y Guest

    "n my opinion RTTY has more general adoption and can support basically the same types of QSOs that PSK can support with similar equipment. Some radios even support direct FSK modulation for RTTY."

    It would be interesting, would it not, to have a radio that can generate a carrier that could be directly frequency modulated by a varying DC voltage representing the analog equivilent of the digital protocol, without going through audio stages, sideband filters, etc. etc. It would seem to me that all that would be required would be replacing the simple "on-off" frequency shifter that is used in direct FSK of these radios, with one that was capable of true FM. A small shift voltage controlled oscillator that could be enabled for the digital modes would do the job, I would think, and there would be your modulation chain eliminator !

    Years ago, I had a Collins KWM-2 and was heavy into the RTTY mode. I sold the rig and bought an Icom 720a. ( which I still use almost daily ! ) One of the main reasons I did this was because the 720 DID HAVE direct FSK.

    Then, years later, the soundcard modes came along, and the external boxes for RTTY faded away. I still have a homebrew TU or two up in the attic, but I have not directly FSKed this rig for YEARS.

    I'm not so sure, Harry that your statement about RTTY is entirely correct. Certainly RTTY is very popular during contests, but in day to day, keyboard to keyboard contacts, PSK-31 seems to be the "top dog". I highly suspect the reason for this is its ability to be decoded correctly when the signal producing the data can hardly be heard in the speaker! Due to the close spacing of the tone stream, there is hardly any "selective fading" and errors due to that phenomena. Of course, phase shifting due to propagation can and does cause errors to creep in to DX signals, but they are usually no more of a problem than what would be expected on any signal coming from many thousands of miles away over a poor propagation path.

    No machine readable code is going to be decoded as well as a CW signal being "decoded" by a highly skilled operator with years of experience. But PSK-31, and certainly some othe more aggressive digital modes such as "Olivia" and other MFSK modes, come very close!

    Edit: Rob you are very welcome ! Yes, it is fun to get out the old boxes and see how well they can do, sometimes. No one says you HAVE to use only the LATEST and GREATEST! Have FUN !

    73, Jim
  2. KR2D

    KR2D Ham Member QRZ Page

  3. WB3BEL

    WB3BEL Ham Member QRZ Page

    Regarding your comment about being able to "print" a PSK31 signal that can't be heard in the speaker...

    I have a slightly different perspective on this one. PSK31 is not some kind of super mode that works with 0 dB SNR in its symbol bandwidth. It has no error correction and that gets back to one of the original questions of the thread.

    Yes, if you listen via the speaker to a receiver with a wide bandwidth that has lots of noise and or interferers, you may find that you see spectrum traces that you can not hear by ear. But if you reduce the bandwidth of the receiver to match the target waveform, you will find that you can always hear the PSK31 signal. Notwithstanding a disclaimer about hearing impairments...

    As far as which mode is more popular PSK31 or RTTY; Maybe I am wrong here, but I find, I can almost always find a RTTY signal if I want to operate that mode or if I call CQ I will get a reply. DXpeditions frequently have RTTY but rarely use PSK31. I think that by number of callsigns, there are more stations that use RTTY than PSK31. Buts thats really just a guess based on casual digital mode operating. I am not really a digimode fanatic so this could just be my perspective.

    Harry WB3BEL
  4. AG3Y

    AG3Y Guest

    Certainly no argument from me on your last post, Harry. It would be interesting to see how many prefer Baudot ( rtty ) to PSK-31, and why ?

    I may have to post a thread about that one! I do know that when there is a RTTY contest, almost the ONLY type of signals you hear are RTTY ones!

    But there are times during the week, when if there are any digital signals to be heard, they are almost all PSK-31.

    I don't really know if it is all that important an issue, though. I will generally use any mode that is necessary to carry on a qso or make a contact. Certainly if I hear a rare one working PSK-31, I am NOT going to go back to him in CW or Baudot! I'm going to match his mode!

    I have my favorites for my own reasons. For instance, I am a good typist, and find PSK-31 to be VERY S L O W W W ! So I prefer a faster mode such as MT-63. But it is very difficult to find someone else using that mode, so I make very few contacts with it.

    Of course, there are purists that insist that CW is the ONLY mode to use. That's OK for them, and more power to them, but there is an expression that comes to mind, "That's why there are horse races! "

    Thanks for your thoughts on this , Harry. This has been an interesting thread!

    73, Jim
  5. AB0WR

    AB0WR Ham Member QRZ Page

    I don't know what you think I am missing. The bandwidth of a CW signal is usually taken to be wpm X 4. This gives a bandwidth of about 200hz for a 50wpm CW signal -- far wider than the 31hz of a psk31 signal. That CW bandwidth *is* made up of odd harmonics of the basic keying speed *and* those odd harmonics can cause all kinds of IMD products in a non-linear amplifier chain.

    "Noise" and harmonic distortion, as you pointed out yourself, are typically far, far below audibility on a psk31 signal. Driver/final amplifier IMD causes significantly more impairment on most psk31 signals than any kind of audio chain noise or harmonic distortion.

    Those "dozens" of stages you mention are *COMMON* to both the CW and the psk31 signal. The very same IMD will be produced in both cases. The IMD from a CW speed equivalent to psk31 (50wpm) will actually be *farther* from the carrier frequency than with psk31!

    Now you are back to trying to confusing the issue. The transmitted dynamic range on psk31 and CW, or even on your DSP generated psk31, WILL BE EXACTLY THE SAME since it is dominated by the capabilities of the driver/final amplifier chain.

    The problem with most high-power psk31 signals is NOT because of audio chain distortion -- it is because of driver/final amplifier chains being overdriven and causing all kinds of IMD.

    A high speed CW signal *can* cause exactly the same results, as you know darn well!

    A psk31 signal doesn't *NEED* to be as powerful as a CW signal for good communications over the same distance. That *is* one of its pro's. No one that knows to turn the AGC off on their receiver is going to holler about a strong signal on psk31 on an adjacent frequency, at least they won't holler any louder than someone trying to work CW in the same situation.

    Turning off the AGC is good practice for *any* digital signal. I have always done the same thing for RTTY. I always ran my Digital-NTS station with the AGC off when running Pactor for the exact same reason. If the AGC is off then the DSP in the psk31 program (or whatever other mode is being used) can do its best at filtering out the separate signals on the baseband signal it gets from the receiver.

    psk31 through the audio chain of SSB transceiver *is* one of the most functional digital modes we have available today. It has a very high spectrum efficiency and operational utility. In the typical amateur transmitter, its power output capabilities is determined almost exclusively by the IMD performance of the driver/final amplifier chain -- just as it would be by any other mode or any other method of generating the psk31 signal.

    tim ab0wr
  6. KI4NGN

    KI4NGN Premium Subscriber QRZ Page

    Exactly! I don't know how many times I've asked an op to back down his audio, that he was driving it way too hard.

    The rule of thumb that all should use is to ensure that there is NO ALC being triggered by their PSK signal. If there is, then that's the kind of signal you see being very wide on that spectrum or waterfall display.

    In addition to those who just don't know how to use their receivers for PSK, there are those who don't know how to properly set their levels for transmit, just cranking up that audio gain to get max signal.

    I often operate PSK at 70 watts, I know that I have a clean signal and have been told that many times, and using that power has enabled a VK contact once in a while that would not have occurred at lower power. I know...we tried....that 3db difference when I cut the power in half also lost the connection.

  7. KR2D

    KR2D Ham Member QRZ Page

    Same here. I've seen guys claiming to be running 400 watts with perfectly clean signals. They were chatting (Midwest US to Europe), and very few others were on because the band was "closed".
  8. AG3Y

    AG3Y Guest

    One of the biggest "urban legends" about PSK-31 is that it HAS to be a "low power" mode. NOT TRUE ! ! ! But the transmitter HAS to be driven in a linear fashion. I regularly operate at around 60 watts or so. Going any higher causes the pilot lights in the rig to flicker, and I KNOW that it is a result of the power supply maxing out.

    A few "flat tops" on SSB are hardly consequential, but they will cause serious problems with PSK and other digital modes such as DRM picture transmissions!

    You shouldn't run flat out in the digital modes, but it isn't necessary to run 10 watts, as some would tell you, either !

    73, Jim
  9. G0GQK

    G0GQK Ham Member QRZ Page

    PSK 31 is very popular in Europe. Why is it popular ? Because Europe has many countries and only four of them speak English. PSK allows amateurs who who not speak the language to make contacts all around the world and be active when the band conditions are poor. The higher speeds are too fast for a relaxing afternoon using PSK, and probably far too fast for the average European who does not read English.

    Its like driving a car. If one wishes to enjoy the drive and the scenery, its better to drive a little slower, and not at 80 mph. I find that it is only on rare occasions, and under extremly poor propagation conditions that I do not receive 100% copy. Before PSK was invented, with those conditions, people used to listen, switch the rig off and go and do something else because it was absolutely impossible to have a conversation with anyone.

  10. W8JI

    W8JI Ham Member QRZ Page

    That's just wrong Tim.

    The bandwidth is set by the rise and fall times, not the speed.

    It's easy to prove that.

    73 Tom
Thread Status:
Not open for further replies.

Share This Page

ad: AlphaRF-1