Re: [wsjtgroup] JT65/JT4 - Max sync values + frequencies
- Hi Bo,
> I wonder what is the highest possible "sync value" in JT65 and JT4 respectively? Even if I connect two PCs directly to each other I have not seen values above say 28. E.g. shouldn't it be 63 for JT65?The sync value is a relative indicator only. It's not calibrated in any
particular units, and is necessarily linear. Its sole purpose is to
help establish whether a signal of the relevant mode is present at
particular time and frequency offsets DT and DF.
> FrequenciesYour numbers are correct. The figure 711 Hz is a good approximation for
> I have seen several values for the highest tone in JT65. For JT65C I have seen values from 685 Hz to 711 Hz. However, using the math from "The JT65 Communications Protocol, March 9, 2005" I cannot get anything else than ~700 Hz (JT65C) above the sync tone.
> Tone_N = Multiplier * (N+2) * 11025/4096, were
> Multiplier: ~A = 1, ~B = 2 and ~C = 4
> N: 0 to 63
> Thus for JT65C:
> Tone_0 = 4 * 2 * 11025 / 4096 = 21,533203125 Hz above sync
> Tone_63 = 4 * 65 * 11025 / 4096 = 699,8291015625 Hz above sync
the total occupied bandwidth of a JT65C signal, which is 66*4*11025/4096
= 710.5957... Hz (one tone spacing more than the figure 699.829... that
-- 73, Joe, K1JT
- Hi Joe et al,
Thanks for the sync clarification and confirmation. Also OK about the bandwidth.
The reason for asking was an abnormal behavior I experienced that caused some head scratching. I am on the process of finalizing the S/W for the Next Generation Beacon http://rudius.net/oz2m/ngnb
When I made the JT65_2 modes I, for some reason, tested with only a call (OZ7IGY/A) and I could not get any decode despite perfect sync. When I added the square (JO55) or used the raw call (OZ7IGY) and locator (JO55WM) it all went well.
For beacon purposes the message is sent in freetext format. But, it seems as if a to short a message, that is padded with spaces, is interpreted as formatted text and thus is not decoded in WSJT. Not a big thing though as this is probably only seen when testing with only a callsign.
However, I also saw a somewhat similar behavior in WSPR if power is not ending in 0, 3 or 7. I sent "OZ7IGY JO55 44" and it comes out as B5L7/OZ7IGY 43 in the WSPR program. As far as I can see the WSPR protocol is "power transparent" with 6 bits? Or am I wrong?
Finally, one question: Do you know how robust JT65C's are to distortion like aurora and rain scatter? Empiric data show that JT4G works with rain scatter. I guess the 65 tones will be blurred and overlapping. Four tones with adequate spacing are probably less prone to this.