Possible new radar, 7735, 8050, and 16110 kHz

[Token]

A signal was reported in the #wunclub IRC channel on the 25th of October. It appears to be a new Radar or a previously unreported mode of an existing Radar. My guess at this point is that it is a new system.

I have seen this signal so far on 3 frequencies, 7735, 8050, and 16110 kHz. I assume more frequencies will be found as time goes on. The signal appears to set on one frequency for an extended time, sometimes for hours, before moving on to the next. So far it has only been seen on one frequency at a time.

The signal appears to be a simple pulsed type Radar. It has a PRF (Pulse Repetition Frequency) of about 9.1 Hz (110 millisecond PRI, Pulse Repetition Interval). Although I have not yet gotten a good recording to determine Pulse Width, the PW appears to be about 900 microseconds (based on the spectrum peak to first null measurement of about 1.1 kHz).

So far I have not seen an indication that the system beam steers.

The signal appears to be from Europe or that general area.

This signal is very similar to the old Russian Woodpecker except it has a slightly narrower pulse width and does not appear to use pulse compression.

Spectrogram capture showing PRI of 110 msec here: http://www.pbase.com/token/image/146959548/original.jpg

50 kHz span spectrum capture of the 7735 kHz signal, using Max Hold, shows the general spectrum shape and the peak to null delta: http://www.pbase.com/token/image/146959634/original.jpg

25 kHz span spectrum of the 16110 kHz signal, again using Max Hold, showing the same general peak to null delta, indicating approximately the same pulse width: http://www.pbase.com/token/image/146968669/original.jpg

Video of the signal on 16110 kHz here, with several different demodulation modes used to show what it sounds like in AM, USB, and CW: Radar, Unknown Pulsed Type, 16110 kHZ, October 27, 2012, 1345 UTC - YouTube

T!
Mojave Desert, California, USA

 

[zl2taw]

hearing it now @ 1145utc on 16110khz here in Australia

 

[jim202]

One possibility is that your hearing a system used to find the max usable frequency. These are used by a number of research facilities in plotting the HF propagation around the world.

When I worked at the M.I.T. facility many years back, in Westord MA on Milstone Hill, there was a system there that did this. Normally they use a vertical RHOMBIC antenna so it doesn't bother anyone very far from the transmitter location. The transmitter would hop in frequency those channels that had known communications on it to prevent causing problems.

Not saying this is the type of transmission your hearing, but it does sound like it.

 

[Token]

jim202, I am pretty sure this signal is not a sounder (ionosonde, the signals that you are talking about to find MUF and paths). However, if you look at my Youtube channel you will see that I have a few videos of sounders up there. There are several different ionosonde networks in use, many of them have data accessible online. I assume you are talking about the ionosonde associated with the Haystack Observatory, Millstone Hill? I work with, and have for a couple decades, one of the gentlemen who built the 150 footer there (he worked on it in the early 60’s, before it was moved to Millstone Hill).

What is my rational for saying this is not a sounder?

First is that it sets on frequency for far too long, typically hours on a given freq. Most (all that I am aware of) modern sounders shift frequency fairly regularly and quickly, to sample a range of propagation paths. In fact one way to tell certain kinds of sounders from a radar is the cycle time for a given frequency. A sounder is really just an atmospheric, or ionospheric, radar, but propagation does not change quickly so a revisit time of a few times an hour, say less than 10 times an hour, is really all that is needed. But a radar tracks things that are more dynamic, so requires more rapid revisit times.

Next would be the power levels involved. Sounders do not need huge power and this signal looks like it might be at significant power levels. Radars need high power to increase energy on small, distant, returns. Of course high power also increases the clutter issue, but there are ways to address that.

Last would be the waveform. A simple pulsed waveform like this is not typical of a sounder. The pulses are too fast for global coverage and not encoded or compressed in any way so that nth time data can be derived or discriminated/detected. I know of no modern sounder that uses such a waveform.

T!

 

[Token]

A 4th frequency has been found for this radar, 12105 kHz. So now it has been seen on 7735, 8050, 12105, and 16110 kHz. Always for hours at a time on each freq and so far always only one frequency at a time.

T!

 

[majoco]

Found it - 8050kHz 1930z S0 though.

 

[Token]

Yes, it was on 8050 at 1930z, left that freq at 2000z.

Rough schedule, still being refined:
2230-0200 7735
0230-0600 8050
0630-1100 12105
1130-1400 16110
1430-2000 8050

Note that all the start times appear to be XX30, and there is nothing from the stop at top of hour to the next start at bottom of hour. Actually there is something in that "dead time". Often I have seen the transmitter tunning up during that dead time, and generally the XX30 start time has a few seconds of carrier before the pulses start.

T!

 

[Token]

I realize this is a year old thread, but this signal appears to have returned to the air, after almost a year of not being seen.

Starting October 17, 2013, this signal was seen on the same frequency set as 350 days earlier, using 7735, 8050, and 16110 kHz. It has been seen pretty much daily since then (as of today, October 25, 2013).

The habits are very similar to last year, up at about ~XX29 and off at XX00 most hours (example up at 0028, off at 0100, up at 0128, off at 0200). The start time is a bit variable, but the stop time seems to be within a few seconds each hour. It works for several hours on one frequency and then changes to one of the others, keeping the same schedule. The day to day schedule seems to be repeating, although different from last year.

Rough schedule right now (remember, ~31 minutes on and 29 minutes off each hour):
2200 - 0200 8050 kHz
0200 - 0500 7735 kHz
0500 - ???? 8050 kHz
1300 - 2000 16110 kHz

Yeah, there are big holes in that sched, but it has been busy at work and I have not had a chance to really keep track of this thing. This weekend I will try to flesh in the sched.

Last year I was not sure what it was, but felt a good possibility of it being a radar of some kind. This year I am pretty sure it is not, so it probably was not last year as well

The signal has shown several different transmission formats. There has been a CW format, where it is CW the entire ~31 minutes of air time. There is a mixed format, some CW and some pulsed for the air time. And there is pulsed for the entire air time.

When pulsed the pulse repetition rate has been pretty solid on 10 Hz every time I have seen it, 100 msec pulse repetition interval. There appears to have been several pulse lengths used, with 10, 20, 30 and maybe 32 milliseconds per pulse seen, but I have not seen it change during a transmission window, it stays with a given pulse width for any window. Sometimes the pulses have modulation on them, other times not.

So, why do I say no radar now? I cannot say it is absolutely not a radar, but I can say it does not seem like one to me, I am pretty sure in my own mind it is not one. The pulse width and bandwidth are inconsistent with radar like operations. The long pulse width with no modulation to provide pulse compression results in a near useless radar waveform with regards to range resolution. The CW is also not radar like (yes there are CW radars, but taken as a whole this just does not fit well here).

When this signal is CW sometimes several weak modulation lines can be seen. These lines fall on 50, 100, 150, 200, 250, and 300 Hz. This is probably 50 Hz power mains hum. That could indicate the transmitter is located someplace with 50 Hz power. Also, it appears the line frequency is not exactly 50 Hz.

T!