Ssb-fdm

[zguy1243]

Not sure what forum to drop this in but here we go.

I have played around listening to the wide band signals from the E-4's and E-6's for a while. There is a post currently in the Military monitoring forum called Orderwire, this is what I am referring to. I have in the past just used my SSB capable receivers to tune the voice channels. I understand that this system is a FDM signal with the SSB voice channels on the approx. 8Khz spaced channels. I am still doing alot of reading about FDM and the inner workings of it. I recently stumbled across a Watkins Johnson 9518 FDM demodulator on ebay. Here is link to a description of the unit. http://watkins-johnson.terryo.org/Documents/Manufacturers/WJ/Data sheets/WJ-9518BE-data-sheet.pdf. I am curious if this could be used to monitor the wideband system we are talking about. I would assume this demodulator would need the IF of another receiver feeding it. Anyone on the forum that can clear the fog for me about this device it would be appreciated.

 

[mancow]

Have you ever had any success with the SSB mode? I tried but never could get anything. It's like the filtering just wasn't right.

 

[zguy1243]

Yes. You can dependably copy the voice channels tuning in the USB mode. Its not super easy. Takes some practice to get a ear for it.

 

[mancow]

What kind of traffic do they carry?

I hear those carriers all the time. They are strong enough to trip the sweeper on the PSR in my work car using an 800 mag mount. With the mil discone at home they are heard for hours on end until they eventually fade. I have noticed that they seem to carry that high pitched whine and I hear what sounds like a telephone line tone and muffled voices at times.

 

[kb2vxa]

Frequency division multiplexed RTTY is NOT SSB, it's FSK, but requires a product detector to receive it. Eh, maybe I should explain that's an integral part of an SSB receiver. (;->) This from another source and I think it states my case extremely well.

"Please note that this device demands a very stable, high quality receiver and considerable teletype knowledge by the operator."

Methinks you're in just a bit over your head.

 

[zguy1243]

Methinks you're in just a bit over your head.

Maybe so. I plan not to be. Break down some details for me. How should I take your comment there? Give me something I can work with not those kind of comments. I do think the subject is not beyond me, obviously you think you know it all already, enlighten me.

 

[kb2vxa]

If I knew it already I would have told you, I'm a technician, not a magician. I refer you back to my quote, this particular brand of RTTY is for commercial, government and military applications and is well beyond the ken of the average ham or SWL. I'm not being a smart alec, only informing you that this is a rather advanced issue not easily understood and difficult to master.

About now you should be mulling over whether or not it's worth all the trouble just to satisfy your curiosity and wind up reading a lot of boring routine traffic, much of it encrypted. If you have a serious application for this advanced equipment I say go for it, if not you have a decision to make.

 

[benbenrf]

Vast majority of it is FSK modulated - (with practise you can learn to hear the diff between the two).

So long as its not encrypted, a WJ FSK (or if its FDM) demodulator will deal with it.

There are a bunch of HF freqs in used by the US defense sector daily to the hour to transmitt FSK data.

Most of it is pretty narrowband, so long as the reciever has suitable filters and bandwidth settings, yes - you can tap off the IF and then demod through a standalone unit such as one of the many WJ boxes - and just hope its not encrypted. Alot of it is, some is not.

But forget consitant good quality results with your typical consumer type scanner - most havent got the "guts" to deal with the parameters/variables (bandwidth settings and selectivity) required to deal with and demod these signals.

 

[fineshot1]

Maybe so. I plan not to be. Break down some details for me. How should I take your comment there? Give me something I can work with not those kind of comments. I do think the subject is not beyond me, obviously you think you know it all already, enlighten me.

Just do as I did - put him on your ignore list - problem solved.

 

[benbenrf]

Maybe so. I plan not to be. Break down some details for me. How should I take your comment there? Give me something I can work with not those kind of comments. I do think the subject is not beyond me, obviously you think you know it all already, enlighten me.

Zguy

Your 8617 is fine for the job - in fact its a very good/excellent choice for the job - and add to it a good quality antenna (very important because noise only makes demod'ing FDM and FSK more difficult - so the better the antenna the better the raw signal you'll have to work with).

Use the 8617 as the front end i.e. tune it in to the desired freq, and extract what you want to listen to via the IF's at the rear - inputting the IF into a FDM or FSK demod unit. So long as the signal is analogue and not encrypted, the audio will be produced, if its audio.

That is for lack of a better description, a "block diagram" of how to go about it, an over simplification: some detail is important to understand:

First and foremost your FDM and/or FSK demod unit will have to be able to tune to/across 21.4Mhz - as you will be taking one or other of the 8617 IF outputs (via coax) and inputting it into the the antenna input on the rear of the demod unit - that means you will be putting a 21,4Mhz rf signal into the demod box via its antenna input - or, depending on what demod unit you use, it may well have a 21,4Mhz IF input - you could use that as well (IF / RF = same thing - both are radio frequencies at 21,4Mhz)

If your 8617 has WBO take note that it is 4Mhz wide on the 21,4Mhz IF output, whereas the other standard 21.4Mhz IF bandwidth output (the one on the TNC plug) changes with the selected bandwidth setting on the front panel.

Which of these IF's you settle on using will come down ultimately to experimentation, but I'd start off with the standard TNC output (so you can play with bandwidth settings easily).

Thats the basics of it - and so long as noise doesn't overwhlem the signal (so use the narrowest bandwidth setting on the front panel you can get away with using), and the signal is not encrypted - you are going to get your audio (if audio is been carried).

But do yourself a favour: go to Wiki ( Radioteletype - Wikipedia, the free encyclopedia ) and read up on FSK and FDM - which comes in quite a few different forms, and find some sites on the net with examples of recorded FSK and FDM un-demodulated - listen to them and get a feel for how they sound, so you become familiar with them and do not purchase a demod unit that is not compatible with the modulation form you wish to demodulate. You'll be suprized how easy it is (with practise) to id one modualtion type from another - I know guys I have worked with over the years who can recognise over a hundred different types of rf modulation by listening to a 3 - 10 second playback of an un-demodulated recorded sample at audio frequency.

Good luck

 

[zguy1243]

Here is a little more information on the FDM system that I am talking about.

This system is also used to provide full duplex telephone grade voice circuits
for the VIP aircraft such as Air Force One. The airborne radio equipment
( made by ECI), transmits a 1kw (or two selectable lower powers) signal to
blade antennas on the aircraft which are more or less omnidirectional. The
ground sites use a phased array of four UHF broadband traveling wave antennas
arranged in a square to give some gain and jamming rejection. The signals are
vertically polarized.

The modulation is FM-FDM-SSB with up to 14 voice channels and a baseband order
wire. Unlike most FDM systems, the channels are paired with a lower sideband
4 khz voice channel sharing a carrier frequency with an upper sideband voice
channel immediately above it. The carrier frequencies used are 8, 16, 24, 32,
40, 48 and 56 khz. A 0 TLP tone on a channel is supposed to deviate the
transmitter about 30 some odd khz (although the actual level on real signals
appears to be less than this).

The aircraft and ground multiplex equipment use the old fashioned 2600 hz sf
signalling for supervision, this means that idle channels carry a -17 db TLP
2600 hz tone which drops when the channel is seized. Dialing is DTMF using
the Autovon standard phone tones. Since the circuits terminate in Autovon
switches, they are fully four wire.

The aircraft transmits a 120 khz pilot tone, ground sites use a lower frequency
pilot. There is considerable provision for air to air relay of communications
and use of relay aircraft is an organic part of the system design. Many of the
command post aircraft are equiped to relay several links at once and relay
operation is quite often tested both to communicate with other command post
aircraft and with VIP aircraft such as Air Force One. The relay aircraft
(particularly the command post aircraft) have manual switchboards that allow
selected channels from an incoming signal to be cross connected to different
channels on one or more outgoing signals. VIP communications circuits are
thus often routed on certain channels of command post links that carry military
traffic on other channels.

The frequency from 0-4 khz on the signals is used for a orderwire. The radio
equipment has appropriate provisions for conference bridging on this circuit
so the order wire at any point has most all of the stations on it. [The order
wire on this system has been code named ADVENT for at least twenty years.] The
order wire is used to coordinate circuit switchover between ground stations
and/or relay aircraft and to coordinate circuit test and maintainence.

There are a number of Ground Entry Point sites scattered throughout the US
used with this system, most seem to be at hardened blast resistant AT&T
microwave sites with deep underground bunkers on springs. These sites are
probably used because they represent points where hardened underground cables
connect to each other and radio systems. The ground entry sites can be
recognized by the distinctive square pattern formed by 4 vertical pole antennas
(quite long and thick, unlike most UHF antennas) on top of an AT&T microwave
tower (above the microwave horns) spaced about 8-12 feet apart. A random sample
of sites I am aware of includes Green Hill Rhode Island, Waldorf Md, Hillsboro
Mo, and Pensuco Florida. There are several more.Maximum range from an aircraft
to a ground site is typically 210-230 miles depending on altitude, air to air
range is closer to 400 miles. With powers of 1 kw or more EIRP, the signals are
very strong on the ground when the aircraft is closer. The high power is
supposed to be intended to ensure that communications can penatrate nuclear
fireballs and other propagation disruptions during a nuclear attack.

The system is used to carry clear voice traffic (including traffic from Air
Force One), and also various forms of digital transmission of a sort that fits
in a 4 khz voice channel. This includes slow speed 75 baud clear and encrypted
tty, 1200/2400 baud data, and full duplex 9600 baud data, fax and secure voice.
The VIP aircraft using the system use only 4 Autovon circuits per signal and
only use 16 and 24 khz channels from the aircraft (but the ground sites
transmit a signal with sf tones or traffic on all the upper 12 channels).
The command post aircraft usually use all 14 channels.

 

[N_Jay]

Just like analog microwave.

 

[mancow]

This?

AT&T Long Lines: Hillsboro, MO

 

[zguy1243]

Thats exactly what I was thinking.

Just like analog microwave.

 

[benbenrf]

Hell ... now why didn't you just say so right at the start - with the degree of insight into the tech spec of the link, you've either copied it verbatim from some reference, or you have a damn detailed understanding of the subject. Must be the former - else you'd sure not be asking for help (tongue in cheek comment). It's pretty old tech that - is it still in use today?

You dont say what the exactr freq's are (?) - as you'll know max input for the 8617 is 1200Mhz (with FE3). But its not an obstacle because you can always down convert.

Anyhow, yes that is indeed FDM.
I don't see any obstacles (assuming encryption is not been used - if it is forget it without the key, or digital recording and a lot of post processing in your spare time!).

Youre going to need a fairly good selection of bandwidths in the 8617, ideally dual freq boards - the 3.2/6.4Khz FM demod and IF Amp boards in slot 1 (because when you select SSB/Lower or Upper the receiver automaticaly defaults to the card in slot 1 - which should be your lowest bandwidth card). This of course is using the standard IF output. If you have WBO you're going to get a full 4Mhz out through the WBO BNC plug, and will then have need to cut that back in the FDM box down to as close to 4Khz as you can.

Which WJ FDM to choose?
Get yourself a 9548 - it has DSP facilities built in (nice)

Just one issue to deal with - youre going to have to halve the 8617's, 21,4Mhz IF output down to 10,7Mhz - because the 9548 only tunes up to 20Megs. But thats straightforward enough.

The rest of the job the 9548 will whistle through.

You can check the full spec for the 9548 on:

http://watkins-johnson.terryo.org/Documents/Manufacturers/WJ/Data sheets/WJ-9548-data-sheet.pdf

Hope this helps.

 

[zguy1243]

You dont say what the exactr freq's are (?) - as you'll know max input for the 8617 is 1200Mhz (with FE3). But its not an obstacle because you can always down convert.

The freqs are scattered in the 225-400Mhz range so not a problem for the 8617.

Youre going to need a fairly good selection of bandwidths in the 8617, ideally dual freq boards - the 3.2/6.4Khz FM demod and IF Amp boards in slot 1 (because when you select SSB/Lower or Upper the receiver automaticaly defaults to the card in slot 1 - which should be your lowest bandwidth card). This of course is using the standard IF output. If you have WBO you're going to get a full 4Mhz out through the WBO BNC plug, and will then have need to cut that back in the FDM box down to as close to 4Khz as you can.

I have heard about the 8617's defaulting to the IF in slot 1 when SSB is selected. My receiver has a 20Khz filter in slot 1 but when SSB is selected I get a 3.2 Khz bandwidth. I have been told the it has some onboard filters with the SSB option. Does it still use the filter in slot 1 like a roofing filter in this case? So using the standard IF out and not the WBO would be the right thing to do. My receiver does have the WBO option too. With the standard IF out it should pass the 3.2 Khz SSB bandwidth selected while in that mode.

This wideband system system that I am talking about can be tuned with the 8617 alone by finding these MUX channels of the sides of the main carrier every 8Khz or so. Tuning these in very fine steps in USB will yield the voice phone calls taking place but tuning them in isnt so easy. Getting right on freq with 8617 alone is hard, being just a few Hz off and the voice is not intelligible. I hope the FDM box will offer a more precision method of tuning them in.

On the note of the 8617's IF out, It seems the 9518 box that I ahve found on ebay now tunes from 300Khz to 15 Mhz. If I understand this correctly this is the frequency range of the IF input it can accept. I wil have to do something about my 21Mhz if from the 8617. I thought I was reading that the 9518 box required a base band input not IF input? This doesnt make sense to me.

you have been very helpful, thanks.

 

[zz0468]

On the note of the 8617's IF out, It seems the 9518 box that I ahve found on ebay now tunes from 300Khz to 15 Mhz. If I understand this correctly this is the frequency range of the IF input it can accept. I wil have to do something about my 21Mhz if from the 8617. I thought I was reading that the 9518 box required a base band input not IF input? This doesnt make sense to me.

The pdf spec sheet says 300 Hz-15 MHz input, not 300 KHz. If your receiver has a wideband demod or baseband output, that's what you'd connect to.

You can also use a frequency selective level meter, or one of the marine grade commercial receivers that goes down to 15 KHz to demux. I used to demux analog microwave signals with a home build down converter and FM demod to an ITT 3041 SSB receiver.

 

[benbenrf]

The freqs are scattered in the 225-400Mhz range so not a problem for the 8617.




I have heard about the 8617's defaulting to the IF in slot 1 when SSB is selected. My receiver has a 20Khz filter in slot 1 but when SSB is selected I get a 3.2 Khz bandwidth. I have been told the it has some onboard filters with the SSB option. Does it still use the filter in slot 1 like a roofing filter in this case? So using the standard IF out and not the WBO would be the right thing to do. My receiver does have the WBO option too. With the standard IF out it should pass the 3.2 Khz SSB bandwidth selected while in that mode.

This wideband system system that I am talking about can be tuned with the 8617 alone by finding these MUX channels of the sides of the main carrier every 8Khz or so. Tuning these in very fine steps in USB will yield the voice phone calls taking place but tuning them in isnt so easy. Getting right on freq with 8617 alone is hard, being just a few Hz off and the voice is not intelligible. I hope the FDM box will offer a more precision method of tuning them in.

On the note of the 8617's IF out, It seems the 9518 box that I ahve found on ebay now tunes from 300Khz to 15 Mhz. If I understand this correctly this is the frequency range of the IF input it can accept. I wil have to do something about my 21Mhz if from the 8617. I thought I was reading that the 9518 box required a base band input not IF input? This doesnt make sense to me.

you have been very helpful, thanks.

It cant be 300Khz - it has a 3dB passband on the Video at 300Hz - so it has to be able to tune down to there - and yes, any freq in the input range (300Hz thru to 15Mhz) can be used to input - so a 21,4/2 (10,7Mhz) IF will be fine. Off the top of myhead the dB level is plenty plenty sufficient.

Baseband inputs(?) only - nope, the outputs can be baseband though.

You are going to have to drop the 8617 IF output - as I said in my earlier comments - best way to do it; halve it to 10,7Mhz, and tune the 9518 to 10,7Mhz.

Does use slot 1 as a roofing filter - good question: I dont know - I'llhave to read up on that point - I've always had 3.2/6.4 Khz in slot 1, and that was what my comment was based on.

Tuning resolution of the 9518? - far superior to the 8617 - keep in mind the very rational of the 9518 is to deal with the signal types you are talking about. Thats its job. Its a bugger with the 8617 - its not really made for that kind of resolution.

Zguy - get a 9518 user manual - its the only way to get to grips with just what WJ models are capable of.
Add to it a scope if you can afford - its a lot easier to adjust/fine tune/demux/demod if you can see a decent spectrum in front of your eyes. I have a 9206 in spot on condition (the tube was replaced in 2002 since when its had no more than about 50hrs use) - drop me an email if interested - it will help a lot.

Lastly, zz0468's comments are not without merit - I have no experiance with the methodolgy he is suggesting, but I see no reason why it wouldn't work - its just that I am a WJ nut and am familiar with the equipment.

 

[zguy1243]

benbenrf, check your private messages....

 

[zguy1243]

The WJ-9518BE Demux unit arrived today, here is a photo of the unit that we have been talking about,