What Does 75 Meters Look Like?

[Steve - WB3HUZ]

http://www.amwindow.org/misc/gif/75m3apr05annotated.gif


If you get a 404 Error, just hit reload on your browser.

[w3jn]

A Fine Business shot indeed, Steve.

There's a lot of information in that spectrogram.  Notice the "brick wall" edge of the DRM.  Some of the slopbuckets have a really nice transient when they key up, and you can really notice who is running processing and who isn't.

And you can see one of those so-called "hifi" slopbuckets around 3900 or so with WAY too much bass (and probably the attendant growly intermodulation distortion).

If the instrument is fast enough and the sweep is narrow, you can actually see an xmitter's synthesizer snap into lock (after a few swings +/- the center frequency), and see how much the carrier drifts from turn-on, as well as see how much synthesizer noise is on the carrier.  Xmitter fingerprinting is child's play with this sort of instrument.

Also cool is being able to see some idiot VFOing around the band then settling down on a particular freq, then calling "CQ".  Makes IDing these knuckleheads easy.

TNX for the post.  73 John

[W2VW]

My vote is for Steve for OO.

[W8ER]

Very impressive display! I can't say that I've ever seen anything quite like it. What piece of equipment was used to make that snap and how well does it work to identify activity in the CW band? The display of carriers being turned off and on may not show that well. I can only imagine a megabuck piece of gear. The scan rate appears extremely fast.

--Larry W8ER


OUTSTANDING DISPLAY!

[W2INR]

YES - - - - - I want one TOOO

[WA1GFZ]

Cool stuff,
I wish I could bring our new R&S ESIB7 RX home on weekends but I would never finish the new QTH and end up spending all my time building scan tables to monitor the band.
Poor receivers are just as bad as poor transmitters.

[WD8BIL]

One Question Steve.....

Are you qualified to use that piece of equipment ??

[K1JJ]

A Fine Business shot indeed, Steve.

There's a lot of information in that spectrogram.  Notice the "brick wall" edge of the DRM.  Some of the slopbuckets have a really nice transient when they key up, and you can really notice who is running processing and who isn't.

And you can see one of those so-called "hifi" slopbuckets around 3900 or so with WAY too much bass (and probably the attendant growly intermodulation distortion).

If the instrument is fast enough and the sweep is narrow, you can actually see an xmitter's synthesizer snap into lock (after a few swings +/- the center frequency), and see how much the carrier drifts from turn-on, as well as see how much synthesizer noise is on the carrier.  Xmitter fingerprinting is child's play with this sort of instrument.

Also cool is being able to see some idiot VFOing around the band then settling down on a particular freq, then calling "CQ".  Makes IDing these knuckleheads easy.

TNX for the post.  73 John

Thanks for the FB summary, John...

Could you please describe some more of what's going on in that shot - in basics?  ie, What is the significance of the totem pole look of each signal?  I see the AM signal has some brighter colors in it showing more amplitude in some areas.

I can see what you mean about some the SSB signals having a fake "carrier/bass" produced by the audio being near the edge of their filter bandpass, etc.  Is this correct?  Where is the transient in some SSBers as you said?  

The sweep says "100ms"... but I see "29.21 sec" on the left corner. What is the total integration time of this shot?

What are some of the finer points you see there? Is that AM station saying, "caw mawn", or "hola" in that particular sweep?

Some signals are darker than others - is this pure signal strength or the density of audio as you mentioned about some ssb processors?

TNX, OM.
T

[w3jn]

It's a Hallicrafters S-20R with the very rare SooperDooperSpectrumScope option, and an outside speaker.  

Either that or an Agilent 89600 VSA.

73 John

[WA1GFZ]

Gee I thought it was one of dose new sound cards with spectran.

Steve Wins the coolest toy of the day award!

[w3jn]

It's a spectrogram much like the hammy hambone audio card software that's drifting around the net, except it's RF as opposed to audio.  

Picture a spectrum analyzer with a 100 mS sweep and 100 hZ RBW (which is what Steve used here IIRC).  Instead of showing the signal intensity in vertical deflection it shows the sig intensity in color - blue is essentially no signal and red is -30dBm.  Each sweep is saved in a "line" of pixels, then the next sweep goes below it, etc.  There are 29.21 seconds of sweeps presented, or 29.21/.1 sec = 292 sweeps.  Essentially you have a 3-d display: freq vs amplitude vs time.  

A VERY strong signal will be red; weaker signals will be various shades according to the color bar on the left side of the dicksplay.  CW signals show up as the dots and dashes you would expect; RTTY shows up as a jiggly line.  FM appears as a zigzaggy mess.  A strong AM signal modulated by a pure tone would show up as three lines; say a yellow carrier and two weaker (say green) sidebands either side of it.  Johnny would show up as a crappy yellow mess blocking about 50 KHz.

You can see the LSB signals as the higher freq audio out to the left and the lower freq out to the right.  See the signal that is more intense at the right - that's an indication of way too much bass.  Denser signals show the effect of audio processing.

You can see one of those carriers that kind of fades in and out - a good representation of ionospheric phading.

You can see on some of them a horizontal line when they start/stop xmitting.  This means a nice audio (or oscillator) transient... kind of like the big "THUNK" my Valiant makes on keyup.

That thing might work on 6 meters but I suspect all you'd get would be TV horizontal lines and a few little pissweak sparklies, caw mawn....

73 John

[K1JJ]

Wow - that's quite a message! [picture]

So, each horizontal pixel stream is a separate time snapshot/sweep of the band... that's really something.

You can see up around 3950 or so where a couple of stations appear too close - maybe 2.5kc spacing. Whereas, down around 3830 those guys appear to be using 2.7kc audio with 3kc channel spacing.

I see a few horizintal lines that may indicate transcient key up as you describe.

I'll bet with the horizontal baseline blown up X10 you could give out some really detailed bandwidth checks.

Yep, we all need a set up like this. Adding visual to anything gives a whole new dimension of data input efficiency.

Tnx again for the tutorial, JN.

T

[Steve - WB3HUZ]

Glad you boyz liked that. It's a different and really cool way to look at the spectrum or a signal. The spectrogram I posted is but one of many ways of "viewing" a signal with the analyzer I used. I can also show instantaneous phase, PSD, PDF, amplitude (more or less what you would see on an oscilloscope) and more. It can show one or all of these simultaneously. Neat stuff.

JN did a great job explaining the image. The URL below is another shot I made of Brent, W1IA on 3 April. He had just put his new Class E rig back on the air after blowing it up the previous night. The shot shows the spectrum 10 kHz above and 10 kHz below Brent's center frequency, so there is much more detail of the AM signal on this shot.

Same as before, if you get a 404 Error, just hit reload.

Enjoy!

http://www.amwindow.org/misc/gif/w1ia75m3apr05annotated.gif

[WA1GFZ]

I heard Brent for the first time last night. He sounds very good
and strapping 70 to 80 dBuV.

[W8ER]

What a neat display Steve. It tells most of what anybody would ever need to know about signals and conditions etc. I want one!

--Larry W8ER

[K1JJ]

Man... you can see Bob's third BA hanging there in the low frequencies compared to Brent.

They both look pretty clean for hi-fi AM.

So, how long will you have use of it, Steve - just last weekend?   If longer, maybe take a bunch of shots of the regulars and post them...

An interesting example would be a stock 3kc audio boat anchor that is overdriven and WIDER than a clean "wide body" hi-fi AM transmitter.

T

[w3jn]

For more info on this cool instrument, click here:

http://www.home.agilent.com/USeng/nav/-12086.536881170/pd.html

Here's a pic of the bad boy:

[W8ER]

I didn't see a price posted.

Should we take bets?    uh   $12K for my guess.

--Larry

[w3jn]

Try $35-70K depending on options.

[Steve - WB3HUZ]

That's it. I wonder why Agilent used such a crummy looking IQ pattern on the upper left display?


The unit is at work, Tom. I will take shots of various interesting signals and post as time/conditions permit. Should be fun.

For more info on this cool instrument, click here:

http://www.home.agilent.com/USeng/nav/-12086.536881170/pd.html

Here's a pic of the bad boy:

[WA1GFZ]

The Tom Vu model comes with the handy 20 dB pad to protect the front end. Not a bad price the guys next door bought the 40 GHz. R&S ESIB7
for about $150K and couldn't afford the tracking generator option.

[Steve - WB3HUZ]

Two different animals really. The R&S is an EMI receiver, albeit a very sophisticated one. Although the VSA can do pre-compliance EMI measurements, it's really for nitty-gritty signal measurement and analysis. The biggest difference is that the VSA can perform vector measurements and analysis. I don't think the R&S can.

[K1JJ]

Not a bad price the guys next door bought
the 40 GHz. R&S ESIB7 for about $150K and couldn't afford
the tracking generator option.

Real World prediction:
Fifteen years from now you see one in a shit pile of test junk
on the ground at Hosstraders - no one knows what it is or even
wants it for  free...  :lol:   [Your home computer does it in year
2020 with a free software download]


Bizzaro World prediction:
In 2020 analyzers become more valuable than moon rocks
because everyone forgot to make any more.

T

[WA1GFZ]

Tom you are pretty close today. All this stuff is running windows based control software. It is a PC with a RF front end. The fancy case is a front
for keeping the cost high. Heck there was a vector network analyzer project in QEX last year you could build for about $200. It functioned to
200 mHz.  All this stuff runs windows and you can connect it to a network.
Testing can be set up over the net so remote control is pretty easy.
Many things have key board and mouse connections.
Yup the ESIB7 is not vector but you can't beat that S meter option
with any RX I have ever seen. You can even monitor FM while setting up
with the handy 2 inch front panel speaker.  Biconical antenna in the hall and you get tunes. No stereo though.

[w3jn]

I saw the 89600's predecessor, the 89441, at Dayton last year for about $5K.  This was an $85K analyzer and was current til about 2000.  Has much the same capability of the 89600 plus has an arbitrary source.  What's cool about that is you can capture a signal to memory (max bandwidth = 8 MHz) then replay it back thru the arbitrary source into a receiver or whatever other instrument.  You can even play it backwards. ANd show the signal in time doman, freq domain, demodulated time domain, and demodulated freq domain SIMULTANEOUSLY if you want.

The 89441 is stand-alone, no PC needed.  5K for this is a hell of a lot better buy than those IFR service monitors, IMHO.

73 John