[HamWAN PSDR] HamWAN Lab expansion + garage sale

[Bart Kus]

Hello,

_*HamWAN Lab Garage Sale Items*_

 1. I've managed to bring back to life an HP 8757A I've had for 6 months
    in a non-working state.  Did it last night actually. I'll be selling
    this on eBay since it's been obsoleted by another item, so if anyone
    local is interested in it, let me know.

 2. Also managed to verify function of a 10MHz-26.5GHz detector (HP
    85025B) that I picked up 7 months ago.  It actually works! I'll be
    selling this too (works with the 8757A), so again if you're local
    and interested, drop me a line.

 3. Finally, I'll be doing a couple more tweaks to an HP 8566A
    100Hz-22GHz spectrum analyzer and selling it.  Right now it still
    has a problem with one set of sweep speeds, which I suspect is a
    constant-current ramp generator circuit failure. An IF filter also
    seems misaligned by about 31Hz, so I'll be touching that up and
    re-calibrating levels.  Let me know if there's any local interest in
    this unit, otherwise on eBay it goes.  They're beautiful machines,
    but I don't need 2 of them.


_*HamWAN Lab Expansion*_

 1. Newly acquired HP 11720A Pulse Modulator.  This thing can take a
    2-18GHz microwave signal and either pass it through or dissipate it
    internally.  Doesn't sound exciting, right?  Until you realize it
    can do the full on+off sequence in a span of less than 50
    nanoseconds!  Rise and fall times are spec'd as less than 10ns. 
    On-off power ratio is spec'd as >80dB, but I just verified it's
    actually >95dB on this particular unit @ 4GHz.

    So what is this good for?  Let's assume the minimum pulse width is
    actually 40ns.  That corresponds to an RF signal in space that's 12
    meters long, traveling at the speed of light.  Let's call it an RF
    packet.  An interesting object, but how is it useful?

    We are doing high-performance (read: high dynamic range) antenna
    radiation pattern measurements.  When you shoot an RF beam at a test
    antenna, that signal is not confined to the antenna itself.  It also
    goes into the surrounding environment and then can bounce back at
    the antenna.  These bounces add to the power received by the
    antenna, and destroy the accuracy of power readings you're trying to
    take.  RF absorbers help the problem, but do not eliminate it since
    they don't absorb 100% of the RF energy scattering back and
    distorting measurements from weird angles.

    Now imagine you take an RF-silent environment (go to the mountains)
    and strike the same antenna now with an RF packet. You then take a
    power reading 10ns after initial RF impact, and for no more than
    30ns after impact (avoid rise+fall times).  As long as there are no
    sources of reflection within a certain radius, there will be ZERO
    power in the RX DUT antenna which comes from reflections!  Your
    dynamic range just increased dramatically, and so did your
    measurement accuracy.

    What is this "certain radius" in this example?  The RF packet has a
    10ns rise time, so that signal's already gone 3 meters past your
    antenna before you start your readings.  Your reading window is 6
    meters (20ns) long.  The head of the rise envelope would have to
    meet your readings no more than 30ns (9 meters) from when it passed
    the DUT antenna to register and distort the experiment.  This means
    it would have to hit a reflector 4.5 meters away and then travel
    back.  So there you have it, a radius of 4.5 meters around the DUT
    clear of reflection sources will guarantee no measurement distortion
    from reflections.

    One problem remains.  Even though this instrument provides a means
    to generate the necessary RF packets to perform such testing, I
    don't yet understand how to receive a precisely timed 20ns window of
    RF and determine its power.  If anyone has any suggestions, I'm all
    ears.  I suspect RF engineers familiar with radar systems will have
    a lot of good input here.

 2. Newly acquired 1W / 35dB gain microwave amplifier (CTT
    APM/060-3032).  It runs off 15V @ 1A, and is really small.  So small
    in fact that we can climb with it.  What is this good for?  We often
    have a hard time aligning dishes.  It can take an hour in an awkward
    position on a tower tweaking with the noisy measurements the modems
    give us, if we can find a modem signal at all.  The problem with the
    modem TX beacons is that while they're 1W signals, they're spread
    out over at least a 5MHz bandwidth.  This reduces their power
    spectral density and makes them harder to detect.  They're also not
    continuous signals. Sending a single frequency continuous 1W signal
    from a remote dish would really help alignment.

    There is still the problem of how do you portably generate the
    required precise 5.9GHz signal to feed the amp, and is this only
    applicable when two sites are being worked on simultaneously. But
    it's an interesting bit of equipment that can possibly solve the
    alignment problems we face.

 3. Corresponding to a single frequency transmission, you need a single
    frequency receiver on the other end.  The modems will not pick this
    up.  Actually, I have to verify if they'll pick it up as noise floor
    fluctuation or not!  That'd be an interesting option if it works. 
    Anyway, a portable receiver capable of 6GHz work is really
    expensive.  We're talking Agilent FieldFox or Anritsu SiteMaster
    type of stuff.  We already have spectrum analyzers capable of 6GHz
    work, but they're too large to take on a tower.  So they have to
    live on the ground.  Running LMR400 down a tower will eat a lot of
    the received signal (20dB for a tall tower), and you may not hear
    the remote site.  This calls for a pre-amp of sorts!  So I bought a
    JCA JCA48-4111B1 34dB gain amplifier.  It's also tiny and can be
    battery-powered, so it can be installed @ an RX antenna to feed
    200ft of LMR400 before it hits a proper receiver.  I'm not sure if
    this will work out or be worth it.  A better LNA-type amplifier
    might be needed, and the LMR400 may prove too bulky in field work,
    but the amp was cheap, so might as well have it on hand.  It might
    also help with signal measures of circuits on the lab bench if
    nothing else.

 4. And then I bought a big amp.  :)  A 10W Traveling Wave Tube
    amplifier made by Hughes, model 1177H13F000.  It covers 3-8GHz at
    this power level.  It will be useful for such experiments as "Hey
    Bob, go stand in front of that dish and tell me if you feel warm." 
    And, "Do I have a death ray yet?"  :)

    But seriously, some of the experiments deal with RF leakage from
    antennas, specifically near-field measurements from the rear of
    antennas.  Finding the sources of this leakage can be tricky since
    tiny probes have to be used to maximize spatial resolution, so this
    means low gain RX.  The attenuation through the back is already
    high, so there's very little there to be heard in the first place. 
    But if you blast the antenna with lots of energy, you greatly
    increase your chances of picking up the hot spots!  +40dBm, here we
    go.  :)

    While I'm not sure yet, this amplifier might be useful in pulsed
    mode to possibly drive 100W (@ <10% duty cycle).  Will have to
    research this more and see how TWTs feel about pulsed RF. Don't
    wanna cause internal arcing.

    The other useful application for this is as a reference amplifier as
    we try to develop our own 10W cheap silicon amp. When/if we do a
    100W cheap silicon amp design, having 10W on hand will be handy as a
    first stage.  These amps would be useful in more challenging links,
    like Cascades to Spokane.  Amazingly, the signal path doesn't hit
    the ground, but does have to travel 256km.

    Finally, if we end up doing filter design, and the filters end up
    being good, we might need a lot of incident power to measure their
    true attenuation.  Having this bad boy on hand solves that problem. 
    Plus I like amps.  And this is my first TWT amp.  Joy! :)

 5. Oh, there is one more thing...

    HamWAN Lab will soon feature an HP 8573E Vector Network Analyzer
    with options 006 (6GHz extension) and 011 (direct access to S/R/A/B
    channels), along with an HP 87050A option H47 S-Parameter Test Set. 
    This instrument features 110dB dynamic range @ 6GHz and is useful
    down to 300kHz with the test set. The special (rare!) test set that
    comes with this is extra nice because it allows you to make
    measurements at higher than normal power levels for this instrument
    by in-lining an amplifier. Direct measurements at up to 1W can be
    made.  It also allows the switching in and out of up to 3 external
    stimulus/measurement systems.  The very broad frequency range of
    300kHz-6GHz will allow this instrument to characterize all sorts of
    devices, including transistors from HF to HamWAN microwave.  These
    transistor characterizations are necessary when designing amplifier
    input and output matching networks (for example), and are not always
    provided by device manufacturers.  Measurements can also be made at
    power levels higher than 1W by using some additional external
    components, so work @ 10W and 100W should be possible, albeit with
    degraded accuracy compared to the native 1W TX / 0.4W RX system.


Just as a general note, HamWAN Lab is my own private collection of 
stuff, not funded in any way by HamWAN.  Although the work done in this 
lab is a key to HamWAN's success.  If you have any projects which might 
benefit from some lab time, feel free to get in touch with me.  The (out 
of date) inventory is here: 
https://www.hamwan.org/t/tiki-index.php?page=Labs&structure=HamWAN .  
Should really update that page.  It's months out of date.


--Bart

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.hamwan.net/pipermail/psdr/attachments/20131005/ff17bcef/attachment.html>