[hpsdr] Mercury Spurious Emissions

Sat Sep 10 10:29:53 PDT 2011

Joe,

You're absolutely right to be careful...  I was apparently tuning to an image of the spur on my R7000.  The actual signals were tens of dB stronger -- S3-S4.  I didn't *think* the calibration was off that far... but had tuned down from the target frequency to the first spur (instead of up, where it was...) and found the other spurs where expected harmonically, so didn't check more carefully.

I've put a spectrum analyzer picture at http://gallery.me.com/k9srb#100047 .  Centered at 120MHz, 10MHz/division.  The input is a 2M whip antenna sitting near the HPSDR rack-mount box (positioning didn't matter much, so it's likely radiating from the power/signal wiring behind the cabinet) so relative signal strength is meaningless, but the FM Broadcast band is clearly visible in the 90-110MHz range.  You can clearly see the comb of 3.072MHz spurs.  The relative amplitudes change with reflections and such as you move things around and touch them, but you do see the 1-in-4 effect Jeremy reported.  When you turn off the power, the spectrum to the right of centerline basically goes flat, along with spurs between 100 and 120MHz that are barely showing up on this particular picture, but are strong in other orientations of the antenna.  I don't see anything below that, but my antenna probably doesn't work well there so it doesn't prove much.  On a wider span, similar-sized spurs are visible up to 240MHz or so (probably antenna-limited).  Taking the lid off the box made no visible difference.  Connecting or disconnecting the antenna from Penny or Mercury made no visible difference, nor did terminating vs. non-terminating unused connectors.  Touching anything inside the box around Atlas with the lid off the box makes a huge difference, different depending on what you touch.  The signals are likely being radiated directly from Atlas and anything it's conducting or inducing RF currents into, like my power supply wiring.

I didn't try to measure the output of Penny at its antenna port, but I doubt that much of this makes it out that path.

Honestly, this is not unexpected.  I once had a lab on the 3rd floor of a building that also hosted a hotel on the top stories, and had Multibus experimental systems on the bench with 16MHz and 10mHz clocks running on exposed, unshielded backplanes.  A test equipment company employee was in town to demo some equipment at the nearby university and stopped by to chat.  He told me he had turned on a spectrum analyzer in his hotel room 10 floors away and had no trouble telling me what frequencies I was using.  Busses running sharp edges generate signals, it's just the way things are.  If you don't want them, you have to shield them.

On the plus side, this does explain the spurs I'd never bothered to track down on 2M... ;-)

Steve, K9SRB

On Sep 10, 2011, at 11:04 AM, Joe Martin K5SO wrote:

> ***** High Performance Software Defined Radio Discussion List *****
> 
> Jeremy,
> 
> Be careful not to confuse signals that you may "see" by using a VHF receiver to look for signals coming out of Mercury with actual RF radiating from the Mercury connector.  What you may be proving is that the VHF receiver you're using as a test receiver is a poor performer in eliminating/reducing any of a variety of image frequencies, not that Mercury itself has dramatic spurious emission problems.
> 
> True, the Mercury board is at least providing a 122.88 MHz source for the "test" receiver to mix with its local oscillators to produce what appears to be emissions coming from Mercury but those "signals" you see are probably not actually radiating from Mercury itself, instead they may actually be generated within your test receiver by mixing with the 122.88MHz and multiples thereof that are being radiated from Mercury.
> 
> 73,  Joe K5SO
> 
> On Sep 10, 2011, at 9:45 AM, Jeremy McDermond wrote:
> 
>> ***** High Performance Software Defined Radio Discussion List *****
>> 
>> On Sep 10, 2011, at 12:00 AM, Helmut Oeller wrote:
>> 
>>> Hi Jeremy,
>>> 
>>> It's possible that you receive the 122,88 Mc sampling clock due to a bad
>>> image rejection of the FM rig, if its if is around 10.7 Mc. In particular
>>> hand-held FM rigs have bad selection in the frontend.
>> 
>> I have checked on the IF frequency of the TH-D7A, and the first IF is supposed to be at 38.85MHz on 2m.
>> 
>> From Steve's dataset, it seems like it's harmonics of 3.072 that are being radiated somehow.
>> 
>> I'm seeing signals at:
>> 
>> Signal -- Harmonic (predicted frequency)
>> ------------------------------
>> 138.245 -- 45th (138.240)
>> Not Hrd -- 46th (141.312)
>> 144.385 -- 47th (144.384)
>> 147.455 -- 48th (147.456)
>> 150.525 -- 49th (150.528)
>> Not Hrd -- 50th (153.600)
>> 156.675 -- 51st (156.672)
>> 159.745 -- 52nd (159.744)
>> 162.815 -- 53rd (162.816)
>> Not Hrd -- 54th (165.888)
>> 168.960 -- 55th (168.960)
>> 172.030 -- 56th (172.032)
>> 
>> 402.435 -- 131st (402.432)
>> 405.505 -- 132nd (405.504)
>> 408.575 -- 133rd (408.576)
>> Not Hrd -- 134th (411.648)
>> 414.720 -- 135th (414.720)
>> 417.795 -- 136th (417.792)
>> 420.865 -- 137th (420.864)
>> Not Hrd -- 138th (423.936)
>> 427.005 -- 139th (427.008)
>> 430.080 -- 140th (430.080)
>> 433.150 -- 141st (433.152)
>> 436.225 -- 142nd (436.224) -- Weak
>> 439.295 -- 143rd (439.296)
>> 442.365 -- 144th (442.368)
>> 445.440 -- 145th (445.440)
>> Not Hrd -- 146th (448.512)
>> 451.585 -- 147th (451.584)
>> 455.655 -- 148th (454.656)
>> 457.725 -- 149th (457.728)
>> 460.800 -- 150th (460.800) -- Weak
>> 463.870 -- 151st (463.872)
>> 466.945 -- 152nd (466.944)
>> 470.015 -- 153rd (470.016)
>> Not Hrd -- 154th (473.088)
>> 476.160 -- 155th (476.160)
>> 479.235 -- 156th (479.232)
>> 
>> So, it appears as though the theory of 3.072MHz harmonics is correct, although it appears as though every 4th harmonic is being suppressed by something.
>> 
>>> 73 Helmut, DC6NY
>> 
>> --
>> Jeremy McDermond (NH6Z)
>> Xenotropic Systems
>> mcdermj at xenotropic.com
>> 
>> 
>> 
>> _______________________________________________
>> HPSDR Discussion List
>> To post msg: hpsdr at openhpsdr.org
>> Subscription help: http://lists.openhpsdr.org/listinfo.cgi/hpsdr-openhpsdr.org
>> HPSDR web page: http://openhpsdr.org
>> Archives: http://lists.openhpsdr.org/pipermail/hpsdr-openhpsdr.org/
> 
> _______________________________________________
> HPSDR Discussion List
> To post msg: hpsdr at openhpsdr.org
> Subscription help: http://lists.openhpsdr.org/listinfo.cgi/hpsdr-openhpsdr.org
> HPSDR web page: http://openhpsdr.org
> Archives: http://lists.openhpsdr.org/pipermail/hpsdr-openhpsdr.org/

 1315675793.0