EMRFD Message Archive 1223
Message Date From Subject 1223 2007-12-10 20:54:01 ve7ca2 Spectrum Analyzer - Tracking Osc. Mod.
by adding 6 db pad after the tracking Osc. mixer as noted on Wes's web
site SA notes 31 Aug06/5 Jan 07 and recommended by KG4ARN. The 6 db
pad did reduce the baseline rise by approx 15 db so that the baseline
now averages only 10 db above the noise floor of the SA.
Later I built a 80 Mhz LP filter as suggested by KG4ARN but find the
extra filter does nothing to lower the baseline any further. All my
SA modules and the tracking osc. are enclosed in Hammond boxes however
the they do share the same power supply. I suspect that there is a
little leakage from the tracking Osc. along the the power supply lead
even though it includes a 12 uH choke between the feed thru cap. and
the PC board and a 12 uH choke on the outside of the box with a
further .1 cap. Also maybe the
SA Swept VCO line has a very small amount of leakage on the Coax going
back to the VCO. All coax used to connect the various boxes are
doubled shielded.
I am wondering if anyone else has tried built the 80 MHz Filter and
if it was any help in reducing the baseline any further? I checked
the response of my 80 MHz filter with my HP8640 and Power meter and
the loss is less than 0.3 db up to 80 Mhz and then falls significantly
thereafter.
73 Markus VE7CA1224 2007-12-11 06:49:20 Robert Cerreto Re: Spectrum Analyzer - Tracking Osc. Mod.
I have had the same exact experiences as you have had. My SA is constructed the same way yours is. I tried to isolate the power supply by running the tracking generator from a battery. That had no effect on the noise floor.
I know the hammond boxes are good, but, I read somewhere that someone "double boxed" their tracking generator and it reduced the noise floor. I haven't tried that yet.
Bob WA!FXT
_,_._,___
---------------------------------
Never miss a thing. Make Yahoo your homepage.
[Non-text portions of this message have been removed]1234 2007-12-14 10:53:43 Ed Almos Re: Spectrum Analyzer - Tracking Osc. Mod.
solution is expensive:
1) Either house each module in double boxes or fit a grounded shield
over the pins of each connecter entering the first box.
2) Each lead not carrying RF should be dealt with as follows. 1nF feed
through on outside of box, series 1mH RF choke, 1nF feedthrough on
shield or second box, 50 ohm coax from the feedthrough to the PCB
(grounded at the shield or second box), 100nF wired across PCB connection.
3) All connections to the feedthrough capacitors should be shielded.
This can be done by placing heatshrink sleeving over the connection,
then a short piece of coax braid grounded at one end, then a second
piece of heatshrink.
4) All connections carrying RF should be shielded in a similar manner
to step (3).
5) Ferrite beads