I decided to rip the add-on board apart, and go right to the oscilloquartz 8633
output.
no surprise, >0dBm level, but alas, 20Mhz harmonics are only -35 dB down, and
there's one at 30, etc.. hmm, what does this really mean? well, when in doubt,
go back and look at the data:
http://www.sungwhatech.com/product/pdf/aging/8663.pdf
rats, *the spec is only -25dBc for harmonics*. phase noise looks good, as does
stability, but
now I understand why I am seeing these spurs, it's in spec! it does look like
my level is out of spec though, I cannot get +4dBm even right from the module,
perhaps that's why these units became surplus. the date code on my OCXO 8663-XS
is 2008. I cannot identify what the X part of the dash number means, maybe
lower output with my luck?
to make the board a bit safer to use, I made up a resistive splitter with 2 51R
carbon resistors, and slipped ferrite beads on the leads to help control high
frequency harmonic content. I can now run the counter and SA at the same time,
and get maybe -2dBm outputs, closer to 0dBm if I use only one.
if I add my outboard 10MHz bandpass filter, harmonics drop into the grass at
-60dBm, but level goes to -5dBm.
clearly not as nice as the FE5680A Rb unit, but I will leave it in there just
for comparison. I think I will filter one output, and leave one raw, since I
have 2 jacks already at the front for outputs. but this was very instructive, I
really did not expect to see so much harmonic content from a precision OCXO
sinewave crystal oscillator. a quick cross check on some of my other hp/FE
ovenized standards in the lab does not show this kind of unwanted content.
what kind of results does everyone else get?
all the best,
walter
--
Walter Shawlee 2, President
Sphere Research Corporation
3394 Sunnyside Rd., West Kelowna, BC
V1Z 2V4 CANADA Phone: (250) 769-1834
[email protected]
WS2: We're all in one boat, no matter how it looks to you.
Love is all you need. (John Lennon)
But, that doesn't mean other things don't come in handy. (WS2)
_______________________________________________
time-nuts mailing list -- [email protected]
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
