开云体育

John,

I definitely concur that making changes to the loop filter is the riskiest
approach. If the loop tunes the YTO only through the FM coil and not through
a combination of FM and main coil, then compensating for too much FM coil
sensitivity is probably best done in the FM coil driver. FM coils typically
have less than one ohm DC resistance, so it would take a low value resistor
to shunt it. Usually there is a single resistor in the FM driver that sets
its gain or there is a resistor in series with the FM coil that sets the
overall sensitivity of the FM coil and its coil driver. This series resistor
is a whole lot greater than the FM coil resistance so it pretty much sets
the TC of the combination. Besides, as you point out, the loop can take care
of any residual temperature related errors. I wish I had a copy of that
portion of the schematic so I could be a little more definitive.

Good luck,

Bob Dildine
W6SFH


________________________________

From: hp_agilent_equipment@...
[mailto:hp_agilent_equipment@...] On Behalf Of John Miles
Sent: Thursday, November 23, 2006 13:18
To: hp_agilent_equipment@...
Subject: RE: [hp_agilent_equipment] HP8568B



Some good points there. My thinking was that the loop was probably
not
pretuned at all, or it would not have locked up properly with the
incorrect
coil-sensitivity figure. But checking with the 8568 manual, it
appears they
do pretune the main coil with a DAC to make sure it locks to the
expected
20-MHz comb tooth.

That suggests that any sensitivity discrepancy is limited to the FM
coil,
rather than the main coil. Not surprising, since main-coil
sensitivity
seems to be 20 MHz/mA in many YTOs. Any temperature-related errors
introduced by shunt resistance across the FM coil should be
corrected by the
PLL without difficulty.

At any rate, the objection I have to altering the loop filter is
that it's
very hard to be sure you're not causing harm elsewhere. To make the
original PLL circuit work seamlessly with a different YIG, you need
to
change the loop gain without altering its bandwidth, which can be
tricky
considering that the gain and bandwidth normally changes over the
tuning
range and, often, during the acquisition process itself.

Conversely, to fool the PLL into thinking it's controlling the old
YTO, you
only have to change one parameter, the YTO's tuning sensitivity. If
you're
lucky enough to have a more-sensitive YTO, adding shunt resistance
across
the (non-pretuned) FM coil seems relatively safe...

-- john, KE5FX