[TenTec] 9Mhz 500 Hz filter

[Bill Tippett]

N4VZ wrote:
 >The loss of crystal ladder filters increases as bandwidth
decreases because the narrower bandwidth filters are
more heavily loaded by the capacitors to ground at the
nodes between the crystals.  Thus, a 4 pole 600 Hz filter
will for sure have more loss than a 4 pole 1000 Hz filter
using the crystals with the same Q.

         Terry, try making some measurements.  Apply a signal
source and set DSP = 6000.  Now scroll through the roofing
filters.  On mine, the 20k-1k filters have virtually identical
losses.  Of course TT could be compensating in the DSP.
Also note that Ten-Tec's own specs for the 2400 Hz #220 is
2.2 dB loss and the 1800 Hz #218 is 2.0 dB loss.  The
percentage between 1800/2400 (75%) is not too far off 612/1070
(57%) so I would not expect a huge difference.  Inrad's 600
Hz shows ~4 dB loss and I doubt the 1000 Hz would be any
better than ~3 dB, so the difference, if any, is slight IMHO.

         Regarding the PBT and LCW/UCW issues, simply try plugging
your 400 or 500 Hz filter into your 1000 Hz slot and tell me
if you can find the signal in both UCW & LCW using the full
range of PBT.  I recall I could not find it in LCW which is
my preferred passband for CW.  Once you have found it, now
switch the roofing filter to 250 or 500 Hz and tell me what you
hear.  I heard hiss on mine until I reset PBT back to zero.
Not exactly a good situation for contests unless you don't
plan to ever switch roofing filters, and it would definitely
eliminate the usefulness of the Auto filter selection.

 >Equalizing the gain with the pad on the filter board will not
degrade the IMD performance if the additional IF amplifier
after the filter is well designed.

         No pun intended, but your "if" following "performance" is
the key issue.  In a perfect world with perfect amplifiers, we
should not need narrow roofing filters (as certain un-named
manufacturers still mistakenly believe).

 >The 4 pole will be somewhat inferior to the usual 6 pole 500Hz filter

         The #217 is 8-pole if that is what you are referring to.
When I can buy a used #217 for $60, why would anyone bother with
building a 4-pole substitute for ~$40?  Surely not to save $20!
The reason I want a 4-pole 600 Hz for Orion at 9.001500 is to
improve IMD and BDR performance at closer signal spacings, not
because I am trying to save money.

                                                 73,  Bill  W4ZV