Roofing Filters

[Bill W4ZV]

K0WA:
 >How many roofing filters do you need in a K3?
I would suggest one for SSB and one for CW.

         That's basically correct, although it
depends on how you plan to operate (i.e. serious
low band DXing or contests versus casual operating).
Remember that a roofing filter's purpose is to
prevent IMD and BDR products from passing through
the IF chain to the final DSP stage where the final
filtering is done.  That stage will have variable
filter bandwidths over a wide range (every 50 Hz if
I recall correctly).  The DSP stage really does the
heavy lifting after the roofing filter ensures that
unwanted products are eliminated at the first IF.

         If you do casual operating where adjacent
signals are seldom above S9+30, it is very unlikely
you need more than the stock 5-pole 2.7 kHz filter.
Remember that (excluding Orion) few modern rigs use
roofing filters narrower than 3 kHz (this includes
the IC-7800, FTDX9000, FT-2000 and most of the Inrad
roofing filter add-ons for older rigs (Ten-Tec's
Omni VI excluded which can be fitted with a 600 Hz).

         Since the filters have not yet been fully
defined or characterized, I would be tempted not to
order ANY optional filters at this point.  As ARRL
discovered in their Orion tests, narrower is not always
better.  This resulted in Ten-Tec redesigning the
Orion II with ONLY 4-pole roofing filters, instead
of using the older 8-pole 500 Hz and 6-pole 250 Hz
filters used in the original Orion.  By "characterized"
above, I mean published IMD and BDR specs for each filter.
As Ten-Tec has discovered with Orion, roofing filters are
somewhat "black magic" and there is no substitute for
actual measurements before making a decision about which
may be best.  A good discussion of roofing filters by
George W2VJN of Inrad is here:

http://www.qth.com/inrad/roofing-filters.pdf

         Filling K3 with filters as Toby suggested is
both unnecessary and expensive IMHO.  I think he is
forgetting that there is very good DSP filtering in
the final DSP IF that is already doing what he is
attempting to do with a wide range of roofing filters.
If you are a serious contester, likely to operate in
an environment of S9+30 signals spaced every 500 Hz
on 160 meters, then you probably do need a narrow
filter like the 500 Hz or 400 Hz.  Going narrower
is questionable IMHO because:

1.  You would not hear off-frequency callers with
narrower bandwidths (I've found even 500 Hz is too
narrow in some cases.

2.  At 500 Hz signal spacing, which a 500 Hz roofing
filter will handle very well (i.e. +/- 250 Hz excludes
the +/- 500 Hz interfering products), you are probably
already limited by issues like transmitted phase noise,
key clicks, etc. which will override even a perfect
receiver with infinitely good IMD and BDR performance.
Even a perfect receiver cannot prevent an adjacent TX
signal's defects!

         Regarding 5-pole versus 8-pole, note W2VJN's
comment from page 6 of the article above:

"5. If 6 poles work so well, why not 8 poles?
The most important part of the filter
characteristic is from the pass-band on down
to about –30 dB on either side of center.
Eight poles would provide much better
stop-band isolation, but it’s not required in a
roofing filter and would make no
noticeable improvement in IMD performance."

         Bottom line:

1.  Narrower is not always better
2.  8-poles is not always better than 5-poles.
3.  Let IMD and BDR measurements be your guide.

                                 73,  Bill  W4ZV




_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Eric Swartz - WA6HHQ]

Here is a link to our 8-pole, 8.215 MHz, crystal filter plots. It has
now been added to the K3 page.

http://www.elecraft.com/K3/K3_8_pole_plots.htm
http://www.elecraft.com/news.htm

Also note that we follow these filters with the DSP for even more out of
band rejection and skirt steepness.

73, Eric  WA6HHQ
---------

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[dj7mgq]

Zitat von Bill Tippett <[hidden email]>:

>          Filling K3 with filters as Toby suggested is
> both unnecessary and expensive IMHO.  I think he is
> forgetting that there is very good DSP filtering in

<< many lines removed >>

> Even a perfect receiver cannot prevent an adjacent TX
> signal's defects!

>From my layman's view, I am not sure that I fully agree with you. Here in the
EU the CW segment of the 40m band is (still) only 40kHz wide, in other words,
assuming that in a major contest there are 200 european stations on 40m, then
you have an average spacing of 200Hz. In SSB the bandwidth is 60kHz giving an
average spacing of 600Hz - assuming 100 stations. (To be honest the number of
stations calling CQ is an educated guess.)

As you may be able to imagine, the 40m band can be a real zoo, and every bit of
(clean) selection, as early as possible, helps. Some of the major contest
stations go so far as to use a bank of switchable crystal filters in front of
the mixer <http://www.yt6a.com/en/front-end-filters/index.php>.

As I see it the main job for the roofing filter in the K3 is to help "protect"
the ADC and prevent AGC pumping. If I am listening to a weakish signal (S5)
between two strong (9+40) stations with a 200Hz bandwidth DSP and a roofing
filter of say 500Hz on 40m then either the AGC will densense the RX (AGC
pumping) or the strong stations will "overpower" the ADC.

I contest mostly in SSB, so I need good early selection in SSB. At out contest
QTH, where my K3, may end up living, there are many avid CW operators and they
will need good selection in CW. Not to mention the RTTY contests. Also, when
the K3 is not at our contest QTH, e.g. if I take it with me on vacation, then I
do want to be able to listen to broadcast stations or when using a transverter
work over relais (FM).

This why my personal dream(!) filter setup is so rich. It might be expensive
but I doubt that it is totally unnecessary.

As to the side effects of poor signals (TX phase noise etc.), I agree with you
fully. It DOES make things much more difficult.

But the again, what do I know... ;-)

vy 73 de toby
--
DD5FZ, 4N6FZ
K2 #885, K2/100 #3248, K3 #<200


_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com