Discussion:
[Elecraft] Roofing Filter Usage
pastormg2
2010-10-28 10:41:26 UTC
Permalink
I purchased a K3 with the standard 2.7 Khz SSB filter plus the 1.8 Khz SSB filter. I did not purchase a cw filter because I operate very little cw. But that may change. The 2.7 is in slot #1 and the 1.8 is in slot #2. I know that if I push the FILT button that I can switch filters. But I was experimenting around listening to a weak station that was rather close to a real strong station and when I activated the 1.8 Khz filter I was still getting some carryover from the strong station on the weak station. Are there other controls that I need to use that would enable me to knock out the real strong station so that I could clearly hear the weak station? I'm sure there is help in the manual but I certainly can't find it. Any help would be much appreciated. Oh and by the way, I love the radio. It's like a breath of fresh air compared to my Kenwood TS0480s. Mark KB3Z
Stephen Prior
2010-10-28 10:48:22 UTC
Permalink
Hi Mark

Try reducing the bandwidth further using the width control, then the DSP
filtering will be responsible for narrowing the bandwidth. You could also
try adjusting the shift control to slide the passband one way or the other.
Although I have all my filter slots filled, I never use the xfil button to
change width, just the width control. The effect is the same.

Yes, it's a lovely radio, and I came from a TS480 too, via a K2 that is :-)

73 Stephen G4SJP
Post by pastormg2
I purchased a K3 with the standard 2.7 Khz SSB filter plus the 1.8 Khz SSB
filter. I did not purchase a cw filter because I operate very little cw. But
that may change. The 2.7 is in slot #1 and the 1.8 is in slot #2. I know
that if I push the FILT button that I can switch filters. But I was
experimenting around listening to a weak station that was rather close to a
real strong station and when I activated the 1.8 Khz filter I was still
getting some carryover from the strong station on the weak station. Are there
other controls that I need to use that would enable me to knock out the real
strong station so that I could clearly hear the weak station? I'm sure there
is help in the manual but I certainly can't find it. Any help would be much
appreciated. Oh and by the way, I love the radio. It's like a breath of fresh
air compared to my Kenwood TS0480s. Mark KB3Z
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:Elecraft at mailman.qth.net
This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
ab2tc
2010-10-28 12:37:40 UTC
Permalink
Hi,

I agree with forgetting about the XFIL button and use the passband controls
exclusively. But using SHIFT/WIDTH on SSB is very awkward as you constantly
have to touch up both controls. Use HiCut/LoCut instead. Leave LoCut at
300Hz and use HiCut as the QRM fighting tool. If you have the 1.8kHz filter
you will see it kick in automatically when HiCut is at or blow 2.1kHz. If
your cut frequencies are on 350, 450 etc boundaries, pop back to SHIFT/WIDTH
mode and adjust the SHIFT one click, then go back to HiCut/Locut and they
should be on even 100Hz boundaries (unless of course you prefer it the other
way).

AB2TC - Knut
Post by Stephen Prior
Hi Mark
Try reducing the bandwidth further using the width control, then the DSP
filtering will be responsible for narrowing the bandwidth. You could also
try adjusting the shift control to slide the passband one way or the other.
Although I have all my filter slots filled, I never use the xfil button to
change width, just the width control. The effect is the same.
Yes, it's a lovely radio, and I came from a TS480 too, via a K2 that is :-)
73 Stephen G4SJP
<snip>
--
View this message in context: http://elecraft.365791.n2.nabble.com/Roofing-Filter-Usage-tp5682095p5682375.html
Sent from the Elecraft mailing list archive at Nabble.com.
George A. Thornton
2010-10-28 14:30:05 UTC
Permalink
Go back to filter settings and set the bandwidth of the 1.8 filter a bit
wider, say 2.1. Then use the hi/lo adjustments during operation.
Typically I keep the low full and gradually lower the hi settings. When
you get the filter narrow enough the K3 will automatically switch from
the 2.7 to the 1.8 filter. Several experienced users recommended this
to me and it works well.

-----Original Message-----
From: elecraft-bounces at mailman.qth.net
[mailto:elecraft-bounces at mailman.qth.net] On Behalf Of
pastormg2 at verizon.net
Sent: Thursday, October 28, 2010 3:41 AM
To: elecraft at mailman.qth.net
Subject: [Elecraft] Roofing Filter Usage


I purchased a K3 with the standard 2.7 Khz SSB filter plus the 1.8 Khz
SSB filter. I did not purchase a cw filter because I operate very
little cw. But that may change. The 2.7 is in slot #1 and the 1.8 is
in slot #2. I know that if I push the FILT button that I can switch
filters. But I was experimenting around listening to a weak station
that was rather close to a real strong station and when I activated the
1.8 Khz filter I was still getting some carryover from the strong
station on the weak station. Are there other controls that I need to use
that would enable me to knock out the real strong station so that I
could clearly hear the weak station? I'm sure there is help in the
manual but I certainly can't find it. Any help would be much
appreciated. Oh and by the way, I love the radio. It's like a breath of
fresh air compared to my Kenwood TS0480s. Mark KB3Z
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:Elecraft at mailman.qth.net

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
David Woolley (E.L)
2010-10-28 22:30:35 UTC
Permalink
The purpose of the roofing filter is to reduce receiver generated
distortion products, not to provide selectivity. Are you sure that that:

1) the strong signals are completely out of band;
2) the only distortion products are those generated by the K3 itself,
and not the transmitter.
Post by pastormg2
I purchased a K3 with the standard 2.7 Khz SSB filter plus the 1.8
+ I was experimenting around listening to a weak station that was rather
+ close to a real strong station and when I activated the 1.8 Khz filter I
+ was still getting some carryover from the strong station on the weak
+ station. Are there other controls that I need to use that would enable
+ me to knock out the real strong station so that I could clearly hear the
+ weak station? I'm sure there is help in the manual but I certainly can't
--
David Woolley
"we do not overly restrict the subject matter on the list, and we
encourage postings on a wide range of amateur radio related topics"
List Guidelines <http://www.elecraft.com/elecraft_list_guidelines.htm>
Ron D'Eau Claire
2010-10-28 23:19:38 UTC
Permalink
That's why the Elecraft doesn't call the first I.F. crystal filters "roofing
filters". Unlike some receivers, the first I.F. crystal filters in the K3
are there for selectivity to isolate the desired signal. That is further
enhanced in the second conversion with the variable DSP filter, but the
crystal first I.F. filters are a key part of the selectivity equation.

I suspect in this situation it is a case of the signals actually overlapping
(assuming he's listening to SSB). As you know, an SSB signal occupies a
swath of the spectrum - typically about 2.7 kHz but may be more if it's
someone running ESSB or someone overdriving their rig.

There's no effective way to separate the two signals completely if they
occupy the same bit of spectrum. Narrowing the selectivity down to exactly
what the desired signal needs will help by reducing any signal outside the
essential pass band but the overlapping energy is there.

There's one other obvious possibility. If the interfering station is very,
very strong, it will still pass through somewhat. The Elecraft filters are
very good, but no filter is really a "brick wall". Any filter only
attenuates out of pass band signals. So if an signal outside the pass band
is strong enough, what remains may still be strong enough to compete with a
weak signal inside the pass band. But, in the Elecraft design, that takes a
huge signal. I suspect it's a matter of two signals occupying some of the
same spectrum.

Ron AC7AC



-----Original Message-----
The purpose of the roofing filter is to reduce receiver generated
distortion products, not to provide selectivity. Are you sure that that:

1) the strong signals are completely out of band;
2) the only distortion products are those generated by the K3 itself,
and not the transmitter.
Post by pastormg2
I purchased a K3 with the standard 2.7 Khz SSB filter plus the 1.8
+ I was experimenting around listening to a weak station that was rather
+ close to a real strong station and when I activated the 1.8 Khz filter I
+ was still getting some carryover from the strong station on the weak
+ station. Are there other controls that I need to use that would enable
+ me to knock out the real strong station so that I could clearly hear the
+ weak station? I'm sure there is help in the manual but I certainly can't
--
David Woolley
"we do not overly restrict the subject matter on the list, and we
encourage postings on a wide range of amateur radio related topics"
List Guidelines <http://www.elecraft.com/elecraft_list_guidelines.htm>
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:Elecraft at mailman.qth.net

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Wes Stewart
2010-10-28 23:37:44 UTC
Permalink
Unfortunately, they do.? Personally I dislike the term.

For example:

A "A narrow roofing filter -- compatible with the communications bandwidth

required -- will protect "downstream" stages. The K3's

shift/width/hicut/locut controls automatically select and properly

position the optimal roofing filter. You'll see the FL1-FL5 icons

selected as you rotate these controls."


http://elecraft.365791.n2.nabble.com/Re-roofing-filters-when-to-use-them-td446665.html#a446666

--- On Thu, 10/28/10, Ron D'Eau Claire <ron at cobi.biz> wrote:

That's why the Elecraft doesn't call the first I.F. crystal filters "roofing
filters". Unlike some receivers, the first I.F. crystal filters in the K3
are there for selectivity to isolate the desired signal. That is further
enhanced in the second conversion with the variable DSP filter, but the
crystal first I.F. filters are a key part of the selectivity equation.
Ron D'Eau Claire
2010-10-29 00:03:36 UTC
Permalink
Good point Wes, but clearly Wayne used it in that context answering
someone's query who called the first I.F. filter by that name. He's picked
up that name (in parenthesis) in the latest Owner's manual as well in the
interest of communicating with people who think "roofing" simply means the
first I.F. filter.



But you can't order a "roofing filter" for the K3 from the Elecraft page
under that name :-)



The term "Roofing" isn't wrong in the way the K3 uses filters, but it's
misleading to those who have worked with receiver architectures that
up-convert to a low-VHF frequency, pass the signal through a "roofing
filter" whose purpose is to block some mixing products. Adequate filters for
taking final selectivity simply aren't available for the high first I.F.
those receivers use so the final selectivity must be taken in subsequent
stages at a lower second or third I.F. It was in these receivers that the
term "roofing" first appeared to designate that wide, first I.F. filter.



While that architecture has some advantages in a receiver that covers a very
wide bit of the RF spectrum, it also has significant disadvantages that
Wayne chose to avoid in the K3.



So the K3's first I.F. filter is at a pretty typical low I.F. for these days
-- just above 8 MHz. At that frequency filters capable of isolating the
desired signal can be made, and that's why Elecraft offers a selection of
crystal filters with bandwidths to suit the different modes and preferences.




In the K3, the DSP filter in the second I.F. is usually a secondary
function. The main selectivity to eliminate unwanted signals is taken at the
first I.F. crystal filter.



Ron AC7AC










Unfortunately, they do. Personally I dislike the term.

For example:

A "A narrow roofing filter -- compatible with the communications bandwidth
required -- will protect "downstream" stages. The K3's
shift/width/hicut/locut controls automatically select and properly
position the optimal roofing filter. You'll see the FL1-FL5 icons
selected as you rotate these controls."

http://elecraft.365791.n2.nabble.com/Re-roofing-filters-when-to-use-them-td4
46665.html#a446666

--- On Thu, 10/28/10, Ron D'Eau Claire <ron at cobi.biz> wrote:



That's why the Elecraft doesn't call the first I.F. crystal filters "roofing
filters". Unlike some receivers, the first I.F. crystal filters in the K3
are there for selectivity to isolate the desired signal. That is further
enhanced in the second conversion with the variable DSP filter, but the
crystal first I.F. filters are a key part of the selectivity equation.
Bob Cunnings
2010-10-29 00:58:07 UTC
Permalink
Yes, they do. Elecraft presents it's position on the term "Roofing Filter" here:

http://www.elecraft.com/K3/Roofing_Filters.htm

where the "protective" function is emphasized.

Bob NW8L
Post by Wes Stewart
Unfortunately, they do.? Personally I dislike the term.
Ron D'Eau Claire
2010-10-29 02:04:47 UTC
Permalink
That paper explains why Elecraft has begun to call it "roofing" when it is
not what most designers think of when they speak of a "roofing filter".

Call it what you want, the primary function of the first I.F. crystal filter
in the K3 is selectivity.

The term "roofing filter" came into use when receiver designers started
up-converting to a first I.F. in the VHF range. There are no practical
narrow filters at those frequencies but it was necessary to filter out
mixing products far from the desired frequency, so very wide (several kHz
wide) filters were used to define the general range to be further processed.


That filter set the "roof" on the frequency range of interest as David noted
earlier in his post.

The VHF I.F. was then down-converted to a lower frequency were selectivity
was taken with narrow filters.

The K3 does not use this scheme but early on many people who did not
understand the origin of "roofing filter" were confused, so Wayne has
written that paper and added "roofing filter" in parenthesis in places in
the Owner's manual to help those people understand.

It appears the meaning of "roofing filter" is becoming lost through careless
use. It won't be the first time - or the last time - that happens.

Ron AC7AC

-----Original Message-----

Yes, they do. Elecraft presents it's position on the term "Roofing Filter"
here:

http://www.elecraft.com/K3/Roofing_Filters.htm

where the "protective" function is emphasized.

Bob NW8L
Post by Wes Stewart
Unfortunately, they do.? Personally I dislike the term.
Geoffrey Mackenzie-Kennedy
2010-10-29 10:37:47 UTC
Permalink
With respect Ron, as far as I know the term "Roofing Filter" was in use many
years before up-conversion superhet receivers with a fist IF at VHF started
to appear in the market. It was at least fifty years ago when I first ran
across the term, which was used to describe the first IF filter used in a
voice/data multichannel Independent Sideband down conversion superhet
receiver manufactured by our company at the time. In this particular case
the bandwidth of this Roofing Filter was wide enough to allow both the upper
and lower sidebands of the incoming signal to pass, and further downstream
in the IF two filters described as USB and LSB IF filters were used to
separate the signal's sidebands for further processing.

It could be argued that the input bandpass filters of a receiver act as
Roofing Filters, and that all filters which follow should be described as IF
filters, Audio filters or whatever. However the applicable "rules" of
terminology as I have understood them since those ancient times restricts
the use of the term Roofing Filter to the first IF Filter, but the term
should only be used if a second and narrower *IF* filter follows the first
( which would include DSP filters, but only if working at IF not audio).

The use of relatively wide bandwidth Roofing Filters at VHF in up-conversion
receivers is actually not such a problem that it might appear to be, because
the use of a very narrow filter behind the mixer can be counter productive
in terms of close in dynamic range, especially if the filter is followed by
a well designed decent IF. The same is true of down-conversion. LO phase
noise is a problem when up-converting to VHF.

73,

Geoff
GM4ESD
Post by Ron D'Eau Claire
The term "roofing filter" came into use when receiver designers started
up-converting to a first I.F. in the VHF range. There are no practical
narrow filters at those frequencies but it was necessary to filter out
mixing products far from the desired frequency, so very wide (several kHz
wide) filters were used to define the general range to be further processed.
Bill W4ZV
2010-10-29 11:45:35 UTC
Permalink
Post by Geoffrey Mackenzie-Kennedy
With respect Ron, as far as I know the term "Roofing Filter" was in use many
years before up-conversion superhet receivers with a fist IF at VHF started
to appear in the market.
Correct Geoff. The first up-conversion ham rigs were introduced in the
early 1980s (TS-930S I believe). The Japanese manufacturers got the idea
that everyone wanted general coverage in their receivers, which then
required up-conversion and an IF at VHF. Rob Sherwood had been marketing
his 600 Hz "roofing filter" mod for the down-conversion Drake R4C several
years before that, and as you said the term had already been in use many
years prior.

73, Bill
--
View this message in context: http://elecraft.365791.n2.nabble.com/Roofing-Filter-Usage-tp5682095p5686154.html
Sent from the [K3] mailing list archive at Nabble.com.
Bob W7AVK
2010-10-29 12:57:32 UTC
Permalink
Bill and group - Can't say it was first, but has to be the first US
designed and produced ham rigs to use up conversion. The DRAKE TR-7
has no RF Amplifier and uses a diode ring "up converter" to 43 mhz as
the first IF. The design was in the late 60s, early 70s. Bob Drake and
his company have always been know for innovative design. In those days
the term "Roofing Filter" hadn't be coined by the marketing types and
the TR-7 has only one fixed filter at the high IF. BTW - The TR-7
second IF with the switchable mode filters is at 8.2 MHz.

73 Bob W7AVK
Post by Bill W4ZV
Post by Geoffrey Mackenzie-Kennedy
With respect Ron, as far as I know the term "Roofing Filter" was in use many
years before up-conversion superhet receivers with a fist IF at VHF started
to appear in the market.
Correct Geoff. The first up-conversion ham rigs were introduced in the
early 1980s (TS-930S I believe). The Japanese manufacturers got the idea
that everyone wanted general coverage in their receivers, which then
required up-conversion and an IF at VHF. Rob Sherwood had been marketing
his 600 Hz "roofing filter" mod for the down-conversion Drake R4C several
years before that, and as you said the term had already been in use many
years prior.
73, Bill
Jack Smith
2010-10-29 14:57:45 UTC
Permalink
Racal's RA-17 was an up-converting receiver, with a first IF at 40 MHz.
However, the 40 MHz IF filter was wideband, with the real selectivity
applied at 1.6 MHz and 100 KHz. With the Wadley Loop tuning system, it
was an extremely advanced design for 1955, all done with vacuum tubes,
of course.

Jack K8ZOA
Post by Bob W7AVK
Bill and group - Can't say it was first, but has to be the first US
designed and produced ham rigs to use up conversion. The DRAKE TR-7
has no RF Amplifier and uses a diode ring "up converter" to 43 mhz as
the first IF. The design was in the late 60s, early 70s. Bob Drake and
his company have always been know for innovative design. In those days
the term "Roofing Filter" hadn't be coined by the marketing types and
the TR-7 has only one fixed filter at the high IF. BTW - The TR-7
second IF with the switchable mode filters is at 8.2 MHz.
73 Bob W7AVK
Post by Bill W4ZV
Post by Geoffrey Mackenzie-Kennedy
With respect Ron, as far as I know the term "Roofing Filter" was in use many
years before up-conversion superhet receivers with a fist IF at VHF started
to appear in the market.
Correct Geoff. The first up-conversion ham rigs were introduced in the
early 1980s (TS-930S I believe). The Japanese manufacturers got the idea
that everyone wanted general coverage in their receivers, which then
required up-conversion and an IF at VHF. Rob Sherwood had been marketing
his 600 Hz "roofing filter" mod for the down-conversion Drake R4C several
years before that, and as you said the term had already been in use many
years prior.
73, Bill
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:Elecraft at mailman.qth.net
This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Ron D'Eau Claire
2010-10-29 17:16:24 UTC
Permalink
Thanks for the clarification Geoff.

The fellow with the original question asked what why he was still hearing
QRM even when using the 1.8 kHz crystal filter (presumably listening to SSB)
and David responded, "The purpose of the roofing filter is...not to provide
selectivity."

I was respectfully disagreeing with David when speaking of the K3. In the K3
the first I.F. filter (what some call the "roofing filter") provides the
main selectivity. That's why a variety of bandwidths are provided for
various modes and why so much attention was paid to the design of those
filters to provide excellent stop band attenuation. The DSP filter at the
2nd I.F. provides additional filtering but it's the first I.F. filter in the
K3 that is responsible for providing the bulk of the selectivity.

Personally, I don't use the term "roofing filter" for such a filter. Rather
when I see "roofing filter" I think of a filter such as David described that
cleans up unwanted mixer products so that the final selectivity can be taken
later in the signal path. That's how I've always seen "roofing filter" used
in the commercial equipment I've worked with.

IIRC the term roofing filter didn't come up with respect to the K3 until
after it was introduced and Hams were trying to compare its architecture
with Ham rigs that up-convert to VHF for the first I.F. Wayne answered many
questions here on the reflector and finally wrote the paper on the web site
explaining it further. He has added "roofing filter" in parenthesis in
places in the Owner's manual to help those Hams understand the K3 Crystal
Filter selected is the first I.F. filter. However, it's clear that applying
the name to the K3 skews the definition of roofing filter from what Dave and
I have always known in the past.

Ron AC7AC



-----Original Message-----

With respect Ron, as far as I know the term "Roofing Filter" was in use many

years before up-conversion superhet receivers with a fist IF at VHF started
to appear in the market. It was at least fifty years ago when I first ran
across the term, which was used to describe the first IF filter used in a
voice/data multichannel Independent Sideband down conversion superhet
receiver manufactured by our company at the time. In this particular case
the bandwidth of this Roofing Filter was wide enough to allow both the upper

and lower sidebands of the incoming signal to pass, and further downstream
in the IF two filters described as USB and LSB IF filters were used to
separate the signal's sidebands for further processing.

It could be argued that the input bandpass filters of a receiver act as
Roofing Filters, and that all filters which follow should be described as IF

filters, Audio filters or whatever. However the applicable "rules" of
terminology as I have understood them since those ancient times restricts
the use of the term Roofing Filter to the first IF Filter, but the term
should only be used if a second and narrower *IF* filter follows the first
( which would include DSP filters, but only if working at IF not audio).

The use of relatively wide bandwidth Roofing Filters at VHF in up-conversion

receivers is actually not such a problem that it might appear to be, because

the use of a very narrow filter behind the mixer can be counter productive
in terms of close in dynamic range, especially if the filter is followed by
a well designed decent IF. The same is true of down-conversion. LO phase
noise is a problem when up-converting to VHF.

73,

Geoff
GM4ESD

Continue reading on narkive:
Loading...