Cheap 455kHz SSB filters?
Russ.Shaw
I'll make an 80m SSB transceiver, and possibly make it a kit,
after i get a source of parts that don't cost too much.
Is there a source of 2-3kHz BW 455kHz SSB filters that doesn't
cost a lot?
Apart from Amidon, are there any other sources of readily
available toroids/IF-transformers?
I'm aiming for 1-10W output, 455kHz single conversion, and
battery-portable.
--
*******************************************
* Russell Shaw, B.Eng, M.Eng(Research) *
* Electronics Consultant *
* email: rus...@webaxs.net *
* Australia *
*******************************************
Michael Black
<rus...@webaxs.net> wrote:
> Hi all,
>
> I'll make an 80m SSB transceiver, and possibly make it a kit,
> after i get a source of parts that don't cost too much.
>
> Is there a source of 2-3kHz BW 455kHz SSB filters that doesn't
> cost a lot?
>
> Apart from Amidon, are there any other sources of readily
> available toroids/IF-transformers?
>
> I'm aiming for 1-10W output, 455kHz single conversion, and
> battery-portable.
>
It strikes me that if there were suitable inexpensive 455KHz filters,
we would have heard about them long ago.
At 455KHz, most construction articles used either Collins mechanical
filters, which of course tend to be expensive, or home made crystal
lattice filters made with surplus crystals, preferably in FT-243
cases so they could be ground to the needed frequency.
The obvious "cheap" filter would be a ceramic filter. But in reading
ham magazines from the past forty years, I've only seen very vague
references about using them for transmitting purposes. The articles
that come to mind were more idea articles, to hint at something or
to point to such a filter and suggest some uses for them. But
hard information wasn't included so it's hard to tell whether there
was merit behind the suggestion, and nobody went and took the idea
further into a construction article. The filters discussed may
not have been really great for transmitting purposes, or perhaps
the only reason they were considered "inexpensive" was in comparison
with Collins mechanical filters.
Of course, some of the Japanese rigs seem to use ceramic filters at
455KHz for transmitting, but I've never seen prices for the filters
alone, so they may not be inexpensive.
Even commonly available ceramic filters, such as the type you find
in a lot of AM broadcast radios, don't seem to have been given much
experimental treatment. Those would probably be too wide, but I wonder
if you could get a sharp enough skirt for SSB if you cascaded some
filters. You can get narrower ceramic filters, I've seen them in
CB sets and maybe http://www.digikey.com carries them along with
wider ceramic filters. Again, the question is whether cascading
will give you a decent filter overall.
I've not seen the actual article, but I have seen mention of an
piece about cascading ceramic filters. From a previous post of
mine:
"In an old article in 73 (December 1977) about the Yaesu
FRG-7, reference is made to an "by Ron Risher VK3OM in the March 1977
issue of "Amateur Radio". He describes... an alternate filter, consisting
of four cascaded SFD-455-B solid state filters linked by small coupling
capacitors". It might be interesting to someone who can get ahold of the
magazine, which I assume is Australian."
It sounds like an idea to look into, though I have no idea of the
specs of that filter.
Some people have speculated, and done some testing, in
rec.radio.amateur.homebrew about using 455KHz ceramic resonators,
which are cheap and fairly readily available, in ladder filters.
Doing a search at http://www.dejanews.com will turn up some old
posts about the idea.
Of course, once we are talking about ladder filters, one could
abondon the 455KHz IF. A lot of work has been done on ladder filters
in homebrew equipment, and the interest has arisen because the crystals
are cheap, and you only need them on one frequency (unlike lattice
filters), and surely because inexpensive commerically made filters
are non-existent. A lot can be said for putting the IF somewhere
in the HF range. The crystals may be even easier fo find than
a ceramic resonator at 455KHz, and you won't have to worry about
image rejection.
QST has published a number of articles about ladder filter design,
and they've been reprinted in various ARRL books. And I've seen
some information about them in RSGB publications.
Michael VE2BVW
Dale Parfitt
Russ.Shaw wrote:
> Hi all,
>
> I'll make an 80m SSB transceiver, and possibly make it a kit,
> after i get a source of parts that don't cost too much.
>
> Is there a source of 2-3kHz BW 455kHz SSB filters that doesn't
> cost a lot?
>
> Apart from Amidon, are there any other sources of readily
> available toroids/IF-transformers?
>
> I'm aiming for 1-10W output, 455kHz single conversion, and
> battery-portable.
>
Why not pick a higher IF frequency? This would allow for a lower freq
VFO, and better image rejection. Then you could use inexpensive color
burst or computer xtals for the filter.
Dale W4OP
NANCY GIANAKOPOULOS
Murata. The cheapest filter is the CFWS-455K, a 6 element ladder filter for
about 4 dollars. A better choice would be the CFS-455K, an 11 element
ladder filter for about 22 dollars. The problem I have had over the years
is buying them in small quantities. The only way I was able to procure them
was by ordering samples from Stan Clothier Associates, one of Murata's reps.
Murata didn't even want to consider small orders; I know we used to use
quite a few of them when I worked at Rockwell, but for critical
applications, we used the Rockwell-Collins torsional mode mechanical
filters, at about 90 dollars apiece in small quantities.
One final note; Murata assigns a letter code, denoting the bandwidth of
their filters. Example:
a CFWS-455E would be an NBFM bandwidth, while a CFWS-455H would be a 6KHz
bandwidth for AM, a CFWS-455I would be a 4KHz bandwidth, and a CFWS-455K
would be a 2.4KHz bandwidth for SSB. The CFWS series of filters are
typically the ones that you see used in the 455KHz IF strip in most of your
commercially available radios. The SSB filter in many radios such as the
FRG-100, the AOR-7030, and others use something similar to the CFS series.
Normally, I don't usually make a whole group posting, but I have seen this
question about filters come up several times on the newsgroups.
One more request I have seen is about product detectors for some of the
boatanchors; if anybody is interested, I would more than happy to send a
schematic of the circuit I have been using in my homebrew designs. Also, if
anybody needs schematics of front-end filters for the 0 to 30MHz range, I
can supply schematics for filters that will break this range up into 7
segments. I also have a PIN diode switching configuration that does not
require a -V supply to reverse bias the diodes. Too bad I didn't think of
this one!
If anybody would like a copy of my latest 0-30MHz receiver design, which
uses a DDS for the LO, feel free to give me a shout, and I will be glad to
send you the schematics, free of charge.
Pete Gianakopoulos
Chicago, Il.
Michael Black <blac...@cam.org> wrote in message
news:blackm00-300...@usrh-5-119.hip.cam.org...
> In article <390BC27F...@webaxs.net>, "Russ.Shaw"
> <rus...@webaxs.net> wrote:
>
> > Hi all,
> >
> > I'll make an 80m SSB transceiver, and possibly make it a kit,
> > after i get a source of parts that don't cost too much.
> >
> > Is there a source of 2-3kHz BW 455kHz SSB filters that doesn't
> > cost a lot?
> >
> > Apart from Amidon, are there any other sources of readily
> > available toroids/IF-transformers?
> >
> > I'm aiming for 1-10W output, 455kHz single conversion, and
> > battery-portable.
> >
John
A easy to get source for 455 filters would be from Yaesu , Kenwood or
Icom. They all use murata filters in there 455 IF . Good radios to look at
would be , Kenwood 930 & 940. Yaesu Ft -102 , 757 (1000MP has a 2.7K wide
murata filter in it's sub receiver ) and many Icom's. You could buy them
from the companies as parts for those radio's.
73's John
Eamon Skelton
>
>I'll make an 80m SSB transceiver, and possibly make it a kit,
>after i get a source of parts that don't cost too much.
CFJ455K5 2.4KHz at -6dB, 4.5KHz at -80dB.
Details from the Cirkit catalogue for 1997.
73, Ed. EI9GQ.
ign...@num.ads.uga.edu
>Details from the Cirkit catalogue for 1997.
The filter has good specs but I saw a big difference after replacing it
with a crystal filter. I got my Kenwood crystal filter (new) for $60.
Ignacy, NO9E
Rick Haub
455kHz that has a BW of 4kHz. P# TK2330-ND, $2.65 quantity one.
Rick KC0GIX
Russ.Shaw wrote:
> Hi all,
>
> I'll make an 80m SSB transceiver, and possibly make it a kit,
> after i get a source of parts that don't cost too much.
>
> Is there a source of 2-3kHz BW 455kHz SSB filters that doesn't
> cost a lot?
>
> Apart from Amidon, are there any other sources of readily
> available toroids/IF-transformers?
>
> I'm aiming for 1-10W output, 455kHz single conversion, and
> battery-portable.
>
J M Noeding
<rh...@midwestmicro-tek.com> wrote:
>As Michael suggested, if you can get away with ceramics Digi-key has
>455kHz that has a BW of 4kHz. P# TK2330-ND, $2.65 quantity one.
>
>Rick KC0GIX
>
>
>Russ.Shaw wrote:
>
>> Hi all,
>>
>> I'll make an 80m SSB transceiver, and possibly make it a kit,
>> after i get a source of parts that don't cost too much.
>>
>> Is there a source of 2-3kHz BW 455kHz SSB filters that doesn't
>> cost a lot?
>>
by the way, have you considered making a ladder filter using the cheap
color TV burst xtals - available on 3.58 or 4.43361875MHz?
73
Jan-Martin
LA8AK
Russ.Shaw
doing a webpage for receiver designs.
NANCY GIANAKOPOULOS wrote:
>
> If anybody would like a copy of my latest 0-30MHz receiver design, which
> uses a DDS for the LO, feel free to give me a shout, and I will be glad to
> send you the schematics, free of charge.
>
> Pete Gianakopoulos
> Chicago, Il.
>
--
Russ.Shaw
Does anyone know what the unloaded Q and frequency tolerance is like?
I figured that if i made a 2.5kHz BW, 1dB ripple cauer-tchebyshev
filter with 3 poles and two zeros using isolated resonators, the
highest unloaded Q i need is 1200. To make the same filter as a
passive ladder network, i'd probably need a Qu of 5000-10000. The
other problem is that the tolerance of resonator resonant frequency
is quite small.
I've thought of making my own resonator units using a glass marble
with ferrite slugs glued to opposite 'sides', coupled with dc-biased
coupling coils. Three separate resonator units would be lightly coupled
using small shunt inductances. However, its really too fiddly to use for
kit constructors. Also, difficult frequency trimming would be needed.
Michael Black wrote:
>
> Some people have speculated, and done some testing, in
> rec.radio.amateur.homebrew about using 455KHz ceramic resonators,
> which are cheap and fairly readily available, in ladder filters.
> Doing a search at http://www.dejanews.com will turn up some old
> posts about the idea.
>
--
Russ.Shaw
and more gain with fewer stages can be got at lower frequencies. I
can get 35-40dB gain from a BC548 transistor at 1mA using transformers,
and they're easier to get and have a nicer agc characteristic than ics
i know of.
Dale Parfitt wrote:
>
> Why not pick a higher IF frequency? This would allow for a lower freq
> VFO, and better image rejection. Then you could use inexpensive color
> burst or computer xtals for the filter.
>
> Dale W4OP
--
Russ.Shaw
Eamon Skelton wrote:
>
> >Hi all,
> >
> >I'll make an 80m SSB transceiver, and possibly make it a kit,
> >after i get a source of parts that don't cost too much.
>
> CFJ455K5 2.4KHz at -6dB, 4.5KHz at -80dB.
>
> Details from the Cirkit catalogue for 1997.
>
> 73, Ed. EI9GQ.
Michael Black
rh...@midwestmicro-tek.com wrote:
> As Michael suggested, if you can get away with ceramics Digi-key has
> 455kHz that has a BW of 4kHz. P# TK2330-ND, $2.65 quantity one.
>
> Rick KC0GIX
>
>
Of course, the slope may be more immediately important than the actual
bandwidth.
If it was just a transmitter, one could use a relatively wide bandwidth
filter, indeed SSB isn't inherently narrow band, and keep the bandwidth
low by good audio filtering. Just as long as the filter is good enough
to chop off the unwanted sideband.
But since the filter is wanted for a transceiver, reception would be
affected by the relatively wide bandwidth. It probably wouldn't be
a consideration for limited use on the VHF bands, to get more SSB
activity going, but on HF it won't be so great. Again, one could
add good audio filtering on receive, but that will mess up the AGC
as in-channel signals affect the gain. Transmitting is a much more
controlled environment.
If you look in old articles about SSB rigs, there was a time when
LC filters were common for filtering. Using the IF coils out of
a Command Set (the one with the 85KHz IF) or some other pre-wound
coils, you could get good enough rejection if the IF was low enough.
That leads to problems with image of course.
I wonder how do-able it would be nowadays with toroids or pot-cores,
(or with terribly improved op-amps, how about an active filter around
50KHz or so?) and a conversion up to a frequency where a cheap and readily
available crystal filter would sit. I can put my hands on all
kinds of crystal filters in the HF range, but NBFM wide; all they
have to do is be narrow enough to get rid of the image from the
really low IF.
Michael VE2BVW
Michael Black
<hy...@fx2.com> wrote:
This isn't a new idea. Years ago, you'd see articles where the author
suggested buying a replacement filter for a homebrew project, or any part
for that matter (I recall someone buying all the parts for an HW-101 VFO
for
their own use).
But about 1974 or '75, when I needed an SSB filter for something, I called
up the local Heathkit outlet to find the price of a crystal filter used
in their SB line. It was fairly expensive for the time, and no bargain
over buying a filter intended for construction, like a KVG.
And then you look at the price of add-on filters from the company
selling the rig, and either they are giving it a tremendous markup, or
we have to expect to pay an awful lot for good filters.
On the other hand, since some people do replace the original filter with
a better one, one source of filters might be finding people who have
upgraded, though that is hardly a source of a lot of filters.
It's a shame that one of the smaller companies selling relatively low cost
equipment or kits doesn't offer SSB (and even CW) filters on the side
to the homebrew market. Since they are buying in quantity, presumably
the price would be low. Sales to the homebrew market would hardly cut
into sales of their complete rigs, but it would be a source of filters
for those who want to build their own.
Or talk a company already selling parts, such as Dan's Small Parts, into
carrying some sort of relatively low cost filter.
Michael VE2BVW
Russ.Shaw
ceramic resonators with suitably high Q and frequency tolerance.
The main reason for a 455kHz if is so that 'standard' 455kHz transformers
can be used, and you can get more gain with less stages. I'm making the
receiver compact/battery-portable, but without compromises.
Also, it'll have 10Hz step synthesized vfo, but not using a
power-hungry/noisy dds.
J M Noeding wrote:
>
> by the way, have you considered making a ladder filter using the cheap
> color TV burst xtals - available on 3.58 or 4.43361875MHz?
>
> 73
> Jan-Martin
> LA8AK
--
ms...@nmia.com
I've tried suffer from poor skirt rejection. Fooling with
termination doesn't seem to help much.
A cheap solution is using the small ones in series.
Example. Using the simple 5 pin 2 element SFD455 in series
makes a pretty good filter if the element coupling caps are
around 10pF. Two in series is better than some of the multi
element filters and four in series gets quite good.
More gets proportionally better.
Since this is a multiplication process, the passband response
[0db] stays the same but the skirt selectivity improves remarkably.
If a skirt frequency is down 10db for one then it will be down
20db for two and 40db down for four in series.
Downside is insertion loss which also multiplies but this may
be compensated with a simple bipolar gain stage at the end
to recover signal level.
73 - John W7ZFB
1400 Catron SE Albq, NM 87123
***********************************************************
* Homebrewer since 1947 CW BoatAnchors Norcal #930 *
* BA CW freqs - 3578, 7050, 7147, 14050, 21050, 28050 *
***********************************************************
Dave
going to 10.7 mhz with a ceramic filter for image rejection and then
down to 455 for the final sideband filtering?
* Sent from RemarQ http://www.remarq.com The Internet's Discussion Network *
The fastest and easiest way to search and participate in Usenet - Free!
Russ.Shaw
Unfortunately, filters for 10.7MHz are usually wider band, so strong
adjacent signals are let thru to the second mixer. So to get good
two-tone dynamic range, two high-level mixers are needed.
With a 455kHz single IF, i only need one ring mixer and less parts.
--
J M Noeding
<dlawryN...@norfolk-county.com.invalid> wrote:
>There are the tradeoffs. Higher If and image rejection. How about
>going to 10.7 mhz with a ceramic filter for image rejection and then
>down to 455 for the final sideband filtering?
>
>
you probably need two filter stages and a 10.7MHz IF amplifier or two
for sufficient ultimate rejection
J M Noeding
wrote:
>I wanted to make the if strip using standard 455kHz if-transformers,
>and more gain with fewer stages can be got at lower frequencies. I
>can get 35-40dB gain from a BC548 transistor at 1mA using transformers,
>and they're easier to get and have a nicer agc characteristic than ics
>i know of.
>
BF199 is better for tuned stages considering stability
NANCY GIANAKOPOULOS
transformers at the IF input, and provide around 80dB of gain. The
demodulated audio is so clean, that when I was demonstrating one of my
radios to a fellow the other day, he thought I was using synchronous
detection!
Pete Gianakopouloos
J M Noeding <LA...@online.no> wrote in message
news:391d0fdf....@news.online.no...
Jim Weir
shared these priceless pearls of wisdom:
->The TDA1072 or the TDA1572 have very good AGC characteristics, require no
->transformers at the IF input, and provide around 80dB of gain. The
->demodulated audio is so clean, that when I was demonstrating one of my
->radios to a fellow the other day, he thought I was using synchronous
->detection!
Anybody got a handle on a good clean device that will do this up at 70 MHz.?
Most of these devices fall off VERY fast above 500 kHz.
Jim
Jim Weir, VP Eng. RST Eng. WX6RST
A&P, CFI, and other good alphabet soup
NANCY GIANAKOPOULOS
chip is a good bet. Be advised, though, that it requires a fair amount of
effort to get the '607 balanced for AM demodulation. I fiddled around with
this combination about a year ago, and went to the TDA1072/1572 series of
chips; I operated these devices at 455KHz though, so I didn't run into the
gain-bandwidth issues.
Pete Gianakopoulos
Jim Weir <j...@rst-engr.com> wrote in message
news:9b7uhsk8vs2ak6dlt...@4ax.com...
Jim Weir
shared these priceless pearls of wisdom:
->The Analog Devices AD600, followed by the AD607 quadrature IF/demodulator
->chip is a good bet.
Nope. The 600 GBW is 3 dB down at 30 MHz. I don't know what it is like at 70,
but I tend to shy away from operating a device above the knee. Mostly, they
turn unstable in this domain.
Russ.Shaw
BC548 works good.
J M Noeding wrote:
>
> On Mon, 01 May 2000 11:00:28 +1000, "Russ.Shaw" <rus...@webaxs.net>
> wrote:
>
> >I wanted to make the if strip using standard 455kHz if-transformers,
> >and more gain with fewer stages can be got at lower frequencies. I
> >can get 35-40dB gain from a BC548 transistor at 1mA using transformers,
> >and they're easier to get and have a nicer agc characteristic than ics
> >i know of.
> >
> BF199 is better for tuned stages considering stability
--
Russ.Shaw
NANCY GIANAKOPOULOS wrote:
>
> The TDA1072 or the TDA1572 have very good AGC characteristics, require no
> transformers at the IF input, and provide around 80dB of gain. The
> demodulated audio is so clean, that when I was demonstrating one of my
> radios to a fellow the other day, he thought I was using synchronous
> detection!
>
> Pete Gianakopouloos
>
> J M Noeding <LA...@online.no> wrote in message
> news:391d0fdf....@news.online.no...
Andrew Steven Dixon
transformer cans have one hell of a selection if used "backwards"
(tapped side as output, but not using tap) and driven from a high
impedance source (10-500K resistor) combined with a high impedance on
the output
(I used 33pf coupling to next transistor).
On Sun, 30 Apr 2000 06:27:21 GMT, blac...@cam.org (Michael Black)
wrote:
>In article <390BC27F...@webaxs.net>, "Russ.Shaw"
><rus...@webaxs.net> wrote:
>
>> Hi all,
>>
>> I'll make an 80m SSB transceiver, and possibly make it a kit,
>> after i get a source of parts that don't cost too much.
>>
>> Is there a source of 2-3kHz BW 455kHz SSB filters that doesn't
>> cost a lot?
>>
>> Apart from Amidon, are there any other sources of readily
>> available toroids/IF-transformers?
>>
>> I'm aiming for 1-10W output, 455kHz single conversion, and
>> battery-portable.
>>
>Some people have speculated, and done some testing, in
>rec.radio.amateur.homebrew about using 455KHz ceramic resonators,
>which are cheap and fairly readily available, in ladder filters.
>Doing a search at http://www.dejanews.com will turn up some old
>posts about the idea.
>
Russ.Shaw
black ones. The red is the autodyne converter, white is for intermediate
IF stages, and black drives the envelope detector. The white and black
ones had an unloaded Q= 30.
Andrew Steven Dixon wrote:
>
> I´m no HF expert, but I recently noticed that standard AM IF
> transformer cans have one hell of a selection if used "backwards"
> (tapped side as output, but not using tap) and driven from a high
> impedance source (10-500K resistor) combined with a high impedance on
> the output
> (I used 33pf coupling to next transistor).
>