10.7 Mhz IF and HF
Paul Rowe
What are the pros and cons compared to 455Khz ? Any response
appreciated. Thanks.
Peter Rachow
the higher you choose your IF, the fewer problems you will face with the so
called "image" frequency. This the second frequency that you're always
receiving unwantedly with a superhet. Due to the fact that your superhet
receives f1 + f2 *and* f1-f2 you will hear two signals.
Calculation sample:
fRX = 14.100 MHz
fIf = 455 kHz
required fOsz. = 14.555 Mhz
Image frequency = 14.100 MHz - 0.455 kHz = 13.645 MHz
So your superhet theoretically is able to receive 2 separate frequencies.
The higher your if is the more the unwanted signal will be filtered by your
input selective circuits (LC-filters) becaus the band distance doubles with
each kHz the if is higher.
Next Problem: A high if requires good IF-filters, because it is not so easy
to filter high frequencies with standard LC-circuits (as they are common for
455 kHZ). Therefore crystal filters are commonly preferred here. A double
superhet sometimes is a good solution but has a seriuos shortcoming: More
self generated intermodulation products that are falling into the received
band.
hth es 73!
Peter
DK7IH
--
http://www.peter-rachow.de
xpyttl
But better receivers today typically have 3 IFs ... usually the first is in
the 60 MHz range. I'm not so sure I know why that is. The second is
generally in the 10 MHz neighborhood, and the third is typically 455 or so.
Peter explained why a first IF of 455 is a problem. It's tough to get good
image rejection at that frequency. But there is another side to the problem
that makes the 10 MHz IF a problem. While crystal filters are possible, and
even necessary in many cases, for a 10 MHz IF, getting the right shape can
be a problem. This is especially a problem for SSB on the ham bands, where
you need a nice flat passband with steep skirts. It's much easier to make
this sort of filter for a low IF than for a high. So many receivers use a
10 MHz (or so) IF to get rid of signals that are a little way from the
desired signal, as well as images, then the 455 to whack out the close in
interference.
This is less of a problem for CW. On CW, you really don't need the flat
passband, a pointy-topped filter is pretty acceptable. This is easy to get
by stringing together a few matched crystals. So many receivers intended
for CW only have only a single IF, typically in the 10 MHz neighborhood,
with a multipole crystal filter that gets you both image and near signal
rejection.
It would seem like the availability of parts for 10.7 MHz IF's would make
that a popular frequency. However, what is available is often for FM
receivers that need a very wide passband, so selecting 10.7 specifically
really doesn't turn out to be a big advantage. As a result, frequencies
like 10, 9, and 4.9152 are a lot more popular for amateur band receivers.
..
"Paul Rowe" <pa...@paulrowe.com> wrote in message
news:bbd01c1e.02022...@posting.google.com...
John Miles
>
> Peter explained it pretty well, I think.
>
> But better receivers today typically have 3 IFs ... usually the first is in
> the 60 MHz range. I'm not so sure I know why that is.
Saves money on front-end filtering. With (say) a 75-105 MHz LO and 70
MHz 1st IF, a fixed low-pass filter is enough to handle the whole HF
spectrum without any image responses. No preselection required.
Of course, VHF local oscillators are almost inevitably noisier than HF
ones, so reciprocal mixing can be a problem. And without any
preselection at all, you need a much stouter mixer with more LO drive to
keep IMD under control. No free lunches in the radio business!
-- jm
------------------------------------------------------
http://www.qsl.net/ke5fx
Note: My E-mail address has been altered to avoid spam
------------------------------------------------------
Michael Black
At this point in time, I think you'd rather go with an IF in
the MHz range, unless there was a specific reason not to. Remember,
that 455KHz IFs date from when it was easier to go low in frequency
to get the required bandwidth, and when crystal filters in the HF range
came along, they gave the needed bandwidth. And you benefitted from
much better image rejection. If you read reviews of low end shortwave
receivers from the sixties, they'll often note very low image rejection
on the highest band. Going to an HF range IF means you don't have
to put a lot of effort into the front end in order to reduce the image
sufficiently.
Some people will say that a Collins mechanical filter which would be
at 455Khz or 500KHz, are a better choice than a crystal filter. But
at this point in time, you'd want to go with a double conversion scheme,
which is what's happening in commercial receivers that still use 455KHz
IFs. In those, the low IFs are often for some extra feature, like
passband tuning, where the added conversion is already needed.
(Note that most consumer AM broadcast band receivers and low end
Shortwave receivers still use 455KHz IFs, and one can assume that
it's due to reasons of economics. Certainly enough time has been put
into designing AM BCB receivers, so they know how to work around the
image problem, and 455KHz is such a common IF that ceramic filters
are plentiful for AM-bandwidth.)
One significant problem of an IF in the HF range is that if the receiver
is general coverage, you'll have a hole in the coverage. You're going
to have problems with reception right around the IF. But if you are
only covering some segments of the HF range, such as the hambands,
then an HF range IF gives you better image rejection, and an easier
design.
Nowadays, a lot of receivers have the first IF above the HF range, so
there isn't that hole in the tuning range. Of course, these tend to
be general coverage receivers. The disadvantage is that you need
a much higher local oscillator frequency for that first mixer,
and so there'll be complication there.
Unless you are just using the 10.7MHz IF as a first IF, and a second
conversion to some other frequency for ultimate selectivity, you might
want to reconsider the frequency. I don't think there is a wide selection
of filters on that center frequency. Most common, you'll find
FM-bandwidth filters, which would only be useful for FM. Certainly
you're more likely to find AM, SSB and CW bandwidth filters at
9Mhz than 10.7.
(And they are likely to be expensive, unless you can find surplus
filters. That is an advantage of 455KHz (or the similarly
used 450KHz); you can probably scrounge up a number of filters for
less than doing the same in the HF range. Certainly AM-bandwidth
filters are easy to come by, and I think you can find narrower filters
in CB sets. For a good CW or SSB filter, you could find a used mechanical
filter. So the choice of IF may depend on whether you can get by with
one filter, or need different bandwidths.)
A lot of people seem to be building their own ladder filters, which
seem to be easier to build than the old lattice filters from the early
days of SSB. Since they depend on a good number of cheap crystals,
the IF frequency is chosen because the crystals are available. For
a ham band receiver, the exact IF frequency won't matter so long
as it's not in a ham band, and you do some calculations to make sure
don't end up with some weird response when the IF and local oscillator
mix together.
Michael VE2BVW
Clyde R. Shappee
energy cannot be created nor destroyed.
Clyde
KA1CRV
P. S. Not an original statement.... Not sure of the source.
Pete Gianakopoulos
I use a doubly-balanced diode mixer, the spur at 10.7MHz is right around
.25uV. When an antenna is connected to the receiver, the spur is masked in
the antenna noise.
Now, what Mr. Black did say was correct, especially with some of the older
mixer designs. If this approach was attempted with some of the older
topologies, there would be a very high level spur at the I.F. That is one
big reason that the Hallicrafters SX122, and some of their other receivers
with that 1650KHz 1st I.F. omitted that range just above the MW band. The
Drake '4 series was another example of this omission, except in this case,
it was in the 5.465KHz range.
If you run a spur simulation, you do see the benefits from using a higher
I.F.; much fewer inband spurs. For anybody interested in seeing what I've
done with a successful implementation of a 10.7MHz 1st I.F., feel free to
contact me.
Pete KE9OA
Chicago, Il.
Michael Black <blac...@cam.org> wrote in message
news:6447bcd3.02022...@posting.google.com...
J M Noeding
<n.giana...@worldnet.att.net> wrote:
>I've been using a 10.7MHz 1st I.F. for some of my receivers, and as long as
>I use a doubly-balanced diode mixer, the spur at 10.7MHz is right around
>.25uV. When an antenna is connected to the receiver, the spur is masked in
>the antenna noise.
Interesting to read
>Now, what Mr. Black did say was correct, especially with some of the older
>mixer designs. If this approach was attempted with some of the older
>topologies, there would be a very high level spur at the I.F. That is one
>big reason that the Hallicrafters SX122, and some of their other receivers
>with that 1650KHz 1st I.F. omitted that range just above the MW band. The
>Drake '4 series was another example of this omission, except in this case,
>it was in the 5.465KHz range.
well, not quite. Drake 2-A .... 2-C have 455kHz and 50kHz IF for 80m
reception, and uses 80M as first IF for other bands. I have never
experienced spurs on 80m with any of my 2-B's. Using a 4645kHz xtal
filter (as for R-4) you do not need 455kHz in the chain to convert to
50kHz.
I don't see mentioned which coverage is intended, so I will still say
455kHz is an optimum IF - at least for frequencies below 4MHz, and the
solution used is usually sought considering the available sources of
components. In our case - having douzens of SSB/FM/CW/AM xtal filters
around for 5.2, 9 and 10.7MHz we might choose different, but still
application of 4.433, 4, 8 or 3.579MHz xtals in ladder type filters
might be interesting to try in some applications
At the moment I am experimenting with 60kHz xtal filter for 136kHz RX
73
Jan-Martin
--
remove ,xnd to reply
Pete Gianakopoulos
line as an example. I remember one of the old Allied catalogs talking about
the tuning range of the R4 receivers, stating that the unit would cover from
160 to 10 Meters, "save for the range from 5.5 to 6.0Mhz". Anyway, the
whole point of this is that manufacturers would generally avoid the I.F.
range on their receivers, except for some exceptions as the Collins units
that had the variable, and some others. Interesting about the Drake '2
series; I didn't realize that they used a variable I.F. tuning scheme. I've
had a few of those units over the years, wondering what the I.F. lineup was,
in order to get the triple-conversion tuning scheme. Those were good
receivers! Thanks for your input!
Pete KE9OA
Chicago, Il.
J M Noeding <la8a...@online.no> wrote in message
news:3c8019d8...@news.online.no...