Wow. Thanks Don!!
john-n4dsp
----- Original Message -----
From: "Don Wilhelm" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>; <[EMAIL PROTECTED]>; <elecraft@mailman.qth.net>
Sent: Tuesday, August 08, 2006 8:39 PM
Subject: RE: [Elecraft] Why no IF shift ?
Folks, (long philosophical response - delete if not interested in my
opinion).
Some very good points mentioned in this thread, but the reality of all of
it
is compromise.
For any given price target, some tradeoffs must be made. If I can
interpret
the K2 design goals loosely, the dynamic range was considered uppermost,
and
good sensitivity and IP3 performance running a close second. To achieve
that in a kit product, single conversion IMHO is the only realistic way to
go.
Yes, there ae very good designs out there for multi-conversion receivers
and
they have their costs and limitations, but the K2 is a compromise of all
that. It is single conversion because that is the way to contain the
dynamic range and good IP3 characteristics within the chosen price range.
It is well known that one must get into the $3K+++ transceivers to achieve
those receive parameters that K2/100 owners have achieved for less than
half
that price.
Certainly, IF shift is not an easy acomplishment with a single conversion
transceiver. Answering one question posed in the thread which asked why a
BFO frequency knob would not do the deed - the answer is YES, BUT - if
only
the BFO frequency is changed, the displayed frequency would no longer be
correct - the firmware currently corrects for both the BFO and VFO
frequencies and displays the correct carrier frequency, so while passband
tuning might be accomplished simply by changing the BFO, the VFO would
have
to be altered manually to compensate for the BFO shift, and the resulting
frequency on the dial would be incorrect. We used to do exactly that on
receivers with a variable BFO, but then we did not have digital dials that
were good to the nearest 10 Hz (but I digress). When CAL FIL is run, all
that is taken into consideration in the firmware, and the EEPROM values
contain the result of that calibration run which produces correct carrier
frequency dial readings - you have to give up one thing to gain another in
any design process unless 'price is no object'.
If the world were perfect, we would all have receivers with a 120+ dB
dynamic range, straight sided selectivity curves (with perfectly flat pass
band shapes and no change in group delay across the passband), MDS figures
in the -160 dB range and all that at a cost of less than $100 - of course
I
am dreaming based on today's technology and price/performance criteria
that
would be within the budget of the majority of hams. There is no sense in
developing a superior design for production that would sell only one unit
because of the cost involved - that is the stuff that extreme designs can
produce, but those are currently the stuff of advanced homebrewers - they
are the designs of tomorrow when the price of such designs comes down to
an
affordable level suitable for production.
73,
Don W3FPR
-----Original Message-----
From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] Behalf Of [EMAIL PROTECTED]
Sent: Tuesday, August 08, 2006 7:27 PM
To: [EMAIL PROTECTED]; elecraft@mailman.qth.net
Subject: Re: [Elecraft] Why no IF shift ?
In a message dated 8/8/06 6:41:46 AM Eastern Daylight Time,
[EMAIL PROTECTED] writes:
> One problem with single down conversion receivers is that their
> "stronger" spurious responses which, depending on the IF used,
can be close
> to or even in one or more of the frequency bands covered by the
receiver.
How?
In a single conversion superhet, the most important spurs are the
image and
IF feedthrough. In the case of a K2, the image is always about
9.830 MHz from
the desired signal, and the IF is around 4.915 MHz. The bandpass
filters take
care of those spurs very well, in my experience.
If
>
> bandpass filters are used between the antenna connector and the
mixer, their
>
> selectivity might offer some degree of protection against
signals coming in
> at spurious response frequencies outside but close to the bands
covered but
> obviously no protection against anything coming in at an
"in-band" spurious
> response frequency.
What in band spurs exist in the K2?
The choice of IF that reduces this problem in a single
>
> down conversion receiver is quite limited.
>
> In a double conversion receiver, up and then down, assuming
that sensible
> design and construction practices are followed, the close or in-band
> spurious responses (if they exist) are considerably reduced.
>
That may be the case in a general-coverage receiver, but in a
ham-bands-only
design, the spurious responses are easily handled by good input filters.
> Then there is the internal birdie problem, usually created by
one or more
> of
> the receiver's oscillators and /or their harmonics getting together to
> produce a signal at some spurious response frequency of the receiver.
Again, this is dependent on the design. For a ham-bands-only
receiver, the
birdies can be placed outside the ham bands.
If the
>
> Front End, LO and IF are not properly shielded in separate
"boxes" with all
> associated DC and control lines filtered, then expect birdies
in a single
> down conversion receiver. The same method of construction
should be used in
> a double conversion receiver. I suspect that commercial double
conversion
> amateur receivers have received a bad press because for reasons
of cost this
>
> is usually not done.
>
There are some very good up/down double conversion amateur receivers. But
they all suffer from the same problem: The signal has to go
through several
stages and conversions before it gets to the sharp filters. In a
single conversion
design like the K2, the number of stages and conversions between
antenna and
sharp filter is minimized.
> It is true that every conversion degrades the performance, but
for several
> years the technology has been available that allows a double conversion
> receiver to be built which exhibits an IIP3 of +40dbm
> at an offset of 2 kHz while running at full gain, Noise Figure
of 8db on
> 10m. I have one here. The downside is that each one of the
three VHF roofing
>
> filters / embedded amplifiers selected draws 240 mA.
>
There's also the issue of price....
> With double conversion, in addition to true "IF Shift" a form
of continuous
>
> bandwidth control can also be introduced.
>
_______________________________________________
Elecraft mailing list
Post to: Elecraft@mailman.qth.net
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
--
No virus found in this incoming message.
Checked by AVG Free Edition.
Version: 7.1.405 / Virus Database: 268.10.7/411 - Release Date: 8/7/2006
_______________________________________________
Elecraft mailing list
Post to: Elecraft@mailman.qth.net
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
