EMRFD Message Archive 2144

Message Date From Subject
2144 2008-09-28 20:44:32 timshoppa Critique my hypothetical multiband LO setup
You guys have seen my progress here the past few weeks... a nice 500Hz
Gaussian crystal filter at 4.43 MHz which I built in a PCB box and
measured and it agreed with XLAD/GPLA remarkably. Then I had a hybrid
cascode (deadbugged) in its own PCB box. Then I built a BFO with TUF-1
mixer as product detector and audio amp in its own PCB box (I had
originally wanted to put it in the same box as the hybrid cascode but
you'll remember my BFO leakthrough).

Right now I've got a simple WA6OTP-style VFO with a TUF-1 mixer and a
2N5109 post-mixer-amp that I use as the front end. A soldering iron to
switch out coils and caps is the way that I switch bands :-). The brass
slug tuning is pretty stable on 40M (VFO frequency 2.6MHz) but even in
a box the stability on 20M (VFO frequency 9.6MHz) leaves a lot to be
desired. It's not unusable... and indeed I've been very happy with the
whole lashup-of-PCB-boxes-connected-with-coax cables receiver in how it
works.

But I'd like to step up to a better way to switch bands. I've got a
small stash of reduction drives/dials and a wide range of variable caps
from the 50's and 60's, and know I can do something clever! But I'm
worried that I may be too clever for my own good so I'm looking for a
critique of my mixing scheme.

So here's the scheme I've come up with, working within the values of
commonly available microprocessor crystals, and with intended coverage
of the bottom 100kc (CW portion I hang out in) 40M, 30M, and 20M bands.
I haven't thought much about 80M yet... lotsa fun there in
wintertime... but I think some additional cleverness might get me there
as an afterthought:

F1: A VFO that tunes from 3.93 to 3.83 MHz. Build this in a nice stiff
metal box with one of those nice dials I've been hoarding for so long.

F2: Three microprocessor crystal oscillators, at 6.5, 9.6, and 13.5
MHz. These are not the most common microprocessor crystal
frequencies... but they are in stock at Digikey.

I take F2-F1 and end up with 2.57 to 2.67MHz on 40M, 5.67 to 5.77 on
30M, and 9.57 to 9.67 on 20M for the LO. Apply particular bandpassing
and it looks like I'm in business to handle 7.0-7.1, 10.1-10.2, and
14.0-14.1 bands.

80M will probably have to tune in reverse and with a semi-random dial
offset unless I order a custom crystal. But I could wait a couple weeks
for 80M to start getting interesting while they grind the crystal for
me :-)

So what are the gotchas? It's looking like I'll need multiple poles of
filtering after the F2-F1 mixer... am I going to be beset by birdies
everywhere?

Oh, speaking of birdies, I was asking about low-pass filtering for
BFO's and VFO's, and then went and listened and discovered that my
unfiltered BFO/VFO scheme gave quite a birdie at 14.043MHz. There the
LO was at 9.61MHz, and the 13th harmonic of the BFO was beating against
the 6th harmonic of the LO (evidently coming through the crystal
filter) to give that one. Add a little low-pass filtering and it was
gone! So now I'm a believer in a simple low-pass filter after each
oscillator :-).

Tim N3QE
2153 2008-09-30 17:15:19 timshoppa Re: Critique my hypothetical multiband LO setup
2154 2008-09-30 19:46:38 brainerd@wildblue... Re: Critique my hypothetical multiband LO setup
On 30 Sep 2008 at 20:21, timshoppa wrote:

> And a silly question, but one I'm wondering about: Most
> microprocessor crystals are obviously a power of 2 or related to
> baud rates or other nice round numbers. I happened to get a bunch of
> 4.433 MHz crystals for my IF frequency.. what is 4.433MHz used for
> in the wide world? Noodling around on my calculator it's not
> obviously related to powers of 2 of baud rates or anything. Some
> foreign TV (PAL, SECAM, ???) colorburst equivalent?
>
> Tim.
>
4.43 MHz is the PAL colorburst frequency.

Dave - WB6DHW
<http://wb6dhw.com>
2155 2008-09-30 20:26:52 John Kolb Re: Critique my hypothetical multiband LO setup
It's the color subcarrier freq for PAL.

http://www.epanorama.net/circuits/vga2tv/conversion.html

John

At 01:21 PM 9/30/2008, you wrote:


>And a silly question, but one I'm wondering about: Most
>microprocessor crystals are obviously a power of 2 or related to
>baud rates or other nice round numbers. I happened to get a bunch of
>4.433 MHz crystals for my IF frequency.. what is 4.433MHz used for
>in the wide world? Noodling around on my calculator it's not
>obviously related to powers of 2 of baud rates or anything. Some
>foreign TV (PAL, SECAM, ???) colorburst equivalent?
>
>Tim.



--
No virus found in this outgoing message.
Checked by AVG.
Version: 7.5.524 / Virus Database: 270.7.5/1700 - Release Date: 9/30/2008 11:03 AM
2157 2008-10-01 15:59:55 wimmie262000 Re: Critique my hypothetical multiband LO setup
Others already provided you with an answer for the 4.433MHz frequency.
In general I found this list to be helpful to identify the original
purpose of crystals that sometimes are offered in bulk.

http://en.wikipedia.org/wiki/Crystal_oscillator#Commonly_used_crystal_frequencies

Joop