"Netgeek" writes:

Len,

Thanks for the input and observations! Comments follow:


Agreed - not that interesting to listen to, but the NAV
channels are exactly what I'm interested in - (both VOR
and LOC). The "experiment" I have in mind is a processor
controlled scanner of sorts that tracks multiple VORs
and performs as a backup to GPS. Hence the need for
fast/agile tuning (either PLL or DDS). As you may have
gathered - I'm not an "RF guy" by any stretch - but I'm
attempting to learn (out of sheer necessity 8-).....

HAH! I did **EXACTLY** that project back in 1996 or so.

Used the 145151 PLL that someone else reccomended.

The RX was the ... crud .. can't remember the part number ... That
motorola 3300 part that was so popular. 3356 maybe?

Used a cheap 567 to track the FM part of the signal, and a simple
envelope detector for the AM part. It was barely adequate for
close-in stuff, and for long range stations, the noise performance was
terrible.

The outputs of the two detectors went into a two-channel 8-bit ADC and
were read by the CPU.

Used a 68HC000 for the CPU, and had a database of VOR stations burned
into the ROM.

The cleverest piece of code was the CW decoder that was used to ID the
stations.

It would tune around, listening to IDs and take a
best-guess at where it was, then build a list of stations to scan,
stopping on each one for a second to get a RMI fix. (There was a
little crude LPF to help with the noise performance. Not that it did
much good... )

I found some nice 4-line x 20 char LCD displays at a local junque
shop, with ENORMOUS characters. 0.6" I Think .. Just HUGE for LCDs
.... It would do a scrolling display of the form

SJC 025 nm @ 036
OAK 044 nm @ xxx

etc.

All in all it was a fun project -- with lots of cool learning
experiences. In terms of practicality, the performance was never
spectacular, RF noise from the CPU section would get into the IF
(21.4 MHz)... it made a neat cockpit toy, but I would **NEVER** have
considered it even a backup aid to navigation.

Just writing about this is making lots of little details come back to
me. Feel free to e-mail and I'll bend your ear.

-- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
Lawrence Statton - m s/aba/c/g
Computer software consists of only two components: ones and
zeros, in roughly equal proportions. All that is required is to
sort them into the correct order.

From: "Lawrence Statton N1GAK/XE2"

HAH! I did **EXACTLY** that project back in 1996 or so.
snip
Used a 68HC000 for the CPU, and had a database of VOR stations burned
into the ROM.
All in all it was a fun project -- with lots of cool learning
experiences.

Sounds like a cool project! *MANY* years ago (circa 1984 or so?)
I had a plan to tear off the front panel of an RST radio I built and
replace it with a display, some controls and a zillion parallel bits to
replace all the thumbwheels switches - all driven by a Z80. I
eventually decided it would take way too long and didn't
accomplish much, besides just being a generally stupid idea 8-)..
So I sold the radio...

Now, twenty years later I'm playing with *this* project! All of
which proves that I now have way too much time on my hands and
I'm getting more stupid as I age 8-)......

Would like to hear more about your experiences. So, what would
you do differently if you were starting over?

I'm thinking that something like a cheapo DSP (or one of the new
dsPIC widgets) might be real handy. But first - I have to get past
the part where I have a reasonable RF front end! I really doubt that
the $40 Ramsey kit is going to do much but I bet I'll learn a few
things 8-)........

I'm not locked in to the MC1350 - it just happens to be
what's used in the "ultra-cheap" kit I ordered. As it turns
out, there's a company (Lansdale.com) that seems to be
forming a business model around buying "obsolete" IP
from Motorola et.al. and keeping the parts available.
Interesting idea. As it happens, NTE has a drop-in
replacement part for the MC1350 in their "NTE746" -
available from Mouser. May be of interest to some on
this group???

The absolute minimum operating voltage for the MC1350 is 12 volts. Not
11.9, 12. If this is a mobile application, then figure on a switching power
supply to get you up to 15 volts or so.


Among that stuff is an HP8654A
good to about 520 MHz that I've never used (and for the
life of me can't figure why I bought 8-)........


Oh, I'll take it off your hands for $20 or so {;-)


Jim

"RST Engineering" wrote in message

The absolute minimum operating voltage for the MC1350 is 12 volts. Not
11.9, 12. If this is a mobile application, then figure on a switching
power
supply to get you up to 15 volts or so.

Thanks for that tip - it will be the first "improvement" to the cheapo
Ramsey receiver (which runs off 9 volts). The power supply is something
I *can* handle easily 8-)...

Among that stuff is an HP8654A
good to about 520 MHz that I've never used (and for the
life of me can't figure why I bought 8-)........

Oh, I'll take it off your hands for $20 or so {;-)
Jim

Well, give it a few weeks! Once the frustration level gets high
enough and I realize that I'm really a bit-banger and not a radio-head
I may be tempted to toss it (and anything else RF related) out the
window - in which case I'll pack it up and donate it to you 8-)...

In the meantime, any other suggested improvements or hints are
most appreciated (e.g. "up the IF from 10.7"?)...!

Thanks,
Bill (Who still thinks the universe *IS* digital - but infinite resolution
8-)

I'd probably jack the IF up to 45 MHz. (use TV parts; this is the TV IF
frequency) or 70 MHz. (satellite IF frequency), and then downconvert to
either 21.4 or 10.7...there are cheap crystal filters at both frequencies.

Jim


In the meantime, any other suggested improvements or hints are
most appreciated (e.g. "up the IF from 10.7"?)...!

Firstly, 108 to 118 MHz is the international civil
aviation radionavigation band. It's not all that
interesting to listen to unless a local tower is
also repeating voice comms over a VOR or Localizer
radionav transmitter nearby on the ground. The
civil aviation voice band is 118 to 137 MHz.

Firstly, the tower does NOT repeat voice comms over a VOR. The local Flight
Service Station MIGHT, but in the days of crystal controlled navcoms, the
amount of voice traffic on a VOR is next to nothing.

Secondly, the tower/FSS will never in HELL repeat something on a localizer
frequency.

You are correct; the civilian aviation voice band is 118.000 to 136.975 MHz.



Jameco sells the MC145151 PLL IC (On Semiconductor
the Motorola spin-off still makes them) which, with
a prescaler, can make a good, stable LO that is
channelized at 50 KHz increments for precise tuning.
MC145151 is parallel-load for division, no extra
IC needed to get the right division ratio as in
some serial-input PLL or DDS chips.

The 145151 is OK if you don't mind spurs every 25 kHz. from dc to daylight.
The 145152 is a much better dual-modulo prescaler that gets rid of a lot of
trash and garbage from single modulo prescaling that you probably don't
want.

Jim

From: "RST Engineering" Mon, Mar 14 2005 8:11
pm

Firstly, 108 to 118 MHz is the international civil
aviation radionavigation band. It's not all that
interesting to listen to unless a local tower is
also repeating voice comms over a VOR or Localizer
radionav transmitter nearby on the ground. The
civil aviation voice band is 118 to 137 MHz.

Firstly, the tower does NOT repeat voice comms over a VOR. The local
Flight
Service Station MIGHT, but in the days of crystal controlled navcoms,
the
amount of voice traffic on a VOR is next to nothing.

Now, now, Jim. They do.

My residence is a mile and a half from BUR, roughly
eight miles from VNY here in the San Fernando Valley
area of Los Angeles.

Agreed, tower operators SELDOM repeat their transmissions
over the VOR but it is there in case it is needed. I've
heard them often enough.

VOR has an almost enormous bandwidth between 30 Hz and
the 9.96 KHz subcarrier phase reference...which was
INTENDED to carry voice as a conventience to the tower.

BUR, now the Bob Hope Airport, USED to carry the taped
weather broadcasts over their VOR but stopped several
years ago.

Secondly, the tower/FSS will never in HELL repeat something on a
localizer
frequency.

Calm down. I may still vote for you next election,
but not if you act like Arnie... :-)

Hokay, I may have spoken hastily on the voice over
Localizer. Color me "probably wrong" there.

You are correct; the civilian aviation voice band is 118.000 to
136.975 MHz.

Thank you. Sigh, I was only in the business of making
civil avionics and their test sets once. :-)

Jameco sells the MC145151 PLL IC (On Semiconductor
the Motorola spin-off still makes them) which, with
a prescaler, can make a good, stable LO that is
channelized at 50 KHz increments for precise tuning.
MC145151 is parallel-load for division, no extra
IC needed to get the right division ratio as in
some serial-input PLL or DDS chips.

The 145151 is OK if you don't mind spurs every 25 kHz. from dc to
daylight.

Sorry, I don't agree there. "Spurs" with an ordinary
PLL happen when the loop filter component values are
incorrect...and/or a higher frequency pole is used
(via an extra R and C in loop filter)to reduce higher
frequency components out of the PFD.

I've made a few PLLs with that MC145151 for homebrew
projects and not had any spurs from "DC to daylight"
or within the band of interest.

The 145152 is a much better dual-modulo prescaler that gets rid of a
lot of
trash and garbage from single modulo prescaling that you probably
don't
want.

I've not tried the 145152 but, back a number of years
before On Semi split from Motorola Semi, a Motorola
factory person said the 152 is essentially the same as
the 151 except for the serial data interface. I can't
vouch for that but that's what I remember.

Single modulus prescaling (putting a simple divide by
8 or divide by 10 in series with the VCO and PLL IC
signal input) doesn't produce any more #$%^&!!! stuff
than going direct into the PLL IC signal input. That
is said PROVIDED the loop filter output line to the
VCO it is controlling is "clean" and doesn't pick up
other circuit signals. Such garbage pickup is the
common cause of "spurs" and is layout dependent, NOT
dependent on whether or not any prescaling is done.
There's some dependency on proper supply rail decoupling
for the phase-frequency detector and any active op-amps
used between the loop filter and VCO control input.

By the way, I've used the MC1350 down to 10 V supply
rails with no problem although I do agree with it
running optimally (for gain and noise figure) at 12 V
supply. There's a lot of internal constant current
sources on that IC and that causes the dependency on
supply voltage. The same with the MC1349 which I'm
working with now as both gain blocks and as mixers.
I've worked with the MC1590 metal can original 34
years ago and the plastic package 1350 since 31 years
ago. I like it as a little block of gain which has
low distortion when running balanced input to balanced
output...on up to 70 MHz with hardly any gain rolloff.

On Sun, 13 Mar 2005 16:42:18 -0500, "Netgeek"
wrote:

I'd like to build some homebrew VHF-AM receivers - specifically a receiver
for the VHF 108-118 Mhz band.

Note that the FM broadcast band is just below that band with lot of
strong transmitters, so any configuration with low side local
oscillator injection and the more or less standard 10.7 MHz first IF
is most likely going to give image frequency problems from FM
broadcast stations in the 86.6 .. 96.6 MHz range. To avoid this, a
good narrow tunable filter is required in front of the first mixer,
which can be a problem if good frequency agility is required.

One approach would be to use a much larger first IF or put the local
oscillator above the desired band (which also swaps the sidebands)
with image responses in the rest of the aviation band.

One idea would be to use a fixed downconverter e.g. with a 98 MHz
crystal frequency, mixing the VOR band down to 10-20 MHz, filter out
the strong mixing products from the FM broadcast band that is on
frequencies below 10 MHz and use a DDS with I/Q outputs to get I/Q
demodulation of the signal.

Paul OH3LWR

"Paul Keinanen" wrote in message

One idea would be to use a fixed downconverter e.g. with a 98 MHz
crystal frequency, mixing the VOR band down to 10-20 MHz, filter out
the strong mixing products from the FM broadcast band that is on
frequencies below 10 MHz and use a DDS with I/Q outputs to get I/Q
demodulation of the signal.

Thank you Paul - I'd like to look into this. Can you point me to some
practical examples or reference materials to start with? Quite some
time ago I ran across the articles from the flex-radio.com guys and
was very interested in their approach. It's my understanding (and I
mean a fairly fuzzy understanding) that direct conversion has many
benefits but is limited to lower bands (unless you're the military with
a big budget)??? What are the trade-offs in doing a downconversion
followed by DDS-based conversion?

On Tue, 15 Mar 2005 12:49:01 -0500, "Netgeek"
wrote:

It's my understanding (and I
mean a fairly fuzzy understanding) that direct conversion has many
benefits but is limited to lower bands (unless you're the military with
a big budget)??? What are the trade-offs in doing a downconversion
followed by DDS-based conversion?

My suggestion of using a DDS for direct conversion in the 10-20 MHz
range is based on the assumption that DDS chips running with 50-60 MHz
clock frequency should be available at a quite a reasonable price,
compared to similar chips running at 400 - 500 MHz, which would be
required for direct synthesis in the VHF band.

However, translating the whole band down to HF requires a strong down
converter, especially due to the nearby strong signal broadcast band.
Some flimsy NE602 type converter will not survive very well in such
environment, but a high current preamplifier followed by a diode ring
mixer might be a better converter.

Paul OH3LWR