This brings back fond memories - pulling car radios out of dusty, cramped
places, fixing them, and putting them back. The owners always loved to
see their radio working again!
I can think of a couple of reasons why car radios were built this way:
1. The difference between the local oscillator and the rf is less, so
tracking is easier. Most car radios have a RF stage to help overcome the
short antenna. So you have three resonant circuits to keep aligned,
instead of two as in a common household radio.
2. Permeability tuning can be quite linear. The inductance is directly
related to the core position. This would help tracking over the 3:1
frequency ratio needed for the broadcast band.
3. There is a lot of vibration in cars. A variable capacitor would have
to be big and sturdy to survive. This would make the radio larger and
make it harder to fit in tight spaces behind the dash. Although I have
worked on variable-capacitor car radios. Real old ones!
Best Regards,
Mike
The two things are connected. Permeability tuning was used because it is
easier to provide preset push-button tuning without frequency drift
problems over the wide temperature range experienced by car radios.
Now, if you have a permeability-tuned mixer stage (whether or not there
is an RF stage before it), with the IF at 455kHz, it tends to go
seriously unstable when tuned to the lower frequency end of the MF band
(520 kHz?), because base and collector circuits are tuned to near-enough
the same frequency. The use of a 262 kHz IF prevents this. I suppose 262
kHz was chosen to minimise problems (whistles) due to image reception
and IF harmonics.
--
Regards, John Woodgate, Phone +44 (0)1268 747839 Fax +44 (0)1268 777124.
OOO - Own Opinions Only. You can fool all of the people some of the time, but
you can't please some of the people any of the time.
Are you sure that a lower IF eases tracking? The 3-point tracking
equations are so complex that it isn't easy to check (I must consider
writing a program, someday!).
>
>2. Permeability tuning can be quite linear. The inductance is directly
>related to the core position.
Only over a limited range of core positions. Normally, tuners are not
restricted to this linear region.
>This would help tracking over the 3:1
>frequency ratio needed for the broadcast band.
But you don't actually want the oscillator inductance to follow the
signal inductance exactly linearly.
>
>3. There is a lot of vibration in cars. A variable capacitor would have
>to be big and sturdy to survive. This would make the radio larger and
>make it harder to fit in tight spaces behind the dash.
Well, true for air-spaced caps, but not for solid dielectric or 'pip'
types. (Pip types have tiny plastic studs inserted in the plates to
define the air-gap and prevent vibration.)
> Although I have
>worked on variable-capacitor car radios. Real old ones!
Yes, really old, and BIG with it!
455 kHz could give you fits if the local radio station was on 910 kHz,
too! Been there, done that!
Roy
John Woodgate wrote:
> Are you sure that a lower IF eases tracking? The 3-point tracking
> equations are so complex that it isn't easy to check (I must consider
> writing a program, someday!).
> >
I have aligned many of these but never had to design one. Yet that appears to be
a solution to a current problem. Could you point me to more information? With
or without a program.
Charlie
Well, we are really showing the gray. I haven't fixed one of
these radios since the early 50's, so memory may fail in some
areas and make me say stupid things. Just ignore them.
[...]
>> 1. The difference between the local oscillator and the rf is
>> less, so tracking is easier. Most car radios have a RF stage
>> to help overcome the short antenna. So you have three
>> resonant circuits to keep aligned, instead of two as in a
>> common household radio.
> Are you sure that a lower IF eases tracking? The 3-point
> tracking equations are so complex that it isn't easy to check
> (I must consider writing a program, someday!).
My copy of the Radiotron Designer's Handbook is in storage with
the rest of the books, so I can't put the equations on screen
either. But you are probably right - the tracking for
permeability tuning has to use a different concept.
>> 2. Permeability tuning can be quite linear. The inductance
>> is directly related to the core position.
> Only over a limited range of core positions. Normally, tuners
> are not restricted to this linear region.
>> This would help tracking over the 3:1 frequency ratio needed
>> for the broadcast band.
> But you don't actually want the oscillator inductance to
> follow the signal inductance exactly linearly.
Agree - I wonder if they changed the spacing of the turns at
different points along the coil.
I do recall that adjusting the cores was quick and easy, much
easier than bending the tabs on variable-capacitor radios. Just
turn the oscillator slug so the radio station came in at the
proper point on the dial, and tweak the rf slugs for maximum
AGC voltage.
I also seem to recall checking the tracking at different points
on the dial and finding it either perfect or very close.
Of course, if you didn't watch what you were doing, the slug
could come out of the coil form. The fit was so snug that it
was a real problem getting it to go back in without breaking
something.
>> 3. There is a lot of vibration in cars. A variable capacitor
>> would have to be big and sturdy to survive. This would make
>> the radio larger and make it harder to fit in tight spaces
>> behind the dash.
> Well, true for air-spaced caps, but not for solid dielectric
> or 'pip' types. (Pip types have tiny plastic studs inserted
> in the plates to define the air-gap and prevent vibration.)
I don't recall seeing the solid dielectric versions until they
came out in transistor radios. I have never seen the 'pip' type
you mention. Wouldn't the studs wear and change the gap over
time?
>> Although I have worked on variable-capacitor car radios.
>> Real old ones!
> Yes, really old, and BIG with it!
I recall one radio made for post-war Chevrolets. It took up the
entire center portion of the dash!
So, I guess the main reason for permeability tuning was the
smaller size. But we still haven't answered the question about
the 262 KHz IF.
Best Regards,
Mike
I thought I had actually given the reason for the 262 kHz IF; mixer
stability.
->seriously unstable when tuned to the lower frequency end of the MF band
->(520 kHz?), because base and collector circuits are tuned to near-enough
->the same frequency.
You aren't as old a fart as I gave you credit for. That's the GRID and
PLATE circuits, sonny.
{;-)
Jim
> Regards, John Woodgate, Phone +44 (0)1268 747839 Fax +44 (0)1268 777124.
Well, I saw that, but I didn't have time to think about it at the time.
Here is a copy of your post, and my response.
John Woodgate <j...@jmwa.demon.co.uk>
Tue, 8 Sep 1998 12:25:07 +0100
> Now, if you have a permeability-tuned mixer stage (whether or not
> there is an RF stage before it), with the IF at 455kHz, it tends
> to go seriously unstable when tuned to the lower frequency end of
> the MF band (520 kHz?), because base and collector circuits are
> tuned to near-enough the same frequency.
The RF and IF stages each have their base and collector tuned to
exactly the same frequency (or grid and plate, if we are talking old
stuff.) Why should the mixer have problems with this?
> The use of a 262 kHz IF prevents this. I suppose 262 kHz was
> chosen to minimise problems (whistles) due to image reception and
> IF harmonics.
An IF of 262 KHz would give poorer image rejection (not by much,
though) The image is not so far down the skirt of the RF response as
it is with a 455 KHz IF.
Also, with a 262 KHz IF, more of the broadcast band is enclosed by
the image frequency, so there are more stations to create problems.
For example:
455 KHz IF
RF LO IMAGE
690 1145 KHz 1600 KHz
262 KHz IF
RF LO IMAGE
1070 1332 KHz 1594 KHz
With an IF of 455 KHz, the range of frequencies susceptible to
images is 550 to 690 KHz.
For 262 KHz, the range is 550 to 1070 KHz.
It seems 455 KHz is a better IF than 262 KHz. So why pick 262?
Best Regards,
Mike
Actually I beleave it had more to do with mechanics than electronics.
Permeability-tuning is a lot more stable and robust that capacitance
tuning. No mechanical connections to corrode and wear. However it is
more difficult to design so that it would tune in a linier fassion.
That is to say, to tune so that the stations are not "bunched up" at
one end of the band or another.
If you looked at the inductors, you would notice that they are not
wound in a liner fassion.
I have repaired a lot of auto-radios, and had to replace or repair
the inductors in them. They a wround in a very particular way.
Definately not linear.
Since permeability-tuning has been in use since ...well I as far back
as I can remember. From the 50's on, and maybe before, when
vacume tubes (valves for those not in the states) where used.
No computers to model designs on, just slide-rules.
Getting the tuning mechanism to track was probably no easy task.
I expect the choice of 262khz IF was probably a compromise.
To make the tracking of the inductors easyer to design while
keeping the image rejection and the size of the IF transformers
reasonable.
Getting a variable capacitor to track with a higher freq. IF is
a lot easyer, than to get permeability-tuned stage(s) to track.
As a aside, I had to repair the FM section of an AM/FM auto-radio
that used a varible cap in the FM section. Had to replace the entier
FM front end because the cap was so worn. The least little bump or
jar would throw it off frequency.
Wlll just my 2 cents.
Chris
--
========================================================================
Chris Maukonen "Whenever we say to others
Sr. Systems Programmer 'you shouldn't behave in such a way'
Univeristy of Central Florida we adopt a position of superiority."
chris@.ucf.edu (407) 823 5460
I think that the 262.5 Khz IF was used to obtain more selectivity before the days
of filters. Car radios with an RF stage were more sensitive than AC/DC sets with no
RF stage and loop antennas and more selectivity was necessary to reduce adjacent
station interference. Comments?
Ed Woodbridge
It's not age: it's diffidence. I KNOW it happens with transistors, but I
DIDN'T know from first-hand experience that it happened inn exactly the
same way with valves, which don't have as much internal feedback as
transistors. But I suspected that it did: wiring would provide enough
feedback.
Valves have anodes, of course. I believe tubes have plates. (;-)
--
Regards, John Woodgate, Phone +44 (0)1268 747839 Fax +44 (0)1268 777124.
[...]
That would make a lot of sense. Given the same Q, a 262 KHz IF would be
twice as good as 455 KHz.
But I wonder which came first. The famous All-American 5 used 455 KHz,
and there certainly were car radios before World War 2.
Did they use 262 KHz in these radios, and simply continued using it after
the war?
Best Regards,
Mike
You make a lot of good points. Can you answer some questions?
> In article <199809080716...@ladder01.news.aol.com>,
> Mjzuccaro <mjzu...@aol.com> writes
> [...]
> Actually I believe it had more to do with mechanics than
> electronics. Permeability-tuning is a lot more stable and
> robust that capacitance tuning.
> No mechanical connections to corrode and wear.
Good point!
> However it is more difficult to design so that it would tune
> in a linear fashion. That is to say, to tune so that the
> stations are not "bunched up" at one end of the band or
> another.
> If you looked at the inductors, you would notice that they
> are not wound in a linear fashion.
> I have repaired a lot of auto-radios, and had to replace or
> repair the inductors in them. They a wound in a very
> particular way. Definitely not linear.
Can you describe how they were wound?
I posted a similar comment earlier. Figuring out how to wind
them using a slide rule must have been fun.
I wonder if it would be easier to get the windings correct when
there is not so much difference between them, as there would be
with a 455 KHz IF?
It shouldn't make any difference doing the calculations, but
perhaps the physical winding may have been more difficult to
achieve.
> Since permeability-tuning has been in use since ...well I as
> far back as I can remember. From the 50's on, and maybe
> before, when vacuum tubes (valves for those not in the
> states) where used. No computers to model designs on, just
> slide-rules. Getting the tuning mechanism to track was
> probably no easy task. I expect the choice of 262khz IF was
> probably a compromise. To make the tracking of the inductors
> easier to design while keeping the image rejection and the
> size of the IF transformers reasonable.
Presumably, the RF stage helped a bit with image rejection.
Given the performance of ferrites available at that time, the
size issue may have been the key. But that would have
encouraged the use of 455 KHz instead of 262.
> Getting a variable capacitor to track with a higher freq. IF
> is a lot easier, than to get permeability-tuned stage(s) to
> track.
Can you explain more? Intuitively, I think you are right. But
why do we both think so?
I suppose you can shape the capacitors to just about any tuning
curve needed - this may be more difficult with variable-spaced
windings.
> As a aside, I had to repair the FM section of an AM/FM
> auto-radio that used a variable cap in the FM section. Had to
> replace the entire FM front end because the cap was so worn.
> The least little bump or jar would throw it off frequency.
The user must have been a nervous type that couldn't stay on
the same station more than a few seconds. The modern version
has a channel remote and a 6-pack.
> Well just my 2 cents.
> Chris
Regards, Mike
ROFL!
Not to mention the Vibrators that powered them!
Harry C.
p.s. The profit from replacing Vibrators in car radios alone made a
major dent in my college costs! ;-)
They were wound using a highly non-linear turns-spacing, the exact
nature of which remains a well guarded secret even to today. Collins
Radio employed the same technology on their tunable oscillator modules
(Used in the R390 military communications receiver here in the US, and
also by the manufacturers of competing products.) In fact, the
frequency linearity of this module became Collins' initial claim
to fame!
While in college, I worked for a firm named Barker & Williamson.
They were most anxious to duplicate the linear frequency performance
of the Collins oscillator, but never succeeded, despite the fact that
we took apart and carefully analyzed many of the Collins units. I
was one of the people that took them apart.
The tuning coils of the oscillator were non-linearly wound, and
the variation in pitch had no obvious analytical basis. Collins
combined this with a set a finely spaced, capacitive tuning fingers
that closely paralleled the coil and, based on unit to unit
comparisons, were obviously aligned on a unit by unit basis.
Barker and Williamson's replication of this product worked well,
but never was able to duplicate the performance of the Collins
product. As a result, Collins owned this market until digital
frequency synthesis took over.
Sorry to detract from the thread, but I thought that this tid-bit
of information might be interesting to you history buffs!
Harry C.
Huh. So that's what those were for. I remember wondering about them.
> Barker and Williamson's replication of this product worked well,
> but never was able to duplicate the performance of the Collins
> product. As a result, Collins owned this market until digital
> frequency synthesis took over.
>
> Sorry to detract from the thread, but I thought that this tid-bit
> of information might be interesting to you history buffs!
>
> Harry C.
>
One idea would have been to scale the problem up to a much higher
frequency -- instead of running the PTO around a few MHz, build a
smaller version for the same magnitude of coverage at several dozen MHz,
with precisely scaled inductor dimensions and strays. That would allow
the turn-to-turn winding pitch to be manipulated by hand to achieve the
required linearity (i.e., you'd have 1/10 the number of turns to fiddle
with). I guess differences in slug permeability might shoot this idea
down in flames, though.
It's hard to believe that an analytical first approximation would be
that hard to come up with. Some hefty calculus would probably be
necessary, but you'd only have to go through the pain once. What were
some of the reasons for giving up on this approach?
-- jm
------------------------------------------------------
Note: My E-mail address has been altered to avoid spam
------------------------------------------------------
And they were a joy to align! R390 manuals all but told you which way
and how far to turn the adjustment screw. Just wish I could afford to
buy a good one now.
Roy
The Rf and IF stages are designed to work in those conditions. The mixer
normally has *widely different* tuning at input and output, except at
the l.f. end of the m.f. band.
>
> > The use of a 262 kHz IF prevents this. I suppose 262 kHz was
> > chosen to minimise problems (whistles) due to image reception and
> > IF harmonics.
>
> An IF of 262 KHz would give poorer image rejection (not by much,
> though) The image is not so far down the skirt of the RF response as
> it is with a 455 KHz IF.
Not what I meant. If one can't use 455 kHz, what then is the next best
frequency? I assumed that 262.5 kHz was a reasoned choice, on those
grounds.
>
> Also, with a 262 KHz IF, more of the broadcast band is enclosed by
> the image frequency, so there are more stations to create problems.
>
> For example:
>
> 455 KHz IF
>
> RF LO IMAGE
>
> 690 1145 KHz 1600 KHz
>
> 262 KHz IF
>
> RF LO IMAGE
>
> 1070 1332 KHz 1594 KHz
>
> With an IF of 455 KHz, the range of frequencies susceptible to
> images is 550 to 690 KHz.
>
> For 262 KHz, the range is 550 to 1070 KHz.
>
> It seems 455 KHz is a better IF than 262 KHz. So why pick 262?
Again, I say, if you *can't* use 455, you *have* to pick another
frequency. I don't know enough about US frequency allocations to go
through the process of determining the next best frequency. Did the RMA
do it alone or did they co-operate with FCC to make a choice? Is it
written up in IEEE or ProcIRE or somewhere?
Well, 262.5 kHz requires less Q for a given bandwidth, certainly, or
gives less bandwidth for a given Q. But how prevalent was adjacent-
channel coverage? Did the FCC normally allocate only alternate channels
to the same geographical area? That wouldn't be wholly effective at
night, of course.
I always thought it was for selectivity, in as few IF stages as possible.
Really (ham or maritime) narrowband radios of the pre-crystal filter era
often were triple conversion down to 100 or 50 kHz.
Mark Zenier mze...@eskimo.com mze...@netcom.com Washington State resident
An analytical first approximation was indeed not difficult to come up
with. Can radio manufacturers used this. It is a design that provides
somthing like 0.5% frequency linearity over the entire operating range
that is difficult.
: Some hefty calculus would probably be
: necessary, but you'd only have to go through the pain once. What were
: some of the reasons for giving up on this approach?
Better, cheaper technology emerged in the form of digital frequency
synthesizers. In the face of the new technology, even Collins abandoned
the business.
You can still find these units in use on the Collins 75A4 ham receiver,
and the military R390 and T368 transmitters (although today considered
pretty obsolete technology with most retired from service).
Harry C.
p.s. I'd still like to own a Collins 75A4, and will buy one in the
right condition and right price if offered. Still, it's like
wanting to own a 1956 Chevy convertable in mint condition --
the only justification is its romance! ;->
p.p.s. Owning a 75A4 does have a practical side -- it's all tubes.
After the "big one", when radiation has zapped all the Kenwoods
out there, with their semiconductor guts, performance of the
75A4 will not even be dented! ;->
[...snip good info]
> Sorry to detract from the thread, but I thought that this tid-bit
> of information might be interesting to you history buffs!
>
> Harry C.
No detraction whatsoever - this is very good info. Thanks!
Best Regards,
Mike
> p.p.s. Owning a 75A4 does have a practical side -- it's all tubes.
> After the "big one", when radiation has zapped all the Kenwoods
> out there, with their semiconductor guts, performance of the
> 75A4 will not even be dented! ;->
>
Who will have a working transmitter you can listen to?
Best Regards,
Mike
I believe in the very early days of car radios, a popular feature was
being able to tune below 535 KHz (and even 455 KHz) to hear police radios.
If that recollection is right, then it's easy to see why a 455 KHz IF
frequency wasn't a good choice. Once the makers made the heavy investment
in designing a 262.5 KHz IF transformer, it wouldn't hurt to use them in
all the radios; and once it became a standard and 262.5 KHz transformers
were available off the shelf, no one wanted to change frequencies. My
guess would be that 455 KHz came into wide use when miniaturization
became a factor; even with today's improved magnetic materials, it's still
cheaper to design a small 455 KHz transformer than a 262.5 KHz unit in the
same size (not to mention the price of copper today).
Note this is all wild-ass guessing unhindered by facts or evidence, just
something conjured by a loose synapse making noise in an unexercised
memory.
--
+---------------------------------------------------------------------------+
| Cliff Sharp | Hate spam? Join The Great American Pink-Out! |
| WA9PDM | http://www.ybecker.net/pink/ |
+---------------------------------------------------------------------------+
>Note this is all wild-ass guessing unhindered by facts or evidence, just
>something conjured by a loose synapse making noise in an unexercised
>memory.
Well, I can add to the noise too. I seem to recall that there was yet another
IF frequency at one time - around 90 KHz - presumably to get higher
selectivity. No idea if this was a single or double conversion thing.
Alan
> In article <35F7CE...@urlfor.addr>, Mike <ch...@urlfor.addr>
writes
>> The RF and IF stages each have their base and collector
>> tuned to exactly the same frequency (or grid and plate, if
>> we are talking old stuff.) Why should the mixer have
>> problems with this?
> The Rf and IF stages are designed to work in those
> conditions. The mixer normally has widely different tuning at
> input and output, except at the l.f. end of the m.f. band.
They seem to have had no problems solving this with household
radios at the time. 455 KHz was also standard IF frequency and
these radios were made in large numbers. They worked fine.
[...]
>> It seems 455 KHz is a better IF than 262 KHz. So why pick 262?
> Again, I say, if you can't use 455, you have to pick another
> frequency. I don't know enough about US frequency allocations
> to go through the process of determining the next best
> frequency. Did the RMA do it alone or did they co-operate
> with FCC to make a choice? Is it written up in IEEE or
> ProcIRE or somewhere?
I don't think that's the problem, John. It's not selecting a
particular IF frequency - with the availability of standard
components at 455 and 262 KHz at the time, the frequencies were
already selected.
I think the question is why did permeability-tuned car radios
use 262 KHz instead of 455? Are these facts related - did the
permeability-tuning affect the choice of IF frequency?
The answers being posted are quite good. It seems to be related
to the difficulty of winding the inductors needed to achieve
good tracking.
Best Regards,
Mike
Some extremely lucky guy...hopefully one owning an all tube rig with
maybe something like a 4-250 or 4-400 as a final (and a generator).
Isn't there an emergency communications contingent within
the ham community that maintains equipment like this for just such
a contingency (AREC or something like that)? Years back, the
amateur radio "field day" exercises were conducted to demonstrate
preparedness for emergency situations, with many participants retaining
and using old, all tube rigs preserved entirely for this purpose.
Harry C.
I suppose the exact choice of frequency had to do with minimizing birdies, etc.
I seem to remember reading some time ago the following (I may not have it
all exactly right; it's been a long time):
Capacitively tuned radios didn't work very well in autos due to vibration
and the resulting microphonics. It was the inductive tuner ("inductuner"?
was/is that a trademark?) which made them viable.
The inductive tuner was invented by Bill Lear (Lear eight-track, LearJet,
Lear Steam Car -- *that* Lear), and it was his first big invention. He sold
it to a struggling company which became successful selling car radios, and
derived their name from their product: Motorola.
One of Bill Lear's daughters is named "Shanda". Really.
Isaac
On Thu, 10 Sep 1998, Mark Zenier wrote:
> In article <35F7CE...@urlfor.addr>, Mike <ch...@urlfor.addr> wrote:
> >
> > It seems 455 KHz is a better IF than 262 KHz. So why pick 262?
> >
>
> I always thought it was for selectivity, in as few IF stages as possible.
> Really (ham or maritime) narrowband radios of the pre-crystal filter era
> often were triple conversion down to 100 or 50 kHz.
>
Selectivity would have been the answer I gave, and I was just figuring out
where to jump into the thread.
The average radio is used for local reception. Selectivity isn't much of
an issue because if you're in Boston and listening to WBZ at 1030KHz, it's
strength will be way stronger than any distant station, and local stations
are never placed close together. But if you're out in the boonies, WBZ's
strenght diminishes. You may find three stations close together of
close to equal strength, say CFRB 1010 in Toronto, KDKA 1020 in Pittsburg,
and WBZ in Boston at 1030. None of them are local, because there are no
local signals. But if the selectivity of the radio is not good, none of
the three will be particularly listenable.
If you're in a car, the strength of the individual signals may vary
according to terrain and which direction you turn. If the selectivity is
not good, when you turn the next station over may increase in strenght,
and lousy selectivity will mean it overrides the station you are listening
to.
It's easier to get selectivity the lower in frequency you go. At the time
when car radios first arrived on the scene, the selectivity would have
come from IF transformers. The manufacturer's want to make the radio
as cheap as possible, so for the same number of IF transformers, 262KHz
is better than 455KHz.
I have no idea why it's 262 and not 260 or 270 KHz. It might have been
arbitrary at the time. But probably there was some selection, based on
actual signals at the possible IF frequency and probably the rest of the
design. I've never seen an explanation of why 455KHz (or 10.7MHz for that
matter) were picked as "standard" IF frequencies, other than general
comments about selecting a good IF frequency. But at some point these
did freeze into fairly standard frequencies.
Now, if you go to a lower IF, it's harder to get good image rejection.
I though all (or most) car radios had an RF amplifier stage ahead of the
mixer, something you didn't see in the average AM radio with a loopstick
(be it the all-American five or a transistor radio). An RF amplifier
stage means some more tuned circuits at the front end, which helps to
get rid of image responses (I understand the National HRO series, with
a 455KHz IF, still had reasonable image rejection long after other
receivers had gone to double cnverstion because it had a great set
of tuned circuits before the mixer). So the manufacturer's could
make up lower image rejection caused by the lower IF.
The question remains whether the RF stage was added because of the low
IF, or for other reasons. I tend to think for other reasons, with image
rejection a secondary consideration. Car radios use a whip antenna,
and they aren't really long at that frequency, so the RF stage
compensates. Moving around gives you a wide range of signals, of varying
strength, so the increased front end selectivity might have been
necessary. If you've got enough selectivty in the IF, but lack gain,
there are good reasons for spreading the gain to the front end, rather
than add another IF stage.
By the way, I looked up the topic in the Radiotron Designer's Handbook
lastnight (mine only cost me two dollars at a used booksale two years
ago, so there is always hope), and it didn't mention 262KHz IF's. It
does have a section on car radios, but concerns itself with noise
from the engine, rather than problems of RF design.
Also, in the original post the point was made that 262KHz was used right
up until recent times. I'm sure I've seen a car radio with a 262KHz
ceramic filter, though I can't remember the exact radio right now. But
if there were ceramic filters for the frequency, it would indicate some
sort of desire to stick with the "standard" even when bringing in
new components.
However, I think all the digitally tuned car radios go to 455KHz IFs,
or more likely 450KHz. The synthesizer IC's are set up for that
frequency, and there doesn't seem to be a desire to create two versions,
one for cars and one for homes.
For that matter, one wonders why radios stayed with 455 or 450KHz as
new designs came in.
Michael
Miniaturization had nothing to do with the rise of 455KHz. If you look in
an old National HRO receiver, it had fairly huge IF transformers, and they
were at 455KHz. And I'm fairly certain it was a common IF, at least for
home receivers, at the time.
The small IF transformers you see in radios nowadays, both car and home,
certainly were a significant shrinkage from the IF transformers used
before that. But I've seen no difference in the size of 455KHz and 262KHz
If transformers from before that.
Michael
On Fri, 11 Sep 1998, Clifton T. Sharp Jr. wrote:
> Mjzuccaro wrote:
> > From the postings Ive read on this NG I know there are a lot of sharp RF men
> > out there.Maybe one of you can help me put this question to rest once and for
> > all- why was 262 kc the standard IF frequency in almost every car radio made,
> > well into the 70's, (and maybe 80's-and maybe still!) when every other AM radio
> > in the USA used 455 kc?
>
> I believe in the very early days of car radios, a popular feature was
> being able to tune below 535 KHz (and even 455 KHz) to hear police radios.
> If that recollection is right, then it's easy to see why a 455 KHz IF
> frequency wasn't a good choice. Once the makers made the heavy investment
> in designing a 262.5 KHz IF transformer, it wouldn't hurt to use them in
> all the radios; and once it became a standard and 262.5 KHz transformers
> were available off the shelf, no one wanted to change frequencies. My
> guess would be that 455 KHz came into wide use when miniaturization
> became a factor; even with today's improved magnetic materials, it's still
> cheaper to design a small 455 KHz transformer than a 262.5 KHz unit in the
> same size (not to mention the price of copper today).
>
> Note this is all wild-ass guessing unhindered by facts or evidence, just
> something conjured by a loose synapse making noise in an unexercised
> memory.
>
>I have no idea why it's 262 and not 260 or 270 KHz. It might have been
>arbitrary at the time.
True. But I sometimes saw the frequency expressed as "262.5 KHz" -- double
that and you get 525 KHz, near the bottom end of the AM BCB. (Hmmm. Half
of 530 KHz is 265 KHz -- why not use that?)
>Now, if you go to a lower IF, it's harder to get good image rejection.
>I though all (or most) car radios had an RF amplifier stage ahead of the
>mixer, something you didn't see in the average AM radio with a loopstick
>(be it the all-American five or a transistor radio). An RF amplifier
>stage means some more tuned circuits at the front end, which helps to
>get rid of image responses (I understand the National HRO series, with
>a 455KHz IF, still had reasonable image rejection long after other
>receivers had gone to double cnverstion because it had a great set
>of tuned circuits before the mixer).
I used to own an early-model HRO. It had _two_ tuned RF stages and a tuned
mixer stage. Those, plus the local oscillator, resulted in a four-section
variable capacitor -- very impressive to watch! If I used a signal
generator set for a high amplitude 910 KHz above the tuned-in frequency, I
could hear a _weak_ image. (BTW, the IF was actually 456 KHz -- apparently
455 hadn't fully locked in as a standard.)
>For that matter, one wonders why radios stayed with 455 or 450KHz as
>new designs came in.
I remember hearing somewhere that 455 KHz was kept as sort of a "clear
channel" and no one (as least here in the States) was ever licensed to use
that exact frequency. Anyone know if this is true?
- Michael
455 and 262 KHz are used for Tunnel Broadcasting. You hear it
when in the tunnel no matter what station your receiver is tuned to.
Boston used it in both the Sumner and Callahan tunnels for many years,
but I haven't heard it in use recently.
I don't know if this is an FCC licensed service. (I would suspect that
the FCC restricts the use of these frequencies in the open space, since
it could take the form of broadcasted SPAM and seriously disrupt
commercial broadcast reception.)
Harry C.
: > >Note this is all wild-ass guessing unhindered by facts or evidence, just
: > >something conjured by a loose synapse making noise in an unexercised
: > >memory.
: > Well, I can add to the noise too. I seem to recall that there was yet another
: > IF frequency at one time - around 90 KHz - presumably to get higher
: > selectivity. No idea if this was a single or double conversion thing.
The old military BC-453, which covered ~190-550kHz, had two IF
stages at 85kHz. Made for a really nice CW filter when coupled and
tuned to the 455kHz IF of a cheap receiver. Even included a BFO
at 85kHz for CW reception.
: > Alan
L
Good post! Thanks for taking the time.
Radio station CFRB was local when I was fixing these radios. In
fact, it was just down the road. It served as a convenient strong
signal source while troubleshooting radios that appeared completely
dead - if CFRB showed up, at least the power supply, local
oscillator, IF, and audio stages were working. If so, it was usually
a bad RF tube or a broken center conductor in the antenna coax.
One day, I went on my bike to find out where this station was, and
why it was so strong. It didn't take long. To my amazement, the
station engineers welcomed me in and were happy to answer my
questions.
I used to spend all my free time at the station. The engineers would
walk me through the design on paper, then show me the actual
sections of the transmitter that performed those functions.
They ran 50 killowatts. The glow of the mercury-vapor high voltage
rectifiers was fascinating, especially when it flickered with the
announcer's voice.
They showed me how they used ordinary vaseline to keep the antenna
feedlines from corroding. They were copper bus bar, about 2 inches
wide and 1/8 inch thick.
Once, they had problems with VSWR - the antenna wouldn't tune
properly. They shut down the transmitter, unbolted all the bus bars,
sandpapered the joints, and applied a coat of vaseline.
When they put it back together, the tuning problems disappeared and
never came back. This old trick has modern applications - if you are
having problems connecting with your modem, you might take a look at
http://www.csolve.net/~add/system/modemtip.htm
One engineer encouraged me to get my ham license and gave me a 7.1
MHz crystal. I scrounged a 6AG7 from an old radio IF stage, and made
my first contact with him with about 4 watts into the antenna.
I visited the station recently, after 40 years travelling around the
world working in electronics.
The station had moved, the site was abandoned, and they were taking
the antennas down.
I reflected that modern stations are unattended and completely
automatic. There are no resident engineers to help young kids that
show up on bikes with lots of questions.
I wish to thank these wonderful engineers, now long gone, who took
the time to answer interminable questions, and who helped me along
the road to a fascinating hobby and career.
Perhaps we can return the favor when a newcomer appears in these
newsgroups with badly-worded and poorly thought out questions.
Instead of giving them a hard time for violating newsgroup protocol,
or attempt to show our superior knowledge for all to admire, maybe
we could try to give them a helping hand.
There's not many places they can go to ask questions these days.
Best Regards,
Mike
I'd once again say (1) it's about as high as you can get without getting
so close to BCB that a station on 540 KHz will start leaking through the
IF filtering, (2) higher means less wire, (3) off-the-shelf means less
design cost.
He may have invented a particular mechanical system, but the
permeability tuner is as old as radio itself. Early systems using really
low radio frequencies could use iron cored coils: thin laminations or
filings in a glass tube.
But they are not permeability tuned. The stability problem does not
occur to the same extent with capacitively-tuned circuits, because the
L/C ratio is lower at the l.f. end of the band.
>
> [...]
>
> >> It seems 455 KHz is a better IF than 262 KHz. So why pick 262?
>
> > Again, I say, if you can't use 455, you have to pick another
> > frequency. I don't know enough about US frequency allocations
> > to go through the process of determining the next best
> > frequency. Did the RMA do it alone or did they co-operate
> > with FCC to make a choice? Is it written up in IEEE or
> > ProcIRE or somewhere?
>
> I don't think that's the problem, John. It's not selecting a
> particular IF frequency - with the availability of standard
> components at 455 and 262 KHz at the time, the frequencies were
> already selected.
Yoiu have utterly begged the question. (;-) WHY was 262 kHz a standard
frequency, if 455 is so much better?
>
> I think the question is why did permeability-tuned car radios
> use 262 KHz instead of 455? Are these facts related - did the
> permeability-tuning affect the choice of IF frequency?
>
> The answers being posted are quite good. It seems to be related
> to the difficulty of winding the inductors needed to achieve
> good tracking.
No, I don't believe that. The winding problem was not solved
analytically, which would be very time-consuming without a computer, as
a finite-element method is probably necessary, but heuristically - by
winding LOTS of coils. That's how we 'designed' ferrite rod antennas in
the late 50's, where similar tracking problems occur, coupled with the
need for enough coil adjustment to allow for variations in rod
properties.
I am sure that the stability problem is real, from first-hand
experience. Some European car radios, even in the 70s, used 422 kHz
instead of 455, 460 or 470 kHz for the same reason. 422 kHz is low
enough to avoid the problem.
Such receivers were particularly good on the h.f. bands, for they could
receive every station at two places on the dial(;-).
>Miniaturization had nothing to do with the rise of 455KHz.
>If you look in
>an old National HRO receiver, it had fairly huge IF transformers, and they
>were at 455KHz. And I'm fairly certain it was a common IF, at least for
>home receivers, at the time.
>
Early National, Hallicrafters and Hammarlund receivers have 465kHz, this
was changed to 455 around 1940, probably because one could not change IF
for every different local situations, see below. Early Eddystone and
RCA ACR111 receivers have 450kHz IF.
In the thirties IF frequencies were sought for minimum interference
product for main transmitters, different standards applied for different
countries in Europe, usually 445-475kHz
See: RSGB; Pat Hawker: Amateur Radio Techniques, sixth edition 1977
pg.330-332 , the book shows IF's for almost every western communication
receivers/ transceivers.
>
---
J.M. Noeding, N-4623 Krsand
Eu VHFDX, wave...@onelist.com, GM4PLM 硬ave News',
UK-Vintage-Radio, VUSHF@SK7DO
There are explanations, but they are far too long to post to a
newsgroup. Basically, there are five criteria that have to be met:
1. IF must not be in a band where there are strong radiated signals.
2. Sufficient gain and selectivity can be obtained economically.
3. Image rejection and whistles, e.g. due to harmonics of IF falling
within a tuning band, must be at acceptable levels.
4. If the LO frequency can fall within a tuning band, it must not
interfere with transmissions reasonably receiveable in the same
geographical area, when radiated from the antenna at levels compatible
with the requirements of the EMC standard CISPR 13.
5. The above criteria have to be met, for a single value of IF, over the
whole geographical area in which the product is to be marketed.
The RF stage is also permeability tuned in these receivers. It is
designed for maximum available gain when running wide open on weak
signals. The input and output circuits resonate at the same frequency.
Why does it not show the same problem as the mixer, which presumably has
lower gain, and different input and output resonant frequencies?
[...]
> Regards, John Woodgate, Phone +44 (0)1268 747839 Fax +44 (0)1268 777124.
> OOO - Own Opinions Only. You can fool all of the people some of the time, but
> you can't please some of the people any of the time.
Best Regards,
Mike
Regards,
Mike
Agreed, the auto radio folks either needed more selectivity with the same
number of IF stages, or if they were trying to save power, then at least
the SAME selectivity with fewer tuned IF stages. The only way was with a
lower IF frequency.
And, of course, you don't want to have to buy a SPECIAL transformer when
all those cheap dime store radio transformers were already available, so...
The standard capacitor for the primary (tuned) winding on almost all those
small transformers was 180 pf. I have absolutely no idea why that number
was chosen, but I've replaced my share of the capacitors and I do remember
the value.
What capacitor do we have on the shelf that will cut our IF in half or
increase our selectivity 2:1? If 470/500 pf isn't the most common value
around there, I'd like to know what is. Tons of them in stock.
See what happens to the frequency of a transformer resonant at 455 kHz. if
you replace the 180 pf with a 470 pf. How about that.
Jim
455 x sqrt(180/470) = 281.6 455 x sqrt(180/500) = 273. I don't see no
262.5, nowhere. In any case, changing the cap value to get a lower
tuning frequency also reduces the bandwidth, very probably too much in
this case. If we say 10 kHz bandwidth at 455 kHz, then we get 6 kHz with
a 500 pF, actually rather less because Q of the coil is probably higher
anyway at 273 kHz. Now, one tuned circuit at 6 kHz bandwidth may be
tolerable, but several must give very poor audio quality - 3 dB down at
1 kHz, perhaps.
--
What do you mean by 'maximum available gain'? The strict circuit-theory
definition? If so, they aren't/weren't so designed. The gain is/was
limited by stability considerations, as for *any* tuned amplifier.
The mixer simply has more feedback, so that it tends to oscillate when
the input and output tuned frequencies *approach* equality. Of course,
teh details are diferent according to whether we are talking about self-
oscillating transistor mixers, or heptodes or triode-hexodes.
>Such receivers were particularly good on the h.f. bands, for they could
>receive every station at two places on the dial(;-).
They probably called this a "feature" :)
Alan
> Radio station CFRB was local when I was fixing these radios. In
> fact, it was just down the road. It served as a convenient strong
> signal source while troubleshooting radios that appeared completely
> dead - if CFRB showed up, at least the power supply, local
> oscillator, IF, and audio stages were working. If so, it was usually
> a bad RF tube or a broken center conductor in the antenna coax.
We're still here :-).
Mike, you can check out:
I've put some great stuff in the archives that I hope you'll
appreciate, including RealAudio clips from days gone by, and
some incredible pictures, dating back to 1927 when we went
on-air.
Take care.
Iain Grant
Executive Producer
CFRB Radio, Toronto, Canada
-----== Posted via Deja News, The Leader in Internet Discussion ==-----
http://www.dejanews.com/rg_mkgrp.xp Create Your Own Free Member Forum
Amazing - how did you find this thread?
Gee - I hope I didn't get anyone in trouble - this happened a long time
ago :)
Best Regards,
Mike
[...]
I think I found the answer.
There are some schematics on-line at
http://www.nostalgiaair.org/NostalgiaAir/
One of them is the Chevrolet Model 986067 (Brazil), a car radio that
uses permeability tuning and 262 KHz IF. It uses a 6SK7 for the rf
and a 6SA7 for the mixer.
I don't have my books available, so I had to do a web search. There
are some tube specs at
http://plato.phy.ohiou.edu:80/~cigna/tubes/sheets/index.html
where I found the specs for the 6SK7. I could not find anything for
the 6SA7, but I found a short spec for the 12BE6, which is similar:
http://www.duncanamps.simplenet.com/tubedata/
The 6SJ7 spec shows a grid 1 to plate capacity of 0.005 pf maximum.
This has a reactance of 58 megohms at 540 KHz, which is high enough
to eliminate oscillations using a permeable core and 30 or 40 pf to
resonate at 540 KHz. Of course, the wiring and layout needs care.
This explains why the RF stage would not oscillate in a
permeability-tuned radio.
The 12BE6 shows a grid to anode capacity of 0.3 pf.
There are two signal grids. If this spec is the capacity from the rf
signal grid to the plate, it means the feedback from the plate is
sixty times higher for the mixer than it is for the rf stage.
If this is true, it explains the higher susceptibility to
oscillation in a mixer using a 455 KHz IF. Moving to 262 KHz would
solve it.
If someone who has manuals could confirm this, it may explain why.
Best Regards,
Mike
->
->Michael F. Coyle wrote in message <6tbtuu$msj$1...@winter.news.erols.com>...
->>..."262.5 KHz" -- double
->>that and you get 525 KHz, near the bottom end of the AM BCB.
-><(Hmmm. Half of 530 KHz is 265 KHz -- why not use that?)
->Excellent thought. Rectifying the IF to detect AM produces much 2nd
->harmonic
Actually, much more 3rd, 5th, 7th, etc. 455 only had two birdies in the
AM band, 910 and a 5 kc birdie at 1360/1370. 262 has birdies at 790, 1050,
1310, and 1570. And they WERE real.
Jim
I don't know about in those days, but just this year I saw someone talking
about an amateur radio transceiver, and he mentioned that it could receive
the cellular frequencies by tuning to the image frequency. He acted like
that was a feature. Meanwhile, other people are complaining because their
transceivers are prone to a lot of out of band junk, precisely because
the front end ot he receiver is broadly tuned.
MIchael
I really don't know the TRUE reason why 262.5 khz was chosen, BUT
let's look at the rest of the components and the capacitance effects
side of the design of tube gear in harsh environments.
1) Larger value capacitors were needed to resonate the coil
inductance values to resonate at 262.5 Khz as opposed to 455 Khz
used today which use higher inductance and lower capacitance.
Today's 455 Khz systems are higher "Q" for better adjacent channel
rejection. This would mean that the -/+ coeficient of capacitors
in a 262.5 Khz system would not detune circuits very much since a
few picroFarads of change was not a problem with temperature
changes. ie: summer/winter
2) Tube grids operate at really high impedances. Suttle changes
in tube interelectrode capacitances from wear/microphonics
from road vibrations along with stray capacitance caused from
moisture across tube sockets would NOT detune the LO and I.F.
system L/C circuits much at all.
3) Adjacent channels were not a problem back then. So lower "Q"
resonate circuits offered an auto radio a more stable tuning
during varying temperatures and humidity conditions, including
all those back seat fornications that caused steaming up the
inside of the vehicle/radio. Would you not be peed-off if
you were involved in one of those scenes and the radio drifted
off station????
When superhets were first invented, an IF of ~60kHz was common. The term
'supersonic heterodyne' makes much more sense when you consider that
originally the IF was at what we would now refer to a 'ultrasonic'
frequencies.
Also there was originally no standard IF, so different manufacturers would
use whatever frequency they felt like. It took quite a number of years
before 455kHz became universal. I understand 175kHz was quite common in the
'30s.
--
Non wearing?
Polished aluminium VS nylon?
Al2O3 always wins.
--
John Fields, Austin Instruments, Inc.
El Presidente Research, Design, and Development
"I speak for the company" Austin, Republic of Texas
I reported what I was told by the manufacturers. If they don't know, who
does? Is A1203 a grade of nylon? If so, maybe they use another grade.
The friction properties of different nylons are rather complex.
John,
I agree with you that the frictional properties of the various nylons can be rather
complex.
For example, a graphite loaded nylon will generally exhibit lower starting and running
friction than will a glass loaded grade.
Al2O3 is aluminum oxide, and it will always form on the surface of unprotected aluminum,
whether the aluminum surface is polished or not, in the presence of atmospheric oxygen.
Being the second hardest material known, (next to diamond) if it it placed in frictional
contact with nylon, the nylon will wear.
Well, that's the limitations of ASCII. I interpreted that as A followed
by the number 1203.
>
>Being the second hardest material known, (next to diamond) if it it placed in
>frictional
>contact with nylon, the nylon will wear.
Well, *some forms* of aluminium oxide are very hard indeed, but hardness
is not the arbiter of whether wear takes place. Glass is quite hard, but
few things wear as a result of contact with it. The molecular-level
smoothness of a surface, and the 'stickiness' of any free chemical bonds
at the surface are important factors.
> Well, *some forms* of aluminium oxide are very hard indeed, but
> hardness
> is not the arbiter of whether wear takes place. Glass is quite
> hard, but
> few things wear as a result of contact with it. The
> molecular-level
> smoothness of a surface, and the 'stickiness' of any free chemical
> bonds
> at the surface are important factors.
--
I'm in general agreement with you, but for the specific case at
hand; an aluminum oxide / Nylon interface, the aluminum oxide
surface formed as a result of oxidation will, I believe, be porous,
rough, and harder than the Nylon.
In this instance I believe the rheology of the interface will
involve only mechanical forces, with the result that particles of
Nylon will be scraped from the suface by the "hills" an will be
embedded in the "valleys" of the Al2O3 surface.
Thus the Nylon, and not the Al2O3 will wear.
When the valleys fill, the interface will be mainly Nylon/Nylon, and
the frictional characteristics of the interface will change to the
point where analysis would probably become a nightmare!
Glen Walpert
In article <36066147...@fc.net>, John Fields <jfi...@fc.net>
wrote:
--
Say again?