"RCA" coils
Fred Nachbaur
I have a cute little Eico FM tuner, and have been thinking lately it
would be nice to build a stereo demultiplexer for it. Does anyone know
if the coils are still available somewhere for the two-tube (12AX7A and
6CL8A) multiplex adaptor described in the RCA Receiving Tube Manuals?
They are:
2- RF coil, 67-kHz trap, RCA stock No. 111047
1- RF coil, 39-kHz band pass, RCA stock No. 111048
1- RF interstage coupling xfmr, 38-kHz, RCA stock No. 111045
1- RF interstage coupling xfmr, 38-kHz, RCA stock No. 111046
Any other ideas? Is J.W. Miller still in existence?
Cheers,
Fred
Tim Williams
news:3DC42CE3...@netscape.net...
> I have a cute little Eico FM tuner, and have been thinking lately it
> would be nice to build a stereo demultiplexer for it.
Note this will only work if your tuner has the 30 or 60kHz or whatever
it is still in the output signal. (If it does, ok. :)
> 2- RF coil, 67-kHz trap, RCA stock No. 111047
> 1- RF coil, 39-kHz band pass, RCA stock No. 111048
> 1- RF interstage coupling xfmr, 38-kHz, RCA stock No. 111045
> 1- RF interstage coupling xfmr, 38-kHz, RCA stock No. 111046
>
> Any other ideas? Is J.W. Miller still in existence?
Try winding them? Those traps are indicated as a parallel resonant
circuit (BTW which RC-# you have?), so a variable ferrite core and/or
capacitor could be used to find the sweet spot. If you have a
calibrated tone generator, you could find where the circuit is resonant
before installing it.
Same idea should work for the interstage trannies, right?
Tim
--
"Hey, back in high school I was voted most likely to be a Hillbilly,
mental patient or Chimpanzee!"
- Homer Simpson
Fred Nachbaur
Tim Williams wrote:
> "Fred Nachbaur" <fnac...@netscape.net> wrote in message
> news:3DC42CE3...@netscape.net...
>
>>I have a cute little Eico FM tuner, and have been thinking lately it
>>would be nice to build a stereo demultiplexer for it.
>
>
> Note this will only work if your tuner has the 30 or 60kHz or whatever
> it is still in the output signal. (If it does, ok. :)
Yup. It has an MPX output right off the discriminator.
>>2- RF coil, 67-kHz trap, RCA stock No. 111047
>>1- RF coil, 39-kHz band pass, RCA stock No. 111048
>>1- RF interstage coupling xfmr, 38-kHz, RCA stock No. 111045
>>1- RF interstage coupling xfmr, 38-kHz, RCA stock No. 111046
>>
>>Any other ideas? Is J.W. Miller still in existence?
>
>
> Try winding them? Those traps are indicated as a parallel resonant
> circuit
Right, and the bandpass would be a series-tuned resonant circuit.
> (BTW which RC-# you have?),
24, 27, 30
> so a variable ferrite core and/or
> capacitor could be used to find the sweet spot.
Yeah, I'm pretty familiar with that drill. G'Zot springs to mind...
> If you have a
> calibrated tone generator, you could find where the circuit is resonant
> before installing it.
> Same idea should work for the interstage trannies, right?
Yeah, I might have to go that route. Not excited about it, 38/67 kHz is
pretty low frequency. That means lots and lots of turns of really really
fine wire.
But judging by the paucity of responses that NOS ones are cheap and
plentiful, the DIY approach might be the only way to build these circuits.
Cheers,
Fred
Chuck Harris
Winding the coils shouldn't be all that hard. Amidon, Inc. is
quite happy to sell you ferrite cores and forms in any size they
make. Tell them you are a ham.
As the the winding, this is a bit trickier. At these low frequencies,
and high "Q"s, you will need to use Litz wire. This is wire that is
made up of very fine INSULATED strands. It is a bit hard to come by.
A search on Google should result in some, though.
JW miller is still around. They are distributed thru DigiKey, last
I checked.
-Chuck, WA3UQV
Fred Nachbaur wrote:
>
>
> Tim Williams wrote:
>
> > "Fred Nachbaur" wrote in message
> > news:3DC42CE3...@netscape.net...
> >
> >> I have a cute little Eico FM tuner, and have been thinking lately it
> >> would be nice to build a stereo demultiplexer for it.
> >
> >
> >
Patrick Turner
Chuck Harris wrote:
> Fred,
>
> Winding the coils shouldn't be all that hard. Amidon, Inc. is
> quite happy to sell you ferrite cores and forms in any size they
> make. Tell them you are a ham.
>
> As the the winding, this is a bit trickier. At these low frequencies,
> and high "Q"s, you will need to use Litz wire. This is wire that is
> made up of very fine INSULATED strands. It is a bit hard to come by.
> A search on Google should result in some, though.
>
> JW miller is still around. They are distributed thru DigiKey, last
> I checked.
>
> -Chuck, WA3UQV
I dunno whether you guys have seen the schematics of the tubed MPX
decoder at my website at
http://turneraudio.com.au/TA.mpxDecoder98_1of2.gif
http://turneraudio.com.au/TAmpxDecoder98_2of2.gif
But this complex piece of radio junk uses quite a few coils.
I used a few existing coils from an existing circuit for some of the
inductors,
and others I wound myself using short lengths of ferrite rod antenna cores cut
to length,
about 30 m will do, and then I made small bobbins with plastic washers glued
to the ferrite, and wound with 0.2 mm dia solid wire until I had about the
right
inductance to resonate at the F required. I added or subtracted turns to trim
the
resonance, and added caps... quite time consuming....
The tuned transformers are a different challenge, and their working is similar
to 455 kHz IF transformers in AM radios, so old transformers could be used.
But the coils involved need to be rewound with 10 times more inductance,
and the caps ten times larger value to get the resonant F down ten times, to
near
what you need to use, say 38 Khz, 19 kHz, and so on.
Then you have to worry about the mutual coupling ratio of the transformers,
but if the dimensions are about the same as an old IF tranny, perhaps the
coupling will be OK.
The coupling shouldn't be a lot, or critical, like AM radio IFTs, because
you want the 19 kHz and 38 kHz trannies to be selective, not broad bandwidth.
Then a tube LTP can be used for the recombination of the 38 kHz
sub carrier and DSB signal, and there is no restriction on bandwidth.
I found the LTP works better than a tranny.
I didn't find any need for litz wire in the coils I wound, but it don't hurt
to use it.
Its horrid to use. The Q will be sufficient with solid and ferrite cores which
reduce the need for
so many turns, reducing the resistance, and thus raising the Q.
The best way to build an MPX decoder is to strip down a badly designed old
Trio
or some such crap, and re-use the 19 khz and 38 kHz trannies just to create
a pure 38 kHz tone, which is amplitude steady, with a locked oscilator,
from the amplified 19 kHz pilot tone, well filtered out from the rest of the
signal.
Its a labour of love, doing a decoder, it took me a few weeks of learning lots
about how they work, and getting the darn thing to work properly.
I built mine as a learning exercise, and for a shed radio, but when I tried it
in the
lounge, it blew away the SS FM chip based radio, which then went into storage,
and I got another standalone old chip based tuner for the shed.
I have a lot of ideas how to improve on my design, but I have no time for the
project.
Patrick Turner.
Fred Nachbaur
I'd need an awful lot of it, and don't have that many to spare anyway.
Maybe I'll have a look at what JW Miller has to offer, if they have a
suitable equivalent I'll go that route. I'm happy to wind VHF tanks, and
do up little RFC's on 2W resistors and such, but beyond that I'd rather
pay the few bucks and be done with it.
Beyond a certain point, inductors (like audio transformers and speakers)
are just "black magic" anyway. ;-)
Cheers,
Fred
--
+--------------------------------------------+
| Music: http://www3.telus.net/dogstarmusic/ |
| Projects, Vacuum Tubes & other stuff: |
| http://www.dogstar.dantimax.dk |
+--------------------------------------------+
Fred Nachbaur
The "time consuming" bit is what I was hoping to avoid. I prefer to
spend the time on the "productive" stuff like learning how circuits
work, and how to fix 'em when they don't, instead of winding, and
winding, and layers falling off... The last big-ish coil I did was for
G'zot, as mentioned, and I was reaching a point of diminishing returns
where I'd add a pile more wire to bring the frequency down only
marginally, because I was adding distributed capacitance almost as fast
as I was adding inductance.
No, I hadn't seen the schematics of your decoder. Have downloaded for
study. First question, though, why the paralleled 12AT7 buffer?
And using a differential amp (LTP) instead of the combiner transformer
is brilliant! I like that! Hm.. how about this: A 6U8 for input buffer
(triode) and CCS (pentode); another 6U8 (or maybe 6X8) for 19 kHz amp
and locked oscillator; and a 6DJ8 for the LTP.
Cheers,
Fred
Jon Yaeger
The easiest thing would be to find a stereo tube receiver that was in junk
condition and appropriate the coils and circuit from that. I think you'd
save a lot of time!
I just gave my neighbor a complete set of coils from a crappy Kenwood
KW-55 that I am transforming into a guitar / harmonica amp (otherwise I
would gladly send them to you). But surely someone on this N.G. has an
achor that they would part with for the cost of shipping, or something
similar.
Good luck!
- Jon
In article <3DC558EC...@netscape.net>, Fred Nachbaur
Patrick Turner
Fred Nachbaur wrote:
> Thanks, Patrick.
>
> The "time consuming" bit is what I was hoping to avoid. I prefer to
> spend the time on the "productive" stuff like learning how circuits
> work, and how to fix 'em when they don't, instead of winding, and
> winding, and layers falling off... The last big-ish coil I did was for
> G'zot, as mentioned, and I was reaching a point of diminishing returns
> where I'd add a pile more wire to bring the frequency down only
> marginally, because I was adding distributed capacitance almost as fast
> as I was adding inductance.
The coils in MPX decoders are all small things, and should use adjustable ferrite
cores.
Then the coils are easy to wind. small amounts of fine wire, say 0.2 mm dia will do.
No need for litz wire.
Plastic sewing machine bobbins might do for coil formers, but I sometimes use
bits of ferrite rod about 20 mm long, glue on two fibreglass sheet washers each
end and add wire until L is about right, then trim the cap values for the right Fo.
The ferrite cores will give a high Q for the tuned parts.
Building a decent MPX decoder is a long hard road, like building an AM radio.
John Byrns has some interesting Scott circuits, but I think mine work better.
>
> No, I hadn't seen the schematics of your decoder. Have downloaded for
> study. First question, though, why the paralleled 12AT7 buffer?
To make sure the main signal from the disriminator don't get mauled by all the
filters at the input of the MPX. The filters work with the required slopes
only when driven by low impedance. Plenty of 12AT7 around,
but 12 AU7, 6DJ8 will do.
>
>
> And using a differential amp (LTP) instead of the combiner transformer
> is brilliant! I like that! Hm.. how about this: A 6U8 for input buffer
> (triode) and CCS (pentode); another 6U8 (or maybe 6X8) for 19 kHz amp
> and locked oscillator; and a 6DJ8 for the LTP.
Yes, a pentode could be used for the 19 kHz amp, but it's over the top,
a triode will do, since its just a trigger for the following oscillator,
which needs to be a pentode to ensure fierce oscillation, unaffected by
tiny amplitude variations of the 19 kHz input, which do occur, I found.
The resultant 38 kHz is then very free of aplitude variations, yet firmly
synchronised with the pilot tone.
A phased lock loop circuit could be used, but its harder to do with tubes, because
you need
a 19 kHz discriminator, and a DC control voltage amp, and some means of
changing the oscillator F with variable C, enough to lock the phase right.
Patrick Turner.
Fred Nachbaur
Patrick Turner wrote:
>
> [... snip lotsa good advice on MPX decoders]
Thanks for all that, Patrick. Your email has been duly saved in the
archives!
Cheers,
Fred