Some of you were following my progress with re-designing & rewiring
of the stereo multitplex decoder I have within an ancient
Trio receiver.

I finally settled on a design using all tubes that resembles the
method of operation of the first Quad add on MPX unit with 3 transistors

except that I have all triodes.

After the ratio detector, there is a V1 6DJ8 amp which amplifies the
signal
from the RD by about 5 times, then buffers its anode output with a
direct coupled CF, V2.
From this V2CF, a feed to a pair of parallel 19kHz tuned circuits
filters out the 19kHz and V3 triode amplifies it and
powers a tuned 19kHz tranny which has a CT secondary and two diodes
then convert the 19kHz into its full wave rectified form which is mainly
38kHz.
This signal is used to synchronize a 38kHz oscillator with 2 triodes
very like the type
Scott used but with a 6CG7, V4/5.
The oscillator is a PP type, and the double tuned 38kHz tranny
has two CT windings, pri for the anodes, sec for the connection via its
CT to the
composite signal from V2CF, minus the 19kHz pilot tone.
A ring diode demodulator almost identical to the design used by Quad is
then employed
with 4 x 27k R and 4 x 1N9148 diodes. A pair 10k R off the two diode
junction
outputs charges a pair of 900pF caps, one for each channel.
These RC values gave the right amount of de-emphasis.
The signals from each channel are then fed to a first order RC filter to

help remove switching noise, then fed to CF buffers, 12AU7, V6/7,
which drive LC and RC 3rd order filters to further remove any switching
artifacts.

The L&R signals are taken to switch contacts to select stereo or mono.

The mono signal is derived by de-emphasising the signal from the
V2CF after the 19kHz notch filter with a simple RC filter, then to a
switch.

Another pair of CF buffers, 12AU7, V8/9 are used to buffer the signals
recovered
before being allowed out of the box to a preamp or other amp.

I found that with the loudest radio station playing, I was getting
a maximumm of 8vrms of audio output voltage.

Stereo separation is determined by careful setting of the tuned circuits

and correct compensation to ensure the gains at audio and the subcarrier
signals
were identical after the CF, V2, which is loaded with the effects of the

notch filter.
I was able to get over 30dB of separation from 20Hz to 8 kHz, abive
which the
sep dropped to about 20dB by 13 kHz, good enough.
I suspect the separation drifts due to the caps I have used across the
tuning coils, and I have to buy a brace of temperature stable caps.
A few degrees of 38khz carrier phase shift relative to the subcarrier
double side band
signal can reduce sep from over 30db to 15dB, which is a lot.
The compensation is achieved by just the right sized cap across the
cathode R of the V1 amp stage, which has 22k as the anode R
and 3.9k as the cathode R, so hence its approx gain of 5.

I may switch to using 12AU7 instead of the 6DJ8, since there is
probably too much gain.

Noise was strange problem after all the bugs were ironed out.

I have an FM stereo miniature transmitter based on the BA1404 chip
running
off a 1.40 V supply, ( which is the right voltage btw ).

I think the standard emphasis circuit which I have used as per the app
schematic
manages to inject some noise, hiss, into the modulation of the supressed
carrier signa
so that when switching from stereo to mono, the hiss with stereo is a
lot greater.
There is also some hum on both stereo and mono, 100Hz, and I don't know
where that's coming from, but it is in the signal from the ratio
detector,
so it isn't from the MPX decoder. Rails are clean as can be.

If i switch off the FM signal gene and allow it to "run on" from the
store of energy
in the caps in its PS then there is no cessation of the hum, which is of
concern
because its 30mV at the tuner output. Hiss remains constant in stereo.
but at least that's only a few mV.
The max AF signals are up to 8Vrms, so the SNR is barely -50dB,
or about as good as vinyl.

Ah, HUM, the Royal Pain in the Arse when you cannot
find any reason for it to exist.

I can't hear any hum in the background of other stations, but then
that's difficult
to hear with a signal going on.


But its sure sounds well compared to an Audio Reflex tuner I have which
uses all chips,
and which has a better
SNR ratio; more like about -60dB.

Later this week i will remove the SE 6BQ5 audio amps in this unit since
I never ever use
them for listening; they are not good enough for realistic levels in my
lounge.

Then i will tweak the PS further, and provide an all DC supply for
each of the 13 tubes that will be left on the chassis.
2 x 6AQ8 for the 100MHz input/oscillatormixer/ AFC stage,
2 x 6BA6 for the 10.7MHz IF,
1 x 6AU6 for the limiter,
and all the 5 tubes tubes above in the MPX unit.
Then the AM section of the tuner which runs separately
from the FM part has 6BE6, 6BA6 for mixer and IF,
then 6AU6 in triode for detector, so quite a lot of tubes all up.

Some folks say diode ring demodulators are inherently noisy.
hence the need to keep the signal level high.
But the 38khz carrier level is about 20Vrms, with a maximum
of 8Vrms of imposed modulation, surely that's a high enough sig level.
It'd have to be a lot higher than the system Quad used 44 years ago
since their 3 transistor
unit used only a 15V supply, thus limiting voltage swings to about
1/3 of what i am using.
But would the type of diodes make a difference?

Quad achieved a lot with just one transistor used for the synchronized
38khz
oscillator.
When there was no 19khz pilot tone as there sometimes isn't with a mono
transmission,
the bias voltages developed by 19khz amplification and stereo operation
ceased to
be generated so DC biased diodes were used automatically
switch from stereo to mono, so the mono signal
was taken direct from the buffered output of the ratio detector.
DC biased diodes and fets are used routinely now to switch all sorts of
things;
they are considered a sonic pest by many,
Quad was the only one to incorporate such a neat trick as feature at
that time.
Not really needed of course, but had the yanks done the same thing
they'd
have used 3 extra tubes and a lot more gear.
Quad's methods were very ingenious.

Patrick Turner.

"I have an FM stereo miniature transmitter based on the BA1404 chip
running
off a 1.40 V supply, ( which is the right voltage btw ).


There is also some hum on both stereo and mono, 100Hz, and I don't know
where that's coming from, but it is in the signal from the ratio
detector,
so it isn't from the MPX decoder. Rails are clean as can be.

If i switch off the FM signal gene and allow it to "run on" from the
store of energy
in the caps in its PS then there is no cessation of the hum

Ah, HUM, the Royal Pain in the Arse when you cannot
find any reason for it to exist.

I can't hear any hum in the background of other stations, but then
that's difficult
to hear with a signal going on. "

Maybe the hum is in the BA1404 chip circuit? Try listening to it with
another good receiver and see if the hum is present or not. Hum is a
big problem with local homebuilt AM "micro" transmitters, usually it's
the RF signal paths being switched from one path to another thru
rectifier diodes in power supplies, causing undesired amplitude
modulation as seen by a receiver.. FM should be less sensitive to
this, but it may still "leak" thru and be heard at the levels you are
talking about. FM receivers do a good job of rejecting AM, but some
still can get thru. Like the 50db down you are getting. The fact
that you still have hum after shutting the receiver's power supply down
but before the filter caps discharge, and that you can't hear it on
radio stations, would make me think it's the BA1404 chip and or
residual AM from the rectifier diodes of the BA1404 chip power supply.
Try passing that power supply's DC output wires thru a ferrite ring
with a few turns bifilar wound to block off the RF from flowing down
those wires to the power supplie's rectifier diodes. That may
eliminate the undesired AM and the hum.

Duh, now I see it "If i switch off the FM signal gene and allow it to
"run on" from the
store of energy
in the caps in its PS then there is no cessation of the hum". For some
reason I read it to say "If i switch off the FM receiver and allow it
to "run on" from the
store of energy
in the caps in its PS then there is no cessation of the hum"

My bad....

Bob wrote:

"I have an FM stereo miniature transmitter based on the BA1404 chip
running
off a 1.40 V supply, ( which is the right voltage btw ).

There is also some hum on both stereo and mono, 100Hz, and I don't know
where that's coming from, but it is in the signal from the ratio
detector,
so it isn't from the MPX decoder. Rails are clean as can be.

If i switch off the FM signal gene and allow it to "run on" from the
store of energy
in the caps in its PS then there is no cessation of the hum

Ah, HUM, the Royal Pain in the Arse when you cannot
find any reason for it to exist.

I can't hear any hum in the background of other stations, but then
that's difficult
to hear with a signal going on. "





Maybe the hum is in the BA1404 chip circuit? Try listening to it with
another good receiver and see if the hum is present or not.

I have to establish this tonight.

The hum tends to dissapear as you tune off the station
before the noise level rises...


Hum is a
big problem with local homebuilt AM "micro" transmitters, usually it's
the RF signal paths being switched from one path to another thru
rectifier diodes in power supplies, causing undesired amplitude
modulation as seen by a receiver.. FM should be less sensitive to
this, but it may still "leak" thru and be heard at the levels you are
talking about.

yes but when I turn off the well shielded PS in the micro mitter box
where the BA1404 is in its own steel box for shielding, the hum remains!



FM receivers do a good job of rejecting AM, but some
still can get thru. Like the 50db down you are getting. The fact
that you still have hum after shutting the receiver's power supply down
but before the filter caps discharge, and that you can't hear it on
radio stations, would make me think it's the BA1404 chip and or
residual AM from the rectifier diodes of the BA1404 chip power supply.
Try passing that power supply's DC output wires thru a ferrite ring
with a few turns bifilar wound to block off the RF from flowing down
those wires to the power supplie's rectifier diodes. That may
eliminate the undesired AM and the hum.

I will have to do some more tests with my other tuner and
amplify its level up by 25db to match the same output levels of the tubed
tuner
to see if the noise problems are the same.

I suspect there is noise entering the BA1404 audio inputs from the series
R used for pre-emphasis, and that noise modulates the subcarrier and hence
there is more
hiss in stereo than when switched to mono, when the SNR is very good.
When listening in mono direct from the ratio detector, the subcarrier info
with the possible high noise in the L-R modulation is all shunted away by
the
de-emphasis RC filter.
The noise may not be affecting the modulation of L+R.
I cannot see how the use of the very selective 38kHz oscillator circuit
and ring diodes could create what sounds like a crook tube or the sound of
resistance noise amplified that is about 12dB above the L+R mono noise
taken direct from the ratio detector.

There is an answer, but finding it is the bother.

There are NO SNR measurements quoted in the data for the BA1404,
so i assume the makers know it was crap.

Patrick Turner.

Bob wrote:

Duh, now I see it "If i switch off the FM signal gene and allow it to
"run on" from the
store of energy
in the caps in its PS then there is no cessation of the hum". For some
reason I read it to say "If i switch off the FM receiver and allow it
to "run on" from the
store of energy
in the caps in its PS then there is no cessation of the hum"

My bad....

I can switch off EITHER the FM sig gene supply, or the FM receiver supply,
and both run on for seconds until the rails sag.
The hum stays the same.

Turning off/on items nearby made no difference.

Patrick Turner.

in article , Bob at
wrote on 11/13/05 7:40 PM:

"I have an FM stereo miniature transmitter based on the BA1404 chip
running
off a 1.40 V supply, ( which is the right voltage btw ).


There is also some hum on both stereo and mono, 100Hz, and I don't know
where that's coming from, but it is in the signal from the ratio
detector,
so it isn't from the MPX decoder. Rails are clean as can be.

If i switch off the FM signal gene and allow it to "run on" from the
store of energy
in the caps in its PS then there is no cessation of the hum

Ah, HUM, the Royal Pain in the Arse when you cannot
find any reason for it to exist.

I can't hear any hum in the background of other stations, but then
that's difficult
to hear with a signal going on. "

Maybe the hum is in the BA1404 chip circuit? Try listening to it with
another good receiver and see if the hum is present or not. Hum is a
big problem with local homebuilt AM "micro" transmitters, usually it's
the RF signal paths being switched from one path to another thru
rectifier diodes in power supplies, causing undesired amplitude
modulation as seen by a receiver.. FM should be less sensitive to
this, but it may still "leak" thru and be heard at the levels you are
talking about. FM receivers do a good job of rejecting AM, but some
still can get thru. Like the 50db down you are getting. The fact
that you still have hum after shutting the receiver's power supply down
but before the filter caps discharge, and that you can't hear it on
radio stations, would make me think it's the BA1404 chip and or
residual AM from the rectifier diodes of the BA1404 chip power supply.
Try passing that power supply's DC output wires thru a ferrite ring
with a few turns bifilar wound to block off the RF from flowing down
those wires to the power supplie's rectifier diodes. That may
eliminate the undesired AM and the hum.



This isn't an "answer" and it isn't very helpful, but the BA1404 is what it
is . . . an inexpensive, mediocre solution designed to run on a single AA
cell. The circuit has a lot of deficiencies (e.g. frequency drift with
temperature).

There are a lot of considerations when using that device. Layout and
shielding are important. I'd try shielding the circuit as best as possible.

A much better candidate for the task is the ROHM FM transmitter series, the
BH14xx. These feature a PLL frequency synthesizer, lower distortion, etc.

I designed and built a portable stereo FM transmitter based upon the
manufacturer's data, which was very comprehensive. It used SMT parts, and I
needed a microscope to prototype it. LAYOUT WAS CRITICAL; I heard all kinds
of spurious crap until it was done correctly. I got a former RF engineer
from Scientific Atlanta to help me with that.

I'd be happy to forward the schematic to anyone interested. I built it and
it does work . . .

Much

Jon Yaeger wrote:

in article , Bob at
wrote on 11/13/05 7:40 PM:

"I have an FM stereo miniature transmitter based on the BA1404 chip
running
off a 1.40 V supply, ( which is the right voltage btw ).


There is also some hum on both stereo and mono, 100Hz, and I don't know
where that's coming from, but it is in the signal from the ratio
detector,
so it isn't from the MPX decoder. Rails are clean as can be.

If i switch off the FM signal gene and allow it to "run on" from the
store of energy
in the caps in its PS then there is no cessation of the hum

Ah, HUM, the Royal Pain in the Arse when you cannot
find any reason for it to exist.

I can't hear any hum in the background of other stations, but then
that's difficult
to hear with a signal going on. "

Maybe the hum is in the BA1404 chip circuit? Try listening to it with
another good receiver and see if the hum is present or not. Hum is a
big problem with local homebuilt AM "micro" transmitters, usually it's
the RF signal paths being switched from one path to another thru
rectifier diodes in power supplies, causing undesired amplitude
modulation as seen by a receiver.. FM should be less sensitive to
this, but it may still "leak" thru and be heard at the levels you are
talking about. FM receivers do a good job of rejecting AM, but some
still can get thru. Like the 50db down you are getting. The fact
that you still have hum after shutting the receiver's power supply down
but before the filter caps discharge, and that you can't hear it on
radio stations, would make me think it's the BA1404 chip and or
residual AM from the rectifier diodes of the BA1404 chip power supply.
Try passing that power supply's DC output wires thru a ferrite ring
with a few turns bifilar wound to block off the RF from flowing down
those wires to the power supplie's rectifier diodes. That may
eliminate the undesired AM and the hum.


This isn't an "answer" and it isn't very helpful, but the BA1404 is what it
is . . . an inexpensive, mediocre solution designed to run on a single AA
cell. The circuit has a lot of deficiencies (e.g. frequency drift with
temperature).

Yes, I am aware of its deficiencies, but today when i carefully compared the
recieved
signal of my new decoder and a fairly noise free Audio Reflex tuner using chips,

I concluded the AR is only 8dB better.
The AR makes 0.15Vrms output with a certain level of modulation on one channel,
and the decoder of mine makes 2.3Vrms, so when i amplified the AR signal
up to match the tube version, then removed the modulation on the test signal,
it was then I found the AR was quieter. I could live with the tube decoder,
but I am a perfectionist, and like low noise if its possible.



There are a lot of considerations when using that device. Layout and
shielding are important. I'd try shielding the circuit as best as possible.

It seems to be OK; the BA1404 is on a board with an RF buffer amp on its output
with filtering for harmonics, and is in a steel metal box. The box is inside a
slightly larger box
which has two modulation oscillators which give 5 different F for each channel.



A much better candidate for the task is the ROHM FM transmitter series, the
BH14xx. These feature a PLL frequency synthesizer, lower distortion, etc.

I designed and built a portable stereo FM transmitter based upon the
manufacturer's data, which was very comprehensive. It used SMT parts, and I
needed a microscope to prototype it. LAYOUT WAS CRITICAL; I heard all kinds
of spurious crap until it was done correctly. I got a former RF engineer
from Scientific Atlanta to help me with that.

But for basic set up and channel sep adjustment, the BA1404 will do.

At the end of the day if better signals come from the broadcast stations then i
should hear it better
once the basic technical set up of the receiver has been established properly.




I'd be happy to forward the schematic to anyone interested. I built it and
it does work . . .

Much

I followed a link from this site to extract the information below,
http://members.tripod.com/~transmitters/index.htm

Patrick Turner.

The copied notes re various chips....


20th Oct '99
My look at the NJM2035 Stereo Multiplexer chip by New Japan Radio
( To be read in conjunction with my "Stereo for Dummies" page )

The Rohm BA1404 tried to do too much on one chip, what with a Stereo
Multiplexer, Pilot Generation Circuitry, RF Oscillator and RF Buffer all on
board.

On top of that, Rohm's Spec Sheets did not show the basic filters on the input,
pilot and MPX that we all know are totally necessary. Low-end Kit manufacturers
did not bother to do a bit of basic design homework themselves. They decided to
do away with such "niceties" and just rake in the money. Most BA1404 kits are
exactly a copy of the Rohm application note with maybe an extra RF stage for
little more power.

BA1404 kits started a rash of FAQ's, newsletters, FTP sites. They all
concentrated on how to increase power or make it a bit more frequency-stable.
Nobody decided to get to the root of the Audio problems - Add the filters.
Wavemach came close by using only the Stereo Generator of the BA1404, adding a
Compressor/Limiter to prevent clipping/aliasing and ignoring the RF parts of the
BA1404.

New Japan Radio's NJM2035 (Specs PDF here) seems to be the nearest contender to
the market addressed by the BA1404. But even though the BA1404 is not being
produced any more for a long time, Low-end Kit manufacturers did not start
making a NJM2035 kit. I think the reason is obvious : The NJM2035 Spec Sheet
does not have a suggested PCB while the BA1404 one did

OK, that's enough venom for a day. Let's get technical again.

Can I just unplug my BA1404 and put in a NJM2035 ??????

No, you cannot. The NJM2035 does not have an RF Oscillator and RF Buffer stage.
Maybe that is wise as now you have no chance of leakage of the MPX signal to the
RF and vice versa. That was one of the internal design flaws of the BA1404.

Is the NJM2035 any good ?????

The NJM2035 is just a Stereo Multiplexor chip using TDM chopping at 38KHz. It
does not use oversampling techniques.

The NJM2035's basic electronics seem to give a channel separation of about 35-40
dB, out of which the suggested MPX filter makes you loose about 5-10 dB in the
3-10 KHz range. Pro equipment would aim at 55dB separation.

The NJM2035's distortion figures are about 0.4% -0.6% in the modulation range
you will want to use it. Pro equipment would aim for distortion much below 0.1%

The PDF claims Signal to Noise figures of around 67 dB. That's quite fair
indeed.

The NJM2035 is made by a respectable company with many other high end Audio
processing chips like DNR, Surround Sound, Digital Equalisers.

So all in all, quite a nice chip for the low end. It will definitely outperform
the BA1404 and might even sound similar to the cheaper 38 KHz TDM switching kits
out there ( But for 20% the cost).

Of course, you should not even remotely consider it for serious broadcasting. It
is a toy, but a much better toy than the BA1404 and maybe some of the other
simpler kits out there. I would say as a rule of thumb, If all you can afford is
1-7W of RF Power, the rest of your Audio equipment would be such that the
NJM2035 would be ideal.

Mods

Don't know if I should call them Mods, as there are no NJM2035 kits out there to
begin with. Anyway, here are some engineering notes -

A) The PDF uses hard to obtain values for the MPX filter. Using SPICE, it's a
matter of minutes to redesign the MPX filter to original specifications, using
more easily available parts. ( A little tad better actually, but what's 0.1 dB
amongst friends)

The Goals are that the new filter should remove harmonics centered around 114
KHz ( Third harmonic of 38 KHz) atleast as good as existing circuit. Also, the
phase shift should be lower/same as existing circuit.

B) The original PDF shows a pre-emphasis stage with very slight roll-off of the
pre-emphasis curve at the high end. We would have loved the pre-emphasis to stop
emphasising signals above 15 KHz. Infact, it would be nice if there was a
brickwall that stopped signals after 15 KHz altogether. It is also nice to notch
out the Audio signals around 19KHz as they interfere with the Pilot Tone and
create beats.

in article , Patrick Turner at
wrote on 11/14/05 12:30 AM:


I followed a link from this site to extract the information below,
http://members.tripod.com/~transmitters/index.htm

Patrick Turner.


One complaint was that the documentation cited lacked filtering info.

Not so with the Rohm "LSI Technical Note" series. For example, the technote
covering the BH1415S/F & BH1416S/F is 37 pages, with lots of info about
filters, sample PCBs with layout requirements, and just about anything you'd
need to know about the devices.

I haven't seen the application info for the BA1404, but I suspect that Rohm
has something comparable in its archives. Contact the local sales office,
and they can probably assist.

Jon

Jon Yaeger wrote:

in article , Patrick Turner at
wrote on 11/14/05 12:30 AM:

I followed a link from this site to extract the information below,
http://members.tripod.com/~transmitters/index.htm

Patrick Turner.

One complaint was that the documentation cited lacked filtering info.

Not so with the Rohm "LSI Technical Note" series. For example, the technote
covering the BH1415S/F & BH1416S/F is 37 pages, with lots of info about
filters, sample PCBs with layout requirements, and just about anything you'd
need to know about the devices.

I haven't seen the application info for the BA1404, but I suspect that Rohm
has something comparable in its archives. Contact the local sales office,
and they can probably assist.

There a pdf page on BA1404 if you search for it.

But this is a superceded chip I think, no longer made i thought.

I'll do a search on the BH14xxxxx.

Thanks for the numbers,

Patrick Turner.


Jon

Hi Patrick,

It only took you 9 vacuum triodes to do what QUAD did with 3 PNP
transistors? I have a stereo radio from 1962 that uses only the two
triodes of a 12AX7 to implement the multiplex decoder, and also uses a
"diode ring demodulator". Also take a look at the stereo decoder used in
the H.H.Scott 345 receiver, which has a tube decoder circuit virtually
identical to the QUAD decoder topology right down to the "diode ring
demodulator", and it uses only a single pentode and a single triode in the
decoder proper, with another triode used simply to control a neon stereo
indicator light based on the output of the 19 kHz "rectifier"/doubler
circuit. The pentode and the two triodes are all contained in a single
6N11 compactron tube. The schematic is available on the web.

I have four questions about your changes to the QUAD circuit.

1.) When will we have a schematic?

2.) At several points in the description of your circuit you refer to a
"19 kHz Notch filter", this is a feature that the original QUAD design
didn't include. What topology did you use to implement the "19 kHz notch
filter"?

3.) The original QUAD design included a low pass filter with a 55 kHz
cutoff at the input to the demodulator "ring", does your circuit include
this low pass filter, or did you leave it out? This filter doesn't just
eliminate potential SCA beats, but is important to minimize the hiss and
noise on the stereo audio outputs even when SCA sub carriers aren't being
used. Without low pass filtering, filtering noise at the output of the
ratio detector around 114 kHz will be translated down to the audio range,
degrading the stereo signal to noise ratio more than need be. This noise
can be the result of weak signals, or it can be due to interference from
distant adjacent channel signals, especially in areas that use 100 kHz
channel spacing in the FM band.

4.) It isn't clear how you are compensating for the fact that the fact
that the L-R sub carrier signal is effectively transmitted at a lower
level than is the main channel L+R signal? QUAD used an auxiliary matrix
to perform this function, feeding inverted L+R audio into the outputs of
the decoder to cancel some of the L+R signal. There are at least two
other was to perform this function, are you following the QUAD lead, or is
the "compensation" capacitor in the cathode circuit of the input amplifier
a crude solution to this problem.?

Did the folks that "say diode ring demodulators are inherently noisy" give
any explanation for this claim? Diode rings are frequently used as low
noise RF mixers, so this claim seems counter intuitive. What kind of
demodulators did they suggest might have lower noise than a diode ring?

QUAD was not the only manufacturer to use this sort of diode ring
demodulator. This type of "diode ring" was the standard topology almost
universally used in discrete component stereo decoders, both tube and
transistor. Even Scott switched to this design in their later tube
decoders as mentioned above, and I would assume they carried it through to
their discrete component transistor designs too.

You mentioned that the QUAD decoder circuit uses four additional diodes to
switch between the monophonic and stereo reception modes. I can see
circuitry that looks like it automatically controls the diodes in the
stereo audio path to interrupt that path when a 19 kHz stereo pilot tone
is not being received. I also see the two diodes in the monophonic audio
path, but I can't see how they are automatically controlled? It looks to
me like you have to manually turn off one of the QUAD II power amplifiers
to get a monophonic output from the decoder when a stereo signal is not
being broadcast. Can you explain what I am missing here?


Regards,

John Byrns


In article , Patrick Turner
wrote:

Some of you were following my progress with re-designing & rewiring
of the stereo multitplex decoder I have within an ancient
Trio receiver.

I finally settled on a design using all tubes that resembles the
method of operation of the first Quad add on MPX unit with 3 transistors

except that I have all triodes.

After the ratio detector, there is a V1 6DJ8 amp which amplifies the
signal
from the RD by about 5 times, then buffers its anode output with a
direct coupled CF, V2.
From this V2CF, a feed to a pair of parallel 19kHz tuned circuits
filters out the 19kHz and V3 triode amplifies it and
powers a tuned 19kHz tranny which has a CT secondary and two diodes
then convert the 19kHz into its full wave rectified form which is mainly
38kHz.
This signal is used to synchronize a 38kHz oscillator with 2 triodes
very like the type
Scott used but with a 6CG7, V4/5.
The oscillator is a PP type, and the double tuned 38kHz tranny
has two CT windings, pri for the anodes, sec for the connection via its
CT to the
composite signal from V2CF, minus the 19kHz pilot tone.
A ring diode demodulator almost identical to the design used by Quad is
then employed
with 4 x 27k R and 4 x 1N9148 diodes. A pair 10k R off the two diode
junction
outputs charges a pair of 900pF caps, one for each channel.
These RC values gave the right amount of de-emphasis.
The signals from each channel are then fed to a first order RC filter to

help remove switching noise, then fed to CF buffers, 12AU7, V6/7,
which drive LC and RC 3rd order filters to further remove any switching
artifacts.

The L&R signals are taken to switch contacts to select stereo or mono.

The mono signal is derived by de-emphasising the signal from the
V2CF after the 19kHz notch filter with a simple RC filter, then to a
switch.

Another pair of CF buffers, 12AU7, V8/9 are used to buffer the signals
recovered
before being allowed out of the box to a preamp or other amp.

I found that with the loudest radio station playing, I was getting
a maximumm of 8vrms of audio output voltage.

Stereo separation is determined by careful setting of the tuned circuits

and correct compensation to ensure the gains at audio and the subcarrier
signals
were identical after the CF, V2, which is loaded with the effects of the

notch filter.
I was able to get over 30dB of separation from 20Hz to 8 kHz, abive
which the
sep dropped to about 20dB by 13 kHz, good enough.
I suspect the separation drifts due to the caps I have used across the
tuning coils, and I have to buy a brace of temperature stable caps.
A few degrees of 38khz carrier phase shift relative to the subcarrier
double side band
signal can reduce sep from over 30db to 15dB, which is a lot.
The compensation is achieved by just the right sized cap across the
cathode R of the V1 amp stage, which has 22k as the anode R
and 3.9k as the cathode R, so hence its approx gain of 5.

I may switch to using 12AU7 instead of the 6DJ8, since there is
probably too much gain.

Noise was strange problem after all the bugs were ironed out.

I have an FM stereo miniature transmitter based on the BA1404 chip
running
off a 1.40 V supply, ( which is the right voltage btw ).

I think the standard emphasis circuit which I have used as per the app
schematic
manages to inject some noise, hiss, into the modulation of the supressed
carrier signa
so that when switching from stereo to mono, the hiss with stereo is a
lot greater.
There is also some hum on both stereo and mono, 100Hz, and I don't know
where that's coming from, but it is in the signal from the ratio
detector,
so it isn't from the MPX decoder. Rails are clean as can be.

If i switch off the FM signal gene and allow it to "run on" from the
store of energy
in the caps in its PS then there is no cessation of the hum, which is of
concern
because its 30mV at the tuner output. Hiss remains constant in stereo.
but at least that's only a few mV.
The max AF signals are up to 8Vrms, so the SNR is barely -50dB,
or about as good as vinyl.

Ah, HUM, the Royal Pain in the Arse when you cannot
find any reason for it to exist.

I can't hear any hum in the background of other stations, but then
that's difficult
to hear with a signal going on.


But its sure sounds well compared to an Audio Reflex tuner I have which
uses all chips,
and which has a better
SNR ratio; more like about -60dB.

Later this week i will remove the SE 6BQ5 audio amps in this unit since
I never ever use
them for listening; they are not good enough for realistic levels in my
lounge.

Then i will tweak the PS further, and provide an all DC supply for
each of the 13 tubes that will be left on the chassis.
2 x 6AQ8 for the 100MHz input/oscillatormixer/ AFC stage,
2 x 6BA6 for the 10.7MHz IF,
1 x 6AU6 for the limiter,
and all the 5 tubes tubes above in the MPX unit.
Then the AM section of the tuner which runs separately
from the FM part has 6BE6, 6BA6 for mixer and IF,
then 6AU6 in triode for detector, so quite a lot of tubes all up.

Some folks say diode ring demodulators are inherently noisy.
hence the need to keep the signal level high.
But the 38khz carrier level is about 20Vrms, with a maximum
of 8Vrms of imposed modulation, surely that's a high enough sig level.
It'd have to be a lot higher than the system Quad used 44 years ago
since their 3 transistor
unit used only a 15V supply, thus limiting voltage swings to about
1/3 of what i am using.
But would the type of diodes make a difference?

Quad achieved a lot with just one transistor used for the synchronized
38khz
oscillator.
When there was no 19khz pilot tone as there sometimes isn't with a mono
transmission,
the bias voltages developed by 19khz amplification and stereo operation
ceased to
be generated so DC biased diodes were used automatically
switch from stereo to mono, so the mono signal
was taken direct from the buffered output of the ratio detector.
DC biased diodes and fets are used routinely now to switch all sorts of
things;
they are considered a sonic pest by many,
Quad was the only one to incorporate such a neat trick as feature at
that time.
Not really needed of course, but had the yanks done the same thing
they'd
have used 3 extra tubes and a lot more gear.
Quad's methods were very ingenious.

Patrick Turner.


Surf my web pages at, http://users.rcn.com/jbyrns/

John Byrns wrote:

Hi Patrick,

It only took you 9 vacuum triodes to do what QUAD did with 3 PNP
transistors? I have a stereo radio from 1962 that uses only the two
triodes of a 12AX7 to implement the multiplex decoder, and also uses a
"diode ring demodulator". Also take a look at the stereo decoder used in
the H.H.Scott 345 receiver, which has a tube decoder circuit virtually
identical to the QUAD decoder topology right down to the "diode ring
demodulator", and it uses only a single pentode and a single triode in the
decoder proper, with another triode used simply to control a neon stereo
indicator light based on the output of the 19 kHz "rectifier"/doubler
circuit. The pentode and the two triodes are all contained in a single
6N11 compactron tube. The schematic is available on the web.

The 3 transistor decoders were not wonderful sonic performers,
and this is true of many early decoders.
Even the RCA type shown in the RCA tube manuals is hopeless, and it uses a few
tubes.

But i have a gain triode to raise the levels of the signal from the ratio
detector
so that the later detection in the ring demod is well above the diode turn on
threshold voltage,
so distortion from dioes is kept low.

This is buffered by a triode CF which many decoders leave out.

Then I have a selective 19khz filter with triode tuned amp to sift out the
19khz free
of any stray audio products which do occur on some stations.

two triodes are used in the locked oscillator, like Scott's version, because
such an oscillator is fairly clean of harmonics and it has two grid inputs one
of
which is easily driven by a doubled 19khz for very good locking
even on a tiny 19kHz signal.

There are 4 triodes used as cathode followers for buffering the output from
the
ring demod, and powering LC filters to remove the 38kHz switching artifacts,
and to then buffer the final outputs after the selector switch for AM, FM
mono, FM stereo.







I have four questions about your changes to the QUAD circuit.

1.) When will we have a schematic?

Very soon.

I just ripped out the two 4 watt audio amps out of the chassis because I want
this unit to be a
dedicated tuner only. I am converting the heaters to all DC excet for the AM
section.





2.) At several points in the description of your circuit you refer to a
"19 kHz Notch filter", this is a feature that the original QUAD design
didn't include. What topology did you use to implement the "19 kHz notch
filter"?

It is a bridged T LC notch filter.

It has tunable coil of nominally 25mH.

Then there are two 0.0062uF caps in series across the coil to make a parallel
tuned LC circuit.

The input is from a CF at one side of the tuned circuit, and the output from
the other.
From the join of the two caps, ie, the centre tap of the capacitance, I have
11.5k to 0V.
This resistance must be adjusted for the best null; and -60dB is possible.
The L's core is adjusted for the best null at 19kHz.
The BW of the null is about 1.5kHz, with negligible phase shift
occuring away from the null.

Thus the 19khz is filtered from the composite signal *before* the
composite signal is added to a carrier and applied to the
diode ring.
So the L&R outputs are free of 19kHz signals, and no IMD occurs in the
diode ring between 19khz and other F.

3.) The original QUAD design included a low pass filter with a 55 kHz

cutoff at the input to the demodulator "ring", does your circuit include
this low pass filter, or did you leave it out? This filter doesn't just
eliminate potential SCA beats, but is important to minimize the hiss and
noise on the stereo audio outputs even when SCA sub carriers aren't being
used. Without low pass filtering, filtering noise at the output of the
ratio detector around 114 kHz will be translated down to the audio range,
degrading the stereo signal to noise ratio more than need be. This noise
can be the result of weak signals, or it can be due to interference from
distant adjacent channel signals, especially in areas that use 100 kHz
channel spacing in the FM band.

Good that you have mentioned the 55kHz filter.
When I looked at the Quad decoder I didn't spot this filter.
I will try such a filter to see if it reduces the much larger hiss level in
stereo
that I am getting compared to mono, even when no modulation is applied
at my transmitter.





4.) It isn't clear how you are compensating for the fact that the fact
that the L-R sub carrier signal is effectively transmitted at a lower
level than is the main channel L+R signal? QUAD used an auxiliary matrix
to perform this function, feeding inverted L+R audio into the outputs of
the decoder to cancel some of the L+R signal.

Indeed you are correct about the Quad additional matrix.

But I saw that as a compensation network because separation levels of L and R
modulation
recovered tend to be poor because the amplifying of the ratio detector signals

tends to lessen separation; the composite signal envelope is mauled by the
RC couplings and gain variations between the audio frequencies and the
subcarrier F.


There are at least two
other was to perform this function, are you following the QUAD lead, or is
the "compensation" capacitor in the cathode circuit of the input amplifier
a crude solution to this problem.?

The compensation cap isn't as crude as you think; it tweaks the phase of the
supressed carrier double sideband signal and slightly increase the gain at
around 38kHz.

Then you should find the separtaion from the diode ring are either L or R,
without any need for Quad's method of compensation.

This is understood if you just modulate the FM signal on R audio channel only
with nothing on the L channel.
The magnitude of the L+R signal = that of the modulation carried by the DSB
signal,
which is L-R.
The signal applied to one side of the diode ring will then appear to be
an AM 38khz carrier but which has the top of the envelope modulated with the R
channel
signal, and the bottom should be the L channel modulation, so it should be a
flat line of mod
because there is no L channel mod in tis test case.

The other side of the diode ring has an inverted version of the other side,
flat line on top, R mod on the bottom.

The levels of L-R transmitted at the station must be fixed to be the same for
all stations for all tuners to give the same separation.



Did the folks that "say diode ring demodulators are inherently noisy" give
any explanation for this claim? Diode rings are frequently used as low
noise RF mixers, so this claim seems counter intuitive. What kind of
demodulators did they suggest might have lower noise than a diode ring?

Multi element tubes like 6BE6 etc were regarded as very noisy.
Triode mixers were quieter.
But to make diode mixers quiet, you needed to have a fair sized signal.
But I could be wrong because I have not much experience with RF
mixers.

I think the extra noise in stereo that I get compared to mono could be
because the BA1404 is just plain noisy, and with no audio modulation,
there is just the noise of the chip modulating the stereo information.
The BA1404 works of only 1.4V supply, so the levels of internal signals must
be quite tiny
compared to what noise is present.
Its really a "toy" of a chip, but 6 years ago its all I could find that was
cheap,
and a radio station here used it in some of their modules, and I bought one
from a dude at the station because it had an RF amp buffer and filter on it
so the signal was large and antenna connection didn't sway the
approx 100Mhz oscillator F.





QUAD was not the only manufacturer to use this sort of diode ring
demodulator. This type of "diode ring" was the standard topology almost
universally used in discrete component stereo decoders, both tube and
transistor. Even Scott switched to this design in their later tube
decoders as mentioned above, and I would assume they carried it through to
their discrete component transistor designs too.

Scott used a much more complex array of Si diodes in their tubed decoders.
Si diodes had become cheap and reliable by that time.
The Scott method means that the 38khz oscillator signal from a single tuned
oscillator tank coil
can be applied by RC coupling, rather than having a pair of tuned circuits
to add to the composite signal so one of the pair has the composite signal
applied to the CT
of the LC circuit to get the summing and subtraction of L+R and L-R done
before diode
+C demodulation.




You mentioned that the QUAD decoder circuit uses four additional diodes to
switch between the monophonic and stereo reception modes. I can see
circuitry that looks like it automatically controls the diodes in the
stereo audio path to interrupt that path when a 19 kHz stereo pilot tone
is not being received. I also see the two diodes in the monophonic audio
path, but I can't see how they are automatically controlled?

Look at the bias voltages for the diodes.
This is varied automatically in such a way so that the diodes either conduct
when forward
bias, or don't conduct when reverse biased.

It looks to
me like you have to manually turn off one of the QUAD II power amplifiers
to get a monophonic output from the decoder when a stereo signal is not
being broadcast. Can you explain what I am missing here?

That I cannot answer.

But to me it looked like an automatic DC signal was sent to the diodes to get
a mono signal for both L&R outputs when using the stereo decoder.
Not all stations transmitted stereo. At the 22Control unit you could make the
selection
of whether you wanted pure mono and thus only one amp and one speaker
operated,
or whether you had the same signal in both amps and both speakers.

Today, ppl wouldn't bother just using 1/2 their system for mono recordings or
broadcast;
they use both speakers all the time including for mono.

Patrick Turner.



Regards,

John Byrns

In article , Patrick Turner
wrote:

Some of you were following my progress with re-designing & rewiring
of the stereo multitplex decoder I have within an ancient
Trio receiver.

I finally settled on a design using all tubes that resembles the
method of operation of the first Quad add on MPX unit with 3 transistors

except that I have all triodes.

After the ratio detector, there is a V1 6DJ8 amp which amplifies the
signal
from the RD by about 5 times, then buffers its anode output with a
direct coupled CF, V2.
From this V2CF, a feed to a pair of parallel 19kHz tuned circuits
filters out the 19kHz and V3 triode amplifies it and
powers a tuned 19kHz tranny which has a CT secondary and two diodes
then convert the 19kHz into its full wave rectified form which is mainly
38kHz.
This signal is used to synchronize a 38kHz oscillator with 2 triodes
very like the type
Scott used but with a 6CG7, V4/5.
The oscillator is a PP type, and the double tuned 38kHz tranny
has two CT windings, pri for the anodes, sec for the connection via its
CT to the
composite signal from V2CF, minus the 19kHz pilot tone.
A ring diode demodulator almost identical to the design used by Quad is
then employed
with 4 x 27k R and 4 x 1N9148 diodes. A pair 10k R off the two diode
junction
outputs charges a pair of 900pF caps, one for each channel.
These RC values gave the right amount of de-emphasis.
The signals from each channel are then fed to a first order RC filter to

help remove switching noise, then fed to CF buffers, 12AU7, V6/7,
which drive LC and RC 3rd order filters to further remove any switching
artifacts.

The L&R signals are taken to switch contacts to select stereo or mono.

The mono signal is derived by de-emphasising the signal from the
V2CF after the 19kHz notch filter with a simple RC filter, then to a
switch.

Another pair of CF buffers, 12AU7, V8/9 are used to buffer the signals
recovered
before being allowed out of the box to a preamp or other amp.

I found that with the loudest radio station playing, I was getting
a maximumm of 8vrms of audio output voltage.

Stereo separation is determined by careful setting of the tuned circuits

and correct compensation to ensure the gains at audio and the subcarrier
signals
were identical after the CF, V2, which is loaded with the effects of the

notch filter.
I was able to get over 30dB of separation from 20Hz to 8 kHz, abive
which the
sep dropped to about 20dB by 13 kHz, good enough.
I suspect the separation drifts due to the caps I have used across the
tuning coils, and I have to buy a brace of temperature stable caps.
A few degrees of 38khz carrier phase shift relative to the subcarrier
double side band
signal can reduce sep from over 30db to 15dB, which is a lot.
The compensation is achieved by just the right sized cap across the
cathode R of the V1 amp stage, which has 22k as the anode R
and 3.9k as the cathode R, so hence its approx gain of 5.

I may switch to using 12AU7 instead of the 6DJ8, since there is
probably too much gain.

Noise was strange problem after all the bugs were ironed out.

I have an FM stereo miniature transmitter based on the BA1404 chip
running
off a 1.40 V supply, ( which is the right voltage btw ).

I think the standard emphasis circuit which I have used as per the app
schematic
manages to inject some noise, hiss, into the modulation of the supressed
carrier signa
so that when switching from stereo to mono, the hiss with stereo is a
lot greater.
There is also some hum on both stereo and mono, 100Hz, and I don't know
where that's coming from, but it is in the signal from the ratio
detector,
so it isn't from the MPX decoder. Rails are clean as can be.

If i switch off the FM signal gene and allow it to "run on" from the
store of energy
in the caps in its PS then there is no cessation of the hum, which is of
concern
because its 30mV at the tuner output. Hiss remains constant in stereo.
but at least that's only a few mV.
The max AF signals are up to 8Vrms, so the SNR is barely -50dB,
or about as good as vinyl.

Ah, HUM, the Royal Pain in the Arse when you cannot
find any reason for it to exist.

I can't hear any hum in the background of other stations, but then
that's difficult
to hear with a signal going on.


But its sure sounds well compared to an Audio Reflex tuner I have which
uses all chips,
and which has a better
SNR ratio; more like about -60dB.

Later this week i will remove the SE 6BQ5 audio amps in this unit since
I never ever use
them for listening; they are not good enough for realistic levels in my
lounge.

Then i will tweak the PS further, and provide an all DC supply for
each of the 13 tubes that will be left on the chassis.
2 x 6AQ8 for the 100MHz input/oscillatormixer/ AFC stage,
2 x 6BA6 for the 10.7MHz IF,
1 x 6AU6 for the limiter,
and all the 5 tubes tubes above in the MPX unit.
Then the AM section of the tuner which runs separately
from the FM part has 6BE6, 6BA6 for mixer and IF,
then 6AU6 in triode for detector, so quite a lot of tubes all up.

Some folks say diode ring demodulators are inherently noisy.
hence the need to keep the signal level high.
But the 38khz carrier level is about 20Vrms, with a maximum
of 8Vrms of imposed modulation, surely that's a high enough sig level.
It'd have to be a lot higher than the system Quad used 44 years ago
since their 3 transistor
unit used only a 15V supply, thus limiting voltage swings to about
1/3 of what i am using.
But would the type of diodes make a difference?

Quad achieved a lot with just one transistor used for the synchronized
38khz
oscillator.
When there was no 19khz pilot tone as there sometimes isn't with a mono
transmission,
the bias voltages developed by 19khz amplification and stereo operation
ceased to
be generated so DC biased diodes were used automatically
switch from stereo to mono, so the mono signal
was taken direct from the buffered output of the ratio detector.
DC biased diodes and fets are used routinely now to switch all sorts of
things;
they are considered a sonic pest by many,
Quad was the only one to incorporate such a neat trick as feature at
that time.
Not really needed of course, but had the yanks done the same thing
they'd
have used 3 extra tubes and a lot more gear.
Quad's methods were very ingenious.

Patrick Turner.

Surf my web pages at, http://users.rcn.com/jbyrns/

In article , Patrick Turner
wrote:

But i have a gain triode to raise the levels of the signal from the ratio
detector
so that the later detection in the ring demod is well above the diode turn on
threshold voltage,
so distortion from dioes is kept low.

This statement doesn't make sense, if anything I would expect the opposite
to be the case. What is important for low distortion is to be sure that
the locally regenerated 38 kHz sub carrier applied to the diodes is large
in amplitude. The signal applied to the diodes from the ratio detector
should be lower in amplitude relative to the local 38 kHz sub carrier that
is applied to the diodes.

2.) At several points in the description of your circuit you refer to a
"19 kHz Notch filter", this is a feature that the original QUAD design
didn't include. What topology did you use to implement the "19 kHz notch
filter"?

It is a bridged T LC notch filter.

It has tunable coil of nominally 25mH.

Then there are two 0.0062uF caps in series across the coil to make a parallel
tuned LC circuit.

The input is from a CF at one side of the tuned circuit, and the output from
the other.
From the join of the two caps, ie, the centre tap of the capacitance, I have
11.5k to 0V.
This resistance must be adjusted for the best null; and -60dB is possible.
The L's core is adjusted for the best null at 19kHz.
The BW of the null is about 1.5kHz, with negligible phase shift
occuring away from the null.

Thus the 19khz is filtered from the composite signal *before* the
composite signal is added to a carrier and applied to the
diode ring.
So the L&R outputs are free of 19kHz signals, and no IMD occurs in the
diode ring between 19khz and other F.

Thanks for the description of your 19 kHz pilot filter.

3.) The original QUAD design included a low pass filter with a 55 kHz

cutoff at the input to the demodulator "ring", does your circuit include
this low pass filter, or did you leave it out? This filter doesn't just
eliminate potential SCA beats, but is important to minimize the hiss and
noise on the stereo audio outputs even when SCA sub carriers aren't being
used. Without low pass filtering, filtering noise at the output of the
ratio detector around 114 kHz will be translated down to the audio range,
degrading the stereo signal to noise ratio more than need be. This noise
can be the result of weak signals, or it can be due to interference from
distant adjacent channel signals, especially in areas that use 100 kHz
channel spacing in the FM band.

Good that you have mentioned the 55kHz filter.
When I looked at the Quad decoder I didn't spot this filter.
I will try such a filter to see if it reduces the much larger hiss level in
stereo
that I am getting compared to mono, even when no modulation is applied
at my transmitter.

You might want to look into combining your 19 kHz pilot filter with the
low pass filter as was done in the Marantz 10B. A former Marantz employee
said that with proper design the phase shifts introduced by the 19 kHz
pilot filter can be used to compensate for the phase shifts introduced by
the low pass filter into the upper sidebands of the 38 kHz DSBSC sub
carrier.

4.) It isn't clear how you are compensating for the fact that the fact
that the L-R sub carrier signal is effectively transmitted at a lower
level than is the main channel L+R signal? QUAD used an auxiliary matrix
to perform this function, feeding inverted L+R audio into the outputs of
the decoder to cancel some of the L+R signal.

Indeed you are correct about the Quad additional matrix.

But I saw that as a compensation network because separation levels of L and R
modulation
recovered tend to be poor because the amplifying of the ratio detector signals

tends to lessen separation; the composite signal envelope is mauled by the
RC couplings and gain variations between the audio frequencies and the
subcarrier F.


There are at least two
other was to perform this function, are you following the QUAD lead, or is
the "compensation" capacitor in the cathode circuit of the input amplifier
a crude solution to this problem.?

The compensation cap isn't as crude as you think; it tweaks the phase of the
supressed carrier double sideband signal and slightly increase the gain at
around 38kHz.

It is a crude approach to equalizing the differing levels of the L+R and
38 kHz L-R signals.

Then you should find the separtaion from the diode ring are either L or R,
without any need for Quad's method of compensation.

This is understood if you just modulate the FM signal on R audio channel only
with nothing on the L channel.
The magnitude of the L+R signal = that of the modulation carried by the DSB
signal,
which is L-R.
The signal applied to one side of the diode ring will then appear to be
an AM 38khz carrier but which has the top of the envelope modulated with the R
channel
signal, and the bottom should be the L channel modulation, so it should be a
flat line of mod
because there is no L channel mod in tis test case.

The other side of the diode ring has an inverted version of the other side,
flat line on top, R mod on the bottom.

The levels of L-R transmitted at the station must be fixed to be the same for
all stations for all tuners to give the same separation.

The point you are missing is that the peak amplitudes of the L+R audio and
the 38 kHz sub carrier are the same, but the demodulated L-R audio is
lower in level than the L+R audio after it is multiplied by the 38 kHz
switching function. The math to show this is relatively simple.

The reasonable separation you are achieving may be due to some combination
of two factors.
1.) Depending on how you are adjusting your "compensation" capacitor in
the cathode circuit of the input amplifier, you may be considerably
increasing the L-R 38 kHz level relative to the L+R signal level, not just
tweaking the phase a little. Can you post the value of your
"compensation" capacitor, I think you already posted the tube type and the
values of the cathode and plate resistors?
2.) The BA1404 signal generator may be generating a nonstandard stereo
signal that helps to compensate for separation problems in your stereo
decoder. I looked at the BA1404 data sheet many years ago and came to the
conclusion that it had a problem like this although on reflection I now
realize that they may have included a simple fix that might not be shown
in either the block diagrams or the specifications, I will have to take
another look at the BA1404 data sheet and see if I can tell what the true
situation is.

QUAD was not the only manufacturer to use this sort of diode ring
demodulator. This type of "diode ring" was the standard topology almost
universally used in discrete component stereo decoders, both tube and
transistor. Even Scott switched to this design in their later tube
decoders as mentioned above, and I would assume they carried it through to
their discrete component transistor designs too.

Scott used a much more complex array of Si diodes in their tubed decoders.
Si diodes had become cheap and reliable by that time.
The Scott method means that the 38khz oscillator signal from a single tuned
oscillator tank coil
can be applied by RC coupling, rather than having a pair of tuned circuits
to add to the composite signal so one of the pair has the composite signal
applied to the CT
of the LC circuit to get the summing and subtraction of L+R and L-R done
before diode
+C demodulation.

What you are describing is the design used in Scott's early decoders,
check this link for a later Scott tube decoder design:

http://hhscott.com/pdf/345.pdf

This shows one of Scott's later tube decoder designs which uses only a
single pentode and a single triode, along with the same type of "diode
ring" that QUAD used. As I said this "diode ring" approach was virtually
universal in later designs.

You mentioned that the QUAD decoder circuit uses four additional diodes to
switch between the monophonic and stereo reception modes. I can see
circuitry that looks like it automatically controls the diodes in the
stereo audio path to interrupt that path when a 19 kHz stereo pilot tone
is not being received. I also see the two diodes in the monophonic audio
path, but I can't see how they are automatically controlled?

Look at the bias voltages for the diodes.
This is varied automatically in such a way so that the diodes either conduct
when forward
bias, or don't conduct when reverse biased.

Yes, as I said before it is reasonably clear how the stereo signal is
automatically switched off, how the mono signal is automatically switched
on is not at all clear, at least to me.

It looks to
me like you have to manually turn off one of the QUAD II power amplifiers
to get a monophonic output from the decoder when a stereo signal is not
being broadcast. Can you explain what I am missing here?

That I cannot answer.

But to me it looked like an automatic DC signal was sent to the diodes to get
a mono signal for both L&R outputs when using the stereo decoder.

This is the part I can't follow, I don't see where an automatically
generated DC switching signal is applied to the diodes that switch the
mono signal on?


Regards,

John Byrns


Surf my web pages at, http://users.rcn.com/jbyrns/

John Byrns wrote:

In article , Patrick Turner
wrote:

But i have a gain triode to raise the levels of the signal from the ratio
detector
so that the later detection in the ring demod is well above the diode turn on
threshold voltage,
so distortion from dioes is kept low.

This statement doesn't make sense, if anything I would expect the opposite
to be the case.

Some moderate raise of the fairly low signal voltage from the ratio detector
does no harm.

What is important for low distortion is to be sure that
the locally regenerated 38 kHz sub carrier applied to the diodes is large
in amplitude.

20Vrms on each side of the diode-resistor matrix.

The signal applied to the diodes from the ratio detector
should be lower in amplitude relative to the local 38 kHz sub carrier that
is applied to the diodes.

It is lower, lest one conceivably have "over modulation of the envelope",
and serious thd.

I looked at the envelopes with high power stations and it was ok.

But some gain reduction of the composite signal amp from 5 to about
2.5 would probably be ideal, so some further tinkering....



2.) At several points in the description of your circuit you refer to a
"19 kHz Notch filter", this is a feature that the original QUAD design
didn't include. What topology did you use to implement the "19 kHz notch
filter"?

It is a bridged T LC notch filter.

It has tunable coil of nominally 25mH.

Then there are two 0.0062uF caps in series across the coil to make a parallel
tuned LC circuit.

The input is from a CF at one side of the tuned circuit, and the output from
the other.
From the join of the two caps, ie, the centre tap of the capacitance, I have
11.5k to 0V.
This resistance must be adjusted for the best null; and -60dB is possible.
The L's core is adjusted for the best null at 19kHz.
The BW of the null is about 1.5kHz, with negligible phase shift
occuring away from the null.

Thus the 19khz is filtered from the composite signal *before* the
composite signal is added to a carrier and applied to the
diode ring.

Yes.

I saw the filter in the Quad circuit and i don't know what F it is for, 55khz, or
19kHz,
since there is no 19kHz filter indicated.


So the L&R outputs are free of 19kHz signals, and no IMD occurs in the
diode ring between 19khz and other F.

Thanks for the description of your 19 kHz pilot filter.

3.) The original QUAD design included a low pass filter with a 55 kHz

cutoff at the input to the demodulator "ring", does your circuit include
this low pass filter, or did you leave it out? This filter doesn't just
eliminate potential SCA beats, but is important to minimize the hiss and
noise on the stereo audio outputs even when SCA sub carriers aren't being
used. Without low pass filtering, filtering noise at the output of the
ratio detector around 114 kHz will be translated down to the audio range,
degrading the stereo signal to noise ratio more than need be. This noise
can be the result of weak signals, or it can be due to interference from
distant adjacent channel signals, especially in areas that use 100 kHz
channel spacing in the FM band.

Good that you have mentioned the 55kHz filter.
When I looked at the Quad decoder I didn't spot this filter.
I will try such a filter to see if it reduces the much larger hiss level in
stereo
that I am getting compared to mono, even when no modulation is applied
at my transmitter.

You might want to look into combining your 19 kHz pilot filter with the
low pass filter as was done in the Marantz 10B. A former Marantz employee
said that with proper design the phase shifts introduced by the 19 kHz
pilot filter can be used to compensate for the phase shifts introduced by
the low pass filter into the upper sidebands of the 38 kHz DSBSC sub
carrier.

But filtering to separate the L+R from the DSB is not required.
Scott also don't separate the composite.

The whole idea of the Quad circuit and some othr similar ones, and including
the Scott circuits is to keep all the components of the composite signal together,
so that good separation is possible across a wide band.
The basic idea is to simply add the missing 38khz carrier to the existing composite
from the
ratio detector, or whatever detector is used.

The phase of the carrier relative to the subcarrier waves is easily
swayed to line up exactly.
Any filters anywhere affect the line up, and thus the RCA original as it appears in
the
RCA tube handbook is an awful MPX unit.



Any attempt to filter out the DSB and thus gain the L-R signal so that it may be
applied
to LPF filtered phases of L+R, and -L-R results in poor separation at HF due to
inevitable phase shifts caused by what must be more than first orer filters.
I tried all this in many efforts but always the phase shifts prevented more than
12dB sep at 10kHz, and
almost no sep at 13kHz.

The phase effects of the 19kHz notch filter between 23kHz and 53 kHz and between
20Hz and 15kHz is utterly negligible.

It can be shunted out and all that does is in my similar to Quad circuit is dump
a whole lot of 19khz in each L&R output, plus probable IMD products.

Much of what we listen to, especially the HF portions, is at a level below the
pilot tone levels.
IN my case I am able to banaish the damn pilot tone after a cathode follower, and
thus
drive drive the diodes with a low impedance.
There *is* a small effect of the 19kHz filter elements after the follower
but that is easily compensated for with a cap on the cathode R of the gain before
the CF.
We are not looking for much compensation, about 3dB max.



4.) It isn't clear how you are compensating for the fact that the fact
that the L-R sub carrier signal is effectively transmitted at a lower
level than is the main channel L+R signal? QUAD used an auxiliary matrix
to perform this function, feeding inverted L+R audio into the outputs of
the decoder to cancel some of the L+R signal.

Indeed you are correct about the Quad additional matrix.

But I saw that as a compensation network because separation levels of L and R
modulation
recovered tend to be poor because the amplifying of the ratio detector signals

tends to lessen separation; the composite signal envelope is mauled by the
RC couplings and gain variations between the audio frequencies and the
subcarrier F.


There are at least two
other was to perform this function, are you following the QUAD lead, or is
the "compensation" capacitor in the cathode circuit of the input amplifier
a crude solution to this problem.?

The compensation cap isn't as crude as you think; it tweaks the phase of the
supressed carrier double sideband signal and slightly increase the gain at
around 38kHz.

It is a crude approach to equalizing the differing levels of the L+R and
38 kHz L-R signals.

Well my system works very well to give over 30db of separation, and is simple.



Then you should find the separtaion from the diode ring are either L or R,
without any need for Quad's method of compensation.

The diode ring seems to cause some loss of separation and there would only be
perfect
sep if there was perfect detection, as theory suggests it should happen, but it just
doesn't occur.
The CF and the oscillator have to provide the power to the matrix, and
the some losses seem to occur.



This is understood if you just modulate the FM signal on R audio channel only
with nothing on the L channel.
The magnitude of the L+R signal = that of the modulation carried by the DSB
signal,
which is L-R.
The signal applied to one side of the diode ring will then appear to be
an AM 38khz carrier but which has the top of the envelope modulated with the R
channel
signal, and the bottom should be the L channel modulation, so it should be a
flat line of mod
because there is no L channel mod in tis test case.

The other side of the diode ring has an inverted version of the other side,
flat line on top, R mod on the bottom.

The levels of L-R transmitted at the station must be fixed to be the same for
all stations for all tuners to give the same separation.

The point you are missing is that the peak amplitudes of the L+R audio and
the 38 kHz sub carrier are the same, but the demodulated L-R audio is
lower in level than the L+R audio after it is multiplied by the 38 kHz
switching function. The math to show this is relatively simple.

Let me see if I have it straight.
If you modulate just one channel with a 1 kHz sine wave, 1Vrms, then the the L+R
audio wave from the ratio detector
has an amplitude equal to the amplitude of the L-R modulation
contained by the double sideband signal, ie, the envelope modulation shape has an
amplitude of 1Vrms.

The composite signal when viewed will then have an amplitude of 2Vrms, consisting
of the +ve 1/2 1kHz waves filled with 38kHz waves, between its arch and 0V, and
then followed by
the -ve 1/2 waves with 38kHz filling between the 0V line and the arch of the sine
wave.

There is a strong tendency for this signal to try to loose the amplitude
relationships between DSB quantity and L+R audio quantity.

So if the 38kHz wave is accurately applied to the composite,
the resulting mainly 38khz wave form will have the L signal on the top
of it, and R signal on the bottom of it.

Simple diode detection could be used to retrieve the signal but the use of the
matrix
and balancing results in a staircase stepped wave at the caps off the diodes, not a
saw tooth
"ripple voltage" like one sees in a power supply, or diode & RC detector
in an AM set.
Filtering the steps out is easier than filtering out 38kHz of ripple.




The reasonable separation you are achieving may be due to some combination
of two factors.
1.) Depending on how you are adjusting your "compensation" capacitor in
the cathode circuit of the input amplifier, you may be considerably
increasing the L-R 38 kHz level relative to the L+R signal level, not just
tweaking the phase a little. Can you post the value of your
"compensation" capacitor, I think you already posted the tube type and the
values of the cathode and plate resistors?

I figured about 3dB of extra gain occurs between below 15kHz, and above 23kHz.



2.) The BA1404 signal generator may be generating a nonstandard stereo
signal that helps to compensate for separation problems in your stereo
decoder.

Nope, after years of testing other tuners, the BA1404 provides a signal to them all
which
could be separated very well.

If better tuners than mine can get good sep from BA1404, my own designs
must also do this.

I have placed a filter on the pilot tone in the BA1404 because as standard the pilot
tone
is transmitted as a square wave, and tries to appear as one at the ratio detector.


I looked at the BA1404 data sheet many years ago and came to the
conclusion that it had a problem like this although on reflection I now
realize that they may have included a simple fix that might not be shown
in either the block diagrams or the specifications, I will have to take
another look at the BA1404 data sheet and see if I can tell what the true
situation is.

Its not as bad a situation as made out.





QUAD was not the only manufacturer to use this sort of diode ring
demodulator. This type of "diode ring" was the standard topology almost
universally used in discrete component stereo decoders, both tube and
transistor. Even Scott switched to this design in their later tube
decoders as mentioned above, and I would assume they carried it through to
their discrete component transistor designs too.

Scott used a much more complex array of Si diodes in their tubed decoders.
Si diodes had become cheap and reliable by that time.
The Scott method means that the 38khz oscillator signal from a single tuned
oscillator tank coil
can be applied by RC coupling, rather than having a pair of tuned circuits
to add to the composite signal so one of the pair has the composite signal
applied to the CT
of the LC circuit to get the summing and subtraction of L+R and L-R done
before diode
+C demodulation.

What you are describing is the design used in Scott's early decoders,
check this link for a later Scott tube decoder design:

http://hhscott.com/pdf/345.pdf

My god, 4.6Mb.
Can't you reduce the file size of stuff like this?

Its also a little hard to follow, but it seems little different in principles
to what have been used commonly except the 38Hz seems to be
created by amplifiying the rectified 19kHz, rather than have a 38kHz which is
then syncronised. Either will work fine, especially in good signal areas.

The locked oscillator is better imho, because
any amplitude changes in the 19khz, or audio spuriae
at 19kHz will not alter the amplitude of the 38khz so badly.
Its very important for fidelity that the 38khz has a steady amplitude.



This shows one of Scott's later tube decoder designs which uses only a
single pentode and a single triode, along with the same type of "diode
ring" that QUAD used. As I said this "diode ring" approach was virtually
universal in later designs.

Its not a bad way of doing it.

Other methods certainly exist; there are many transistors at work on the
signal in a chip MPX, and it may be possible to emulate
their methods well enough with a few more twin triodes.
But in 1965, nobody would have wanted the extra heat and weight of
the extra tubes...

Anyway, a friend just won a bid on E-bay for a Scott tuner, and no doubt
I will have the tormentous pleasure of working on it to align it and bring it to its
best,
as well of course as setting the de-emphasis for 50uS to suit Oz.

The front end of my set is ex-trio with 2 x 6AQ8, 2 x 6BA6(IF), 6AU6 limiter,
and the rest are in the MPX.

It could have another limiter; some FM sets did have two limiters,
and in my case the signal from the gene is only strong enough
to make -4V of AVc voltage without direct connection to the set's antenna terminal.
Strong locals here produce -13V, much more healthy.
When there is a strong signal, and good limiting, the stereo signal
is not much noiser than the mono signal.
So I don't think its the BA1404 that is at fault like I suspected.

If there isn't sufficient limiting, then amplitude noise gets in because the
bandwidth need for the stereo allows it, and somehow it
makes it through and audible noise is heard more when in stereo mode.






You mentioned that the QUAD decoder circuit uses four additional diodes to
switch between the monophonic and stereo reception modes. I can see
circuitry that looks like it automatically controls the diodes in the
stereo audio path to interrupt that path when a 19 kHz stereo pilot tone
is not being received. I also see the two diodes in the monophonic audio
path, but I can't see how they are automatically controlled?

Look at the bias voltages for the diodes.
This is varied automatically in such a way so that the diodes either conduct
when forward
bias, or don't conduct when reverse biased.

Yes, as I said before it is reasonably clear how the stereo signal is
automatically switched off, how the mono signal is automatically switched
on is not at all clear, at least to me.

Well, I think that when no 19kHz is present, there is a bias voltage
present which switches the outputs to the composite signal rather than from the
L or R.


It looks to
me like you have to manually turn off one of the QUAD II power amplifiers
to get a monophonic output from the decoder when a stereo signal is not
being broadcast. Can you explain what I am missing here?

That I cannot answer.

But to me it looked like an automatic DC signal was sent to the diodes to get
a mono signal for both L&R outputs when using the stereo decoder.

This is the part I can't follow, I don't see where an automatically
generated DC switching signal is applied to the diodes that switch the
mono signal on?

Its all in the operation of the transistor detector amp for 19kHz.
If you look carefully, it all gels.
There are two pairs of 330k R biasing the diodes, and look where the voltage applied
to the
330K comes from.

The other problem I had was with hum.
I applied rectified heater power to all tubes and it made utterly no difference.
So it wasn't from the heaters.
I changed all the tubes to fresh ones.
One 6AQ8 was down a little so i swapped it with the very similar
12AT7, after altering the heater power.
Hiss and hum diminished, and hum dissappeared
when tuned for maximum AVC voltage, and not for 0V at the centre
of the ratio detector.
So some serious re-aligning has to be done and then i expect to
get noise 5mV in stereo, with max Vrms from any radio station
at about 8Vrms, so that will be a fine SNR.

These things take a lotta work.

Patrick Turner.





Regards,

John Byrns

Surf my web pages at, http://users.rcn.com/jbyrns/

John Byrns wrote:

In article , Patrick Turner
wrote:

I saw the filter in the Quad circuit and i don't know what F it is for, 55khz,
or 19kHz, since there is no 19kHz filter indicated.

It is a low pass filter, presumably cutting off starting somewhere around
55 kHz, with a null tuned to 67 kHz according to the alignment
instructions in the QUAD manual. The 67 kHz null was necessary in the US
to eliminate the 67 kHz SCA sub carrier components, plus it gives the
filter a sharper cutoff, although at the expense of ultimate attenuation
at higher frequencies. The QUAD stereo mpx design does not include a 19
kHz filter, 19 kHz filters were uncommon in FM stereo tuners prior to the
introduction of Dolby noise reduction into cassette recorders.

We don't have 67kHz subcarriers in Oz, so no need to filter out what isn't there.
But its looks like a bandstop filter, not a low pass filter.
The parallel LC circuit is in series with the composite signal going to the matrix.

Probably you are right about no 19kHz filters.

Often they are placed at the output of each channel.
But early tuners with onboard mpx units with 19kHz notch filters on each channel
were rare.
There is an unmeasurable amount of 19kHz in the outputs of my MPX unit.



Good that you have mentioned the 55kHz filter.
When I looked at the Quad decoder I didn't spot this filter.
I will try such a filter to see if it reduces the much larger hiss
level in
stereo
that I am getting compared to mono, even when no modulation is applied
at my transmitter.

You might want to look into combining your 19 kHz pilot filter with the
low pass filter as was done in the Marantz 10B. A former Marantz employee
said that with proper design the phase shifts introduced by the 19 kHz
pilot filter can be used to compensate for the phase shifts introduced by
the low pass filter into the upper sidebands of the 38 kHz DSBSC sub
carrier.

But filtering to separate the L+R from the DSB is not required.
Scott also don't separate the composite.

Yes, both the Scott circuits and the QUAD circuit keep the composite
signal together and do not separate the L+R from the DSB, however it is
still necessary in the US to eliminate the 67 kHz SCA sub carrier, as well
as high frequency random noise resulting from the FM demodulation
process. Today audiophiles feel that 19 kHz filters are a necessity least
the residual 19 kHz bother their golden ears. The 19 kHz filter is
commonly inserted in the two audio paths coming out of the decoder, but
some times it is placed before the decoder as you have done. In the
design of the famed 10B Marantz combined the 19 kHz null with the low pass
function and the 67 kHz SCA null into a single integrated filter design,
which has been claimed to offer improved phase compensation.

The whole idea of the Quad circuit and some othr similar ones, and including
the Scott circuits is to keep all the components of the composite signal
together,
so that good separation is possible across a wide band.
The basic idea is to simply add the missing 38khz carrier to the
existing composite
from the
ratio detector, or whatever detector is used.

Yes that is how it is usually, but not always, done. The low pass and 67
kHz null filters used between the ratio detector and the mpx decoder cause
phase shifts or time delays that are not constant across the band up to 53
kHz occupied by the stereo composite signal. This nonlinear time delay
causes a loss of stereo separation, keep in mind that modern audiophiles
are not content with the 20 to 30 dB of separation that you are willing to
settle for, and demand separation figures in the 60 dB range least the
"sound stage" be compromised.

Marantz simply observed that by combining the 19 kHz null with the 67 kHz
null, and the 54 kHz low pass in a single coordinated package, they were
able to make the phase shifts due to the 19 kHz null filter compensate for
the phase shifts introduced by the 67 kHz null and low pass filter, or so
a former Marantz employee has claimed. I haven't gone through the math to
verify this, but I assume that the idea is that the phase shift of the 19
kHz null at the upper end of the L+R audio band is designed to complement
the phase effects on the demodulated L-R audio resulting from the
combination of the phase shift of the 67 kHz null and LPF on the upper
sideband of the 38 kHz sub carrier and the 19 kHz null on the lower
sideband of the 38 kHz sub carrier. The idea being that the L+R and L-R
audio would line up correctly in the decoder ring and subsequent auxiliary
matrix. This would not however imply that the time delay would be
constant vs. frequency for the demodulated left and right audio signals.
This is just my speculation as to what Marantz was trying to do with their
filter, I have not gone through the analysis, so my speculation could
easily be all or partly wrong. Comments about the design of the Marantz
composite filter from anyone that knows more would be welcomed.

I am getting very good separation which is over 23dB at 1kHz, and which improoves
up to 16kHz.
It can actually be increased to 35dB if one fiddles with it and there is a double
null situation;
I go go for the middle between the nulls to get the best stabilisation of the 38kHz
oscillator; tuning the 38kHz oscillator secondary away from 38kHz provokes
synchronization to stop, and a beat not is heard off lock.
25dB sep at 5kHz is OK.
And the sep tends to drift a little with temp of the coils, and adjustment is fine
with only
1/8 of a turn of the tranny slug able to reduce the sep from 25dB to 10 dB.

So once I am done I will wax the cores.



The phase of the carrier relative to the subcarrier waves is easily
swayed to line up exactly.

Yes, you can sway the phase of the carrier relative to the sub carrier
waves so that it effectively matches the phase of the original suppressed
carrier, but I don't believe that necessarily implies that the phase of
the demodulated L-R audio and the L+R audio will correctly line up in
phase to give maximum separation, especially across the entire audio band,
that depends on the constancy of the time delays in the composite filter
comprised of the 19 kHz null filter, 53 kHz low pass filter, and the 67
kHz null filter.

It works well in my case.

The type of notch filter I use for 19kHz has a very narrow notch,
and very small phase effects on the signals.
Had I used a simpler parallel LC circuit, either a low value R feeding a series type

LC, or high value R feeding a parallel LC filter to get a high Q, there would be
other serious phase effect in the two bands of interest, the AF and the 38kHz band.

The bridged T notch filter with L&C is by far the *only* sort of filter that should
be used
where a single F is to be eliminated.
broader filters to stop a whole band centred on 67 kHz can be the simpler
LC type as Quad used.
It won't have much effect below 53 kHz.



Any attempt to filter out the DSB and thus gain the L-R signal so that
it may be
applied
to LPF filtered phases of L+R, and -L-R results in poor separation at HF
due to
inevitable phase shifts caused by what must be more than first orer filters.
I tried all this in many efforts but always the phase shifts prevented
more than
12dB sep at 10kHz, and
almost no sep at 13kHz.

The phase effects of the 19kHz notch filter between 23kHz and 53 kHz and
between
20Hz and 15kHz is utterly negligible.

I'm not sure what you mean by "utterly negligible", clearly something is
causing your poor high frequency separation even if it isn't your 19 kHz
filter.

You can shunt out the 19kHz notch filter with little effect on the recovered L&R
except that there is 19kHz present in the signals.

Its a large size component if not filtered because the pilot is about
10% of the maximum audio modulation.

In any case the phase effects of the 67 kHz null and 53 kHz low
pass filters are not negligible, I take it what Marantz was trying to do
in the 10B is to also make the phase effects of the 19 kHz null filter non
negligible and then take advantage of the phase effects of the 19 kHz null
to compensate for the effects in the decoded audio resulting from the 67
kHz null and LPF.

I have not seen the Marantz schematic.
Nor have I worked on one. And trying to theorize about
the exact phase response of such things is rather difficult.



It can be shunted out and all that does is in my similar to Quad circuit
is dump
a whole lot of 19khz in each L&R output, plus probable IMD products.

Much of what we listen to, especially the HF portions, is at a level below the
pilot tone levels.
IN my case I am able to banaish the damn pilot tone after a cathode
follower, and
thus
drive drive the diodes with a low impedance.
There *is* a small effect of the 19kHz filter elements after the follower
but that is easily compensated for with a cap on the cathode R of the
gain before
the CF.
We are not looking for much compensation, about 3dB max.

I wouldn't call 3 dB an "utterly negligible" effect! Are you sure it is
the "small effect of the 19kHz filter" that your "compensation" capacitor
is correcting for? That 3 dB figure is suspiciously close to the
theoretical 3.9 dB difference in level between the L+R and L-R signals in
the FM stereo composite signal.

The passage of the composite signal from ratio detector always
gives some reduction of DSB levels.
Perhaps the bandpass shape of the ratio detector tranny
reduce the DSB a little.
Anyway, the addition of a cap across the cathode R of the composite
SET gain stage works well to slightly boost the DSB and tweak its phase.





4.) It isn't clear how you are compensating for the fact that the fact
that the L-R sub carrier signal is effectively transmitted at a lower
level than is the main channel L+R signal? QUAD used an auxiliary
matrix
to perform this function, feeding inverted L+R audio into the outputs of
the decoder to cancel some of the L+R signal.

Indeed you are correct about the Quad additional matrix.

But I saw that as a compensation network because separation levels
of L and R
modulation
recovered tend to be poor because the amplifying of the ratio
detector signals

tends to lessen separation; the composite signal envelope is mauled by the
RC couplings and gain variations between the audio frequencies and the
subcarrier F.


There are at least two
other was to perform this function, are you following the QUAD
lead, or is
the "compensation" capacitor in the cathode circuit of the input
amplifier
a crude solution to this problem.?

The compensation cap isn't as crude as you think; it tweaks the
phase of the
supressed carrier double sideband signal and slightly increase the gain at
around 38kHz.

It is a crude approach to equalizing the differing levels of the L+R and
38 kHz L-R signals.

Well my system works very well to give over 30db of separation, and is simple.

I thought you said the separation was only 20 dB at higher frequencies?

I measured it last night again, and got 25dB at 10kHz.

Its never less than 20dB until you get to about 20Hz when sep has fallen to
about 18dB which doesn't matter since bass F are in-phase anyway.




Then you should find the separtaion from the diode ring are either L or R,
without any need for Quad's method of compensation.

The diode ring seems to cause some loss of separation and there would only be
perfect
sep if there was perfect detection, as theory suggests it should happen,
but it just
doesn't occur.
The CF and the oscillator have to provide the power to the matrix, and
the some losses seem to occur.

There is a theoretical "problem" with the FM stereo signal that you
haven't considered yet that is responsible for most of this loss of
separation, even if the diode ring were perfect you would still see the
effect.

So what is the "theoretical problem" i have not considered?



This is understood if you just modulate the FM signal on R audio
channel only
with nothing on the L channel.
The magnitude of the L+R signal = that of the modulation carried by
the DSB
signal,
which is L-R.
The signal applied to one side of the diode ring will then appear to be
an AM 38khz carrier but which has the top of the envelope modulated
with the R
channel
signal, and the bottom should be the L channel modulation, so it
should be a
flat line of mod
because there is no L channel mod in tis test case.

The other side of the diode ring has an inverted version of the
other side,
flat line on top, R mod on the bottom.

The levels of L-R transmitted at the station must be fixed to be the
same for
all stations for all tuners to give the same separation.

The point you are missing is that the peak amplitudes of the L+R audio and
the 38 kHz sub carrier are the same, but the demodulated L-R audio is
lower in level than the L+R audio after it is multiplied by the 38 kHz
switching function. The math to show this is relatively simple.

Let me see if I have it straight.
If you modulate just one channel with a 1 kHz sine wave, 1Vrms, then the
the L+R
audio wave from the ratio detector
has an amplitude equal to the amplitude of the L-R modulation
contained by the double sideband signal, ie, the envelope modulation
shape has an
amplitude of 1Vrms.

The composite signal when viewed will then have an amplitude of 2Vrms,
consisting
of the +ve 1/2 1kHz waves filled with 38kHz waves, between its arch and
0V, and
then followed by
the -ve 1/2 waves with 38kHz filling between the 0V line and the arch of
the sine
wave.

Yes, if I followed your explanation correctly the signal is as you have
described it. To correctly demodulate this signal without an auxiliary
matrix or "compensation" capacitor to boost the sub carrier gain, it is
necessary to use an impulse sampling function, where as with your high
amplitude sine wave you are effectively using a square wave sampling
function, which results in the L+R component of the signal having a higher
amplitude than the demodulated L-R component, hence poor separation.

I am not sure what you mean by impulse samplig function.
But yes, with a 56 p-p V applied across the diode + R ring,
the adjusments to the cap charge levels cannot be instantaneous.
If you look at the Quad circuit there are 4 x 22k plus diodes in a "square" pattern,
then there
are a pair of 12k taken from the diodes to a cap of 4,000pF, C11, and C12 on the
Quad schematic.
The 4,000pF seems to be an incorrect value.
I set up my diodes + R with R = 4 x 27k and the "build out" R = 10k, and C = 900pF.
which I found just right to give low ripple F or switching noise *and*
the de-emphasis of 50uS.
The outputs from the caps are either L or R, and in my case go
to CF buffers; I found loading the 900pFs with any low R following worsened the
separation.
So there must be an interaction between the caps being charged to make the audio
signal and the
circuit driving it.





There is a strong tendency for this signal to try to loose the amplitude
relationships between DSB quantity and L+R audio quantity.

It is more than just some vague "strong tendency", the lower amplitude of
the recovered DSB quantity is a direct result of the way the FM stereo
signal is mathematically defined.

So if the 38kHz wave is accurately applied to the composite,
the resulting mainly 38khz wave form will have the L signal on the top
of it, and R signal on the bottom of it.

Yes, one channel will appear on the upper peaks of the 38 kHz carrier,
while the other will appear on the lower peaks, but that doesn't imply
that you can completely separate the two channels with a square wave
switching function of the type you are using,

But there are no square waves used anywhere.

for that you need an impulse
type sampler. When you multiply the composite signal by a 38 kHz
switching function as you are, the L+R signal is attenuated by a factor of
1/2, while the L-R signal demodulated from the 38 kHz side bands is
attenuated by a factor of 1/PI. This is a difference in level of 3.9 dB
and is suspiciously close to the approximately 3 dB compensation you say
your "compensation" capacitor introduces.

I am not sure how you figure all that.
Anyway, my system works well, so i plan to stay with it.



Simple diode detection could be used to retrieve the signal but the use of the
matrix
and balancing results in a staircase stepped wave at the caps off the
diodes, not a
saw tooth
"ripple voltage" like one sees in a power supply, or diode & RC detector
in an AM set.
Filtering the steps out is easier than filtering out 38kHz of ripple.

It sounds to me like "ripple" is just another word for "steps"?

There is a vast difference.

In a detector for AM radio at 455kHz, the ripple voltage at a typical 100pF cap
is a saw tooth wave; its ok because its all easily filtered out since
455kHz is well above audio.
But in the diode matrix used in a decoder, the two R and two diodes each 1/2 of the
matrix
both conduct depending which way the flow of current is in the 38kHz.
The cap with AF information is either being charged quickly on +ve going audio peaks
or
discharged quickly on -ve going audio peaks. There is a very slow time constant
used for the discharge of the charge at the caps since there isn't any need for the
caps
to discharge at all between the fast charges applied by the 38kHz waves.
So a stepped wave is the result.

Its very like the D to A process where a stepped wave id first produced which must
be converted to
a sine wave by removing the steps by filtering. Except a CD player does it 44.1kHz,
and the FM decoder does it at 38kHz.
Its would have been cleaner to have the process at say 76kHz,
and the pilot at 38kHz but that
would have meant less multiplexing ability and some more critical
matrixing circuitry.

Its possible to have the creation of the audio via a simple diode & CR detector
but the saw tooth wave resulting would have a higher amplitude than the
step wave does, and a little more phase shift in the detection process.
Either detector process does create a phase lag in the recovered audio of around
90 degrees at 15kHz.

That is a bother with nearly all radio audio signals.
Lotsa extra phase lag occurs in the HF signals.

But its nothing compared to what is done in nearly every studio to music
which has been much manipulated and equalised.
There is almost no recorded phase coherent music to be listened to.
Our ears don't mind.




2.) The BA1404 signal generator may be generating a nonstandard stereo
signal that helps to compensate for separation problems in your stereo
decoder.

Nope, after years of testing other tuners, the BA1404 provides a signal
to them all
which
could be separated very well.

I am sure you are correct here, and that the BA1404 includes a
compensation circuit although the data sheet doesn't mention it. I think
I was confused about the conclusion I had drawn about the BA1404 many
years ago, thinking about it again I think the problem was that the BA1404
puts out a square wave 38 kHz DSB signal, which if feed through a very
wide band tuner, or directly into a mpx decoder without a 53 kHz low pass
filter, will impact the separation because there will be L-R sidebands
around the third harmonic of 38 kHz which if not filtered out at some
point will be demodulated by a square wave decoder increasing the level of
the demodulated L-R signal with a consequent impact on the separation.
This may partly explain why you only required 3 dB of compensation rather
than theoretical 3.9 dB, or even more when potential high frequency losses
in a real circuit are considered.

Since inserting a filter into the BA1404 transmitter circuit to try
to convert the 19kHz square wave to a sine wave, no change to separation occured.

But the DSB waves at the ratio detector show up as sine waves, not square waves.




What you are describing is the design used in Scott's early decoders,
check this link for a later Scott tube decoder design:

http://hhscott.com/pdf/345.pdf

My god, 4.6Mb.
Can't you reduce the file size of stuff like this?

It wasn't me, I just provided the link.

OK.



Its also a little hard to follow, but it seems little different in principles
to what have been used commonly except the 38Hz seems to be
created by amplifiying the rectified 19kHz, rather than have a 38kHz which is
then syncronised. Either will work fine, especially in good signal areas.

The locked oscillator is better imho, because
any amplitude changes in the 19khz, or audio spuriae
at 19kHz will not alter the amplitude of the 38khz so badly.
Its very important for fidelity that the 38khz has a steady amplitude.

I think if you look again you will see that there is positive feedback
from the plate circuit to the grid, making it an oscillator, unless what
you are saying is that there isn't enough positive feedback to make it an
oscillator, and the positive feedback is only there to increase the gain?

OK, maybe there is, as I said its hard to follow, but it would be possible to
make the 19kHz amp which also doubles the F within it to make 38kHz
to also act as an oscillator.



Notice also that in the later Scott designs like this one the out of phase
"compensation" signal for the auxiliary matrix is taken from the "bottom"
side of a floating ratio detector.

The front end of my set is ex-trio with 2 x 6AQ8, 2 x 6BA6(IF), 6AU6 limiter,
and the rest are in the MPX.

It could have another limiter; some FM sets did have two limiters,

Doesn't your set use a ratio detector? The 6AU6 plus the ratio detector
make a total of two limiters.

The ratio detector does tend to reject AM, if that is what you mean.

I did some measurements last night.

The hiss noise from the decoder is dependant on limiting in the
IF part of the set.

Strong stations generate -13V of AVC voltage applied to the 3 IF tubes including the
limiter.

My sig gene only makes enough RF to generate -4V AVC, and the SNR
of hiss to signal level of 4.2Vrms output which is about 1/2 the
maximum daytime audio signal levels received on many stations
is about -50dB.
But when the sig gene is directly connected to the RF input the AVC generated
is about -12V, and the SNR is -66 dB.

I tried comparing it to the Audio Reflex tuner and found that it was better
compared to mine when mine was receiving low power signals, but worse
with high power signals, so as long as the station signal strength
is high my set will have a better SNR than the Audio Reflex.
Hum levels were the same for both tuners, and not much above the hiss levels,
and I think due to hum generated at the BA1404.

I reduced the gain of the composite signal gain amp and
to give 1/2 the gain, so 2.1Vrms is produced instaed of 4.2Vrms of audio at the
outputs
and found the SNR only got worse by slightly more than the gain reduction so
I will go back to having a 6DJ8 with a gain of 5 instead of 2.5.

I will place a resistance divider between the last LPF filter on the L&R
outputs and the final CF output buffers to reduce the signal levels to
about 1Vrms and match the filtered output from the AM section of the set.

The method I am using is producing much more output voltage than the system
I used first in the decoder at my website, which may R.I.P, since it is a
way of doing things which has serious flaws.

Patrick Turner.







Regards,

John Byrns

Surf my web pages at, http://users.rcn.com/jbyrns/

John Byrns wrote:

I wonder why Phil hasn't jumped into this thread as he does most of your
threads?

Be thankfull.

Maybe he is wary to call me too many rotten names when some inevitable
disagreement might occur, or else he agrees with what I have said or he just has
not
bothered to tread through all the details of ancient silly old tubed
MPX decoders that couldn't ever sound as well as the latest digital gee whiz tuner

in a tiny box costing $20.

I don't mind anachronistic persuits; I do resent that ppl
be nasty about it to me or anyone else.



In article , Patrick Turner
wrote:

I am getting very good separation which is over 23dB at 1kHz, and which
improoves up to 16kHz.
It can actually be increased to 35dB if one fiddles with it and there is a
double null situation;
I go go for the middle between the nulls to get the best stabilisation of
the 38kHz oscillator;

The double null is an approach that some recommend to minimize the loss of
separation when the 38 kHz oscillator locks with a finite phase error.
The problem is that to get this benefit you sacrifice the maximum
separation when the oscillator is correctly tuned/locked.

You only have to play around with these circuits to realize just how touchy and
fidly
they are to get working just right, and make them keep working well.

Just 1/8 of a turn on the 38kHz oscillator tuned circuit slugs means
you have 6dB of separation or 26dB.

In the mass produced article i really am not sure how they managed to get
the decoders to work as specified, and I suspect many samples
just didn't.
Not many may have noticed.



The bridged T notch filter with L&C is by far the *only* sort of filter
that should be used where a single F is to be eliminated.

Wouldn't a simple parallel L&C circuit using a high Q inductor serve the
same purpose?

Not a hope.
The 19kHz has a notch shape which is -3db at 18kHz and 20kHz, and rejection of
19kHz is
easily -60dB, and this far surpasses anything you will get with a
parallel or series resonant circuit loaded by appropriate optimal
termination resistances and fed by suitably low driving source resistances
which will not suffer due to low Z effects of the filter.


I thought the only advantage of the "bridged T" over a
simple parallel L&C was the cost advantage of being able to use a cheap
low Q inductor, whose losses at the notch frequency are "tuned" out by the
second balancing R?

Believe me, notched T LC circuits are the best, forget the rest.
BTW, samples made to function at lower F, say 1kHz are terrific
for THD analysis, when removing the fundemental
test tone F from a DUT only the thd and noise remains;
its a simple and effective and very rugged way to
construct distortion filter for testing amps.





I have not seen the Marantz schematic.
Nor have I worked on one. And trying to theorize about
the exact phase response of such things is rather difficult.

The correct thing to do is not to to theorize about the exact phase
response, but to either measure the phase response of the actual filter,
or to simulate it to see what it looks like. The best thing to do is to
first simulate the proposed design until you have tweaked it to the point
you want, then build and measure it to see that it performs as desired.

I first made sure i grasped the mental concept of how the decoder worked,
and what all the wave forms were approximately around the circuit
and what the approximate phase shifts were likely to be.
I expected it to work moderately well because others before me
had tried similar ideas.
Just how well depended on some careful tweaks, but
so far I am quite delighted by the sound.

Its pointless to simulate or calculate too much with such things
because its impossible to take in all the aspects of the device behavious
and stray L and C etc.
And I am hopeless with equations using "j" or anything
with reactance and the square root of minus one.
I have tried to follow some explanatory texts on the subject,
and my mind just seizes up because whoever wrote the text
strays from theory for theory's sake to theory
for getting from A to B in a practical and real circuit
someone is likely to encounter.





I wouldn't call 3 dB an "utterly negligible" effect! Are you sure it is
the "small effect of the 19kHz filter" that your "compensation" capacitor
is correcting for? That 3 dB figure is suspiciously close to the
theoretical 3.9 dB difference in level between the L+R and L-R signals in
the FM stereo composite signal.

The passage of the composite signal from ratio detector always
gives some reduction of DSB levels.
Perhaps the bandpass shape of the ratio detector tranny
reduce the DSB a little.
Anyway, the addition of a cap across the cathode R of the composite
SET gain stage works well to slightly boost the DSB and tweak its phase.

While it is true that you are likely to lose some response at the upper
DSB frequencies, that isn't the main cause of loss, unless maybe you are
using a vestigial sideband type of lowpass/SCA filter.

I am getting a flat response across the band with -3dB at 15kHz, and good
separation,
so that's all i need to get.





What you have not yet recognized is that the specification of the FM
stereo wave form is such that the demodulated L-R audio level will be
lower than the L+R audio level unless you are using an impulse function
for sampling in the stereo demodulator.

I don't understand what you mean by "impulse function"
Care to refer me to a text someplace?

The whole working of a decoder is simply the
manipulation of accurrately enough phased sine waves
between dc and 53 kHz.

"Compensation" in the cathode of
a composite amplifier stage is often used to compensate for the phase
shift and roll off that sometimes occurs at the higher composite signal
frequencies due to effects in the IF stages of the FM receiver, but
without realizing it you seem to be using the "compensation" capacitor to
also raise the general level of DSB signal enough to compensate for the
naturally lower recovered L-R level. The problem with using a
"compensation" capacitor to boost the level of the L-R DSB signal so that
the recovered L-R is equal in level to the L+R signal, is that it doesn't
provide a uniform boost over the entire 23-53 kHz DSB range, hence you end
up with separation problems at some frequencies.

The tests don'y indicate any problems with separation.
Its best at the higher F.
Many poorer MPX units had declining sep as F rises past 1 kHz.

I figured that if Quad could do ok with 3 lousy germanium transistors
working from +15V, then i could do
something at least as good or better with a bunch of triodes
commonly available.



Other than your design
the only stereo decoder I have ever seen that used this approach to
equalizing the L+R and L-R levels is the one used in the Dynaco FM3, if
anyone knows of other examples I would be interested in hearing of them so
I can add them to my list.

When I finally draw up the schematic for posting on the web
you will know what is actually in it.
Very few ppl will understand the subtle aspects of the circuit workings.



I thought you said the separation was only 20 dB at higher frequencies?

I measured it last night again, and got 25dB at 10kHz.

Its never less than 20dB until you get to about 20Hz when sep has fallen to
about 18dB which doesn't matter since bass F are in-phase anyway.

Poor low frequency separation is caused by low frequency phase shift that
results from too high a low frequency cutoff in the coupling circuits
before the "diode ring". To maintain 30 dB of separation down to 50 Hz
requires that the cutoff frequency of the coupling circuit be 2.5 Hz or
lower if there is only a single roll off or coupling capacitor involved.

Correct, so coupling from the ratio detector to the
composite signal amp must have a long time constant.
In my case its quite long enough without passing
excessively low F signals.
The response is about -3db at 10Hz, good enough for an old radio.



Then you should find the separtaion from the diode ring are
either L or R,
without any need for Quad's method of compensation.

The diode ring seems to cause some loss of separation and there
would only be
perfect
sep if there was perfect detection, as theory suggests it should happen,
but it just
doesn't occur.
The CF and the oscillator have to provide the power to the matrix, and
the some losses seem to occur.

There is a theoretical "problem" with the FM stereo signal that you
haven't considered yet that is responsible for most of this loss of
separation, even if the diode ring were perfect you would still see the
effect.

So what is the "theoretical problem" i have not considered?

You have failed to consider that the demodulated audio will be 3.9 dB
lower than the L+R audio for purely theoretical reasons, even before
considering further losses at the DSB frequencies due to practical circuit
considerations.

I don't know about the origin of this 3.9dB.
All I know is that adding a right valued cap across a cathode R of the
composite signal SET amp at the input to the MPX unit
did work.



While you have attempted to fix both problems with your
"compensation" capacitor, that is more appropriate to only the "roll off
at the ratio detector" part of the problem. While the same approach can
be used to provide a crude fix for the more theoretical part of the
problem arising from the specification of the FM stereo system, it is sub
optimal and is rarely used in practice.

Gee, I see other systems that look far more complex and perhaps
more difficult to get right.
So far so good.


Yes, if I followed your explanation correctly the signal is as you have
described it. To correctly demodulate this signal without an auxiliary
matrix or "compensation" capacitor to boost the sub carrier gain, it is
necessary to use an impulse sampling function, where as with your high
amplitude sine wave you are effectively using a square wave sampling
function, which results in the L+R component of the signal having a higher
amplitude than the demodulated L-R component, hence poor separation.

I am not sure what you mean by impulse samplig function.

You may know what I am calling an "impulse function" by the name "dirac
delta function".

Not in my vocab, sorry.

I am a practical man, with an intuively guided soldering iron,
and an uncanny ability to get something working
without overly extensive calculations.

If that doesn't ring a bell you should add a book on
communications theory to your bookshelf right next to your copy of the
RDH4th if you are going to with authority about systems like FM stereo.

I have plenty of auxilliary books on radio and audio and some are
complete wastes of time for me to read because every
second line of text is a big long line of equation,
none of which makes the slightest bit of sense.
But such things as MPX units can be made knowing about
what waveforms occur where, and at what amplitudes and with
what slight imperfections or losses may occur, as witnessed by the
oscilliscope and thd meter.

One could build a television the same way, and ppl
who were tradesmen did just that, and often their maths
ability was quite poor like mine.
I am not university educated.
I am school of life tryer.

I
refer to an old book by John C. Hancock titled "An Introduction To The
Principals Of Communication Theory". This is a relatively small book that
was published in 1961 and concentrates on the thing s we need to know to
truly understand the FM stereo system like sampling, convolution, and AM
DSBSC modulation, without wasting a lot of space on esoteric subjects like
information theory.

I have already got enough material on such basics.



Yes, one channel will appear on the upper peaks of the 38 kHz carrier,
while the other will appear on the lower peaks, but that doesn't imply
that you can completely separate the two channels with a square wave
switching function of the type you are using,

But there are no square waves used anywhere.

What do you think you are actually using as a switching function in your
design, certainly not a sine wave?

It is a bunch of sine waves.
Not a single square wave is to be seen.
There is staircase wave produced at capacitors by the action of rectified sine
waves
produced by the ring diodes, but C&R and L&R LPF filters remove the "steps",
leaving the buried sine wave component.
Its magic what happens.
At 16kHz, there are not many 38kHz peaks to describe the amplitude
of the 16kHz tone, yet there they are, and any left over HF noise artifacts
won't be heard evenn if they make it through the following
audio amps and speakers.
But at least I remove the 19kHz, and other artifacts are
low levels.



The high amplitude sine wave you are
using to drive the "diode ring" is effectively a square wave as far as
switching the diodes goes, the reason you use it is to insure that the
diode switches are either fully conducting, or fully turned off, if this
condition weren't maintained all sorts of problems could arise.

The only time the diodes conduct is when the peak voltage of the sine waves
at 38kHz reach a level where conduction occurs for a short part
of the wave cycle, and whatever voltage is stored in the caps gets changed a
little
from one voltage to another.
The right values of R&C have to be chosen so this occurs quickly enough.



for that you need an impulse
type sampler. When you multiply the composite signal by a 38 kHz
switching function as you are, the L+R signal is attenuated by a factor of
1/2, while the L-R signal demodulated from the 38 kHz side bands is
attenuated by a factor of 1/PI. This is a difference in level of 3.9 dB
and is suspiciously close to the approximately 3 dB compensation you say
your "compensation" capacitor introduces.

I am not sure how you figure all that.
Anyway, my system works well, so i plan to stay with it.

You have it 95% right, one area in your decoder that could use some more
work is the crude way you are compensating for the reduced demodulated L-R
level resulting from the FM stereo specifications.

But the outcome isn't crude or defective compared to othe decoders I have
investigated.

A better way to
compensate for this effect would be to add an auxiliary matrix to your
decoder, as is typically done. One possible way to do this would be to
take some out of phase L+R from the cathode circuit of your amplifier tube
and add it to the outputs of the "diode ring" as QUAD did.

Possibly i could do it that way, but 1 cap seems easier.



Its would have been cleaner to have the process at say 76kHz,
and the pilot at 38kHz but that
would have meant less multiplexing ability and some more critical
matrixing circuitry.

Actually the problem wouldn't have been "less multiplexing ability", it
would have been a serious noise problem for the FM stereo signal due to
FM's triangular noise spectrum which is the reason FM stereo has a worse
signal to noise ratio than monophonic FM. If the stereo sub carrier had
been placed at 76 kHz the noise level would have been intolerable.

Why?
Noise is relative to bandwidth, and the same bandwidth would have been used
for the same ranges of audio and DSB signals.
I could be wrong though.
I think they tried to cram the FM with as much as possible, so
multiplexing could be done with subcarriers at 67kHz and 96kHz,
thus allowing quadrophonic sound+++,
and still have it all fit into about 300kHz of bandwidth
on the 88 to 108 MHz band.


As far
as "multiplexing ability" goes, SCA sub carriers could have been placed in
the regions between 15 kHz and the 38 kHz pilot, and between the 38 kHz
pilot and the lower stereo sideband at 61 kHz, the "multiplex" interests
would have probably greatly preferred this to the placement of the SCA sub
carriers in the region above 53 kHz as was actually done.

Regards,

John Byrns

Surf my web pages at, http://users.rcn.com/jbyrns/

Digital looks set to sweep all this asside, but FM will be around for awhile
until ppl who still see value in the old system die off.

Patrick Turner.

In article , Patrick Turner
wrote:

John Byrns wrote:

Hi Patrick,

Before I get to your current post, let me make a comment on an earlier
post in which you were describing the difference between your version of
the "diode ring" and QUAD's version. You commented with respect to QUAD's
version that "The 4,000pF seems to be an incorrect value." I agree that
4,000pF doesn't seem quite right, by my calculations 4,000pF would give a
de-emphasis time constant of about 68 usec. 3,000pF would seem to be
closer to the correct capacitor value to give a time constant of 50 usec.
The 900 uF capacitor with the resistor values in your circuit look like
they have a time constant of about 42.3 usec.

Quad use 4,000pF, way off the mark imho, maybe its a typo.
In Oz we have 50uS.
I tried 220pF, and got almost no de-emphasis.

I doubt it is a typo, it is close enough to the 3,000pF value required for
proper 50 usec de-emphasis, that it is likely what QUAD used. It was
quite common in the old days to use a de-emphasis time constant somewhat
longer than optimum.

Let's consider an FM stereo multiplex decoder like the QUAD
which effectively uses a train of square pulses repeating at a 38 kHz rate

But strings of pulses don't exist when they encode the waves at the
transmitter;
sine waves are used.
BA4014 does use square waves for the 19&38kHz, but these can be
alterered to sine
waves with no changes
to the receiver function.

This is not historically correct, and in any case what does it have to do
with the decoding method used in a receiver, the IC decoders used in
modern solid state FM tuners are clearly based on square waves, one look
at a typical data sheet will show you that. Early vacuum tube broadcast
stereo encoders like the RCA and Gates/Harris models used diode ring
modulators to generate the 38 kHz DSBSC L-R signal which was mixed with
the L+R audio after first passing though a 38 kHz bandpass filter to
eliminate the sidebands produced around the odd harmonics of 38 kHz, which
result from what is effectively a square wave 38 kHz sub carrier driving
the diode ring modulator. Look here for a schematic diagram of the RCA
circuit:

http://users.rcn.com/jbyrns/pics/BTS-1A-sch.jpg

As an aside note the diode ring modulator used in this circuit. What you
are calling a "diode ring" in the QUAD circuit isn't actually a diode ring
at all, it is two independent circuits, each using two diodes as a switch,
one for the "Left" channel and a second one for the "Right" channel.

A circuit that uses square waves more explicitly is found in the Collins
786M-1 Broadcast Stereo Generator, an early transistor encoder based on
the switching principle. This is the encoder equivalent of the switching
principle used in the QUAD decoder, except in place of the diode switches
used in the QUAD decoder, Collins used transistor switches to alternately
short the inputs from each of the two audio channels to ground, sort of
like the Scott did with the two diode quads in their original tube
decoders. I have never seen a decoder using transistor switches in this
way, I wonder why nobody seems to have tried transistor switches in a
decoder?

Sine waves didn't become popular in Stereo Generators until later. An
example is the Orban Optimod which used a pure 38 kHz sine wave as the sub
carrier, which was multiplied by the L-R audio signal in a linear
multiplier circuit. This approach using a pure sine wave and linear
multiplier eliminated the need for a filter at the output of the stereo
generator, and all the attendant problems the filter created. Note that
the use of sine waves requires a linear multiplier, not the diode switches
you used. There are some vintage Japanese high end FM tuners that have
stereo decoders based on the true sine wave and linear multiplier
principle.

Of course today everything is "digits", we no longer use sine waves, or
even square waves, in stereo generators and the old decoders based on
square waves still work just fine.

I need to see wave forms.

Maybe all you need to know about it is at

http://members.tripod.com/~transmitters/stereo.htm

I would take this site with a big grain of salt. I haven't looked at it
since I had a long running conversation with author on the Yahoo FMTuners
group, and by email, about three years ago. There were originally serious
technical errors on the site, although I believe most of those were
corrected as a result of my suggestions. Many of the simulation results
were also somewhat misleading, presumably as a result of inappropriate
settings for the circuit simulator that was used. Unfortunately I am only
a rank novice on circuit simulators, so I was unable to offer advice on
fixing the problems related to the simulations. The author had a
particular spin he was pushing and so was not too concerned about errors
which might be misleading as long as they were consistent with his view of
the subject. I will have to take another look at the site to see if it
has changed at all. Nevertheless there is much to be learned from this
site, just don't necessarily believe the details, which are important.
Take it with a big grain of salt unless the author has fixed all the
problems, both theoretical, and with the simulations.

I suggest 20 dB of separation is fine for stereo enjoyment
from a radio.

Its a 20dB SPL ratio

Would you settle for 10% distortion in the amplifiers you design?

Nope, it doesn't matter if the composite wave with respect to the 20vrms
of the oscillator is maybe up to 5Vrms.
The waves which are fed to the diode ring are ALL 38kHz sine waves.
The diodes ONLY conduct when the voltage gets above of below the
existing voltage.

"above of below the existing voltage", what are you saying? Are you sure
you understand how the QUAD circuit really works? You spoke in an earlier
post of how you use 'C&R and L&R LPF filters remove the "steps", leaving
the buried sine wave component'. Doesn't the presence of these "steps"
suggest to you that there may be something more than sine waves involved
in the operation of the QUAD decoder?

There is a similar amount of noise between 20Hz and 16kHz and between
60kHz and 76kHz.
OK, there will be more noise between 60khz and 92kHz, which is the
bandwidth for
a 76 kHz sub carrier and the sidebands.
Its only 3dB more.

I can't believe you make this statement, especially after I tried to set
you on the correct path with respect FM noise, your head is truly as thick
as a brick. You will find FM's triangular noise spectrum discussed in
just about any text book that deals with FM, it is a well known fact that
is not under any dispute that I am aware of, other than by you. Some
books that discuss this subject include "An introduction To The Principles
Of Communication Theory" by John C. Hancock that I mentioned in an earlier
post. Hancock's treatment is probably too mathematical for you, so you
might want to look at "F-M Simplified" by Milton S. Kiver which was
written more for the Television serviceman type and has a good totally non
mathematical discussion of the subject with a lot of nice drawings.
Finally you might want to look at Edwin H. Armstrong's paper in the May
1936 issue of the Proceedings of the IRE, titled "A Method Of Reducing
Disturbances In Radio Signaling By A System Of Frequency Modulation".
Armstrong was the inventor of the wide band FM system which he describes
in this paper. Like you Armstrong was a practical type who wasn't much
into math. It's been a while since I read this paper, but IIRC Armstrong
gives a good, largely intuitive, description of the reasons why FM has a
triangular noise spectrum.

You might want to dig a little deeper before you repeat statements like
the above again, all the world's radio knowledge isn't contained in the
RDH4th.


Regards,

John Byrns


Surf my web pages at, http://users.rcn.com/jbyrns/

John Byrns wrote:

In article , Patrick Turner
wrote:

John Byrns wrote:

Hi Patrick,

Before I get to your current post, let me make a comment on an earlier
post in which you were describing the difference between your version of
the "diode ring" and QUAD's version. You commented with respect to QUAD's
version that "The 4,000pF seems to be an incorrect value." I agree that
4,000pF doesn't seem quite right, by my calculations 4,000pF would give a
de-emphasis time constant of about 68 usec. 3,000pF would seem to be
closer to the correct capacitor value to give a time constant of 50 usec.
The 900 uF capacitor with the resistor values in your circuit look like
they have a time constant of about 42.3 usec.

Quad use 4,000pF, way off the mark imho, maybe its a typo.
In Oz we have 50uS.
I tried 220pF, and got almost no de-emphasis.

I doubt it is a typo, it is close enough to the 3,000pF value required for
proper 50 usec de-emphasis, that it is likely what QUAD used. It was
quite common in the old days to use a de-emphasis time constant somewhat
longer than optimum.

But I have used almost the same identical circuit and needed only
a final 800pF.
So the drive impedance to the matrix I have must be higher if that is to be
added into the R which preceeds the C for the de-emphasis.



Let's consider an FM stereo multiplex decoder like the QUAD
which effectively uses a train of square pulses repeating at a 38 kHz rate

But strings of pulses don't exist when they encode the waves at the
transmitter;
sine waves are used.
BA4014 does use square waves for the 19&38kHz, but these can be
alterered to sine
waves with no changes
to the receiver function.

This is not historically correct, and in any case what does it have to do
with the decoding method used in a receiver, the IC decoders used in
modern solid state FM tuners are clearly based on square waves, one look
at a typical data sheet will show you that.

They have to be able to work with either square or sine waves for the received
19kHz and 38kHz, although square wave use isn't needed, but
easily possible with the chip speeds.



Early vacuum tube broadcast
stereo encoders like the RCA and Gates/Harris models used diode ring
modulators to generate the 38 kHz DSBSC L-R signal which was mixed with
the L+R audio after first passing though a 38 kHz bandpass filter to
eliminate the sidebands produced around the odd harmonics of 38 kHz, which
result from what is effectively a square wave 38 kHz sub carrier driving
the diode ring modulator. Look here for a schematic diagram of the RCA
circuit:

http://users.rcn.com/jbyrns/pics/BTS-1A-sch.jpg


Interesting.
The actual "ring" of transformers prior to the diodes looks
like it may be tricky to wind and set up.


As an aside note the diode ring modulator used in this circuit. What you
are calling a "diode ring" in the QUAD circuit isn't actually a diode ring
at all, it is two independent circuits, each using two diodes as a switch,
one for the "Left" channel and a second one for the "Right" channel.

Maybe you are right on the name for the ""ring"" of 4 diodes and four
resistors.

A circuit that uses square waves more explicitly is found in the Collins
786M-1 Broadcast Stereo Generator, an early transistor encoder based on
the switching principle. This is the encoder equivalent of the switching
principle used in the QUAD decoder, except in place of the diode switches
used in the QUAD decoder, Collins used transistor switches to alternately
short the inputs from each of the two audio channels to ground, sort of
like the Scott did with the two diode quads in their original tube
decoders.

The scott method is a real "switcher because" some of the diodes
just act as switches to allow a non floating secondary winding on the
locked oscillator plate circuit.
The single CT winding of the 38khz oscillator can therefore drive the matrix
from caps, and the usual extra tuned circuit is avoided, which would
make the drive impedance of the matrix lower.


I have never seen a decoder using transistor switches in this
way, I wonder why nobody seems to have tried transistor switches in a
decoder?

Sine waves didn't become popular in Stereo Generators until later. An
example is the Orban Optimod which used a pure 38 kHz sine wave as the sub
carrier, which was multiplied by the L-R audio signal in a linear
multiplier circuit. This approach using a pure sine wave and linear
multiplier eliminated the need for a filter at the output of the stereo
generator, and all the attendant problems the filter created. Note that
the use of sine waves requires a linear multiplier, not the diode switches
you used. There are some vintage Japanese high end FM tuners that have
stereo decoders based on the true sine wave and linear multiplier
principle.

I don't see there is a linear multiplier needed.
The 38kHz subcarrier waves look like sine waves to me, and
I see no reason why the simple Quad approach won't work OK.



Of course today everything is "digits", we no longer use sine waves, or
even square waves, in stereo generators and the old decoders based on
square waves still work just fine.

I need to see wave forms.

Maybe all you need to know about it is at

http://members.tripod.com/~transmitters/stereo.htm

I would take this site with a big grain of salt. I haven't looked at it
since I had a long running conversation with author on the Yahoo FMTuners
group, and by email, about three years ago. There were originally serious
technical errors on the site, although I believe most of those were
corrected as a result of my suggestions. Many of the simulation results
were also somewhat misleading, presumably as a result of inappropriate
settings for the circuit simulator that was used. Unfortunately I am only
a rank novice on circuit simulators, so I was unable to offer advice on
fixing the problems related to the simulations. The author had a
particular spin he was pushing and so was not too concerned about errors
which might be misleading as long as they were consistent with his view of
the subject. I will have to take another look at the site to see if it
has changed at all. Nevertheless there is much to be learned from this
site, just don't necessarily believe the details, which are important.
Take it with a big grain of salt unless the author has fixed all the
problems, both theoretical, and with the simulations.

I suggest 20 dB of separation is fine for stereo enjoyment
from a radio.

Its a 20dB SPL ratio

Would you settle for 10% distortion in the amplifiers you design?

Distortion is very different to separation.

Having one speaker producing a sound level that is
10dB lower SPL does give you very decent stereo separation.

But since finalizing and trimming the values, I was able to get
38dB, and that's about the limit for the BA1404 gene.
If the max separation is set at 5kHz, it now drifts up to about 30dB.
That's with the set turned on its side on the bench
so I can access botyh top and bottom for adjusments; maybe there will be less
drift
when the set is laid flat; final tests are due in the next few nights.

Separation remains constant for different levels of modulation signal
so the stereo detection appears to be linear.




Nope, it doesn't matter if the composite wave with respect to the 20vrms
of the oscillator is maybe up to 5Vrms.
The waves which are fed to the diode ring are ALL 38kHz sine waves.
The diodes ONLY conduct when the voltage gets above of below the
existing voltage.

"above of below the existing voltage", what are you saying?

Well without any modulation, there is an oppositely phased 38kHz signal across the

38kHz oscillator secondary.
In my case its about 40Vrms, or 56V p-p, so when
each side of the tranny winding peaks, two resistors conduct through two
of the diodes facing the same direction, and a "centre" voltage exists at the
diode junction.
This voltage establishes a steady voltage in the caps ( 4,000pF )
and where audio modulation occurs, the voltage in the 4,000pF
is changed at a rate of 38,000 times per second.
A staircase wave form results.
With no audio signal, the diodes don't pass much current to the caps.
The cap voltage is either stepped up or down depending
the relative value to 0V, or the voltage at the CT of the 38kHz
winding.



Are you sure
you understand how the QUAD circuit really works? You spoke in an earlier
post of how you use 'C&R and L&R LPF filters remove the "steps", leaving
the buried sine wave component'. Doesn't the presence of these "steps"
suggest to you that there may be something more than sine waves involved
in the operation of the QUAD decoder?

What comes out at the 4,000pF caps are a stepped wave form,
since the discharge resistance discharging the 4,000pF is very high'
no need to have much cap discharge at all because unlike a simple
diode and cap and R in an AM set, the caps are forced to change their voltage
in a series of steps; discharging is done through the diodes.
The caps get charged in a small % of the 38kHz sine wave.
The 38khz could be a square wave but there is no need.





There is a similar amount of noise between 20Hz and 16kHz and between
60kHz and 76kHz.
OK, there will be more noise between 60khz and 92kHz, which is the
bandwidth for
a 76 kHz sub carrier and the sidebands.
Its only 3dB more.

I can't believe you make this statement, especially after I tried to set
you on the correct path with respect FM noise, your head is truly as thick
as a brick. You will find FM's triangular noise spectrum discussed in
just about any text book that deals with FM, it is a well known fact that
is not under any dispute that I am aware of, other than by you.

Well there is SFA extra noise in my stereo decoder.

The only bother is that in stereo, about 1/2 the noise
is 38kHz ripple, and in mono when the signal is merely
the output from the composite signal amp after the 19kHz, there
is a slightly excessive amount of 95kHz because the
bloomin 19kHz pilot tone of the BA1404 is not a sine wave even though I have
placed
an LC filter at the BA1404 gene to convert square wave 19kHz to a supposed
sine wave, according to the mods I faithfully carried out from a site on the
BA1404.
I am getting 20% of the 19kHz as 5th harmonic.




Some
books that discuss this subject include "An introduction To The Principles
Of Communication Theory" by John C. Hancock that I mentioned in an earlier
post. Hancock's treatment is probably too mathematical for you, so you
might want to look at "F-M Simplified" by Milton S. Kiver which was
written more for the Television serviceman type and has a good totally non
mathematical discussion of the subject with a lot of nice drawings.

Lord knows where I would find such books.


Finally you might want to look at Edwin H. Armstrong's paper in the May
1936 issue of the Proceedings of the IRE, titled "A Method Of Reducing
Disturbances In Radio Signaling By A System Of Frequency Modulation".
Armstrong was the inventor of the wide band FM system which he describes
in this paper. Like you Armstrong was a practical type who wasn't much
into math. It's been a while since I read this paper, but IIRC Armstrong
gives a good, largely intuitive, description of the reasons why FM has a
triangular noise spectrum.

I'll assume you are right.
Anyway, I am almost quite happy with results.
I need to try the set with my decent spaekers that are better than the
old workshop monitors i have.



You might want to dig a little deeper before you repeat statements like
the above again, all the world's radio knowledge isn't contained in the
RDH4th.

You don't say eh.

Patrick Turner.



Regards,

John Byrns

Surf my web pages at, http://users.rcn.com/jbyrns/

In article , Patrick Turner
wrote:

John Byrns wrote:

In article , Patrick Turner
wrote:

John Byrns wrote:

Hi Patrick,

Before I get to your current post, let me make a comment on an earlier
post in which you were describing the difference between your version of
the "diode ring" and QUAD's version. You commented with respect
to QUAD's
version that "The 4,000pF seems to be an incorrect value." I agree that
4,000pF doesn't seem quite right, by my calculations 4,000pF would
give a
de-emphasis time constant of about 68 usec. 3,000pF would seem to be
closer to the correct capacitor value to give a time constant of
50 usec.
The 900 uF capacitor with the resistor values in your circuit look like
they have a time constant of about 42.3 usec.

Quad use 4,000pF, way off the mark imho, maybe its a typo.
In Oz we have 50uS.
I tried 220pF, and got almost no de-emphasis.

I doubt it is a typo, it is close enough to the 3,000pF value required for
proper 50 usec de-emphasis, that it is likely what QUAD used. It was
quite common in the old days to use a de-emphasis time constant somewhat
longer than optimum.

But I have used almost the same identical circuit and needed only
a final 800pF.
So the drive impedance to the matrix I have must be higher if that is to be
added into the R which preceeds the C for the de-emphasis.

I don't think the drive impedance to the matrix is a big issue with either
your decoder or QUAD's. The time constant for your 800 pF capacitor, and
resistors as I understand them, comes out to about 37.6 usec., I am sure
the total time constant is right on at 50 usec. when other practical
matters in the circuit are considered. On the other hand QUAD's 4,000 pF
capacitor is a little on the large side, but not nearly as large as you
make out with a time constant of 61.7 usec. based on the resistor values
and without considering other practical matters.

I think the point you are missing about the QUAD decoder circuit is the
effect that the two 27k resistors for the auxiliary matrix have on the
time constant.

Early vacuum tube broadcast
stereo encoders like the RCA and Gates/Harris models used diode ring
modulators to generate the 38 kHz DSBSC L-R signal which was mixed with
the L+R audio after first passing though a 38 kHz bandpass filter to
eliminate the sidebands produced around the odd harmonics of 38 kHz, which
result from what is effectively a square wave 38 kHz sub carrier driving
the diode ring modulator. Look here for a schematic diagram of the RCA
circuit:

http://users.rcn.com/jbyrns/pics/BTS-1A-sch.jpg


Interesting.
The actual "ring" of transformers prior to the diodes looks
like it may be tricky to wind and set up.

Those transformers are just stock off the shelf UTC LS series audio line
transformers, I forget the precise model number, but they are 660 Ohm to
600 Ohm transformers with split primary and secondary windings.

As an aside note the diode ring modulator used in this circuit. What you
are calling a "diode ring" in the QUAD circuit isn't actually a diode ring
at all, it is two independent circuits, each using two diodes as a switch,
one for the "Left" channel and a second one for the "Right" channel.

Maybe you are right on the name for the ""ring"" of 4 diodes and four
resistors.

A circuit that uses square waves more explicitly is found in the Collins
786M-1 Broadcast Stereo Generator, an early transistor encoder based on
the switching principle. This is the encoder equivalent of the switching
principle used in the QUAD decoder, except in place of the diode switches
used in the QUAD decoder, Collins used transistor switches to alternately
short the inputs from each of the two audio channels to ground, sort of
like the Scott did with the two diode quads in their original tube
decoders.

The scott method is a real "switcher because" some of the diodes
just act as switches to allow a non floating secondary winding on the
locked oscillator plate circuit.
The single CT winding of the 38khz oscillator can therefore drive the matrix
from caps, and the usual extra tuned circuit is avoided, which would
make the drive impedance of the matrix lower.

The Quad circuit is also a "real switcher", the difference is that the
Scott design uses shunt switches while the QUAD design uses series
switches. It is also possible to use diode quads, like the ones Scott
used, as series switches in a QUAD style decoder, and the diode pairs that
QUAD used can also be used as shunt switches. I still have a multiplex
decoder that I designed and built 40 years ago that uses a Scott style
diode quad as a series switch. Note that the Scott style diode quad is
not what I would call a "diode ring" as all the diodes don't point in the
same direction around the ring, but that is probably just me.



I have never seen a decoder using transistor switches in this
way, I wonder why nobody seems to have tried transistor switches in a
decoder?

Sine waves didn't become popular in Stereo Generators until later. An
example is the Orban Optimod which used a pure 38 kHz sine wave as the sub
carrier, which was multiplied by the L-R audio signal in a linear
multiplier circuit. This approach using a pure sine wave and linear
multiplier eliminated the need for a filter at the output of the stereo
generator, and all the attendant problems the filter created. Note that
the use of sine waves requires a linear multiplier, not the diode switches
you used. There are some vintage Japanese high end FM tuners that have
stereo decoders based on the true sine wave and linear multiplier
principle.

I don't see there is a linear multiplier needed.
The 38kHz subcarrier waves look like sine waves to me, and
I see no reason why the simple Quad approach won't work OK.

You might try asking Robert Orban, or one of the manufacturers of vintage
Japanese tuners, about it if you don't understand it.

I suggest 20 dB of separation is fine for stereo enjoyment
from a radio.

Its a 20dB SPL ratio

Would you settle for 10% distortion in the amplifiers you design?

Distortion is very different to separation.

Having one speaker producing a sound level that is
10dB lower SPL does give you very decent stereo separation.

Yes, but when we get up into the 60 dB separation range favored by the
audiophile crowd, much of the cross talk from one speaker to the other is
actually distortion caused by the FM IF stages in the tuner! I will admit
though that this is probably not much of a factor when we are talking of
separation in the 20 dB range.

But since finalizing and trimming the values, I was able to get
38dB, and that's about the limit for the BA1404 gene.
If the max separation is set at 5kHz, it now drifts up to about 30dB.
That's with the set turned on its side on the bench
so I can access botyh top and bottom for adjusments; maybe there will be less
drift
when the set is laid flat; final tests are due in the next few nights.

38 dB is much better, it sounds like you are really getting the kinks
ironed out.

Separation remains constant for different levels of modulation signal
so the stereo detection appears to be linear.

I would certainly hope so.

Some
books that discuss this subject include "An introduction To The Principles
Of Communication Theory" by John C. Hancock that I mentioned in an earlier
post. Hancock's treatment is probably too mathematical for you, so you
might want to look at "F-M Simplified" by Milton S. Kiver which was
written more for the Television serviceman type and has a good totally non
mathematical discussion of the subject with a lot of nice drawings.

Lord knows where I would find such books.

Finally you might want to look at Edwin H. Armstrong's paper in the May
1936 issue of the Proceedings of the IRE, titled "A Method Of Reducing
Disturbances In Radio Signaling By A System Of Frequency Modulation".
Armstrong was the inventor of the wide band FM system which he describes
in this paper. Like you Armstrong was a practical type who wasn't much
into math. It's been a while since I read this paper, but IIRC Armstrong
gives a good, largely intuitive, description of the reasons why FM has a
triangular noise spectrum.

I'll assume you are right.

No need to assume, you said in an earlier post that you "have plenty of
auxilliary books on radio". It's hard to believe that at least one of
those books doesn't discuss the nature of the FM post detection noise
spectrum, you don't have to understand the math behind it, just the fact
that it is triangular, hopefully at least one of the books will have a
nice picture or drawing.


Regards,

John Byrns


Surf my web pages at, http://users.rcn.com/jbyrns/

John Byrns wrote:

In article , Patrick Turner
wrote:

John Byrns wrote:

In article , Patrick Turner
wrote:

John Byrns wrote:

Hi Patrick,

Before I get to your current post, let me make a comment on an earlier
post in which you were describing the difference between your version of
the "diode ring" and QUAD's version. You commented with respect
to QUAD's
version that "The 4,000pF seems to be an incorrect value." I agree that
4,000pF doesn't seem quite right, by my calculations 4,000pF would
give a
de-emphasis time constant of about 68 usec. 3,000pF would seem to be
closer to the correct capacitor value to give a time constant of
50 usec.
The 900 uF capacitor with the resistor values in your circuit look like
they have a time constant of about 42.3 usec.

Quad use 4,000pF, way off the mark imho, maybe its a typo.
In Oz we have 50uS.
I tried 220pF, and got almost no de-emphasis.

I doubt it is a typo, it is close enough to the 3,000pF value required for
proper 50 usec de-emphasis, that it is likely what QUAD used. It was
quite common in the old days to use a de-emphasis time constant somewhat
longer than optimum.

But I have used almost the same identical circuit and needed only
a final 800pF.
So the drive impedance to the matrix I have must be higher if that is to be
added into the R which preceeds the C for the de-emphasis.

I don't think the drive impedance to the matrix is a big issue with either
your decoder or QUAD's. The time constant for your 800 pF capacitor, and
resistors as I understand them, comes out to about 37.6 usec., I am sure
the total time constant is right on at 50 usec. when other practical
matters in the circuit are considered. On the other hand QUAD's 4,000 pF
capacitor is a little on the large side, but not nearly as large as you
make out with a time constant of 61.7 usec. based on the resistor values
and without considering other practical matters.

I found 800pF to be just right for de-emphasis.
I tried a range of C values and adjusted the C by 50pF incremants until
the response was 3dB down at 15 kHz and without any peaking below 10kHz.



I think the point you are missing about the QUAD decoder circuit is the
effect that the two 27k resistors for the auxiliary matrix have on the
time constant.

The R required for 4,000pF to get 50uS, or a pole at 3,200Hz
is 12.4k.

I don't have the auxiliary compensationn matrix so
indeed that makes a difference but it is irrelevant in my case
because I am using a slightly different circuit to the exact Quad circuit.





Early vacuum tube broadcast
stereo encoders like the RCA and Gates/Harris models used diode ring
modulators to generate the 38 kHz DSBSC L-R signal which was mixed with
the L+R audio after first passing though a 38 kHz bandpass filter to
eliminate the sidebands produced around the odd harmonics of 38 kHz, which
result from what is effectively a square wave 38 kHz sub carrier driving
the diode ring modulator. Look here for a schematic diagram of the RCA
circuit:

http://users.rcn.com/jbyrns/pics/BTS-1A-sch.jpg


Interesting.
The actual "ring" of transformers prior to the diodes looks
like it may be tricky to wind and set up.

Those transformers are just stock off the shelf UTC LS series audio line
transformers, I forget the precise model number, but they are 660 Ohm to
600 Ohm transformers with split primary and secondary windings.

So what would the circuit be like if used for demodulation?



As an aside note the diode ring modulator used in this circuit. What you
are calling a "diode ring" in the QUAD circuit isn't actually a diode ring
at all, it is two independent circuits, each using two diodes as a switch,
one for the "Left" channel and a second one for the "Right" channel.

Maybe you are right on the name for the ""ring"" of 4 diodes and four
resistors.

A circuit that uses square waves more explicitly is found in the Collins
786M-1 Broadcast Stereo Generator, an early transistor encoder based on
the switching principle. This is the encoder equivalent of the switching
principle used in the QUAD decoder, except in place of the diode switches
used in the QUAD decoder, Collins used transistor switches to alternately
short the inputs from each of the two audio channels to ground, sort of
like the Scott did with the two diode quads in their original tube
decoders.

The scott method is a real "switcher because" some of the diodes
just act as switches to allow a non floating secondary winding on the
locked oscillator plate circuit.
The single CT winding of the 38khz oscillator can therefore drive the matrix
from caps, and the usual extra tuned circuit is avoided, which would
make the drive impedance of the matrix lower.

The Quad circuit is also a "real switcher", the difference is that the
Scott design uses shunt switches while the QUAD design uses series
switches. It is also possible to use diode quads, like the ones Scott
used, as series switches in a QUAD style decoder, and the diode pairs that
QUAD used can also be used as shunt switches. I still have a multiplex
decoder that I designed and built 40 years ago that uses a Scott style
diode quad as a series switch. Note that the Scott style diode quad is
not what I would call a "diode ring" as all the diodes don't point in the
same direction around the ring, but that is probably just me.

In away Quad seems to switch, but in the sense that the detection
currents being fed to the L or R caps are only occuring
at each reveresal of phase of the 38khz waves.
So the pulses of current into the L cap can only come from
+/- across the 38kHz floating winding attatched to the L+R by CT to the emitter
follower,
and the pulses to the R cap come from -/+ across the winding.
there are 76,000 charge pulses per second, 38,000 for L and 38,000 for R.




I have never seen a decoder using transistor switches in this
way, I wonder why nobody seems to have tried transistor switches in a
decoder?

Sine waves didn't become popular in Stereo Generators until later. An
example is the Orban Optimod which used a pure 38 kHz sine wave as the sub
carrier, which was multiplied by the L-R audio signal in a linear
multiplier circuit. This approach using a pure sine wave and linear
multiplier eliminated the need for a filter at the output of the stereo
generator, and all the attendant problems the filter created. Note that
the use of sine waves requires a linear multiplier, not the diode switches
you used. There are some vintage Japanese high end FM tuners that have
stereo decoders based on the true sine wave and linear multiplier
principle.

I don't see there is a linear multiplier needed.
The 38kHz subcarrier waves look like sine waves to me, and
I see no reason why the simple Quad approach won't work OK.

You might try asking Robert Orban, or one of the manufacturers of vintage
Japanese tuners, about it if you don't understand it.

Well, either the Quad ideas and the othesr that are similar work as linear detectors
or they don't.

I did a search on Orban who has long list of equipment credits,
but I could not find a single schematic to dissect.




I suggest 20 dB of separation is fine for stereo enjoyment
from a radio.

Its a 20dB SPL ratio

Would you settle for 10% distortion in the amplifiers you design?

Distortion is very different to separation.

Having one speaker producing a sound level that is
10dB lower SPL does give you very decent stereo separation.

Yes, but when we get up into the 60 dB separation range favored by the
audiophile crowd, much of the cross talk from one speaker to the other is
actually distortion caused by the FM IF stages in the tuner! I will admit
though that this is probably not much of a factor when we are talking of
separation in the 20 dB range.

Its nice to get 60+ dB of separation, and that the channel which is the
low level one has low distortion, which is often does not.
Think of vinyl and the worship it gets from audiophiles, and the
stereo sep isn't 60dB, and there is an absense of Dn either.

I could not tell any difference to the Dn when switching from stereo
to mono in my decoder while listening to one speaker
only.
Many old tuners sound a little worse in stereo because of the
extra processing by the electronics.
You get some imaging, at the expense of fidelity.



But since finalizing and trimming the values, I was able to get
38dB, and that's about the limit for the BA1404 gene.
If the max separation is set at 5kHz, it now drifts up to about 30dB.
That's with the set turned on its side on the bench
so I can access botyh top and bottom for adjusments; maybe there will be less
drift
when the set is laid flat; final tests are due in the next few nights.

38 dB is much better, it sounds like you are really getting the kinks
ironed out.

Separation remains constant for different levels of modulation signal
so the stereo detection appears to be linear.

I would certainly hope so.

Some
books that discuss this subject include "An introduction To The Principles
Of Communication Theory" by John C. Hancock that I mentioned in an earlier
post. Hancock's treatment is probably too mathematical for you, so you
might want to look at "F-M Simplified" by Milton S. Kiver which was
written more for the Television serviceman type and has a good totally non
mathematical discussion of the subject with a lot of nice drawings.

Lord knows where I would find such books.

Finally you might want to look at Edwin H. Armstrong's paper in the May
1936 issue of the Proceedings of the IRE, titled "A Method Of Reducing
Disturbances In Radio Signaling By A System Of Frequency Modulation".
Armstrong was the inventor of the wide band FM system which he describes
in this paper. Like you Armstrong was a practical type who wasn't much
into math. It's been a while since I read this paper, but IIRC Armstrong
gives a good, largely intuitive, description of the reasons why FM has a
triangular noise spectrum.

I'll assume you are right.

No need to assume, you said in an earlier post that you "have plenty of
auxilliary books on radio". It's hard to believe that at least one of
those books doesn't discuss the nature of the FM post detection noise
spectrum, you don't have to understand the math behind it, just the fact
that it is triangular, hopefully at least one of the books will have a
nice picture or drawing.

I don't recall reading of the significance of "triangular" noise spectrum
in FM reception in the books I have.
I may go back to them if I have insurmountable noise problems.

Patrick Turner.




Regards,

John Byrns

Surf my web pages at, http://users.rcn.com/jbyrns/

In article , Patrick Turner
wrote:

John Byrns wrote:

I doubt it is a typo, it is close enough to the 3,000pF value
required for
proper 50 usec de-emphasis, that it is likely what QUAD used. It was
quite common in the old days to use a de-emphasis time constant somewhat
longer than optimum.

But I have used almost the same identical circuit and needed only
a final 800pF.
So the drive impedance to the matrix I have must be higher if that
is to be
added into the R which preceeds the C for the de-emphasis.

I don't think the drive impedance to the matrix is a big issue with either
your decoder or QUAD's. The time constant for your 800 pF capacitor, and
resistors as I understand them, comes out to about 37.6 usec., I am sure
the total time constant is right on at 50 usec. when other practical
matters in the circuit are considered. On the other hand QUAD's 4,000 pF
capacitor is a little on the large side, but not nearly as large as you
make out with a time constant of 61.7 usec. based on the resistor values
and without considering other practical matters.

I found 800pF to be just right for de-emphasis.
I tried a range of C values and adjusted the C by 50pF incremants until
the response was 3dB down at 15 kHz and without any peaking below 10kHz.

Yes, it looks like you have picked your C to be somewhat on the low side
to partly compensate for the effects of your low pass filters that you
mentioned in an earlier post.

I think the point you are missing about the QUAD decoder circuit is the
effect that the two 27k resistors for the auxiliary matrix have on the
time constant.

The R required for 4,000pF to get 50uS, or a pole at 3,200Hz
is 12.4k.

The R in the QUAD design is 15.4k, not too far off from your theoretical 12.4k.

I don't have the auxiliary compensationn matrix so
indeed that makes a difference but it is irrelevant in my case
because I am using a slightly different circuit to the exact Quad circuit.

That was exactly my point, the resistance in the QUAD circuit is quite a
bit lower than the resistance in your circuit, so the QUAD's capacitor is
quite a bit smaller than your capacitor. Plus you went a bit light on the
capacitor and still ended up 3 dB down at 15 kHz because of the effects of
your low pass filters. QUAD's choice of capacitor should put the response
of the QUAD decoder only 1.75 dB down at 15 kHz.

Early vacuum tube broadcast
stereo encoders like the RCA and Gates/Harris models used diode ring
modulators to generate the 38 kHz DSBSC L-R signal which was mixed with
the L+R audio after first passing though a 38 kHz bandpass filter to
eliminate the sidebands produced around the odd harmonics of 38
kHz, which
result from what is effectively a square wave 38 kHz sub carrier driving
the diode ring modulator. Look here for a schematic diagram of the RCA
circuit:

http://users.rcn.com/jbyrns/pics/BTS-1A-sch.jpg


Interesting.
The actual "ring" of transformers prior to the diodes looks
like it may be tricky to wind and set up.

Those transformers are just stock off the shelf UTC LS series audio line
transformers, I forget the precise model number, but they are 660 Ohm to
600 Ohm transformers with split primary and secondary windings.

So what would the circuit be like if used for demodulation?

Pretty much the same, except of course you would need to replace the
crystal controlled 38 kHz oscillator with one locked to the received 19
kHz pilot tone. The inputs and outputs would of course be reversed, the
composite signal from the ratio detector would have to be buffered and
used to drive the "Band Pass Network" and the "Time Delay Network", AKA
low pass filter, with appropriate means provided to adjust the relative
levels for optimum separation. The pre-emphasis networks would need to be
replaced with de-emphasis networks, and the Left and Right channel outputs
would come from what were originally the inputs.

Notice that what you called the "ring of transformers" are in reality the
audio matrix that creates the L+R and L-R audio signals from the Left and
Right channel inputs. If we reversed the whole process and made it into a
decoder the matrix works in reverse, generating Left and Right channel
audio from the L+R signal and the demodulated L-R signal from the "diode
ring modulator".

In away Quad seems to switch, but in the sense that the detection
currents being fed to the L or R caps are only occuring
at each reveresal of phase of the 38khz waves.
So the pulses of current into the L cap can only come from
+/- across the 38kHz floating winding attatched to the L+R by CT to the
emitter
follower,
and the pulses to the R cap come from -/+ across the winding.
there are 76,000 charge pulses per second, 38,000 for L and 38,000 for R.

If I am following you, you are simply describing the basic operation of a
switching decoder, ignoring subtleties like correcting for the fact that
the 38 kHz L-R sub carrier is effectively transmitted at a lower power
level than the L+R signal.

Sine waves didn't become popular in Stereo Generators until later. An
example is the Orban Optimod which used a pure 38 kHz sine wave as
the sub
carrier, which was multiplied by the L-R audio signal in a linear
multiplier circuit. This approach using a pure sine wave and linear
multiplier eliminated the need for a filter at the output of the stereo
generator, and all the attendant problems the filter created. Note that
the use of sine waves requires a linear multiplier, not the diode
switches
you used. There are some vintage Japanese high end FM tuners that have
stereo decoders based on the true sine wave and linear multiplier
principle.

I don't see there is a linear multiplier needed.
The 38kHz subcarrier waves look like sine waves to me, and
I see no reason why the simple Quad approach won't work OK.

You might try asking Robert Orban, or one of the manufacturers of vintage
Japanese tuners, about it if you don't understand it.

Well, either the Quad ideas and the othesr that are similar work as
linear detectors
or they don't.

Certainly that is the hope. The idea of the sine wave and linear
multiplier approach is that it eliminates the harmonic responses at
multiples, mostly odd, of 38 kHz that occur when square wave switching is
used, as in your decoder. These responses, for example the sidebands
around 114 kHz, must be eliminated by a low pass filter before the
composite signal is feed to the FM modulator, otherwise the transmitter is
likely to be in violation of the government regulations. The sine wave
and linear multiplier eliminate the need for the low pass filter. In a
decoder, the effect of square wave switching, of the type you are using,
is to degrade the signal to noise ratio by the demodulation of the noise
around 114 kHz coming from the ratio detector. Due to the fact that the
sine wave drive you are using does not create a perfect 50% duty cycle
square wave, there will also be some second harmonic present at 76 kHz,
which will also demodulate the noise in that area. If you have a strong
noise free signal, and no SCA carriers, this may not be of particular
concern to you.

I don't recall reading of the significance of "triangular" noise spectrum
in FM reception in the books I have.
I may go back to them if I have insurmountable noise problems.

If you have strong signals it probably won't matter much to you.


Regards,

John Byrns


Surf my web pages at, http://users.rcn.com/jbyrns/

John Byrns wrote:

In article , Patrick Turner
wrote:

John Byrns wrote:

I doubt it is a typo, it is close enough to the 3,000pF value
required for
proper 50 usec de-emphasis, that it is likely what QUAD used. It was
quite common in the old days to use a de-emphasis time constant somewhat
longer than optimum.

But I have used almost the same identical circuit and needed only
a final 800pF.
So the drive impedance to the matrix I have must be higher if that
is to be
added into the R which preceeds the C for the de-emphasis.

I don't think the drive impedance to the matrix is a big issue with either
your decoder or QUAD's. The time constant for your 800 pF capacitor, and
resistors as I understand them, comes out to about 37.6 usec., I am sure
the total time constant is right on at 50 usec. when other practical
matters in the circuit are considered. On the other hand QUAD's 4,000 pF
capacitor is a little on the large side, but not nearly as large as you
make out with a time constant of 61.7 usec. based on the resistor values
and without considering other practical matters.

I found 800pF to be just right for de-emphasis.
I tried a range of C values and adjusted the C by 50pF incremants until
the response was 3dB down at 15 kHz and without any peaking below 10kHz.

Yes, it looks like you have picked your C to be somewhat on the low side
to partly compensate for the effects of your low pass filters that you
mentioned in an earlier post.

But there are no low pass filters in the design I am trying now
that are before the matrixing.
In fact the HF part of the composite signal is just slightly
boosted.
I do have low pass filters after the matrixing to remove the 38khz ripple
signals and this is a critically damped LC filter which is about 1dB down at 16kHz,
so it
does not attenuate the audio band.

The Quad method is simple, and there is no other de-emphasis
networks on the stereo outputs.
I just went for whatever value C worked.



I think the point you are missing about the QUAD decoder circuit is the
effect that the two 27k resistors for the auxiliary matrix have on the
time constant.

The R required for 4,000pF to get 50uS, or a pole at 3,200Hz
is 12.4k.

The R in the QUAD design is 15.4k, not too far off from your theoretical 12.4k.

I don't have the auxiliary compensationn matrix so
indeed that makes a difference but it is irrelevant in my case
because I am using a slightly different circuit to the exact Quad circuit.

That was exactly my point, the resistance in the QUAD circuit is quite a
bit lower than the resistance in your circuit, so the QUAD's capacitor is
quite a bit smaller than your capacitor. Plus you went a bit light on the
capacitor and still ended up 3 dB down at 15 kHz because of the effects of
your low pass filters.

Wait until you see the schematic before commenting.



QUAD's choice of capacitor should put the response
of the QUAD decoder only 1.75 dB down at 15 kHz.

Early vacuum tube broadcast
stereo encoders like the RCA and Gates/Harris models used diode ring
modulators to generate the 38 kHz DSBSC L-R signal which was mixed with
the L+R audio after first passing though a 38 kHz bandpass filter to
eliminate the sidebands produced around the odd harmonics of 38
kHz, which
result from what is effectively a square wave 38 kHz sub carrier driving
the diode ring modulator. Look here for a schematic diagram of the RCA
circuit:

http://users.rcn.com/jbyrns/pics/BTS-1A-sch.jpg


Interesting.
The actual "ring" of transformers prior to the diodes looks
like it may be tricky to wind and set up.

Those transformers are just stock off the shelf UTC LS series audio line
transformers, I forget the precise model number, but they are 660 Ohm to
600 Ohm transformers with split primary and secondary windings.

So what would the circuit be like if used for demodulation?

Pretty much the same, except of course you would need to replace the
crystal controlled 38 kHz oscillator with one locked to the received 19
kHz pilot tone. The inputs and outputs would of course be reversed, the
composite signal from the ratio detector would have to be buffered and
used to drive the "Band Pass Network" and the "Time Delay Network", AKA
low pass filter, with appropriate means provided to adjust the relative
levels for optimum separation. The pre-emphasis networks would need to be
replaced with de-emphasis networks, and the Left and Right channel outputs
would come from what were originally the inputs.

Complex to imagine.

I'll leave that for you to doodle up in a sketch...



Notice that what you called the "ring of transformers" are in reality the
audio matrix that creates the L+R and L-R audio signals from the Left and
Right channel inputs. If we reversed the whole process and made it into a
decoder the matrix works in reverse, generating Left and Right channel
audio from the L+R signal and the demodulated L-R signal from the "diode
ring modulator".

In away Quad seems to switch, but in the sense that the detection
currents being fed to the L or R caps are only occuring
at each reveresal of phase of the 38khz waves.
So the pulses of current into the L cap can only come from
+/- across the 38kHz floating winding attatched to the L+R by CT to the
emitter
follower,
and the pulses to the R cap come from -/+ across the winding.
there are 76,000 charge pulses per second, 38,000 for L and 38,000 for R.

If I am following you, you are simply describing the basic operation of a
switching decoder, ignoring subtleties like correcting for the fact that
the 38 kHz L-R sub carrier is effectively transmitted at a lower power
level than the L+R signal.

Maybe it is a lower "power level"
The DSB will always be a lower modulation signal level than the L+R signal.
But say I modulate just one channel of the stereo BA1404 sig gene.
Then the peak voltage of L+R signal is equal in amplitude to the
peak to peak voltage of the DSB signaL.
So presumably the DSB only causes 1/2 the 100MHz F deviation.
The power in the FM signal does not change, since the 100MHz carrier amplitude
remains constant.




Sine waves didn't become popular in Stereo Generators until later. An
example is the Orban Optimod which used a pure 38 kHz sine wave as
the sub
carrier, which was multiplied by the L-R audio signal in a linear
multiplier circuit. This approach using a pure sine wave and linear
multiplier eliminated the need for a filter at the output of the stereo
generator, and all the attendant problems the filter created. Note that
the use of sine waves requires a linear multiplier, not the diode
switches
you used. There are some vintage Japanese high end FM tuners that have
stereo decoders based on the true sine wave and linear multiplier
principle.

I don't see there is a linear multiplier needed.
The 38kHz subcarrier waves look like sine waves to me, and
I see no reason why the simple Quad approach won't work OK.

You might try asking Robert Orban, or one of the manufacturers of vintage
Japanese tuners, about it if you don't understand it.

Well, either the Quad ideas and the othesr that are similar work as
linear detectors
or they don't.

Certainly that is the hope. The idea of the sine wave and linear
multiplier approach is that it eliminates the harmonic responses at
multiples, mostly odd, of 38 kHz that occur when square wave switching is
used, as in your decoder.

But there isn't any square wave switching.
There is just charging of the L&R caps by the peaks of the 30khz sine waves.
These act like pulse charging of the L&R caps and only for a brief
time, or a small % of the 38khz sine wave cycle.
If the 38kHz waves were square waves, there'd be a lot more time
for the charging of the caps which could then be a larger value.
But still the L and R signals will be formed by pulses occuring
that can be no longer than 1/2 the wave cycle period of a 38khz wave, ie 13.1 uS.


These responses, for example the sidebands
around 114 kHz, must be eliminated by a low pass filter before the
composite signal is feed to the FM modulator, otherwise the transmitter is
likely to be in violation of the government regulations. The sine wave
and linear multiplier eliminate the need for the low pass filter. In a
decoder, the effect of square wave switching, of the type you are using,
is to degrade the signal to noise ratio by the demodulation of the noise
around 114 kHz coming from the ratio detector.

I don't see any artifacts at 114khz in my design.

Due to the fact that the
sine wave drive you are using does not create a perfect 50% duty cycle
square wave, there will also be some second harmonic present at 76 kHz,
which will also demodulate the noise in that area. If you have a strong
noise free signal, and no SCA carriers, this may not be of particular
concern to you.

There are few unwanted harmonics in the output from the
38khz demod process.

In my original design that's still on my website i tried to separate
the composite signal into its components with HPF and LPF.

Then the signals above 23kHz was the double sideband signal
containing the modulation information of L-R audio.
This DSB was then fed to an LTP on one side and 38kHz carrier to the other,
and at the output anodes I got two opositely phased versions of the reconstructed
38kHz AM signal modulated bt L-R.
Normal full wave rectifires ( and not a diode ring like Quad used, ) was able to
produce
the audio L-R and -L+R signals, and the ripple frequency in the process is 76 kHz.

As far as i know its impossible to combine the DSB with L-R mod with a 38khz signal
and with the L+R signal included and be able to produce two 38kHz modulated waves
of opposite phase which can be detected by any full wave rectifier to procuce L and
R
audio with 76kHz ripple F.

But some may like to set up a simulation of adding the same phase of
38kHz oscillator carrier to each of two phases of the composite signal,
so examine if the L and R signals can thus be decoded at 76 kHz instead of at 38
kHz,
which effectively makes it more accurate to describe the HF waves of audio.

With the decoders I am using and what was common in 1960,
a tone of 15kHz had to be described by the amplitudes of the peaks of waves of
38kHz,
ie, for 10 waves of 15khz, there are 2.53 waves each with only one peak that can be
used to
maye a stair case wave in which 15khz is buried.
Digital is similar; to describe 18kHz with a 44.1 kHz oscilator wave, there can be
only
2.45 samples taken per 18khz wave.

The more "dots to join" the better the recovered audio wave is likely
to be close to the original.

The other possiblity is that a reconstructed 38kHz AM wave with
L-R mod may be heterodyned with a second higher F to produce
an AM signal at say 100kHz which is then easier to detect.

Patrick Turner.




I don't recall reading of the significance of "triangular" noise spectrum
in FM reception in the books I have.
I may go back to them if I have insurmountable noise problems.

If you have strong signals it probably won't matter much to you.

Regards,

John Byrns

Surf my web pages at, http://users.rcn.com/jbyrns/