UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio and TV Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > General Vintage Technology > Components and Circuits

Notices

Components and Circuits For discussions about component types, alternatives and availability, circuit configurations and modifications etc. Discussions here should be of a general nature and not about specific sets.

Closed Thread
 
Thread Tools
Old 9th Sep 2021, 5:39 pm   #1
AidanCroft
Tetrode
 
Join Date: Feb 2021
Location: Crewe, Cheshire, UK.
Posts: 66
Default Ratio and Foster-Seeley detectors

After hours of research on the above, I've managed to improve my knowledge but I'm still stuck on one or two points.

For both, I understand that both diodes conduct equally and with respect to the secondary winding centre tap, so giving zero output overall. This is the same when the secondary winding polarity changes, I believe.

Above resonance, the primary tank acts as an inductor and so secondary current lags the primary tank voltage. My confusion is - why should this mean that ONLY the sum of secondary voltage e1 and primary voltage ep is higher (so diode Cr1 conducts more)? Surely when the polarity of the secondary winding changes then e2+ep will then be higher?

The opposite of the above paragraph is true below resonance but again, why doesn't that reverse when secondary polarity reverses?

Also, in any case, why should a lagging or leading current cause either e1 or e2 to have a greater output across its diode with respect to ep - surely the same current and voltage is induced either side of the centre tap of the secondary?

Kind regards,

Aidan.
Attached Thumbnails
Click image for larger version

Name:	0261.gif
Views:	180
Size:	5.4 KB
ID:	240922  

Last edited by AidanCroft; 9th Sep 2021 at 5:45 pm.
AidanCroft is offline  
Old 9th Sep 2021, 7:27 pm   #2
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,801
Default Re: Ratio and Foster-Seeley detectors

Slope Detection

The most basic way to receive FM is by slope demodulation. An AM receiver is off-tuned until the centre frequency of the incoming signal is part way up the slope on one side of its selectivity curve. The receiver BW needs to be wide enough to fit in most of the sidebands of the modulated signal into the slope region.

This sort of works, in an awful way. Remember the receiver's peak sensitivity is offset out of the channel being received. It's wide open to extra noise and other signals. But it's a start.

Travis Detector

It's a lie! the Travis is really two detectors. Rather than using the main receiver IF selectivity the Travis uses a single resonator after the IF has done its work. The slope converts FM into AM and a diode detector detects the AM to give the audio. But the Travis has TWO resonators, one pitched above the IF, the other below. They work as a balanced arrangement with the audio taken as the difference between the diode detectors. The differential arrangement then gets simplified by having one detector produce a + going output, the other a - going output. Some variants of the Travis introduce a third resonator to improve linearity a little. Note that the Travis is still an FM to AM converter driving an AM detector. Any AM on the signal gets demodulate just fine. To stop this, it needs a very good limiter function in the IF.

Back in the days of analogue FDM telephony over microwaves, 2600 phone calls got packed together as 18.6MHz of SSB signals and FM'd onto a 140MHz carrier. This got converted to 8 or 11 GHz, squirted along a series of microwave towers and down converted to 140 MHz for a Travis detector to demod it back to baseband... 18.6MHz of SSB signals. Let's say the Travis detector can be made very linear to do this job.

Foster-Seeley detector

This is a simplified version of the Travis, with some common stuff combined and the final subtraction made easier. This is a good linear detector and was used in many early FM tuners.

Ratio Detector

Some references will tell you that this is a derivative of the Foster Seeley. This isn't quite true and is very misleading. The radio detector isn't a slope detector at heart. It adds two IF signals with a frequency-dependent phase shift between them. Vector addition of these two signals gives us an audio signal. The ratio detector was lauded as a way to reduce the Foster-Seeley's susceptibility to AM signals, and allowed a cost reduction by needing a less good limiting IF system, typically saving a stage. There is a price to pay, the linearity isn't quite as goods as the Foster-Seeley or Travis. You can't really understand the ratio detector without going through all the phasor diagrams and understanding the approximations used.

The extra transformer winding needed was seen as worth it for saving an IF stage. Accountancy ruled!

Quadrature Detector

The fancy transformer of the ratio detector wasn't going to fit on a silicon IC, so another technique was tapped. The idea of multiplying the IF signal by its own carrier, phase shifted by 90 degrees. In fact multiplying it by its whole self phase shifted through 90 degrees is not just OK and easier, it's actually better.

The multiplier is easy on a chip (what is erroneously but almost universally called a Gilbert Cell mixer... the most notable exception being Barrie Gilbert!) The phase shift network can be done fairly well with one inductor, but a bit better with two. This can be amde a bit better than the ratio detector. It too is AM-sensitive, but FM IF chips usually includ good limiting stages to handle that. I wouldn't say this is a very low distortion detector, but it's on the way.

Pulse Count Discriminator

Very high quality FM tuner designers wanted something a step better than quadrature detectors. A pulse counter watches the IF signal, and once per cycle it fires off a pulse of carefully controlled amplitude and duration. The average of these pulses is created by a lowpass filter. Distortion can be excellent. Phase linearity of Zenith system stereo signal components also excellent. However, it's usually best to down convert the IF to a lower one where the deviation is relatively wider when seen as a fraction of the centre frequency. You'll find these in some posh tuners, like the Revox ones.

So there you have it. Don't sweat too hard at the ratio detector. It is a bit of a sod to get your head around, but its performance is limited and it has been upstaged a few times.

It was a technique to save one IF limiter stage. There has been a hell of a lot of them made, but people have forgotten the compromises packaged in them, and they're not redily visible without doing the phasor diagrams. I had to study these things as an undergrad student, but I've not got the ratio detector diagrams in my head. They aren't needed.

One thing to remember with the ratio detector is that it needs a large capacitor which is part of the AM removal trick. If it is electrolytic and goes leaky, performance is spoiled.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is offline  
Old 9th Sep 2021, 7:50 pm   #3
AidanCroft
Tetrode
 
Join Date: Feb 2021
Location: Crewe, Cheshire, UK.
Posts: 66
Default Re: Ratio and Foster-Seeley detectors

Quote:
Originally Posted by Radio Wrangler View Post
Slope Detection

The most basic way to receive FM is by slope demodulation. An AM receiver is off-tuned until the centre frequency of the incoming signal is part way up the slope on one side of its selectivity curve. The receiver BW needs to be wide enough to fit in most of the sidebands of the modulated signal into the slope region.

This sort of works, in an awful way. Remember the receiver's peak sensitivity is offset out of the channel being received. It's wide open to extra noise and other signals. But it's a start.

Travis Detector

It's a lie! the Travis is really two detectors. Rather than using the main receiver IF selectivity the Travis uses a single resonator after the IF has done its work. The slope converts FM into AM and a diode detector detects the AM to give the audio. But the Travis has TWO resonators, one pitched above the IF, the other below. They work as a balanced arrangement with the audio taken as the difference between the diode detectors. The differential arrangement then gets simplified by having one detector produce a + going output, the other a - going output. Some variants of the Travis introduce a third resonator to improve linearity a little. Note that the Travis is still an FM to AM converter driving an AM detector. Any AM on the signal gets demodulate just fine. To stop this, it needs a very good limiter function in the IF.

Back in the days of analogue FDM telephony over microwaves, 2600 phone calls got packed together as 18.6MHz of SSB signals and FM'd onto a 140MHz carrier. This got converted to 8 or 11 GHz, squirted along a series of microwave towers and down converted to 140 MHz for a Travis detector to demod it back to baseband... 18.6MHz of SSB signals. Let's say the Travis detector can be made very linear to do this job.

Foster-Seeley detector

This is a simplified version of the Travis, with some common stuff combined and the final subtraction made easier. This is a good linear detector and was used in many early FM tuners.

Ratio Detector

Some references will tell you that this is a derivative of the Foster Seeley. This isn't quite true and is very misleading. The radio detector isn't a slope detector at heart. It adds two IF signals with a frequency-dependent phase shift between them. Vector addition of these two signals gives us an audio signal. The ratio detector was lauded as a way to reduce the Foster-Seeley's susceptibility to AM signals, and allowed a cost reduction by needing a less good limiting IF system, typically saving a stage. There is a price to pay, the linearity isn't quite as goods as the Foster-Seeley or Travis. You can't really understand the ratio detector without going through all the phasor diagrams and understanding the approximations used.

The extra transformer winding needed was seen as worth it for saving an IF stage. Accountancy ruled!

Quadrature Detector

The fancy transformer of the ratio detector wasn't going to fit on a silicon IC, so another technique was tapped. The idea of multiplying the IF signal by its own carrier, phase shifted by 90 degrees. In fact multiplying it by its whole self phase shifted through 90 degrees is not just OK and easier, it's actually better.

The multiplier is easy on a chip (what is erroneously but almost universally called a Gilbert Cell mixer... the most notable exception being Barrie Gilbert!) The phase shift network can be done fairly well with one inductor, but a bit better with two. This can be amde a bit better than the ratio detector. It too is AM-sensitive, but FM IF chips usually includ good limiting stages to handle that. I wouldn't say this is a very low distortion detector, but it's on the way.

Pulse Count Discriminator

Very high quality FM tuner designers wanted something a step better than quadrature detectors. A pulse counter watches the IF signal, and once per cycle it fires off a pulse of carefully controlled amplitude and duration. The average of these pulses is created by a lowpass filter. Distortion can be excellent. Phase linearity of Zenith system stereo signal components also excellent. However, it's usually best to down convert the IF to a lower one where the deviation is relatively wider when seen as a fraction of the centre frequency. You'll find these in some posh tuners, like the Revox ones.

So there you have it. Don't sweat too hard at the ratio detector. It is a bit of a sod to get your head around, but its performance is limited and it has been upstaged a few times.

It was a technique to save one IF limiter stage. There has been a hell of a lot of them made, but people have forgotten the compromises packaged in them, and they're not redily visible without doing the phasor diagrams. I had to study these things as an undergrad student, but I've not got the ratio detector diagrams in my head. They aren't needed.

One thing to remember with the ratio detector is that it needs a large capacitor which is part of the AM removal trick. If it is electrolytic and goes leaky, performance is spoiled.

David
A lovely detailed response, thanks. Do you know what's going on in relation to those couple of things I was asking about? I really want to fully understand its working.

Aidan.
AidanCroft is offline  
Old 9th Sep 2021, 7:55 pm   #4
ms660
Dekatron
 
ms660's Avatar
 
Join Date: Apr 2011
Location: Cornwall, UK.
Posts: 13,454
Default Re: Ratio and Foster-Seeley detectors

Have you seen the vector diagrams?

Lawrence.
ms660 is offline  
Old 9th Sep 2021, 8:00 pm   #5
AidanCroft
Tetrode
 
Join Date: Feb 2021
Location: Crewe, Cheshire, UK.
Posts: 66
Default Re: Ratio and Foster-Seeley detectors

Quote:
Originally Posted by ms660 View Post
Have you seen the vector diagrams?

Lawrence.
Yea I've seen that the two secondary voltages get "rotated", which puts one of them more in phase with the primary voltage (so adding) whilst the other secondary voltage ends more out of phase with the primary voltage.

My confusion though, why should this mean that ONLY the sum of secondary voltage e1 and primary voltage ep is higher (so diode Cr1 conducts more)? Surely when the polarity of the secondary winding changes then e2+ep will then be higher?

The opposite of the above paragraph is true below resonance but again, why doesn't that reverse when secondary polarity reverses?

Also, in any case, why should a lagging or leading current cause either e1 or e2 to have a greater output across its diode with respect to ep - surely the same current and voltage is induced either side of the centre tap of the secondary?

Kind regards,

Aidan.
AidanCroft is offline  
Old 9th Sep 2021, 8:12 pm   #6
ms660
Dekatron
 
ms660's Avatar
 
Join Date: Apr 2011
Location: Cornwall, UK.
Posts: 13,454
Default Re: Ratio and Foster-Seeley detectors

When e1 increases and e2 decreases the phase relationship between e1 and e2 stays the same, they are 180 degrees apart due to the center tapped secondary, the increase and decrease of the voltages across them is due to the increase and decrease of the frequency either side of the rest frequency (the center frequency) as shown by the rotating vector.

Lawrence.
ms660 is offline  
Old 9th Sep 2021, 8:26 pm   #7
AidanCroft
Tetrode
 
Join Date: Feb 2021
Location: Crewe, Cheshire, UK.
Posts: 66
Default Re: Ratio and Foster-Seeley detectors

Quote:
Originally Posted by ms660 View Post
When e1 increases and e2 decreases the phase relationship between e1 and e2 stays the same, they are 180 degrees apart due to the center tapped secondary, the increase and decrease of the voltages across them is due to the increase and decrease of the frequency either side of the rest frequency (the center frequency) as shown by the rotating vector.

Lawrence.
Thanks. When the polarity of the primary changes, won't e1 then decrease and e2 decrease? This is where part of my confusion lies.
AidanCroft is offline  
Old 9th Sep 2021, 8:42 pm   #8
ms660
Dekatron
 
ms660's Avatar
 
Join Date: Apr 2011
Location: Cornwall, UK.
Posts: 13,454
Default Re: Ratio and Foster-Seeley detectors

Quote:
Originally Posted by AidanCroft View Post
Thanks. When the polarity of the primary changes, won't e1 then decrease and e2 decrease? This is where part of my confusion lies.
Changing the polarity on the primary doesn't change the phase relationship only a change in frequency does that.

Lawrence.
ms660 is offline  
Old 9th Sep 2021, 9:31 pm   #9
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,801
Default Re: Ratio and Foster-Seeley detectors

There is a second way of looking at the ratio detector/ The diodes can be viewed as a half-wave rectifier AND a sampling bridge, both at the same time.

The full centre-tapped main secondary of the IFT drives the diodes to charge the electrolytic capacitor. THe capacitor charges to the peak voltage of the IF and then slowly discharges into its load resistors over the rest of the cycle. Note that the rectifier diode amounts to both diodes in series, distributed around the same circuit loop. Note that both turn on at once. over time, the load resistors on the big capacitor discharge it a little and so set the time the diodes have to turn on for to refresh the charge on the capacitor. So the diodes are both on for a fraction of the time of the IF cycle.

THe extra secondary, with its load capacitor gives a phase-shifted version of the IF signal. This phase-shifted signal goes through to the audio output, but during the on-time of the diodes, it is shorted out, back to the centre tap of the main secondary.

So the diodes act as a sampler, running at 90 degrees to the IF signal.... in other words we have a form of quadrature detector. As frequency is changed, the phase relationship of the strong signal driving the sampling diodes is no longer working on a perfect 90 degree offset signal from the third winding. This swinging around of the signal path with respect to the sampling phase gives variable efficiency of the synchronous rectifier effect.

Put a bigger signal into the IF and the big capacitor has to charge to more volts. The resistor loads take more current and the capacitor takes longer to charge to the new full voltage each time. So the diodes are on for longer, and shorting the third winding signal for longer. Thus acting to reduce the output signal when the input signal is longer. Not that the timeconstant of the big capacitor and its load resistors has to be slow enough to not track the lowest frequency AM component you want to reject. Within that limitation, you get a degree of compensation for input voltage at the IFT.

So the real signal is that on the phse shifted third winding. The apparently main secondary of the IFT is there to charge up the capacitor and to hammer the diodes and to make them work as a sampler. On top of it all, the sampler shorts the signal rather than the more obvious process of passing it. This comes out in the wash as equivalent.

There is a secondary effect of the phase shifted signal having some influence on the sampler switching time.

From a pont of view of distortion, the thing can be seen to have similarities to the quadrature detector.... the basic quadrature detector with a single tuned circuit as phase shifter. But the quadrature detector in reasonable grade FM tuners gets tarted up with two resonators in its phase shift network. Something similar could be done to improve the ratio detector and drop its distortion a little, but nobody bothers. Time has changed and they just jump straight to the quadrature detector with two coils and an IC. The CA3089 and CA3090 really did take over the FM tuner world and their offspring kept it up for a lot of years.

There isn't much to do to a ratio detector in a classic radio. Resistors go high, sometimes diodes fail, but mostly it's the bif capacitor ageing and leaking. There's always the attentions of the phantom tweaker, of course. He's particularly drawn to discriminators because they are extra fiddly to correct afterwards. A sweeper/wobbulator and a scope will let you get the S curve wide enough, reasonably straight and centre to suit the IF filtering. Ths transformers just live on so long as the phantom didn't go for extra points and crack a slug...

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is offline  
Old 9th Sep 2021, 10:13 pm   #10
Restoration73
Nonode
 
Restoration73's Avatar
 
Join Date: Jun 2013
Location: Surbiton, SW London, UK.
Posts: 2,801
Default Re: Ratio and Foster-Seeley detectors

Another method of FM demodulation uses a PLL - Phase Locked Loop.

Certainly in consumer designs, the ratio detector and the quad detector are most
prevalent. The quad circuit generates a higher level of white noise in the absence of a
carrier, but as this type uses a custom IC these usually contain a muting or "squelch"
circuit to prevent it being heard.
Restoration73 is offline  
Old 9th Sep 2021, 10:25 pm   #11
AidanCroft
Tetrode
 
Join Date: Feb 2021
Location: Crewe, Cheshire, UK.
Posts: 66
Default Re: Ratio and Foster-Seeley detectors

So what's the relationship between the primary and the two halves of the secondary? How is it even possible for the two halves to have different values on the same coil? Obviously the tertiary is centre tapped to it and a primary behaving inductively or capacitively is what changes the values of the two secondary halves, but how? Via what mechanism? What physics is going on?

If I can understand that I might have some luck!!
AidanCroft is offline  
Old 9th Sep 2021, 10:58 pm   #12
kalee20
Dekatron
 
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 7,061
Default Re: Ratio and Foster-Seeley detectors

Just imagine the voltage across ep is zero, and the voltage across the whole secondary is 20V peak. So, C3 gets charged to 10V and so does C4. The sum is 20V.

The midpoint is at B, and the output is taken from A, the junction of C3 and C4. Since the voltage is equal, there will be zero output.

Now let's suppose there is a small voltage on ep, in-phase with e1. The upper diode now sees 11V peak and C3 gets charged to 11V. But the lower diode sees only 9V peak, so C4 gets charged to 9V. The sum is still 20V, but the junction is offset by 1V, so you get 1V of output.

If the phase on ep is reversed, then C3 sees 9V and C4 sees 11V, so now your output is -1V. Although none of the AC voltages have changed, the phase relations have, and this has given an opposite-polarity output voltage.

It's now a fairly straightforward step to consider what happens if ep3 voltage is not in-phase or out-of-phase, but 90 degrees of shift. Pythagoras tells us that each capacitor will see 10.05V, and the sum will be 20.1V, hardly any change. The mid-point will be at 10.05V, and because C3 and C4 are equally charged, the output will be zero.

If the phase relationship changes from 90 degrees and heads towards zero, yo start to get to the first situation where C3 gets charged more and C4 less, so you see a positive output. And if it changes in the other direction, towards 180 degrees, you start to move towards the other situation, and get a negative output.
kalee20 is offline  
Old 10th Sep 2021, 12:17 am   #13
AidanCroft
Tetrode
 
Join Date: Feb 2021
Location: Crewe, Cheshire, UK.
Posts: 66
Default Re: Ratio and Foster-Seeley detectors

Quote:
Originally Posted by kalee20 View Post
Just imagine the voltage across ep is zero, and the voltage across the whole secondary is 20V peak. So, C3 gets charged to 10V and so does C4. The sum is 20V.

The midpoint is at B, and the output is taken from A, the junction of C3 and C4. Since the voltage is equal, there will be zero output.

Now let's suppose there is a small voltage on ep, in-phase with e1. The upper diode now sees 11V peak and C3 gets charged to 11V. But the lower diode sees only 9V peak, so C4 gets charged to 9V. The sum is still 20V, but the junction is offset by 1V, so you get 1V of output.

If the phase on ep is reversed, then C3 sees 9V and C4 sees 11V, so now your output is -1V. Although none of the AC voltages have changed, the phase relations have, and this has given an opposite-polarity output voltage.

It's now a fairly straightforward step to consider what happens if ep3 voltage is not in-phase or out-of-phase, but 90 degrees of shift. Pythagoras tells us that each capacitor will see 10.05V, and the sum will be 20.1V, hardly any change. The mid-point will be at 10.05V, and because C3 and C4 are equally charged, the output will be zero.

If the phase relationship changes from 90 degrees and heads towards zero, yo start to get to the first situation where C3 gets charged more and C4 less, so you see a positive output. And if it changes in the other direction, towards 180 degrees, you start to move towards the other situation, and get a negative output.

That sort of makes sense but how is that e1 ends up being more or less in phase with ep at a moment in time? Why might these changes not show on e2?

In other words, why when above resonance is it that e1 is more in phase with ep than e2 is? Why when above resonance can't e2 be more in phase with ep than e1 instead?
AidanCroft is offline  
Old 10th Sep 2021, 12:32 am   #14
kalee20
Dekatron
 
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 7,061
Default Re: Ratio and Foster-Seeley detectors

Quote:
Originally Posted by AidanCroft View Post
That sort of makes sense but how is that e1 ends up being more or less in phase with ep at a moment in time? Why might these changes not show on e2?

In other words, why when above resonance is it that e1 is more in phase with ep than e2 is? Why when above resonance can't e2 be more in phase with ep than e1 instead?
Well, e1 and e2 are always 180 degrees apart, because they're the voltages on a centre-tapped winding.

So, if ep comes more in phase with e1, it'll necessarily become less in-phase with e2.

It will depend on the polarity of the windings whether ep comes more in phase with e1 or with e2 above resonance. But once that connection is made, it's there for life!
kalee20 is offline  
Old 10th Sep 2021, 12:37 am   #15
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,801
Default Re: Ratio and Foster-Seeley detectors

Also, don't assume that all windings have coupling factors of 1.0000 to each other. the layout of a ratio detector transformer is, um, interesting. The centre-tapped secondary has two halves which are close-coupled but the other two windings are less associated.

It may be best to think of the device as four inductors flying in formation.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is offline  
Old 10th Sep 2021, 1:30 am   #16
kalee20
Dekatron
 
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 7,061
Default Re: Ratio and Foster-Seeley detectors

Absolutely. If the windings are close-coupled, the voltages will be in-phase (or 180degress out). It's because the coupling between the primary and the secondary is very loose, that the variable phase shift occurs... the tertiary is close-coupled to the primary though, so that the primary-secondary phase shift is the same as the tertiary-secondary phase shift.

The two halves of the secondary are close-coupled together, so that the ends of the secondary are always 180 degrees out of phase.

Last edited by kalee20; 10th Sep 2021 at 1:32 am. Reason: Clarification
kalee20 is offline  
Old 10th Sep 2021, 9:00 am   #17
lesmw0sec
Octode
 
Join Date: Jul 2009
Location: Carmel, Llannerchymedd, Anglesey, UK.
Posts: 1,498
Default Re: Ratio and Foster-Seeley detectors

Quote:
Originally Posted by Restoration73 View Post
Another method of FM demodulation uses a PLL - Phase Locked Loop.

Certainly in consumer designs, the ratio detector and the quad detector are most
prevalent. The quad circuit generates a higher level of white noise in the absence of a
carrier, but as this type uses a custom IC these usually contain a muting or "squelch"
circuit to prevent it being heard.
I used this (PLL) when we were installing some comms. in a military establishment. The control room was fully screened and we were not allowed to use wiring for the phones. The solution was fibre-optic cables to carry the audio. I seem to recall using CD4046s. The ring voltage was signaled via F/O pulses and regenerated in the control room.
lesmw0sec is offline  
Old 10th Sep 2021, 10:19 am   #18
ms660
Dekatron
 
ms660's Avatar
 
Join Date: Apr 2011
Location: Cornwall, UK.
Posts: 13,454
Default Re: Ratio and Foster-Seeley detectors

Here's the neets article again:

http://www.tpub.com/neets/book12/51d.htm

Here's what Mr Seeley had to say about it (page 201):

https://worldradiohistory.com/ARCHIV...w-1947-Jun.pdf

And for the calculator crunchers what Mr Johnstone of the BBC said (page 124):

https://worldradiohistory.com/UK/Wir...7-03-S-OCR.pdf

Continued on page 235:

https://worldradiohistory.com/UK/Wir...ld-1957-05.pdf

Lawrence.

Last edited by ms660; 10th Sep 2021 at 10:32 am. Reason: link added
ms660 is offline  
Old 10th Sep 2021, 11:24 am   #19
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,801
Default Re: Ratio and Foster-Seeley detectors

Be wary of the oft repeated claims that the output of the ratio detector is independent of the signal amplitude. There are two flies in the ointment.

1) If you don't put any signal voltage in, even if that lack of voltage is modulated around the right frequency range, then there is not going to be any audio coming out. So the independence of IF signal level has to break down beyond limits. Conservation of energy insists.

2) If you put in a comfortable IF level and the ratio detector is working happily, If the IF level is not setting the scale of the output audio voltage... then what is? Are you getting a voltage reference out of nowhere?

Dimensional analysis of the parameters frequency, inductance, capacitance and resistance does not allow an absolute voltage to be established.

Alarm bells should be ringing.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is offline  
Old 10th Sep 2021, 11:39 am   #20
G6Tanuki
Dekatron
 
G6Tanuki's Avatar
 
Join Date: Apr 2012
Location: Wiltshire, UK.
Posts: 13,953
Default Re: Ratio and Foster-Seeley detectors

There's also the gated beam demodulator as invented by Zenith in the early 1950s using the 6BN6 valve, which is rather cunning!!

Just search for 6BN6 on r-type for an analysis and some sample circuits.
G6Tanuki is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 7:17 am.


All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at https://www.vintage-radio.net/rules.html.
Failure to heed this warning may result in death or serious injury to yourself and/or others.


Powered by vBulletin®
Copyright ©2000 - 2024, vBulletin Solutions, Inc.
Copyright ©2002 - 2023, Paul Stenning.