Forums :: Resources :: Features :: Photo Gallery :: Vintage Radio Shows :: Archives
Support This Site: Contributors :: Advertise


It is currently Dec Sun 15, 2019 11:38 am


All times are UTC [ DST ]





Post New Topic Post Reply  [ 16 posts ] 
Author Message
 Post subject: Philco FM1000 FM Demodulator
PostPosted: Sep Mon 21, 2015 5:57 am 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 12933
Location: San Jose, CA USA
This is really just an informational post, but comments are of course welcome.

A few years ago, I restored a Philco FM radio from the late 1940s for a customer. When he got it back, he raved about its performance and was particularly surprised that this was achieved, because he had heard that the Philco FM Demodulator using their proprietary FM1000 tube was a "disaster."

Yesterday I restored a Philco 48-482 that I picked up for $20 at a CHRS meet a few months back. It also uses the special demodulator with the FM1000 tube. A quickie attempt at FM alignment that works just fine for FM sets with conventional discriminators or ratio detectors failed miserably. I was aware that the FM1000 is a very special design, so I read up on it a bit.

Indeed, there is plenty on the web calling this type of FM demodulator a disaster. Even some forum discussions (right here on ARF) to the effect that ripping it out and replacing with a ratio detector is pretty much the best way to go.

What I have found now in two cases is just the opposite. These radios are downright hot performers, with very nice FM behavior -- IF you figure out how to align them properly.

The circuit is basically a phase-locked-loop FM demodulator, which -- by the way -- became the most widely used design in the last 35 years or so. You won't find discriminators or ratio detectors in modern FM radios.

Chuck's Philco Repair Bench website has a nice excerpt from Milton Kiver's "FM Simplified" that describes how this circuit works. A good read if you're interested in these kinds of details: http://www.philcorepairbench.com/tube/fm1000.htm .

Unfortunately, the Philco alignment instructions make the alignment job harder than it has to be. Basically all you want to do is the following:

1. Align all the FM IF transformers prior to the final transformer (which is part of the FM1000 tube circuit) for maximum signal on the designated IF, which is 9.1 MHz. Philco has you connect all kinds of loading networks and jumpers, and this gave me nothing but trouble. Just peak them all at 9.1 MHz, cycle through a few times, and you're golden, just as you would be in any other FM radio. The easiest way to peak them is to disable the FM1000 tube by grounding pin 2; then apply modulated 9.1 MHz with a wire loop around the mixer tube, and peak by listening for loudest audio. With the FM1000 disabled, it works as an AM detector, hence the use of modulated RF in this step.

2. Disable the FM1000 oscillator's phase feedback by shorting the red and blue leads of the last transformer (shorts the quadrature coil). Set the trimmer cap on the last FM transformer so that the free-running frequency of the FM1000 oscillator is 9.1 MHz, either by zero beating with a 9.1 MHz signal (same one as used to align the IF) or using a frequency counter.

3. Remove the short above, and then temporarily connect a jumper from pin 6 of the FM1000 to chassis to disable any input from the IF chain. Adjust the coil slug on the last transformer so that the free-running frequency of the FM1000 is 9.1 MHz, again either with frequency counter or by zero beating with a 9.1 MHz signal.

Remove the jumper above and prepare to be surprised at how well this FM demodulator works!

Note that the first time I tried this, it didn't work. I then replaced 5 mica caps (three inside the final transformer, one feeding the grid of the FM1000, and the 1500 pF one on the audio output line). Second attempt worked well. The caps all tested good, so not quite clear why they had to be replaced, but mica cap tests are not always conclusive.

So why did Philco drop this design after shipping just a few models with it? And why did other models in the concurrent model year use more conventional FM demodulators? Some thoughts:

- When it gets out of alignment, it really goes crazy! Massive distortion, etc. And not really as easy to align as other FM demodulators, so the poor serviceman might have had a hard time.

- The experience while tuning the dial is quite different than other FM sets of the time. In those days, a lot of FM radios didn't have full limiter stages, so strong "hash" between stations was probably a weird experience (a decade or two later, it would be normal). On typical FM sets of the day, the dial was almost silent between stations. There can also be sharp distortion right at the edge of the tuning of each station, which sounds a bit nasty. But it is not difficult to tune at all. There is a nice capture range (which probably was quite helpful in these pre-AFC days!).

- What you hear on weak stations is different than other FM sets of the day. Weak stations are just as "loud" as strong stations, but with noise mixed in, and to a limited extent, some distortion. This was probably also a disconcerting experience at that time, although does not seem strange today if you've tuned a lot of analog FM receivers without muting.

- Cost: With a more conventional FM demodulator, the final IF transformer feeds two diodes directly to demodulate FM. The FM1000 circuit requires a more complex tube and another transformer. This was probably the real killer? It seems Philco used it only on its most expensive sets.

In any case, when aligned properly, these radios are beautiful performers. If you like to have things in your collection that are unique from an electrical design point of view, Philco FM sets with the FM1000 would certainly qualify.

[11/24/15 edit: based on a response from a reader, today I corrected item #3 above to refer to pin 6 instead of pin 2. Looks like that was a typo in my original posting.]

_________________
Tom K6VL


Last edited by Tom Albrecht on Nov Wed 25, 2015 1:40 am, edited 2 times in total.

Top
 Profile  
 
 Post subject: Re: Philco FM1000 FM Demodulator
PostPosted: Sep Mon 21, 2015 1:43 pm 
Moderator

Joined: Jan Thu 01, 1970 1:00 am
Posts: 20727
Location: Utica, NY 13502 (USA)
I find that many alignment instructions are far more complicated than they need to be and the procedure for the 48-482 is a good example. It's been a while, but I think I used a similar method as you did for the FM1000 detector. I agree that the 48-482 is a great performer.

Dave


Top
 Profile  
 
 Post subject: Re: Philco FM1000 FM Demodulator
PostPosted: Sep Mon 21, 2015 8:16 pm 
Moderator
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 10983
Location: Latham NY 12110
Thanks Tom, I will try this with the next Philco I repair.


Top
 Profile  
 
 Post subject: Re: Philco FM1000 FM Demodulator
PostPosted: Aug Fri 30, 2019 3:33 am 
Member

Joined: Jan Tue 16, 2007 7:02 am
Posts: 3267
Location: Lexington, KY USA
Since this thread was recently referenced, it might be useful to know that Kiver's "FM Simplified", is now available on the americanradiohistory site. The 2015 link to the Philco Repair Bench seems to be broken.

Chuck's Philco Repair Bench website has a nice excerpt from Milton Kiver's "FM Simplified" that describes how this circuit works. A good read if you're interested in these kinds of details: http://www.philcorepairbench.com/tube/fm1000.htm .

The whole book is here:

https://www.americanradiohistory.com/Archive-Bookshelf/Technology/FM-Simplified-Kiver-2nd-1951.pdf

The Philco FM1000 detector info begins on pg. 230 (pdf pg. 238) of this 1951 edition.

There is a second section on Philco FM beginning on pg. 307 (pdf pg. 317) pertaining specifically to the 46-1213, or to a very similar model.

This book covers FM pretty well, if you need some light reading. Nearly 500 pages.

This and related radio & TV material can be found here:

https://www.americanradiohistory.com/Archive-Bookshelf/Bookshelf_Technical.htm

Ted


Top
 Profile  
 
 Post subject: Re: Philco FM1000 FM Demodulator
PostPosted: Aug Fri 30, 2019 4:41 pm 
Member

Joined: May Thu 14, 2015 4:15 pm
Posts: 1849
Location: Dallas, TX
Thanks Ted.
Those are some very good links!

_________________
Tim
It's not the Destination, It's the Journey.


Top
 Profile  
 
 Post subject: Re: Philco FM1000 FM Demodulator
PostPosted: Aug Sat 31, 2019 1:30 am 
Member
User avatar

Joined: Sep Wed 10, 2014 7:13 pm
Posts: 565
Location: Urbana, Illinois
Fascinating thread here. I’m learning plenty of things about FM radio that I hadn’t paid much attention to in the past.

-EB


Top
 Profile  
 
 Post subject: Re: Philco FM1000 FM Demodulator
PostPosted: Nov Wed 13, 2019 2:16 am 
Member

Joined: Mar Thu 03, 2011 8:53 pm
Posts: 1102
Location: Davis, CA
I am trying to align the FM on a Philco 48-482. I am new at this so please bear with me. I used Tom's instructions to align the FM IF transformers. I am on his second step and have some questions.

1. Is the grounding of pin 2 of the FM 1000 removed?

2. Is wire loop from the signal generator around the mixer tube kept in place?

3. Is the signal generator set for modulation or no modulation?

4. What does zero beating mean?

5. I do have a frequency counter so how is it used to to set the trimmer cap?

Thanks, Bob


Top
 Profile  
 
 Post subject: Re: Philco FM1000 FM Demodulator
PostPosted: Nov Thu 14, 2019 9:12 am 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 12933
Location: San Jose, CA USA
For step 2, pin 2 must not be grounded. When pin 2 is grounded, the oscillator section of the FM1000 is disabled, and that's the oscillator that you're trying to measure the frequency of in step 2.

If you wish to use the zero-beat method to set the oscillator frequency to 9.1 MHz, then keeping the loop around the mixer and feeding it with precisely 9.1 MHz is useful. If you use a frequency counter to check the frequency of the oscillator, then there would be no reason to keep the loop around the mixer tube for this step.

In step 2, you would use no modulation on the signal generator if you are zero-beating. In step 1, you want the modulation, since you are listening for the audio as a measure of signal strength through the IF stages.

Zero beating means to have two RF signals which are very close together in frequency together on the same signal line, and either observing the signals on a scope, or listening to the signals through a detector. When the two signals are far apart in frequency, you won't hear anything, and the scope display will look constant with a complex periodic waveform. As the two signals are brought close together, you'll hear the difference frequency as audio, descending in pitch as the two get closer together. When they get really close together (less than 20 Hz apart), instead of a tone, you'll hear a beating sound, with the beats getting farther apart in time as you approach zero frequency difference. If observing on a scope, you'll see envelope modulation, which will basically stand still when you get to zero frequency difference. To see this on a scope, place the probe close to the plate (pin 4) and observe the signal. May also be OK to connect directly to the plate with a 10:1 probe.

To use a frequency counter, you'll want to couple the signal from the oscillator in a manner that does not load the oscillator or shift its frequency even slightly. Bascially, this means no direct connection. Place your frequency counter probe close to, but not touching, either pin 2 (oscillator grid) or pin 4 (plate) to see if you can get enough signal for the frequency counter to read. I use a scope that has a frequency counter built in, and it works well in this way without touching the probe directly on the circuit. I don't know whether a typical frequency counter input is as sensitive as my scope probe, so not sure how well that works. If you must make a direct connection to get enough signal, couple to the plate through a small capacitor (like 10 pF). Coupling to the plate has less chance of upsetting the oscillator than coupling to the grid, which will definitely shift the frequency.

When you get to step 3, you can disconnect the loop from the mixer, since in step 3 you want no input signal coming from the IF chain. Grounding pin 6 should take care of this, but you might as well not be injecting any signal in the first place. You want to find out what ithe oscillator's free-running frequency is (with its phase feedback enabled) in the absence of any IF signal. Frequency counter would be the best way to go. If you want to use zero-beating, you would need to combine the two signals outside of the radio for that purpose, which is probably more trouble than it's worth.

_________________
Tom K6VL


Top
 Profile  
 
 Post subject: Re: Philco FM1000 FM Demodulator
PostPosted: Nov Thu 14, 2019 1:16 pm 
Member

Joined: Jun Fri 19, 2009 6:34 pm
Posts: 8944
Location: Long Island NY
Great that you are getting good results with the FM1000, but before comparing it favorably to other circuits, you should do some measurements. Frequency response and bandwidth, distortion, hum and noise, limiting, capture ratio, etc. Those radios may sound okay to the ear but a properly designed limiter/discriminator or ratio detector will offer better performance. Yes it is true that many modern FM radios use PLL detectors, but since the mid 1970s most have been based on digital PLL techniques, not analog.

It might also be mentioned that Motorola had its own version of a PLL detector which was developed because Paul Galvin refused to pay Armstrong any royalties on FM (the courts finally found in Armstrong's favor many years later). However, so far as is known, the Motorola PLL detector only went into two way equipment; their broadcast FM radios and TVs used ratio detectors or gated beam detectors.

A lot of the apprehension towards the FM1000 circuit springs from a 1941 paper by Edwin Howard Armstrong which raked Philco over the coals for selling "high fidelity" FM table radios which were anything but. He demonstrated through an exhaustive series of tests that the circuit had no limiting and hence no noise immunity, limited frequency response, and high distortion--the last two exacerbated by the small speakers and class B output stages in Philco's FM models. Performance of the radios Armstrong tested was no better on FM than it was on AM, and he held that this was a big disservice to the public as it would convey the idea that FM was just another gimmick which offered no real improvement.

_________________
"Hell, there are no rules here--we're trying to accomplish something!"

Thomas A. Edison


Top
 Profile  
 
 Post subject: Re: Philco FM1000 FM Demodulator
PostPosted: Nov Thu 14, 2019 9:37 pm 
Member
User avatar

Joined: Sep Wed 10, 2014 7:13 pm
Posts: 565
Location: Urbana, Illinois
The main part of the "Armstrong" FM radio receiver architecture is its "FM discriminator" stage (aka the "FM detector stage."). It contains a specially designed IF transformer and 2 diodes. This FM discriminator stage, along with a "high-gain" IF section, are the main focus of Armstrong's FM patents. The front-end, oscillator/mixer, and IF stages are similar to standard superhet radios, with the exception of using a much higher IF frequency (usually near 10 MHz). Also Armstrong's IF section has very high gain (usually with 3 IF amplifier tubes), so that the IF signal reaching the FM discriminator is "limited" (aka "saturated" or "clipped"). This removes all "AM" (amplitude modulation). It reduces the background noise level and also tends to eliminate "fading in" and "fading out."

I am aware of the following alternate FM radio circuits (such as Philco's FM1000 demodulator tube), which do not contain Armstrong's "FM discriminator" section.
    These alternate circuit designs were developed to avoid infringing on Armstrong's FM patents. As noted above, the main parts of the Armstrong FM patents are the "FM discriminator" circuit and the high-gain IF section. Evidently radios without Armstrong's "FM discriminator" were not required to pay Armstrong's patent royalties.

1) The Philco FM1000 design starts out with an architecture very similar to the Armstrong FM design (and to all standard superhet radios).
    There is a tuned FM front end, an oscillator/mixer stage, and several IF amplification stages. The combined gain of the 3 IF stages in the Philco architecture is large enough to "limit" or "clip" most of the unwanted amplitude modulation, thereby removing it from the FM signal.
    There is one interesting difference in the Philco IF section: Most FM radios use 10.7MHz for their IF frequency. But Philco uses a slightly lower IF frequency of 9.1MHz. Perhaps a different IF frequency was chosen in order to get around the Armstrong FM patents? That said, there isn't any major difference between 10.7MHz and 9.1MHz in terms of performance. Because of its high IF gain, the Philco FM1000 circuit design has sensitivity and noise reduction comparable to the Armstrong FM architecture.
    The main difference is how Philco demodulates the FM IF signal into audio. This occurs inside the FM1000 tube. One part of the FM1000 oscillates at a steady and continuous 9.1MHz frequency which tracks and "locks onto" the exact average "center frequency" of the incoming FM IF signal. The FM1000 oscillator is kept in sync with the "center frequency" of the incoming FM IF signal by a tuned coil which receives its input from the third FM IF amplifier. Because the "Q" of the FM1000 oscillator circuit is extremely high, it doesn't follow the audio signal. Rather, the FM1000 oscillator frequency remains close to the long-term average of the input frequency. The FM1000 oscillator section is able to "track" small/slow (sub-audio) frequency variations which might be caused by drift or slight rotation of the tuning knob.
    Then, in the second portion of the FM1000 tube, the incoming frequency-modulated FM IF signal (containing the audio) is mixed with the constant precise FM carrier frequency coming from the FM1000 oscillator section. This input signal comes directly from the third IF stage and goes into a special grid in the FM1000 tube. There are no tuned coils here. When this input frequency > FM1000 oscillator frequency, the FM1000 plate voltage goes positive. When the input frequency < FM 1000 oscillator frequency, the FM1000 plate voltage goes negative. When input frequency = FM1000 oscillator frequency, then the FM1000 plate voltage does not vary. The 9.1Mhz IF signal component present at the FM1000 plate is then filtered out by a low-pass RC filter, leaving the audio signal. The performance of FM1000 demodulator resembles a modern "phase-locked loop" in some ways: Both share the characteristic that there are very abrupt points above and below the center frequency where the audio totally drops out and only noise is heard. Either there is audio, or nothing but noise.

2) There's another type of early FM radio circuit design that uses only one tube.
    It is directly connected to the FM antenna, and a single tube both amplifies and detects the FM signal. I'm not familiar with this design. It was used in some low-priced radios. It is said that this "1 tube FM" section works only for very strong local stations, and even with a strong signal it still doesn't sound very good. There are some posts on ARF that cover these "1 tube FM sections."

3) During the mid 1950's another type of 2-diode FM demodulator was invented. It is known as the "ratio detector."
    Evidently it didn't conflict with the Armstrong patents. It gradually became the standard type of FM demodulator by 1960. The "ratio detector" circuit design is only a little bit different vs. the original "Armstrong FM discriminator." Both use a special IF transformer and 2 diodes. But these IF transformers are NOT interchangeable. Performance and tuning characteristics of the "ratio detector" are relatively similar to the original Armstrong FM discriminator. The remainder of the radio (front-end, oscillator/mixer, IF stages) are similar regardless of whether the FM demodulator stage is "ratio detector" or "Armstong FM discriminator." Some radios that use the "ratio detector" are able to get by with only 2 FM IF stages. Apparently the "ratio detector" doesn't require a fully "limited" input signal. Nearly all radios that use the "ratio detector" also have AFC (automatic frequency control). Typically there is a 3 position band switch: AM, FM, FM with AFC. As with the Armstrong FM discriminator, radios with the ratio-detector are easier to tune with the AFC switched off. Then the AFC is switched on (after tuning) to hold the signal.


Both the "Armstrong FM discriminator" and the "ratio detector" have a very different tuning characteristic vs. the Philco FM1000:
    The Armstrong FM discriminator has a relatively wide tuning range where audio can be heard, but only one exact tuning point where the distortion becomes very low and all the background noise goes away. Unfortunately the front end local oscillators in most FM radios have enough frequency drift with time/temperature to cause the audio quality to gradually degrade unless the radio user periodically retunes the radio. Therefore, the radio manufacturers developed "AFC" (automatic frequency control) circuits for their Armstrong-style FM radios. The AFC section keeps the radio precisely tuned so that the audio signal won't gradually "lose its quality" and get distorted due to drift or slight mistuning. There is always a user-controlled selector switch for "AFC on" or "AFC off." That's because its easier to tune the radio to a station with the AFC disabled. Then the AFC is switched on to hold the station.

Up to a certain point the Philco FM1000 design automatically provides this "AFC" feature, without needing a switch to turn it on or off. The FM1000 circuit tends to "capture" the desired radio station. Some people think it is easier to tune the FM1000 compared to radios with Armstrong FM discriminators or "ratio detectors."

-EB


Top
 Profile  
 
 Post subject: Re: Philco FM1000 FM Demodulator
PostPosted: Nov Fri 15, 2019 12:02 am 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 12933
Location: San Jose, CA USA
Comments on Chris's input above -- Yes, it would be interesting to see how well the FM1000 performs compared to more conventional discriminators and ratio detectors. When analog PLL FM demodulators were introduced in the 1970s (such as in the Phase Linear 5000 tuner, which we had back then), they were thought to have lower distortion, which makes some sense. The "S" curve of a conventional FM demodulator can have a "pretty straight" line center section, but it is not truly flat, and this necessarily introduces some distortion. In a PLL, if the voltage-controlled oscillator has a linear voltage-to-frequency conversion, the distortion in a PLL demodulator could potentially be better. That being said, the FM1000 circuit used by Philco might actually have been a poor performer from a distortion point of view, and it would be interesting to compare. By ear in a simple table radio, nothing sounds amiss, but real distortion measurements would reveal much more.

Frequency response might well be poor in a primitive PLL demodulator. The carrier frequency-following servo loop of the PLL will have some frequency response limitations, and if these extend down into the audio range, high frequency audio response could be weak. By the time analog PLLs were introduced in the 1970s FM receivers, this problem was solved (and it's fairly easy to imagine a PLL running at 10.7 MHz carrier tracking audio frequency modulation really quite well).

I don't know when digital PLLs would have been introduced in consumer equipment, but it certainly would not have been in the 1970s.

AM suppression (which Chris mentions as a weak point for the FM1000) isn't as big a deal in the VHF for analog FM broadcasting; static on the FM band (even the old 50 MHz) band is pretty minor compared to lower frequencies. So even if AM suppression is not as good as other demodulators, customers would have perceived that FM was "static free" in comparison to typical reception on the AM band.

In any case, my positive comments about the FM1000 circuit were intended more to counter comments by others to the effect that it's not even worth trying to get one to work. They work just fine and are suitable for everyday FM listening.

_________________
Tom K6VL


Top
 Profile  
 
 Post subject: Re: Philco FM1000 FM Demodulator
PostPosted: Nov Fri 15, 2019 12:35 am 
Member

Joined: Jun Fri 19, 2009 6:34 pm
Posts: 8944
Location: Long Island NY
Actually, Armstrong did not have a patent on the discriminator circuit. It was developed by Dudley Foster and Stuart Seeley in 1936. Armstrong recommended it as it gave the highest performance of any FM demodulator then available. It is to be noted that in order to obtain noise immunity from a discriminator, the input must be hard limited by some means, so a limiter stage is usually placed ahead of a discriminator.

The ratio detector was developed by Murray Crosby of RCA Communications in 1940 and represents a simplification of the limiter-discriminator arrangement. Effectively it uses the diodes for both functions. Limiting is adequate but not as good as the limiter-discriminator. However, the circuit does produce a higher audio output with fewer parts, so it was quite popular with radio manufacturers. Philco did in fact swallow its pride and used the ratio detector circuit in its later sets.

It should also be understood that Armstrong's basic FM patent was on a process, not any particular circuit. The process was a mathematical one; by increasing the frequency deviation of FM to several times the frequency of the input audio, a point is reached where the system becomes quite insensitive to noise and interference. Since this was never anticipated nor even predicted by the prior art, Armstrong's patents on wide band FM as a process were upheld. All the companies like RCA, Philco, Motorola, Emerson, and others who thought they were safe because they sold wide band FM equipment which did not use Armstong's circuits ended up paying back royalties and in some cases damages to Armstrong's estate.

_________________
"Hell, there are no rules here--we're trying to accomplish something!"

Thomas A. Edison


Top
 Profile  
 
 Post subject: Re: Philco FM1000 FM Demodulator
PostPosted: Nov Fri 15, 2019 1:17 am 
Member
User avatar

Joined: Sep Wed 10, 2014 7:13 pm
Posts: 565
Location: Urbana, Illinois
Chris108 wrote:
Actually, Armstrong did not have a patent on the discriminator circuit. It was developed by Dudley Foster and Stuart Seeley in 1936. Armstrong recommended it as it gave the highest performance of any FM demodulator then available. It is to be noted that in order to obtain noise immunity from a discriminator, the input must be hard limited by some means, so a limiter stage is usually placed ahead of a discriminator.

The ratio detector was developed by Murray Crosby of RCA Communications in 1940 and represents a simplification of the limiter-discriminator arrangement. Effectively it uses the diodes for both functions. Limiting is adequate but not as good as the limiter-discriminator. However, the circuit does produce a higher audio output with fewer parts, so it was quite popular with radio manufacturers. Philco did in fact swallow its pride and used the ratio detector circuit in its later sets.

It should also be understood that Armstrong's basic FM patent was on a process, not any particular circuit. The process was a mathematical one; by increasing the frequency deviation of FM to several times the frequency of the input audio, a point is reached where the system becomes quite insensitive to noise and interference. Since this was never anticipated nor even predicted by the prior art, Armstrong's patents on wide band FM as a process were upheld. All the companies like RCA, Philco, Motorola, Emerson, and others who thought they were safe because they sold wide band FM equipment which did not use Armstong's circuits ended up paying back royalties and in some cases damages to Armstrong's estate.
This is fascinating. Thanks for sharing it! The history of electronics and radio manufacturing is one of my interests.

-EB


Top
 Profile  
 
 Post subject: Re: Philco FM1000 FM Demodulator
PostPosted: Nov Sun 17, 2019 3:24 am 
Member

Joined: Dec Tue 29, 2015 8:23 am
Posts: 253
Location: Australia
Tom Albrecht wrote:
I don't know when digital PLLs would have been introduced in consumer equipment, but it certainly would not have been in the 1970s.

Define "digital" :D.

RCA's CD4046 was introduced in the early 70's, and its phase/frequency detector is digital (built entirely from digital logic, with a digital logic output) - heck, it's almost all digital apart from parts of the limiter and the analogue VCO (though the latter is mostly digital logic). The main things it's lacking compared to a modern "digital" PLL are an internal divider, and a numerically-controlled oscillator as the VCO.

At ~1Mhz max it was no good for FM, but by the mid 70's it was appearing in things like ham FSK decoders, and consumer equipment like ultrasonic remote control decoders, etc.

There was also at least one Motorola PLL before that, but I'm not sure if it used digital techniques or was purely analogue.


Top
 Profile  
 
 Post subject: Re: Philco FM1000 FM Demodulator
PostPosted: Nov Sun 17, 2019 4:26 am 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 12933
Location: San Jose, CA USA
I thought about that as well, and suspected that may also be where Chris got the idea that "digital" PLLs were available in the 1970s. I think of those chips as analog, since the feedback loop and VCO are analog. Phase detectors are sorta inherently digital with their threshold switching function.

The NE565 was the first widely used PLL chip that I was aware of, and I think of it as an analog chip. Google says it was introduced in 1969.

In any case, I don't mind a bit of a grey area between what is called digital vs analog with regard to PLLs. I imagine today there are PLLs that have very sophisticated digital processors for their feedback and VCO functions.

_________________
Tom K6VL


Top
 Profile  
 
 Post subject: Re: Philco FM1000 FM Demodulator
PostPosted: Nov Thu 21, 2019 4:56 am 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 1173
Location: Detroit, Michigan
Great info. I have a 48-482 that's been on the to restore list for about 20 years. That FM1000 tube had me mystified, and a bit put off from touching it.


Top
 Profile  
 
Post New Topic Post Reply  [ 16 posts ]  Moderators: Marcc, Norm Leal

All times are UTC [ DST ]


Who is online

Users browsing this forum: Google Adsense [Bot] and 14 guests



Search for:
Jump to:  




























Privacy Policy :: Powered by phpBB