UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio 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 Nov 2017, 2:13 pm   #21
G8HQP Dave
Dekatron
 
Join Date: Sep 2008
Location: Solihull, West Midlands, UK.
Posts: 4,331
Default Re: Infinite Impedance detectors.

Quote:
Originally Posted by kalee20
The 'special circumstances,' though, are not mentioned by Sturley! He's an author of pedigree, but it's still appropriate, as good engineers, to challenge and try to understand WHY.
One of the nice things about writing a book (unlike lecturing, for example) is that you can use vagueness to sweep messy details under the carpet. Sometimes this avoids adding many pages (which might upset the publisher), sometimes the author is just not interested in that topic, and I assume sometimes the author realises that he does not know enough about that topic to write anything useful but still needs to mention it in passing.

Quote:
With an infinite impedance detector, AC/DC load is not an issue as long as there is a sufficient cathode current drain to allow the cathode capacitor voltage to decay between HF cycles so that it can always follow downward modulation.
The capacitance is still being discharged by a resistor, so it still suffers from the negative peak problem. The valve cannot do anything further beyond being cutoff. AC/DC load is less of an issue because the II detector cathode circuit can use lower impedances than a diode detector, but it is not a non-issue.
G8HQP Dave is offline  
Old 9th Nov 2017, 2:25 pm   #22
Skywave
Dekatron
 
Skywave's Avatar
 
Join Date: Jun 2006
Location: Chard, Somerset, UK.
Posts: 5,844
Question Re: Infinite Impedance detectors.

Quote:
Originally Posted by kalee20 View Post
So who cares if it's a signal-dependent infinity?
Conceptually, I'm having a struggle to fully understand that statement. As it's written, I understand it to mean that since the signal will be a variable quantity, infinity will be variable also as a consequence. Are we now into the mathematics of Georg Cantor?

Al.
__________________
"Your idea is crazy! But the question is this: is it crazy enough . . . . to be true?" (Niels Bohr)
Skywave is offline  
Old 9th Nov 2017, 2:34 pm   #23
Skywave
Dekatron
 
Skywave's Avatar
 
Join Date: Jun 2006
Location: Chard, Somerset, UK.
Posts: 5,844
Arrow Re: Infinite Impedance detectors.

Quote:
Originally Posted by G8HQP Dave View Post
One of the nice things about writing a book (unlike lecturing, for example) is that you can use vagueness to sweep messy details under the carpet. Sometimes this avoids adding many pages (which might upset the publisher); sometimes the author is just not interested in that topic; and I assume sometimes the author realizes that he does not know enough about that topic to write anything useful but still needs to mention it in passing.
I have witnessed that in some of the books written by F.E. Terman. In places, there are certain topics of high relevance to Radio Engineering that he has not mentioned at all. But he is not alone in making such omissions, but that is no justification for any author to so do.

Al.
__________________
"Your idea is crazy! But the question is this: is it crazy enough . . . . to be true?" (Niels Bohr)
Skywave is offline  
Old 9th Nov 2017, 6:00 pm   #24
kalee20
Dekatron
 
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 4,511
Default Re: Infinite Impedance detectors.

Quote:
Originally Posted by Skywave View Post
As it's written, I understand it to mean that since the signal will be a variable quantity, infinity will be variable also as a consequence. Are we now into the mathematics of Georg Cantor?
No.

An infinite impedance detector isn't really infinite of course, it's only that it can be considered infinite (can be neglected) from a practical point of view.

A diode detector may load a tuned circuit with 250kΩ, pretty constant whatever the signal, whereas the infinite impedance detector, with leakage through the glass etc may be equivalent to a load of 1000MΩ with strong signals and 500MΩ with weaker. That's a big change. But with the tuned circuit itself having a typical parallel loss resistance of 2MΩ, nobody is going to worry as the change will be swamped.
kalee20 is offline  
Old 9th Nov 2017, 6:13 pm   #25
kalee20
Dekatron
 
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 4,511
Default Re: Infinite Impedance detectors.

Quote:
Originally Posted by kalee20
With an infinite impedance detector, AC/DC load is not an issue as long as there is a sufficient cathode current drain to allow the cathode capacitor voltage to decay between HF cycles so that it can always follow downward modulation.
The capacitance is still being discharged by a resistor, so it still suffers from the negative peak problem. The valve cannot do anything further beyond being cutoff. AC/DC load is less of an issue because the II detector cathode circuit can use lower impedances than a diode detector, but it is not a non-issue.[/QUOTE]

With the diode detector, at the modulation trough of 100% modulation is negative audio peak, you have vanishingly small voltage for the resistor to discharge the capacitor, so the capacitor voltage has great trouble following the modulation envelope.

But with the II detector, a typical valve may have its cathode sitting at 5V or so above chassis. So the cathode resistor always has some voltage across it to discharge the capacitor. A signal will take it more positive, and at modulation peak it can be quite a lot more positive, but it will never go less than this quiescent value. Get the values right, and you can be assured of following the modulation envelope. You can even use a constant-current cathode load if you want.

So, yes it's not a non-issue, but while the diode inherently can't be made distortion-free from this point of view, the II can. (It still leaves other forms of distortion, of course!)
kalee20 is offline  
Old 9th Nov 2017, 10:22 pm   #26
Synchrodyne
Nonode
 
Join Date: Jan 2009
Location: Mt. Maunganui, New Zealand
Posts: 2,063
Default Re: Infinite Impedance detectors.

Quote:
Originally Posted by kalee20 View Post
The 'special circumstances,' though, are not mentioned by Sturley! He's an author of pedigree, but it's still appropriate, as good engineers, to challenge and try to understand WHY.
Agreed, but to be fair to Sturley, he did provide a reference to an article by W.T. Cocking, another author of pedigree. I’d expect that article to answer the “why” question, but unfortunately the said article (in “Wireless Engineer” 1935 November) is likely to be unobtainium. Both Sturley and Langford-Smith supplied many references, presumably considered to be readily available in typical equipment maker and institutional libraries at the time when they were writing, but a bit frustrating for the casual observer these days. On the other hand, their books would have been gigantic multi-volume sets had they expanded on all topics. Still, on the odd occasion when one can find the referenced articles they are usually quite helpful.


Cheers,
Synchrodyne is offline  
Old 9th Nov 2017, 10:25 pm   #27
G8HQP Dave
Dekatron
 
Join Date: Sep 2008
Location: Solihull, West Midlands, UK.
Posts: 4,331
Default Re: Infinite Impedance detectors.

Quote:
Originally Posted by kalee20
But with the II detector, a typical valve may have its cathode sitting at 5V or so above chassis.
Good point, which I had temporarily forgotten.
G8HQP Dave is offline  
Old 20th Nov 2017, 10:26 pm   #28
Synchrodyne
Nonode
 
Join Date: Jan 2009
Location: Mt. Maunganui, New Zealand
Posts: 2,063
Default Re: Infinite Impedance detectors.

Another approach to a low-distortion AM demodulator was the combination of a diode direct-connected to a cathode follower.

It was mentioned briefly by Langford-Smith (page 1495 in the “Classic” edition), with more detailed coverage in AWV Radiotronics for 1953 June, pertinent page attached:

Click image for larger version

Name:	AWV Radiotronics 195306 p.79.jpg
Views:	42
Size:	56.4 KB
ID:	152604

A more detailed analysis was provided in AWV Radiotronics 1956 January, page 03ff. This is available at: https://tubedata.altanatubes.com.br/...ics/index.html.

This suggested that the reflex demodulator (another name for the infinite impedance type) did not offer much improvement over the diode, whereas the diode-plus-cathode follower combination did.

I imagine that for the diode-plus-cathode follower combination, one could use a 6BJ8, 6BN8 or similar double diode-triode with separate cathodes for each section. The second diode could then be used as the AGC rectifier with free choice of delay voltage. Perhaps an EABC80 or similar would work, with the independent diode used for demodulation. Whether one of the other diodes could then be used for AGC is debatable, given that its cathode would not then be grounded at AF signal frequencies. On the other hand, extra smoothing in the AGC circuit, as used when the AGC bias is taken from the signal diode, might work. AGC considerations aside, an ECC82 with one triode strapped as a diode would also be suitable.


Cheers,
Synchrodyne is offline  
Old 21st Nov 2017, 11:27 am   #29
turretslug
Nonode
 
Join Date: Nov 2011
Location: Surrey, UK.
Posts: 2,418
Default Re: Infinite Impedance detectors.

That's an interesting and very straightforward approach- and putting RF filtering beyond the cathode follower, so to speak, would certainly help with the criteria re. AC and DC loading. The likes of the 6C4/EC90 with small envelope and 0.15A heater loading would look to be useful here- but to manufacturers conscious of minimising expense, the time-honoured detector arrangement on its own was obviously "good enough" for the money.

Also implicit in the use of a separate triode from the accustomed double-diode-triode common cathode set-up is the need for a detector/AVC rectifier- the 6AL5/EB91 is obvious here, and double diode with separated cathodes would also allow detector and AVC circuits to be optimised for their particular desired minimum distortion or delay characteristics. Again, however, we're running up against the practical need to minimise costs...
turretslug is offline  
Old 21st Nov 2017, 8:15 pm   #30
turretslug
Nonode
 
Join Date: Nov 2011
Location: Surrey, UK.
Posts: 2,418
Default Re: Infinite Impedance detectors.

Also intrigued that this arrangement is discussed in publications dated 1953 and 1956- surely, they would have been tried/researched before then, even in the mid-'30s as AM radio and valve refinement really got into stride? Mid '50s seems to be when FM radio was catching on and the AM side of radios was really getting sidelined by price-conscious manufacturers- certainly as regards any refinement that involved complication. Certainly, if the EABC80 and equivalents could offer a one-envelope good-enough solution to AM/FM demodulation and AF amp, it's difficult to imagine mainstream manufacturers doing this. Even if EB91/ECC83 would offer all-separate cathodes for optimum-performance of delay AVC, low-distortion biased detector, cathode follower and AF amp.

I think that, culturally as much as anything else, AM radio continued to be seen as a parallel option, rather than obsoleted companion, to FM in the US- perhaps there were high-quality tuners/receivers there that continued to try to get the best from the medium with configurations like the above.
turretslug is offline  
Old 29th Nov 2017, 10:21 pm   #31
Synchrodyne
Nonode
 
Join Date: Jan 2009
Location: Mt. Maunganui, New Zealand
Posts: 2,063
Default Re: Infinite Impedance detectors.

Yes, one might expect all of that to have been “old hat” by the 1950s. But perhaps not. Wireless World (WW) ran a two-part article by W.T. Cocking on “Modern Detector Circuits” in the 1939 August 10 and August 17 issues. The opening summary was:

“The characteristics and circuits of the most usual forms of detector are described in this article. Commencing with the diode, the author goes on to deal with the grid, anode-bend and negative feedback detectors, and points out their advantages and their limitations. In the following article, a special method of feeding the diode will be discussed and will be shown to offer important advantages.”

This article did not include mention of a diode direct-coupled to a cathode-follower. But counter-intuitively, the special circuit turned out to be an RF cathode-follower driving a diode demodulator. The argument was that a negative feedback (infinite impedance) demodulator, with its minimal loading on the final RF stage, could be useful for a communications receiver, where high-Q, relatively narrow-band IF transformers (IFTs) were required. But to take full advantage, the final IFT could not feed the separate AGC diode directly, but needed to do so via a suitable buffer stage. If that buffer were a cathode-follower, then conditions were created for the employment of a diode demodulator whose operating environment allowed it to perform in an optimum way.

Returning to the AWV Radiotronics articles, something to bear in mind is that whilst the northern hemisphere the emphasis at the time was moving to FM, that was not the case “down under”. FM did not arrive in Australia and New Zealand until the 1980s, so getting the best out of AM remained important until then. By the 1970s, not only was high quality AM equipment from the north quite scarce, but the AM sections of hi-fi tuners and receivers had generally deteriorated, particularly those from Japan Inc. (To be fair, there were exceptions, of which Sansui was probably the most notable.) Thus, local solutions were required. By way of one example, in Australia, Allen Wright produced a couple of wideband AM tuners that were intended to match or better the performance provided by the Quad AMII and AM3 tuners. Both used precision rectifiers for AM demodulation, and the second model also had a quasi-synchronous demodulator (MC1330). Of these, the designer later said: “Both sounded like GOOD FM, just in mono, from the superb quality non commercial government stations. Compressed commercial stations sounded just that - compressed!” (The compression problem is not new – it has just got a lot uglier since the 1970s and 1980s.)

In the USA, the interest in better quality AM reception has waxed and waned over the years, but did not disappear entirely. Certain events have resulted in peaks of activity. In the late 1950s, experimental FM-AM stereo broadcasting resulted in the production of quite a few tuners with good AM sections. Then with the end, or at least extent limitation of FM and AM simulcasting, there was another upswing in AM interest the late 1960s. In the 1980s, the arrival of AM stereo increased interest in better quality AM. In the 1990s, the introduction of the NRSC curve and the AMAX standards was another high point.

Only the first of those “peak interest” events was pertinent to the valve era. I suspect that for the most part, the makers used diode demodulators. One example was the Heathkit PT-1 FM-AM tuner. This used a push-pull crystal diode AM demodulator, which was AC-coupled to the AM cathode follower. However, the latter was of the bootstrapped type, so would have provided negligible additional AC loading on the demodulator. The overall AM circuit was quite sophisticated considering that it was MF only, with an RF amplifier, two IF stages, with full AGC on the 1st and one third AGC on the second, and a sidechain AGC amplifier feeding a diode-strapped triode AGC rectifier. But the mixer was a self-oscillating 6BE6 The late 1960s/early 1970s solid-state models appear to have used crystal diode demodulators. In the AM stereo era, synchronous demodulators took over. The Motorola C-QUAM decoding ICs used a mix of wideband quasi-synchronous and PLL fully synchronous demodulation, which disposed of many of the problems associated with diodes and other rectifying demodulators. Nonetheless, one can find examples of synchronous AM demodulation in the valve era, more in connection with professional equipment but occasionally in consumer equipment. (Synchronous (quadrature) FM demodulation was much more common in the consumer domain, particularly for TV intercarrier sound channels.)

Outside of the hi-fi arena, some of the American domestic receivers had better-than-average AM sections, albeit not of the wideband type. Several Zenith FM-AM valved table receivers from the late 1940s onwards had MF-only AM sections with RF amplifiers and two IF amplifiers. Mostly these used one diode from a 19T8 or similar for demodulation. That one RF, two IF configuration was sustained even with Zenith’s first FM Stereo-AM table receiver, the MJ1035W, where one might have expected the emphasis on FM stereo to have justified cutting corners on the AM side. The GE “Superradio” transistor portable was said to have a good AM section.

In the UK, the two hi-fi makers who stayed longest in the wideband AM tuner business were Chapman, until c.1969, and Quad, who exited in 1973. Both used diode demodulators, with the output taken from a fraction of the load (about a quarter in the Chapman case, and around one tenth in the Quad case) in order to minimize excess AC loading. The basic Chapman design dated from 1951. The Quad AMII was new in 1960, and quite a bit different from its 1953 Quad AM predecessor, although that had also used a diode demodulator with tapped-down output. The Quad AM3 of 1969 was essentially a repackaged, self-powered AMII with some detail changes to the AF output filters.

In the Chapman design, with its sidechain AGC IF, it would have been quite easy to use an infinite impedance demodulator. And a direct-coupled cathode-follower could have been added to either, avoiding the need for tapping down on the diode load. But at the time, their AF output levels, although low by today’s standards, were fine for feeding the amplifiers of the time.

Valve count insofar as it contributed to total cost was something of an issue, and no doubt mitigated against adding say a 6C4 as an infinite impedance demodulator or as a cathode follower, particularly if it offered little material advantage in situ. Back in the 1950s and into the 1960s, tuners attracted quite high purchase tax in the UK, whereas amplifiers did not. So, there was some inclination not to include buffer stages in tuners, but to use the amplifier input stages as such. And to take HT and LT from the amplifier, rather than a built-in power supply. The Chapman tuners were originally externally powered, with a self-powered option arriving in the mid-1950s, and late production being only self-powered.


Cheers,
Synchrodyne is offline  
Old 30th Nov 2017, 1:03 pm   #32
G8HQP Dave
Dekatron
 
Join Date: Sep 2008
Location: Solihull, West Midlands, UK.
Posts: 4,331
Default Re: Infinite Impedance detectors.

The arrival of FM radio was not a reason for AM to shrink into the background. On the contrary, for some time they were rival systems. VHF wideband AM was seriously considered for the UK but eventually FM was chosen.
G8HQP Dave is offline  
Old 30th Nov 2017, 10:30 pm   #33
Synchrodyne
Nonode
 
Join Date: Jan 2009
Location: Mt. Maunganui, New Zealand
Posts: 2,063
Default Re: Infinite Impedance detectors.

Be that as it may in overall terms, the arrival of FM was used as an excuse for some of the equipment makers, as it were, to “demote” AM (of the MF and LF variety, that is) and “cut corners” on the AM sections of combined FM-AM receivers. Delayed AGC was one of the early casualties, which one may attribute to the widespread adoption of triple diode-triode valves to do both the FM and AM demodulation jobs – these did not have the extra diode required for either a separate AGC rectifier with included delay or a separate AGC delay diode. And in the UK case, one may make a rather extreme comparison between say Murphy’s late AM-only receivers, the A186 and A188C, which had bandpass input tuning, variable selectivity and 9 kHz adjacent channel notch filters, with the less-well-specified AM sections of its subsequent FM-AM receivers. There is an irony there. With the early FM-AM receivers, the setmakers probably paid more attention – within cost constraints – to the distortion and bandwidth performance of the amplifier sections, and also to the speaker and cabinet interaction and performance, as compared with preceding “run-of-the-mill” AM-only units, in order to get the best out of FM. AM RF side performance was going backwards whilst the AF side was improving.
.
Another oddity was that in the USA, in the late 1950s, in the run-up to stereo broadcasting, the NSRC (National Stereo Radio Committee) considered stereo systems for both FM and AM, with quite a bit of work done on the latter by RCA amongst others. (In fact, TV sound stereo systems were also within its ambit.) But the corresponding FCC docket related only to FM stereo. This looked somewhat like an institutional bias in favour of FM.

In respect of the UK VHF-AM proposal, for the comparative AM vs FM tests, Ambassador Radio designed and built a number of FM-AM “Comparator” receivers to the BBC’s specification. This receiver did not look to be particularly constrained on valve count, and was required to have a low-distortion AM section. It is easy to see it as an ideal case for the use of an infinite-impedance demodulator. This could be tapped off the IF strip ahead of the FM limiter whilst imposing negligible load, and AGC could be taken from the FM 1st limiter grid, as was done for some FM-only receivers. But Ambassador chose not to do it that way. Rather it used the diode formed by the grid and cathode of the 1st limiter as an AM demodulator, claiming that the clamping of the limiter valve anode and screen potentials (by voltage dividers) improved diode linearity. The post-demodulator circuitry was arranged to provide an AC-to-DC load ratio of 0.998. Delayed AGC was provided by a separate diode fed from the anode of the 3rd IF stage. Here is the AM demodulator circuit:

Click image for larger version

Name:	Ambassador Comparator AM Demodulator.png
Views:	30
Size:	8.8 KB
ID:	153183

It is almost as if the infinite impedance demodulator was generally avoided if the makers could find other ways of getting the desired results with diodes of some kind.

Some of the early UK FM tuners, from c.1953, are known to have also had provision for VHF-AM reception, including those from Chapman, Lowther and Sound Sales. The prototype Quad FM of 1952 might have had this facility, but that is unconfirmed. What form of AM demodulators they had is unknown. Later, in 1958 Jason used the limiter grid-cathode diode as an AM demodulator in its JTV FM and (AM) TV sound tuner.

Staying with AM demodulators, this Murphy item from 1949 discusses (AM) demodulator distortion and address the AC/DC load ratio issue, but all of this is in terms of the diode type, there being no mention of other types such as infinite impedance.

Click image for larger version

Name:	WW 194906 p.31.jpg
Views:	16
Size:	83.7 KB
ID:	153184


Cheers,
Synchrodyne is offline  
Old 2nd Dec 2017, 12:32 am   #34
Synchrodyne
Nonode
 
Join Date: Jan 2009
Location: Mt. Maunganui, New Zealand
Posts: 2,063
Default Re: Infinite Impedance detectors.

The origins of the infinite impedance demodulator were recorded in a Wireless World article in the 1937 January 01 issue, page 08ff, entitled “New Detector Circuit – Applying Feedback to the Anode Bend Rectifier”, by W.N. Weeden.

Evidently the circuit was announced by Hygrade Sylvania after development in the RCA Licensee Laboratories. It was described thus:

“It is actually a triode anode-bend detector having a high resistance load in its cathode circuit. This load resistance is by-passed for carrier frequencies, and as it is common to both anode and grid circuits, degeneration of negative feed-back plays a large part in its freedom from harmonic distortion.”

The preamble discussed the diode demodulator and in particular the AC-to-DC load ratio issue, including the comment “The ideal condition is obtained when this ratio is unity, which can be realised only with direct coupling.” So quite possibly the idea of direct coupling a diode to a cathode follower was known in those days.

One of the concluding comments was: “..the new detector should prove superior to any but the direct-coupled diode, and even then, should win out because of the freedom from damping of its input circuit.” I think that overlooked the fact that freedom from damping was not necessarily a virtue. In high quality receivers, the final IFT often had the role of infilling the gap of a double-humped IF response developed by the preceding IFT(s), for which a modest loaded Q was often required.

More recently, WW 1978 September carried an article “The F.E.T. as Detector – Improvements Over the Diode Detector”, by Amos the Younger (Roger B.) This discussed various configurations, but most emphasis was placed upon the fet version of the infinite impedance circuit, including mention of its tendency to oscillate if not carefully designed and built. A more general discussion of AM demodulators was provided by Amos the Elder (S.W.) in WW 1980 April page 68ff.


Cheers,
Synchrodyne is offline  
Old 4th Dec 2017, 8:13 am   #35
Synchrodyne
Nonode
 
Join Date: Jan 2009
Location: Mt. Maunganui, New Zealand
Posts: 2,063
Default Re: Infinite Impedance detectors.

The attachment shows the AM demodulator that was used in the McKay Dymek series of HF receivers of the late 1970s:

Click image for larger version

Name:	McKay Dymek DR33C AM Demodulator.jpg
Views:	39
Size:	49.2 KB
ID:	153327

To me, it looks like the bipolar transistor version of the infinite impedance demodulator. In this case AGC bias was taken from the emitter and thence to the AGC DC amplifier. In the solid-state realm, AGC was less of a problem than with valves in that with transistors and ICs, AGC voltages were relatively easily inverted, offset one way or another and amplified as required for application to RF and IF amplifier stages.

But McKay Dymek described the demodulator differently, saying in its product literature: “Class D AM envelope detection is used giving much lower distortion levels than the conventional AM diode detector”.

That seems to be just a fancy description for the infinite impedance demodulator, or am I missing something in that schematic that makes it radically different to the point of justifying a new name?


Cheers,
Synchrodyne is offline  
Old 4th Dec 2017, 6:50 pm   #36
G8HQP Dave
Dekatron
 
Join Date: Sep 2008
Location: Solihull, West Midlands, UK.
Posts: 4,331
Default Re: Infinite Impedance detectors.

Someone in the marketing department did not let ignorance of what Class D actually means get in the way of a good story.
G8HQP Dave is offline  
Old 4th Dec 2017, 11:19 pm   #37
Argus25
Octode
 
Argus25's Avatar
 
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 1,625
Default Re: Infinite Impedance detectors.

Quote:
Originally Posted by Synchrodyne View Post
The attachment shows the AM demodulator that was used in the McKay Dymek series of HF receivers of the late 1970s:

To me, it looks like the bipolar transistor version of the infinite impedance demodulator. ,
When transistors are used in this manner all that happens is the B-E junction is used as the detector diode, but the current in that "diode" is reduced by the transistors hfe, because the bulk of the emitter current is sourced from the transistors collector and power supply, not the base. So its not really an "infinite impedance" it just behaves like a diode where the input impedance is elevated and the output impedance is lower.

In the circuit you attached they have at least biased the B-E junction of the transistor correctly. They have created a voltage reference from the B-E junction of another transistor (which has the same tempco and voltage drop as the detector transistor's B-E), to set up the initial DC conditions for the B-E junction of the detector transistor. This way the B-E junction forward voltage drop is largely negated and the detector will work down to low input signal levels with a lot less distortion.

It is a very old idea though, Sony invented this biasing scheme in 1956, I recently attached the schematic in the Homebrew section in a thread on detectors of Sony's TR-72 radio that biases the detector in this manner, but they cleverly used the B-E voltage drop from one of the IF transistors to do it !

see post #27 here:

https://www.vintage-radio.net/forum/...=132589&page=2
Argus25 is offline  
Old 5th Dec 2017, 12:15 am   #38
kalee20
Dekatron
 
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 4,511
Default Re: Infinite Impedance detectors.

The circuit Synchrodyne put up is not a peak envelope detector, though. It's an average (positive) responding thing.

The emitter of Q811 is loaded by a resistor, unbypassed to RF, with then an RF filter to give the audio output. If there had been a capacitor across the 2.2k emitter resistor, this would have charged up to the RF peaks, giving about 3 times as much voltage (actually π times as much)..

There is also an AC/DC load ratio issue here, because the emitter resistor of 2.2k is bypassed from an AC point of view by the load on the AF output. More extreme is the AGC filter following the 47k resistor, which will effectively take the emitter positive a tiny amount, so at approaching 100% modulation there will be appreciable distortion at the modulation troughs.
kalee20 is offline  
Old 5th Dec 2017, 12:25 am   #39
Argus25
Octode
 
Argus25's Avatar
 
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 1,625
Default Re: Infinite Impedance detectors.

Quote:
Originally Posted by kalee20 View Post
The circuit Synchrodyne put up is not a peak envelope detector, though. It's an average (positive) responding thing.
Yes, its output works on averages not peaks but is still the average of half wave rectification by the transistor's B-E junction, with energy taken from the power supply, via the collector, which has the effect of raising the input impedance and lowering the output impedance. Since the peak voltages at the emitter are not charging a capacitor directly, this further increases the input impedance at the base, on peaks, but its still not infinite.

One thing I could have mentioned, is that when a transistor's B-E is used as the rectifying device (additional power taken from the collector circuit and power supply or not) it is inferior to a typical diode in dynamic range. This is because most signal transistors have a B-E reverse breakdown voltage of around 7V, so if the signal level is too high the transistor detector malfunctions and a simple diode detector doesn't. On the other thread I mentioned a circuit was posted where a diode was added in series with the transistor's base to prevent this from happening.
Argus25 is offline  
Old 5th Dec 2017, 6:17 pm   #40
kalee20
Dekatron
 
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 4,511
Default Re: Infinite Impedance detectors.

I'd not thought of that! And it would be even worse if there was an emitter capacitor to hold-up the emitter voltage.

1V RMS carrier input... that's 1.4V peak... If 100% modulated then at the modulation peaks the voltage will be double, so that's 2.8V, which the emitter capacitor would charge up to... then at the modulation peak but half an RF cycle later the base would be at -2.8V, thus 5.6V reverse Vbe with only 1V RMS carrier in! (yes I know I neglected the forward Vbe drop, but this is only a crude estimate...)

So yes, not much dynamic range for a standard silicon planar transistor.
kalee20 is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 10:38 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 - 2018, vBulletin Solutions, Inc.
Copyright ©2002 - 2018, Paul Stenning.