This is the talk page for discussing improvements to the Tuned radio frequency receiver article. This is not a forum for general discussion of the article's subject. |
Article policies
|
Find sources: Google (books · news · scholar · free images · WP refs) · FENS · JSTOR · TWL |
![]() | This article is rated Start-class on Wikipedia's content assessment scale. It is of interest to the following WikiProjects: | |||||||||||||||||||||||||
|
Disadvantages
editThe list of disadvantages is suspect.
Terman 1943 p 658 characterizes the TRF's disadvantages as "poor selectivity and low sensitivity in proportion to the number of tubes employed. They are accordingly practically obsolete."
For a variable tuned receiver, a primary disadvantage would be the trouble with synchronous tuning.
Also, wide ranges of frequencies require bandswitching. Old variable capacitors might be 40 to 360 pF -- a 9:1 cap ratio that only gives (ignoring pads and strays) a 3:1 frequency ratio. (AM superhets would do 540 kHz to 1600 kHz.)
For the given list of disadvantages:
- For constant bandwidth, the filter loaded QL should increase linearly with center frequency. If L stays constant and C varies, then reactance of L increases linearly with frequency. If load resistance is constant and swamps skin effect, then the loaded QL increases linearly. Skin effect causes unloaded QU to diminish, but skin effect copper losses increase as the sqrt(frequency), but copper losses may be dominated by core losses. In any event, the following stage's load RL on a tuned circuit is often at least the same as the tank's unloaded loss resistance RU to keep the insertion loss low. If , then and the insertion loss is 3 dB. Effectively half the power gets burned up in the tank and half flows to the load. A better reason for selectivity changing with center frequency is when the filter is not constant bandwidth.
- Stagger tuning is primarily to flatten the passband. It does affect the gain (and therefore stability) a little, but there is still plenty of center frequency gain. The TRF confronts the same stability problems as a high gain IF amplifier. The IF amplifier can do a better job of isolating the stages because there aren't tuning connections. Using two IF frequencies means less gain is needed at one frequency. (RF gain usually just compensates for mixer losses.)
- L/C ratios? XL does not set the passband gain/amplitude. Typically L and C cancel each other out, so stage gains are set by the effective loss resistances and loads.
Glrx (talk) 04:03, 14 March 2011 (UTC)
- I agree that this section may have problems and I'm not too inclined to work on it myself but I suspect that no one would object if you rewrote it. Since I wasn't born 100 years ago I don't remember what the shortcomings were perceived to be at the time although it's not hard to see how the superhet was a great improvement with a modest increase in the number of tubes (the most important cost!) and that the regenerative receiver could achieve considerably better performance with fewer tubes but was touchy to operate. Just commenting on the technical issues as you laid them out:
- For a variable tuned receiver, a primary disadvantage would be the trouble with synchronous tuning.
- Absolutely, if you mean "tracking" of the sections, although I'd point out that tracking an RF frequency and a LO in a superhet is a greater challenge since those 2 frequencies aren't the same (or even proportional). But tracking the stages (especially if you were trying to maintain stagger tuning!) would seem difficult and this should be listed as a "disadvantage."
- Also, wide ranges of frequencies require bandswitching. ..... a 3:1 frequency ratio.
- I don't understand what you're driving at. I would have called 3:1 wideband, but in any case an RF stage in a superhet has the same problem and needs switching coils/taps for a multiband receiver too.
- On (1), I can see there are multiple things to consider. As you seem to agree, if the coils loss resistance R_L were constant then the Q would increase with f^1 and the selectivity would be constant, but as the article appears to be saying, and seems reasonable, is that because of the skin effect R_L ~ f^.5 and the bandwidth ~ f^.5. Of course core losses (which you mention) would make it yet worse (but doesn't contradict the point). And then you also throw in the loading of the LC circuit by the connected stages, and there I think I disagree with you. If R_U were dominant then the problem is MUCH worse because this isn't a series resistance but a shunt conductance, and would (since we are tuning the capacitor while L is constant) cause the Q to actually fall: Q ~ 1/f so BW ~ f^2, much worse than we were talking about. BUT, since with vacuum tubes the grid is a high impedance and does NOT load down the tuned circuit, this doesn't become the main problem (however there would be some loading due to the Miller effect I believe which does get worse with frequency. I'm thinking of that old diagram with the triodes, of course.). Anyway, we agree that BW ~ f^.5 or worse, which is what the original text implied so it isn't exactly wrong.
- 2. Stagger tuning is primarily to flatten the passband. It does affect the gain (and therefore stability) a little, but there is still plenty of center frequency gain.
- Yes plenty, but if you think about it, then in order for stagger tuning to work the center gain is reduced considerably, so that the net gain say 3kHz off-resonance will be similar. But the current article is wrong in saying that stagger tuning is used TO reduce the gain (there are many easier ways to do that!). It means you get a more even bandpass using too highly selective tuned circuits (if that had actually been the problem), but then NEED more amplification to compensate for it. So the article is wrong. Whether stagger tuning was actually used in a TRF is doubtful anyway, and would be even more difficult to track across stages (so that the bandPASS stays the same!). I'd say remove mention of it (without a good reference). And we agree that making a tuned amplifier stable is much more difficult if it has to be tuned, whereas with a superhet once you have a stable IF strip working and stable, it works regardless of the tuner frequency EVEN if it's "barely" stable, whereas with the TRF you have to limit the gain so that it NEVER breaks into oscillation across the tuning range.
- 3. L/C ratios? XL does not set the passband gain/amplitude. Typically L and C cancel each other out, so stage gains are set by the effective loss resistances and loads.
- Yes, but you didn't think that through. If the triode (like a pentode) were viewed as a transconductance amplifier, then the voltage it produces across a parallel LC would be proportional to the equivalent parallel resistance Req of the LC at resonance. HOWEVER if that loss is due to a SERIES resistance in the LC (due to the wire resistance, say) and you are tuning the capacitor (not L) for resonance at different frequencies, then if you work it out Req actually increases ~ f^2 (much more gain at higher frequencies) where the Q has increased. Or only as f^1.5 considering the skin resistance, and I'm sure it's more complicated considering other losses, transformer coupling, output impedance of the triode, etc. The main point is clearly correct: you can't make such an amplifier easily maintain a constant gain across a wide frequency range by tuning the capacitor (and as above, the instability at some frequency then requires reducing the gain overall). So "because of the non-uniform L/C ratios" isn't wrong, is one way of (poorly) saying what I did above, nonquantitatively. I'm not sure if there is any better wording while still keeping it simple, but go ahead and try if you wish :-) Interferometrist (talk) 14:08, 14 March 2011 (UTC)
- Tracking RF and LO for a superhet is a simpler problem. See Terman (1943, pp. 649–652) citing Landon (1932) . The LO and IF determine the tuned frequency; RF has less impact. The RF filter is not as narrow as the IF for many reasons. To conserve gain, the RF should have low insertion loss. An unloaded Q of 100 is practical; loaded Q might be 30. At low end of HF (3 MHz), BW is 100 kHz. Although suppressing adjacent channels is nice, the primary purpose of the RF filter is to suppress the image.
- I buried the idea on the 3:1 ration. The skin effect will only change by .
- For (1), the design goal should be that RL is the dominant load. If core loss and/or copper/skin loss were dominant, then the filter would have high insertion loss. In other words, the core and copper loss, when transformed to a parallel load on the tank, should have a much higher resistance than RL. Your comment if RU were dominant is correct but is the wrong regime for a radio design. It basically devotes the power gain of the amplifier to heating up the inductors rather than amplifying the signal.
- For (2), I'm not sure that stagger tuning is a big hit on gain. A Q of 30 at 1 MHz is a BW of 30 kHz. Consequently, one must move the center frequency of the resonator 15 kHz to see a loss of 3dB. For AM voice, the CF shift would probably be less than 3 kHz. In the given schematic, there are only 3 tanks.
- For (3), for low insertion loss, it should still be the case that your transformed Req is much greater than the following stage's RL. Consequently, the output voltage of the stage is approximate gmRL and independent of Req and Q. The higher Q means the gain falls off more rapidly with offsets from the center frequency.
- I have no trouble with the first sentence of article's disadvantage (2); it certainly jives with Terman's statement about low sensitivity; sensitivity requires gain and high gain leads to instability. The other statements seem to be the consequences of poor design practice.
- A more direct criticism is the article's disadvantages do not have a WP:RS.
- Glrx (talk) 18:51, 17 March 2011 (UTC)
- Thanks for your thoughts, which I read through quickly but will take a closer look at when I have more time. But I'm pretty sure we don't have any huge disagreements, and I'm surely not discouraging you from editing the page! I myself did NOT write that section (or any of it); it was done 3 years ago by Mercyjhansi who appears inactive. But I also think the article will look wrong if it doesn't name multiple drawbacks of the TRF, no? You could certainly mention the "parts count" (tubes) especially compared to the regenerative receiver, not to mention the performance shortfall in every respect compared to the superhet. So if you want to edit out some of the questionable claims, I'd suggest adding ones that he missed. Anyway, I was just giving you my thoughts (and indeed I'm interested in circuit theory, less in ancient history) but I don't have any big stake in this! :-) Interferometrist (talk) 23:53, 17 March 2011 (UTC)
- A laundry list of disadvantages without WP:RS runs into WP:NOR. I'm not familiar with the modern usage of TRF, but there are probably applications where a fixed-frequency TRF receiver is practical. Glrx (talk) 23:14, 21 March 2011 (UTC)
- Oh, just one quick question, requiring only a quick answer -- this will give me some more to think about when I reread this. What exactly do you think the input impedance R_L of the next amplifier is, anyway? I understand transistors, but this is 40 years earlier. For a pentode I see it more as an open circuit with a bit of capacitance. For a triode, I'd think the Miller capacitance depending on the plate circuit could lead to an R_L though I seem to remember that more the opposite occurs: negative R_L (and oscillation)! I'd expect if R_L is high then the transformers are quite step-up to take advantage of that. You REALLY think R_L dominates the 1/Q of those tanks?? Interferometrist (talk) 00:03, 18 March 2011 (UTC)
- I'm not a tube guy, and I don't have a simple handle on a common cathode input impedance. Whatever the transformed R_L is, if it doesn't load the tank, then it is throwing away available power. That means more gain will be required. Power transfer and power gain are important. I've seen published designs where a stage has high power gain, but the input and output mismatch losses made the stage an attenuator. Glrx (talk) 23:14, 21 March 2011 (UTC)
- Thanks for your thoughts, which I read through quickly but will take a closer look at when I have more time. But I'm pretty sure we don't have any huge disagreements, and I'm surely not discouraging you from editing the page! I myself did NOT write that section (or any of it); it was done 3 years ago by Mercyjhansi who appears inactive. But I also think the article will look wrong if it doesn't name multiple drawbacks of the TRF, no? You could certainly mention the "parts count" (tubes) especially compared to the regenerative receiver, not to mention the performance shortfall in every respect compared to the superhet. So if you want to edit out some of the questionable claims, I'd suggest adding ones that he missed. Anyway, I was just giving you my thoughts (and indeed I'm interested in circuit theory, less in ancient history) but I don't have any big stake in this! :-) Interferometrist (talk) 23:53, 17 March 2011 (UTC)
Tilted coils?
editThe article says, Inside, [...] will be a series of large coils. These will sometimes be tilted slightly to reduce interaction between their magnetic fields. But, the picture that follows shows three coils, at right angles to each other (which makes sense). Perhaps the wording, sometimes be tilted slightly is too timid? -- RoySmith (talk) 14:55, 18 November 2012 (UTC)
TRF vs regenerative receiver
editI would like to delete the following paragraph, but hesitate to do so as I don't want to appear too aggressive:
- During the 1920s, an advantage of the TRF receiver over the regenerative receiver was that, when properly adjusted, it did not radiate interference.[6][7] The popular regenerative receiver, in particular, used a tube with positive feedback operated very close to its oscillation point, so it often acted as a transmitter, emitting a signal at a frequency near the frequency of the station it was tuned to.[6][7] This produced audible heterodynes, shrieks and howls, in other nearby receivers tuned to the same frequency, bringing criticism from neighbors.[6][7] In an urban setting, when several regenerative sets in the same block or apartment house were tuned to a popular station, it could be virtually impossible to hear.[6][7] Britain,[8] and eventually the US, passed regulations that prohibited receivers from radiating spurious signals, which favored the TRF.
I feel that there is a problem in the distinction between "TRF" and "regenerative" receivers, and also in the usage of "regenerative detector": this problem also exists in the Wiki "Regenerative Circuit" page. In the UK, reaction (also termed "regeneration") was universally applied to TRF receivers up to the early 1930s - that is, throughout their period of vogue. Reaction could be applied to any of the RF stages, or to several, but was most often applied to the detector stage only. Its purpose was to increase gain, and narrow the bandwidth, both of which it does spectacularly well. These receivers were not called "regenerative receivers", and the detector stage with reaction was not called a "regenerative detector" - it was called a leaky grid detector with reaction. When reaction (or regeneration) is applied in this way, it is not intended that the amplifier should oscillate, though if badly adjusted it could do so. If the set were made to oscillate, it ceased to function correctly, producing howls and whistles, so it should not be understood that this type of receiver routinely oscillated continuously; oscillation was an accident that sometimes happened during tuning and adjustment. In contrast to this, there exists a class of receiver known as a "regenerative receiver" which employs a "regenerative detector". This type of receiver is used for reception of SSB and CW signals (but not AM), and is also called an "autodyne". The circuit is almost identical to a TRF set with reaction, but differs in use in that sufficient reaction is applied that the detector stage oscillates continuously. This continuous oscillation provides the carrier which is present in AM transmissions, but is absent in SSB & CW transmissions. The oscillating detector acts as a heterodyne detector, and receivers of this type can also be classified as "direct conversion" receivers, though the term "autodyne" is preferable. It seems to me that the article, as it stands, conflates these two distinct types of receiver. It interests me that in the US (and perhaps elsewhere) reaction was not usually applied to TRF receivers, and I would like to know why. I suspect that it might have been to do with patent restrictions which for some reason did not apply in the UK. The description given in the article seems to be restricted to the types most common in the US, and the author seems not to be aware of the form that TRF receivers took in other lands. G4oep (talk) 20:04, 23 January 2015 (UTC)
- Interesting. Are we perhaps conflating regenerative and superrregenerative receivers here? Superregens definitely oscillate all the time and are notoriously noisy. But what do contemporary sources say about this? Were superregens used as broadcast receivers? Certainly the stereotype sound effect for tuning in an old-timey radio broadcast has various whoops and whistles before getting the station tuned properly; if this was a film trope, it must be derived from common experience with these systems? Let's find out! --Wtshymanski (talk) 22:50, 23 January 2015 (UTC)
- Did you read the two sources cited, from 1922 and 1924? In the US regenerative receivers were called "bloopers" because they radiated so much noise, and there was talk of outlawing them. In the early "tickler coil" types it was easy to make them oscillate. One-tube regenerative sets had to be adjusted very close to oscillation to give the single tube enough gain to bring in distant stations. It said a common method of tuning in a station back then was to increase the loop gain until the receiver oscillated, then adjust the tuning until the audible beat frequency went to zero, then back off the gain. I've done that. The author said as an experiment he connected a microphone in his regenerative set, and using it as a transmitter was able to talk to receivers more than a mile away! As G4oep says, a regen didn't generally radiate once it was properly tuned in, but in a dense city apartment block in the 20s there could be hundreds of regenerative sets all tuning into the same one or two radio stations. And there were also plenty of radio amateurs around listening to radiotelegraphy signals using the oscillating receiver as a heterodyne detector (BFO). The long wire antennas used back then were ideal for radiating the interfering signal. Later in the decade more advanced regenerative sets were made so they couldn't oscillate, but there were still plenty of legacy and homemade "blooper" sets around. These are the sets that TRF sets are being compared to, in the paragraph in question.
- Of course, this is all in the US. G4oep, in the UK receivers had to be licensed, didn't they? And radiating receivers were made illegal in the 20s, and tracked down by the Post Office with DF cars. Maybe interference from regenerative receivers was just much less of a problem in the UK? --ChetvornoTALK 01:41, 24 January 2015 (UTC)
Thanks for the comments... its all very interesting. What is on my mind is the terminology, and clarification of the different types of receivers. In the UK, 1920s receivers which we would now classify as TRF always had reaction fitted (in the UK). But I am not aware that they were termed regenerative receivers. In the amateur radio fraternity "regenerative receivers" always refers to autodyne receivers, which operate on a different principle. These have circuits which are often indistinguishable from single-valve TRF sets with reaction & grid-leak detectors. But in use, the reaction control is set so that the valve oscillates. The oscillation provides the missing "carrier" which is needed for detection of SSB & CW signals. This type of receiver is an autodyne, and could also be classed as a direct conversion receiver. Because the valve always oscillates, the risk of radiating an unwanted signal which can cause interference is constantly present. This is unlike the case with an AM TRF with reaction, where, if properly adjusted there will be no radiation. So I wonder whether there is consensus that "regenerative receiver" should be reserved for the autodyne type; if there is, then the paragraph I refer to should be changed.
I would classify receivers on the basis of how they achieve selectivity: 1) TRF by filtering at the frequency of the received signal; I would make this a definition of a TRF receiver. TRFs include crystal sets and also typical 1920s sets with reaction. 2) superhets; filtering at a fixed intermediate frequency 3) direct conversion; filtering at af (includes autodynes).
Then I would classify them according to the type of detector: For AM: anode bend, grid leak, discreet diode, homodyne, etc For SSB & CW: all the different kinds we know about (endless variety), autodyne FM: ratio, quadrature, etc
Thus a crystal set would be a TRF with diode detector. A "regenerative" would be a direct conversion RX with autodyne detector, or maybe just an autodyne receiver (autodyne & regenerative receivers are different names for the same thing in my mind). A 1920s type would be TRF with grid-leak detector; if it had reaction that would not change the classification, but you could add that detail in the description. A typical FM rig would be a superhet with quadrature detector. Super-regenerative receivers are TRF sets with reaction. The reaction is set so that if left to itself, the signal would gradually increase in magnitude and the set would oscillate, but the receiver is periodically interrupted at a super-sonic frequency so that self-sustained oscillations are never actually able to occur. Would readers like me to get my hands on this article and sort it out along these lines ? G4oep (talk) 16:18, 24 January 2015 (UTC)
- In the US your autodyne set was called a regenerative receiver (mostly, autodyne was occasionally used) regardless of whether it was receiving AM, CW or SSB. I haven't heard "regeneration" called "reaction" but I'll take your word for it. I didn't realize American usage differed so much from British. It seems reasonable to add these two alternate terms to the text to make it understandable by everyone. --ChetvornoTALK 17:38, 24 January 2015 (UTC)
- Comment. Generally, I follow Chetvorno's comments; G4oep's comments gets parts right but confuse some items.
- There were crystal detector radios, but they were neither sensitive nor selective. Adding an audio amplifier (instead of driving the headphones directly) would make them more sensitive, but it would not make them more selective; powerful off-frequency stations would slip through. Adding a passive synchronously tuned multipole filter in front of the crystal detector would make it more selective, but I don't recall seeing that approach used in old receivers; it may be because network theory and filter design had not blossomed yet. (See Network synthesis filters; Wilhelm Cauer.)
- Building multipole filters using active devices was understood, and the TRF is an instance of that: LC filters were cascaded with vacuum tube for gain and isolation. The TRF is a synchronously tuned RF amplifier followed by a detector stage (and possibly an audio amplifier because audio gain is cheaper than RF gain). Multiple stages provide improved gain and selectivity, but vacuum tubes were expensive and synchronous tuning was a challenge.
- The regenerative amplifier is an improved RF amplifier; positive feedback provides more gain and more selectivity; it's more economical than a TRF, but it still expects to have a downstream detector. It could, while being adjusted, oscillate and therefore self-radiate, but that should not be a problem when the receiver is correctly adjusted. The bias point is linear. The radiated signal would be low amplitude (improved only by the gain of the stage) and inject into a poor radiator (short antenna). Terman 1943 p 439: "When tuned radio-frequency amplifiers were first used, regeneration was frequently introduced intentionally in order to increase the amplification and selectivity. The development of improved amplifiers has caused this to be considered poor practice, because the regeneration varies with the resonant frequency, the volumne setting, etc., is critical with respect to tube voltages, and causes oscillations when excessive." (The Radio Handbook, 1940, Ch. 4, page 76, regenerative RF amps (preselectors) would be used in superheterodyne receivers for wavelengths less than 30 meters.)
- The vacuum tube receivers described above would have a separate nonlinear detector stage for amplitude demodulation. That stage could be a crystal detector, but early crystal detectors required adjustment. A more pragmatic detector would be vacuum tube diode (no hunting for a hot spot on the galena crystal), but an even better detector would be a triode biased for nonlinear operation because it would provide gain with a small increment in cost (one more element). Tube characteristics can vary, so an automatically adjusting grid-leak detector would be used (apparently AE grid-leak detector = BE leaky grid detector).
- There are also regenerative detectors. A detector extracts the amplitude modulated signal. Regeneration (positive feedback) can be applied to grid-leak detectors to make them more sensitive. (Terman 1943 p. 574, Regenerative Detectors; Terman also states, "Although regeneration represents an inexpensive means of increasing the radio-frequency amplification, it increases the selectivity excessively, requires critical adjustments that depend on the signal frequency, and introduces oscillations with consequent interference and whistles whenever the regeneration is accidentlly made excessive. Since the development of satisfactory radio-frequency amplifiers, regenerative detectors have found relatively little application.") The detection is still done with the tube's nonlinearity, but regeneration provides more gain and selectivity.
- One could make a one-tube regenerative receiver using a regenerative detector.
- None of the above circuits are operated as frequency mixers (a regenerative detector is not an autodyne). It happens that if a (linear bias point) regenerative RF amplifier is adjusted to oscillate, then it can become an autodyne (a self-oscillating mixer or converter). It oscillates at one frequency, the oscillations have high ampltitude/power, and the circuit can radiate that frequency from even an inefficient antenna. IIRC, simple autodyne CW receivers were effectively outlawed; autodynes could be used if they were isolated from the antenna (e.g., by an RF amplifier stage; the issue is antenna radiation). Due to tube nonlinearity caused by large oscillator amplitude excursions, the autodyne will heterodyne signals received from the antenna. An autodyne is not necessarily a direct conversion receiver; it can be used with a non-zero IF. Consequently, an autodyne can be used to receive CW by beating to an audio frequency (Chetvorno's BFO used at RF rather than IF); such a receiver cannot effectively isolate the LO from the RF input without an RF stage. An autodyne can also be used as a converter in a superheterodyne receiver. (See All American Five; pentagrid converters isolated the oscillator from the antenna; local oscillator emissions were 5 kHz off station centers; the RF input filter would attenuate the LO emissions.)
- The superregenerative receiver is a more complicated beast. The input signal affects self-oscillation start up time of a quenched oscillator.
- To me, any radio with a regenerative circuit would be a regenerative receiver (unless some other moniker (e.g., superhet) would be more appropriate); it doesn't matter if regeneration is in the RF stage, the detector, or both. (Using regeneration in both would be a nightmare of too much selectivity and adjustment.) My viewpoint may be wrong; see suggestive Terman comment above; one online definition says the important point is using a regenerative detector; another just requires using positive feedback. There are many designs on the web, but I do not feel comfortable with them for many reasons. Here's a non-authoritative example of a "TRF receiver" with regenerative detector.[1] Many web designs for regenerative receivers use a common plan of an ordinary RF amp followed by a regenerative detector. However, some designs appear to use detected audio (rather than RF) for the positive feedback. My dim and unreliable memory tells me an early US Army regenerative radio used regen for the RF amp only, but I haven't found the schematic and it was years ago. Consequently, I don't have any solid authority for the definition of "regenerative receiver".
- The "when properly adjusted" comment in the quoted paragraph above is confusing. An improperly adjusted TRF does not radiate (unless it uses regenerative amplifiers); a properly adjusted regenerative receiver may radiate, but the radiation would be coherent and not interfere (save multipath effects).
- Glrx (talk) 06:32, 29 January 2015 (UTC)
Bad definition of TRF in How it works section
editI agree, the definition "...all the bandpass filtering... is done at the RF frequency" is too broad. Crystal radios and reflex receivers would also come under that def. and possibly regenerative receivers depending on your point of view. That was my addition; I don't know how that slipped in. Thanks for catching it, Grlx.
I have some reservations about the other definitions. A lot of sources [2], [3], (Terman 1943, p. 658) define a TRF receiver as: "one or more tuned RF stages, a detector stage, followed by one or more audio amplifier stages". This def bothers me historically, because I think I have read that early TRF receivers without audio stages were called TRFs. Before broadcasting began in 1920, radio was a solitary hobby and there were few loudspeakers marketed, so most radio receivers were designed to be listened to with earphones, which could be driven by the detector and didn't need an audio amp stage. I think the term TRF was even applied to the first one-tube nonregenerative receivers like De Forest made, where the detector provided all the gain. I think back then "tuned radio frequency" was just a name that was used to distinguish a receiver from a regenerative receiver. Of course I need to find these references. What do you think? --ChetvornoTALK 03:14, 27 July 2015 (UTC)
- I never found a definition of TRF that sounded right. I have to give weight to Terman, but I don't like the inclusion of regenerative RF amplifiers in a TRF. You don't like the inclusion of an audio amplifier stage, and I agree.
- A crystal radio has no RF amplifier; it's a tank driving a detector; there may be a audio amplifier after the detector.
- To me, a TRF radio needs at least one tuned RF amplifier stage and not employ regeneration or heterodyning. The goal is a more sensitive receiver, and RF amplifiers were a way to get it. Unfortunately, plain RF amplifiers didn't offer much gain, so sensitive receivers needed to use several RF stages, and that imposes a synchronous tuning requirement.
- If any RF amplifier uses regeneration, then to me it would be a regenerative receiver (even if it used a non-regen diode/crystal/plate detector). Terman allows a TRF to use a regenerative RF stage, but I don't see that as right. If one uses regeneration, then one should have enough gain so that multiple RF stages are not needed. (Well, maybe a regenerative receiver would use a non-regenerative input stage for antenna isolation.) To me, a regenerative RF stage driving either a simple detector or a regenerative detector would be a regenerative radio. However, a regen RF amplifier driving a regen detector doesn't make sense. Or even two regen RF stages.
- If one had several non-regenerative RF stages driving a regenerative detector, then most sources would call that a regenerative receiver. Practically speaking, most of the gain and selectivity would come from the regenerative detector, so I'd expect zero or one non-regen RF stages.
- Glrx (talk) 06:37, 27 July 2015 (UTC)
Tuned radio frequency versus heterodyne, and WWII.
editHello Glrx (talk · contribs). You were concerned about the reliability of the statement regarding the use tuned radio frequency receivers versus heterodyne receivers during WWII. Here is the more information confirmed by another reference, an article from the August 1980 edition of CQ Magazine, pp. 21-22.
New reference: (deleted link)
“You are probably surprised that the Germans used t.r.f. sets. But they certainly knew how to make them in superior form. A great advantage of the straight set is that spurious responses are non-existent, even in the presence of extremely strong signals as in shipboard use, where several transmitters may be active at the same time receivers are operated.
"Another advantage from a military point of view is that the t.r.f. set does not use oscillators and so the chance of location by the enemy using a direction finder on spurious radiation of the set is negligible. This article could only be prepared thanks to the assistance of PA0AOB. Not only did he make the receivers available for photography, he also gave the author the opportunity of using some of the sets in his own shack for a considerable period of time.
"The fact that PA0AOB could provided the original technical manuals, or exact replicas of them, was also of great help in the preparation of this article.”
How about we put the original paragraph back in, and improve it with this latest material? Regards,Desertroad (talk) 16:10, 22 August 2016 (UTC)
- I'm opposed because the sources are still inadequate.
- A three-part article on German radios does nothing to support the claims of Allies using TRF radios to avoid detection or performance specification for Allied receivers. The three-part article does not cite references, so it is not a serious work.
- The three quotations given above all have trouble. Spurious responses do no preclude the use of superhets or dictate that TRFs must be used. Oscillator radiation was a problem with regenerative sets; superhet antenna radiation is minimized with an RF amp stage; efforts to track oscillator radiation from superhets in the 1960s required nearby equipment (Spycatcher); it was not effective at locating; turning on a radio and using it and photographing it does not provide historical perspective. Original technical manuals provide details, but they are not appropriate for broad statements.
- On another front, the 1980 three-part CQ article is presumably still under copyright protection. The author was a resident of the Netherlands, so it is unlikely that the author is posting his work on a Washington state website with the publisher's permission. I've deleted the link as an apparent WP:COPYLINK violation.
- Glrx (talk) 19:22, 22 August 2016 (UTC)
CMOS
editWhy is a CMOS book the reference for neutralization in vacuum tubes? I don't have the book, but I would suspect that there should be more historical references around. Gah4 (talk) 13:33, 21 March 2020 (UTC)