Talk:Autodyne

description

The description of an autodyne mixer is too precise to the extent that it would include configurations that are not, in fact, autodyne mixers. The Pentagrid converter, for example is a mixer where the mixing and oscillation take place in the same active device. But it is not an autodyne mixer. In fact the Pentagrid works by turning a single active device into effectively two active devices by using different parts of the same device to perform the oscillation and mixing.

An autodyne mixer strictly uses the same active device as the oscillator and the mixer. 20.133.0.14 14:39, 22 August 2007 (UTC)

The Superhetrodyne articles have been a problem. The Pentagrid converter article became a bit of an edit war with a fellow who was upset that the americans claim the invention, while the Brits used something along the same lines. The discusion then got to be about BVA rules, and far off topic. I set aside the material I had found as it would have been a lot of work to clear things up. I even found the reference in "radiotron designers handbook" showing the two types of Pentagrid and the Heptode all in one mixers. That is even a neutral source being published in australia.

Best explanation I came up with was an Old Slyvania Technical note, which stated that the autodyne was fairly early, and worked OK but was unable to co-exist with AGC. The 2A7/6A7 came out which COULD live with AGC, and the american industry switched over. The european industry seeming to prefer the two unit converters. (triode-hexode) Later after the War, The americans came back to the autodyne, using (typicaly) a 12AU6 as the conversion gain was higher, and for sets in urban areas, the set could do without AGC on the converter. Transitor radios tened to have more gain stages and so could also avoid the need for AGC on the converter, and thus use an autodyne.

Talking about stable in relation to transistors is a bit NPV as they typicaly came in to the set maker from the transitor maker with a wider range of values, even though they did not change characheristics as much, the circuits had to be designed to compensate for the variablity in what was supplied. Tubes tended to be closer to the specifications when new, and gradualy lost some performance over many years. Tube equipment thus did not need to have as much compensation for device characheristics. As time went on and silicon repleced Germanium, the gain did get better on run of the mill transistors.cmacd 16:25, 22 August 2007 (UTC)

Merge into Regenerative circuit?

Autodyne was the early term for a regenerative radio receiver. The Regenerative circuit article duplicates the content of this article, and the circuit is now known by that name. I think this article should be merged into that one. --Chetvorno^TALK 00:06, 22 April 2012 (UTC)

I'll need to look further, but right now I disagree with the first sentence (and the characterization in the regenerative circuit lede). My sense is that regenerative receiver and Autodyne are different. The Autodyne is essentially a superhet where the tube functions as the LO and mixer (a converter or self-oscillating mixer). (See pentagrid debate above.) This article description says the Autodyne oscillates at a slightly different frequency (LO function); nonlinearity would produce mixer products; CW could mix down to audio to turn CW into on/off tones.

Regenerative receivers are positive feedback beasts that keep the loop gain just below one to get a high-Q and high gain (i.e., 1/(1-AB) is very large when AB near one). In the simple view, the RR is a linear circuit and would not generate mixer products -- there would not be an audio tone output for CW. My view runs up against Regenerative circuit#CW reception, so the (unsourced) RC article section may have problems. I'll look further; I don't remember how RR demodulated CW; AM demod may use a self-mixing power detector (bias point shifts in the tube circuit).

Glrx (talk) 15:35, 23 April 2012 (UTC)

Autodyne was the technique of making a regenerative receiver oscillate, to serve as a BFO to make CW signals audible. (see 1, 2, 3, 4, 5) As you say, the oscillation frequency was slightly offset from the signal frequency to generate an audible heterodyne. But it still uses regeneration to amplify the signal and improve selectivity. The gain $G(j\omega )=A/(1-A\beta (j\omega ))\,$ is increased by regeneration whether the loop gain $A\beta (j\omega )\,$ is slightly below 1 as in a regenerative receiver receiving AM, or slightly above 1 when it is operating in the autodyne mode and also serving as an oscillator. --Chetvorno^TALK 16:42, 23 April 2012 (UTC)

So I think this mode should be explained in the Regenerative receiver article. --Chetvorno^TALK 02:37, 24 April 2012 (UTC)

Thanks for hunting down those references, but I still need to look further. I'd still characterize Autodyne as separate from a regerative receiver/circuit.

Ref 1 is good, but it is part of a larger discussion about heterodyne (not RR). It says that a regenerative receiver may use one tube/valve for both amplification and detection. That doesn't contradict my viewpoint. P. 267 says Fessenden 1908 for heterodyne applied to receivers; use of crystal/tube detectors. (Heterodyne detectors existed before RR.) P. 268 reports Armstrong's Proc IRE October 1917 heterodyne detector results with the "auto-heterodyne" (which is after 1913 RR patent). (U. lib doesn't have early P. IRE; I don't know if it goes back to 1917.)

Ref 2 also separates amplification and detection. The "autodyne circuit" is an oscillator and detector (nothing about regeneration); "Autodyne reception" is a regenerative receiver with an Autodyne detector -- no requirement for the two functions to be in the same tube/valve.

Ref 3 just makes a bare statement.

Ref 4 has "regenerative amplification" at the top of the page (the regenerative receiver issue), and then starts a new section about "Autodyne Production of Beats" (a detector issue). It refers to "heterodyne action". The next page distinguishes Autodyne/"self-heterodyne" from a "separate-heterodyne" circuit using another device/tube/valve for the LO. Early designers were very good at exploiting tubes for multiple functions. It sounds like the Autodyne was a clever merging of the mixer and LO (when was the BFO invented?). We could have an RF amplifier followed by a detector (your first ref talked about Hogan's 1910 detectors; RR is 1913; Audion is 1906; triode 1907); a good CW detector would be the "Autodyne circuit" of ref 2. Now some clever soul says lets merge the RR and the Autodyne circuit into one tube/valve.

Ref. 5 suggests how that merge took place. I don't think it links RR to Autodyne.

I don't see the record as completely clear, but I see regenerative circuit (amplification) and Autodyne (detector) being separate topics that seem merged because designers were exceedingly clever and stuck the two functions into the same tube/valve.

I disagree with your observation about the loop gain near one. At gains less than one, the system is stable with high gain. At gain equal to one, the system can be an oscillator (producing an output without any input) with constant amplitude. At a gain above one, the system is an oscillator with an exponentially increasing output amplitude (which ultimatly crashes into the rails).

I think the better explanation is the Autodyne receiver has a loop gain equal to one at the LO frequency. At any frequency slightly different from that frequency, the gain will be less than one. That's where the Autodyne RR operates -- slightly detuned from the LO but with plenty of regenerative gain. (Actually, the loop gain will average to one at the oscillation frequency due to nonlinearities, but those nonlinearities are providing the mixing.)

I see the Autodyne as a distinct subject; I think the RC article, when it talks about two modes, confuses regeneration and heterodyning.

This whole discussion also makes it clearer why Armstrong would use the term superheterodyne; he needed to distinguish the common heterodyne detector.

Glrx (talk) 19:42, 24 April 2012 (UTC)

Google search for "Autodyne detector" turned up

http://theradioboard.com/rb/viewtopic.php?t=1862&sid=7b90b8da805165338b8dcb803fa81def Notes confusion about Autodyne term; suggests Pressley and Roberts circuits were for first stage converters (reradiation issue).

http://www.radio-electronics.com/info/radio_history/gtnames/round/hjround.php Dunwoody crystal detector in 1906. H. J. Round Autodyne patent in GB ca 1913. Round was looking for amplification, so suggests regenerative.

http://www.one-electron.com/doc_archives/RAL7.pdf Military radio from 1950s with 2 RF amps followed by Autodyne detector; "Regeneration" control affects Autodyne stage.

http://kearman.com/RH1940-ch06.pdf Autodyne front end followed by audio amp.

http://rsi.aip.org/resource/1/rsinak/v31/i8/p909_s1?isAuthorized=no, Review of Scientific Instruments

Glrx (talk) 20:42, 25 April 2012 (UTC)

I took a closer look at my references, as well as the additional ones you gave. In general I have to agree with everything you said. I was not familiar with antique radio terminology. As you say, there seems to be some disagreement about what "autodyne" means today, which is not surprising since it is mostly just of interest to hobbyists and antique radio collectors. But if it had a specific meaning in the 1920s I'd like to pin it down. I agree that the term "autodyne" certainly meant using a local oscillator to create an audible heterodyne with the signal for CW radiotelegraphy reception. But, confining attention to sources from the period, such as 1 and 4 above, it seems to me likely that it referred specifically to using the same tube for the local oscillator and mixer. This would mean regenerative amplification of the signal itself, and thus an autodyne was a form of regenerative receiver. I'm not saying this article should be merged with Regenerative circuit; I guess there's enough here for a separate article, but we should get the definition straight. There are a large number of technical books from that period preserved on Google Books. I'm going to see if there's enough WP:RS's to distinguish which is right. --Chetvorno^TALK 23:37, 9 May 2012 (UTC)

And of course you're absolutely right about the loop gain of a regenerative receiver which is oscillating. The loop gain is one at the local oscillator frequency, while at the signal frequency it is down on the skirt of the gain curve, less than one. I don't know what I was thinking. --Chetvorno^TALK 22:26, 9 May 2012 (UTC)

I've been reading a lot of references in the 1920 to 1925 period, but it hasn't cleared up many issues. De Forest generated a clear tone in 1912, but there were no details on how it was generated; it may have been an audio oscillator or it may have been an autodyne; it was, however, observation rather than understanding. Other sources give the sense that Armstrong and others thought the oscillation was a nuisance. The instructions for the RR were to adjust it to just below instability. The court case seems to be more about who invented the vacuum tube oscillator even though they didn't claim it. I'm not sure what Meissner did, but a source gives him credit for thinking the oscillation was useful. Some sources say about 4 separate groups came up with the autodyne idea. Some ways don't jive with my above notion of merging separate functions into one tube. There's a nice diagram of a separate LO and mixer attributed to Fessenden, but I think that is refined view. A Fessenden coworker pointed out that he didn't have a good LO and didn't get good results. Even some modern sources that look good in some areas go off the deep end with some technical blunders.

I'm even having trouble with what should be simple history/technology. The term audion is not clear. Apparently the first audion had the control element outside the glass envelope; when I first heard that I got one image; later I saw an illustration with the control element being a solenoid at right angles. De Forest later put an electrode in the envelope, but it was not strategically positioned. De Forest thought the gain came from small movements of the filament and plate. Armstrong talks of using an audion, but by that time it was probably a more conventional triode. Maybe the term became generic for time. It's also amusing that resistors and inductors were drawn the same way -- most resistors back then were probably wirewound.

Through all of this, there's the damped wave / continuous wave issue. In 1913, a heterodyne detector may not have been needed because the transmitters were damped wave and had their own "tone". By 1922, CW had pretty much taken over, so the autodyne would have been appropriate. By 1925, the radio community was trying to stamp out the autodyne receiver because of its LO radiation. There's a nice vocabulary for different kinds of interference.

Clearly, an autodyne is a combined mixer and oscillator (also termed a converter). I'm not sure about much else.

Glrx (talk) 01:24, 10 May 2012 (UTC)

You've obviously done more research into the history than I have. I just wanted to find out what autodyne meant, in modern technological terms. I did some reading. I'm just going to summarize it, to get it straight in my mind. Forgive me if I repeat what you already know or stated above, and let me know if I've got anything wrong:

It's clear to me that, in the 20s, an autodyne receiver meant a "heterodyne" receiver with the local oscillation and detection done by the same tube 1, 2, 3. A "heterodyne" receiver, invented by Fessendon in 1901, is what today would be called a receiver with a beat frequency oscillator (BFO). It was a receiver with a local oscillator (LO) that was tuned to produce a heterodyne (beat frequency) in the audio range, to make CW signals audible. It shouldn't be confused with a superheterodyne receiver. which used heterodyning for a different purpose. In a superheterodyne receiver, the heterodyne or intermediate frequency is higher, in the low radio ("super-sonic") range. The radio signal is converted to the intermediate frequency to allow it to be amplified and filtered better than it could be at the original frequency. After this, it is applied to a second detector to convert it to audio. If CW was being received, a second local oscillator was used to produce a second heterodyne frequency in the audio range.

Fessendon invented the heterodyne receiver before the audion was invented in 1906, and I seem to remember seeing circuits of Fessendon's early heterodyne receivers with Poulsen arc or spark gap local oscillators. The crystal or electrolytic detector would serve as the mixer (converter) to combine the RF and LO signals. After the audion was invented, there were TRF heterodyne receivers, such as simple two-tube circuits, with one tube providing (nonregenerative) amplification and (because a vacuum tube conducts in only one direction) detection (mixing) of the RF and LO signals, and the second tube as the LO. Here's where it gets hazy for me. These sources 4, 5 say that in 1913 British captain H. J. Round combined the LO and detector in the same tube, patenting the autodyne. Since the tube had positive feedback it would also have exhibited regenerative amplification of the signal, so why isn't Round also credited as inventor of the regenerative receiver? I guess Armstrong's patent predated his, or else he didn't recognize regeneration was going on in his circuit. Armstrong's 1913 regenerative receiver patent has several circuits, all with a single tube. Did Armstrong use his circuits in oscillation (heterodyne) mode, to create a beat tone to receive CW? If so, it was an autodyne. So if his patent predated Round's I would think he would be considered the inventor of the autodyne. But maybe, if you say he regarded oscillations as a nuisance in his circuits, he used a separate tube as a LO in these first regenerative sets.

Regardless, an autodyne circuit, since it uses the same tube for LO and mixer, exhibits regenerative amplification of the signal, whether that was recognised in the earliest sets or not. Because the beat frequency is in the audio range, the LO frequency is no more than about 10 kHz from the RF signal. So the incoming signal falls within the gain curve of the LO, and is amplified by positive feedback. Certainly within a few years of Armstrong's discovery the value of regeneration was widely understood, and it was recognized that regeneration allowed one tube in an autodyne to have the gain of 3 or 4 tubes in the earlier TRF receivers. So both from the original POV and the modern POV an autodyne is a type of regenerative receiver. I think the introduction should describe it that way. --Chetvorno^TALK 01:16, 13 May 2012 (UTC)

Basic def is right, but it's more complicated as I understand it right now; many details are not clear. I haven't had time to chase the story; secondary sources give some indications but drop the ball on crucial details. See Fessenden patent list at superheterodyne and discussion in one of the footnotes. Fessenden invented heterodyne detection in 1901, but he did not use a local oscillator but rather 2 CW waves were transmitted 1 kHz apart; the two waves would mix in the local detector to get the beat. Fessenden invented his own detector diode. In 1905, Fessenden finally used a local oscillator (but the patent was granted many years later). That's why the sources at superhet disagree on when heterodyne was invented.

I'm not sure of the usage for beat frequency oscillator. To me, BFO is used for an oscillator that is close to the IF frequency; I'm not sure the term applies to an oscillator close to the RF frequency.

Many detector technologies appear.

I think Lodge was before Fessenden, but that detail isn't clear to me right now. In any event, damped wave transmission was prevalent, so Fessenden's heterodyne wasn't the mainstream detector. It had to wait for CW. Advantages of CW were recognized, and Poulson comes along with the arc converter. Now different methods of receiving CW become important, but it is still difficult to make a decent local oscillator. There were no tube oscillators. It's pretty easy to chop the CW wave locally and detect it that way.

Tubes also appear, but they are crummy and improving. De Forest didn't know where to put the grid. The invention of the B battery was significant, but I've forgotten when. The B battery was also low potential, the inter-electrode distances were probably long, so the transit time and frequency response were probably poor. Tubes get better.

I suspect the TRF is king, but I don't really know. I doubt that Fessenden's heterodyne would be popular because it would be trouble to make the LO. A chopper seems simpler.

The RMS Titantic sank in 1912. RMS Titatanic#Technical facilities suggests a second spark gap was used for receiving. http://revolutionsincommunication.wordpress.com/features/radio-and-the-titanic/ claims Marconi equipment was outdated (and no tubes).

Armstrong and others come up with the regeneration scheme to get more gain. It's clear that anyone who played that game right also made an oscillator, but it's not clear that they recognized what it was good for. De Forest could have made an autodyne detector. I haven't seen any clear sources on how the "clear tone" was generated in the DC court case. I haven't even found the US Supreme Court case. But the sources I've seen suggest regenerative amplification and oscillation were separate things. Who invented the autodyne is even less clear. But we've marched into 1913 or 1914.

I've heard big gain numbers for RR amplifiers, but I'm not sure the same claim can be made for autodyne gain. The transmitters might be at 100 kHz; with a Q of 100, the BW is 1000Hz. At 500 Hz offset, the loop gain would be 0.7 instead of one; that means the denominator is about .3 rather than being a really small number. Tuning the grid circuit for RF and the plate circuit for LO can play a double hump filter gain to give high gain at both frequencies. Armstrong's circuit could tune both meshes and had adjustable regen. I need to read more.

Armstrong's superhet goal in 1918 was gain; his tubes were crappy, so he used a lower IF where the tubes had reasonable gain. It's not clear to me that he had to worry about image rejection (or even tuning his IF, for that matter, but I suspect he did do that).

Gotta go. Glrx (talk) 22:14, 13 May 2012 (UTC)

I see lots of confusion here

A "Regen" receiver is an RF circuit which uses positive feedback to dramatically increase gain and Selectivity. For AM or MCW reception it is operated just short of Oscillation, but when receiving Morse or SSB it is purposefully made to oscillate in order to generate an audio beat note.

When a Regen is Oscillating it is essentially the same as an Autodyne. These days we would call it a "Direct Conversion" or "Zero IF" receiver.

A Superhet however is a different thing entirely. The front end of a Superhet is essentially the same as an Autodyne, but the Superhet has an Intermediate Frequency stage (IF) who's purpose is to provide Selectivity with high Gain.

The Autodyne has the Oscillator running close the the incoming signal, while the Superhet has the Oscillator frequency offset by the IF frequency.

Whether the mixer and oscillator functions are combined into one stage is irrelevant. The important thing is that the operating principles of a Regen, an Autodyne or a Superhet are each distinctly different.

Gutta Percha (talk)

Is it appropriate to thank you for this post? I will anyway. Thank you! — Preceding unsigned comment added by Hkiernan (talk • contribs) 16:10, 3 April 2019 (UTC)

Thank you very much for this post. After reading all the confusing stuff in the previous paragraphs, it was good to see what you wrote here.

When Edwin Armstrong was playing with his Audion setup he got the output wires too close to the input. This caused a squeal in the phones. This had probably happened to other people, but they probably cursed and got rid of the squeal. Mr. Armstrong, on the other hand, was a radio nerd and a radio geek before those terms were invented. So he played with it. He invented the feedback amplifier, the oscillator, and the direct conversion receiver. You may call it an autodyne or zero-IF, or anything else, but it’s a direct conversion receiver. It generates an internal signal and mixes it with the incoming signal and produces an audio output. That is a direct conversion receiver. They are used extensively today. Your iPhone probably has three: one each for the WiFi, Bluetooth, and the cell phone. Handheld GPS receivers are direct conversion.

You can make a direct conversion receiver with one tube, but most designs were two or three tubes. They were a huge advance, because they allowed the gain to be at the receive end rather than at the transmit end. The oscillator and the direct conversion receiver allowed for lower transmitter power over the same communication range.

Until recently, superheterodyne receivers were the most common receiver, but I don’t think they are any more. Direct conversion receivers have taken over. Coupled with a computer and appropriate software, a direct conversion receiver can far outperform a superheterodyne receiver. Anyway, Mr. Gutta Percha, thank you for clearing some stuff up.

wa6bjh (talk) 17:17, 3 April 2019 (UTC)

External links modified

Hello fellow Wikipedians,

I have just modified one external link on Autodyne. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:

• Added archive https://web.archive.org/web/20120207233031/http://ieee.cincinnati.fuse.net/reiman/01_2003.html to http://ieee.cincinnati.fuse.net/reiman/01_2003.html

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}} (last update: 5 June 2024).

• If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
• If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—InternetArchiveBot (Report bug) 08:12, 12 July 2017 (UTC)

VT proximity fuse - possibly the most important use of this design

It may be advisable to include the so called VT proximity fuses of WWII in the Application section as these fuses are in the short list of the most important technological inventions decisive for the war, right along with the nuclear weapons. Wikipedia covers that under "Proximity fuze" in detail and mentions the autodyne. JB. --92.193.146.62 (talk) 21:29, 17 May 2023 (UTC)