Talk:Audion

Legal battle

I'm pretty sure that DeForest and Armstrong didn't go to court over FM.

They did have the longest legal battle in US history over the regeneration patent, starting in 1915 and ending at the Supreme Court ( RADIO CORP. v. RADIO ENGINEERING LABORATORIES, 293 U.S. 1 (1934) ). It was a win by a technicality of law for De Forrest. It was in this case that Armstrong's lawer showed that DeForrest couldn't explain how the Audion oscillated. (see [1] ) — Preceding unsigned comment added by 208.59.112.198 (talk • contribs) 04:41, December 9, 2004‎

Differences

The Audion is s discrete variety of vacuum tube device. By the same standard as the suggestion of merging the Audion and Vacuum Tube articles, could you not also include the conventional Incandescent Light Bulb as well? — Preceding unsigned comment added by Zorbatron (talk • contribs) 22:38, June 17, 2006

Well no! A light bulb just provides light! A triode provides amplification of electronic signals. I think they are quite distinct-- dont you?--Light current 22:55, 17 June 2006 (UTC)

Actually , it provides more heat than light. A bit like some WP talk pages! 8-)--Light current 15:07, 23 June 2006 (UTC)

If you were to merge all the articles that are this closely related into the tube article it would be "a mile" long. Perhaps this could be part of an article on deforest himself? The article SHOULD be referenced from the Vaccuum tube article..

cmacd 15:53, 22 June 2006 (UTC)

I don't think this should be merged with "vacuum tube" - for one thing, Audions initially weren't supposed to be under vacuum. The history of this particular device I think makes a sensible stand-alone article that can be linked back to vacuum tube as needed. --Wtshymanski 17:37, 22 June 2006 (UTC)

But the term 'vacuum tube' (or 'valve' as we say over here), covers both hard vacuum and gas filled envelopes does it not?--Light current 15:32, 23 June 2006 (UTC)

Actually Some folks do talk of "electron tubes" with the "class" including sealed devices which depend on interactions of electrons, thus including items such as Glow lamps, Phototubes, Tyratrons, Mercury vapor rectifiers and other tubes which are Not designed with a hard vacuum. On the other hand, vacuum tube is a common term. The brits use thermoonic valve whih of course also excludes Phototubes and cold cathode devices. (The 0Z4 does not have a hot cathode at least when it starts up)

I suppose that all Vacuum tubes tend to be filled with gas at less than 1 Atmosphere, (even the hardest tube has SOME gas)

I am still wanting to know WHY folks think this article is not acceptable here, with the links form the other related subjects. cmacd 16:20, 23 June 2006 (UTC)

Its because Audions perform the same function as modern vacuum triodes andn were their precursor. They were the first vacuum tube. THe fact that they didnt always work right because all the gas couldnt be evacuated is beside the point. But Audions were not intended to be gas filled tubes - were they?--Light current 17:47, 23 June 2006 (UTC)

Actually the folks we under the impression that the gas was a good thing, and so they delibratly made them a bit gassy. The circuits they used did not work if a "hard" tube was used.

See in the article: "The problem was that De Forest's original patents specified that low-pressure gas inside the Audion was essential to its operation (Audion being a contraction of "Audio-Ion"), and in fact early Audions had severe reliability problems due to this gas being absorbed by the metal electrodes."

I agree that this is not clear at the moment, I may have to pull out my copy of Tyne to get more detaisls to add.

If you look at one of DeForest's orignal schematics, it shows the antenna connection capacitively coupled to the grid with no provision for grid bias. This was apparently OK when there was ionic conducion, since the extreme non-linearity is what produced the detection process. As the gas was absorbed though, the device would turn into a linear amplifier, which would mean the radio signals would no longer be audible in the headphones, (although they'd still be present). Also, with no grid leak, a large negative bias would quickly build up on the grid, eventually cutting off the plate current. While the grid leak seems a simple concept to us NOW, the necessary 1 megohm or so resistors would have been very thin on the ground in those days. The important thing is, while the Audion is superficially similar to a vacuum triode, they are really a different device. The biggest difference is that they were not capable of linear amplification, and for that reason, the Audion should have an entry of its own--Elekas 12:51, 17 July 2006 (UTC)

even though this is a major step in the evolution of electron tubes, that should not preclude an article on the sub-topic having more details. Otherwise we would have to fold in Cathode ray tube, Electron Microscope, and a host of otehr topics all of which are "part of the development" of the electron tube.

As it stands, the main article mentions the audion, and links here. just like it mentions the octal base.

cmacd 19:02, 23 June 2006 (UTC)

discussion about posible merge..

I see that LightCurrent REALLY wants to merge this in with Vacuum tube..

In favor of this I can see that the audiotron WAS in some ways the first triode, (and probaly the first thyratron but thats another story)

One actually could also move the information to the pages of the inventors..

Against, this is a "more detailed" deeper pass that can be looked up for more inforamation.

I think that it can be filled out MORE, particularly the story about the hard vs soft vacuum issue. By thinking that the audion needed a soft vacuum, it probaly delayed the real vacuum tube by several years.

A sidelight is McChandless (I probaly have his name misspelt) who made many audions under licence, and probaly did much of the early development.

The other thing that I see as Missing is the whole patent situation, this was one of the patents that prevented ANYONE from legaly building a radio commeraly in the US. It was only after RCA and Western electric agreed to split the market, with RCA having radio, and western having Toll wire communications, and pooling the patents both held/controled that making a radio was posible. This also held up radio for another five years.

Thoughts on where this could go?

cmacd 12:36, 23 June 2006 (UTC)

It seems to me from what's said above that just from a historical point of view, the Audion is unique and notable enough to merit its own article. And if it is to be merged, I think the appropriate article to merge it into would be Triode, not Vacuum tube. --Chetvorno^TALK 13:36, 7 December 2009 (UTC)

Should Audion be called the first triode?

I think the Audion should be defined as "the first triode" rather than "...the forerunner of the triode". In spite of the above-noted differences from later triodes (the gas in the tube), and its slight differences in operation (its nonlinearity, and not needing a grid leak resistor), its principle of operation was that of the triode: the grid voltage controlled the plate current. DeForest invented the triode and it does him a disservice to equivocate about that. --Chetvorno^TALK 14:00, 7 December 2009 (UTC)

The audion should not have been described as a vaccuum tube as it was not. Also, to clarify, I have changed the claim to "...the forerunner of the vacuum triode". This is clearer, because, as noted, the audion was a triode, in that it had three (tri-) electrodes (-ode) but not a vacuum device. The modus operandi of the audion was such that a vacuum would stop it working (as noted in the article). I was going to change the reference to 'grid power' to 'grid voltage' but decided against it, because unlike the vacuum triode, the audion grid voltage caused grid current to flow (for both positive and negative grid voltages) so it really did use power (though a very small quantity of it). I B Wright (talk) 15:34, 14 December 2012 (UTC)

A vacuum made the tube linear, so it only stopped it from functioning as a radio detector, it didn't stop it from amplifying. Both the audion and the later vacuum triodes amplify by the same technique; the grid controls the current through the tube. You won't discover this from De Forest's writings, because De Forest, a technologically uneducated man, didn't realize that it could amplify(!!!). His patent application only covered use as a detector 1. Amplification was the important function, and the audion was soon applied to other applications which did not require detection, such as telephone repeaters 2. The audion was the first three-terminal tube which could amplify, so it was the first triode, as stated in numerous sources: 1, 3, 4, p.9, 5, p.77,6, p.7, 7, p.100 --Chetvorno^TALK 14:50, 2 January 2013 (UTC)

I agree with Chetvorno. Sources show the audion was the first triode despite De Forest dubious history and poor understanding. There were a lot of initial audions -- one even had a solenoid outside of the tube.

The article could use lots of references (e.g., Langmuir).

Glrx (talk) 06:41, 5 January 2013 (UTC)

I B Wright says he is doing a rewrite, so I have been sort of waiting for that. --Chetvorno^TALK 07:57, 5 January 2013 (UTC)

The problem here is that almost anyone can turn up references that state that De Forest invented the first vacuum triode (I can even turn up a dozen or so myself). The principle issue is that, visually at least, a De Forest audion is indistinguishable from a true vacuum triode. But although De Forest's device was indeed a triode, it most certainly did not contain a vacuum. The presence of the low pressure gas (actually air in De Forest's device) completely changes the way in which the device works. Most sources that make the claim (including those kindly provided by Chetvorno above) have not appreciated the important difference that the gas makes. I am not rewriting the article, but am working on a section describing how the audion works (something many Wikipedia articles lack). It is taking a bit of time, because the source material (training notes for a related device) are having to be shortened as they are far too comprehensive for the intended task. My aim is to give a simplified account that the average reader can understand, though some of the finer points will not necessarily be explined due to their complexity.

I have a long expertise in vacuum tube technology. Many may be surprised to discover that although the audion has long been considered obsolete, it operating principle did survive in a more modern device until comparatively recently. As far as I am aware, I am the only person in the world who knows how to determine one of the operational parameters of both the audion and its more modern derivative. Have patience. I B Wright (talk) 17:15, 6 January 2013 (UTC)

I B Wright, the reference you added to the introduction, Armstrong's Operating features of the Audion (the link is broken, BTW) demonstrates just the opposite of what you are using it to support. Armstrong's paper was the first correct explanation of the amplifying principle used in common by the Audion, triode, and other vacuum tubes; the grid voltage controls the current of electrons between the filament and plate.1 The gas wasn't necessary as De Forest thought.

But in calling the Audion a triode, it doesn't matter what the patent or Armstrong's paper said. WP's standard is what contemporary RSs say. The modern engineering community regards the Audion as the first triode 2, 3, 4, 5. I respect your research; the article is greatly improved by your addition of the differences between the Audion and later triodes. That's the important thing, the name is just a matter of convention. --Chetvorno^TALK 17:34, 6 January 2013 (UTC)

You are thinking as every one else does: that the grid voltage controls the anode current (which it does in a true vacuum triode). I appreciate that the device is so obsolete, that there are few refernces as to how it actually worked and so in the absence of information to the contrary, most people just assume that it works the same way as a vacuum tube. The interactions are much more complex in the audion because of the presence of the gas. In an audion, it is the grid current that modulates the anode current (there is no grid current in a correctly biased vacuum tube - and no grid bias in an audion). I suggest a little more patience and await the description of the modus operandi. I have the first draft (actually just printed), but it isn't quite right as there are some aspects that are not clear enough and it is not the only item on my agenda. I am however, very gratified to see someone else taking such a strong interest in the old technology - kind of makes the effort worthwhile don't you think? I B Wright (talk) 19:07, 6 January 2013 (UTC)

Untitled

I'm changing the wording of the sentence, "till the 21st Century, in every television set worldwide as its CRT display" to something less broad. It appears whomever wrote that forgot about the existance of LCD and plasma display-based sets. — Preceding unsigned comment added by 97.115.33.102 (talk) 02:15, 8 July 2012 (UTC)

Requested move

The following discussion is an archived discussion of a requested move. Please do not modify it. Subsequent comments should be made in a new section on the talk page. Editors desiring to contest the closing decision should consider a move review. No further edits should be made to this section.

The result of the move request was: page moved. Binksternet (talk) 03:42, 9 September 2012 (UTC)

---

Audion tube → Audion – "Audion tube" sounds redundant. An audion is a vacuum tube, so it sounds like saying "vacuum tube tube". Google search counts are not authority, but "audion tube" gets 30k and just "audion" gets 281k. Glrx (talk) 00:15, 7 September 2012 (UTC)

Support Makes sense to me. --Wingtipvorte^X PTT ∅ 14:25, 7 September 2012 (UTC)

There are a couple of other "Audions" on WP so it will have to be named Audion (vacuum tube) or something. It might be easier to leave it as it is. --Chetvorno^TALK 15:11, 7 September 2012 (UTC)

No. This article already has "Audion" staked out: currently "Audion" redirects to "Audion tube". Audion (disambiguation) shows other topics, and all of them are qualified. Glrx (talk) 16:09, 7 September 2012 (UTC)

Sorry, you're right. --Chetvorno^TALK 16:15, 7 September 2012 (UTC)

• Support – audion is the common name; never heard it called audion tube. Dicklyon (talk) 03:55, 8 September 2012 (UTC)
• Support. This topic is the first and most prominent Audion, never "Audio tube". Binksternet (talk) 14:57, 8 September 2012 (UTC)
• Support --Chetvorno^TALK 15:08, 8 September 2012 (UTC)
• Support - Seems common enough. Marcus Qwertyus (talk)

The above discussion is preserved as an archive of a requested move. Please do not modify it. Subsequent comments should be made in a new section on this talk page or in a move review. No further edits should be made to this section.

AES presentation by Mike Adams, SJSU

This historical session at the Audio Engineering Society convention was very informative. Adams should be used as a source.

• Adams, Mike (2012). Lee de Forest: King of Radio, Television, and Film. Springer. ISBN 1461404177.
• http://books.google.com/books?id=Y_NMWZFVgCsC
• http://www.springer.com/about+springer/media/pressreleases?SGWID=0-11002-6-1321821-0

Licensed ham radio enthusiast and San Francisco lawyer Bartholomew "Bart" Lee (K6VK, xKV6LEE) wrote the Foreword to the Adams book. Lee says the "Audion" name came from "audible ions". Whether that is true or not I think it would be helpful for this article to draw from new source material. Binksternet (talk) 20:15, 6 January 2013 (UTC)

Lede is no longer a lede

The lede section of the article has undergone a rewrite with the inclusion of many much needed references (no complaints there!). Unfortunately, the lede is no longer a lede as such but an article section in its own right. Also the lede now contains all but one of the article's references.

The lede is supposed to be a summary of the main content of the article. As such, it should not contain any references at all because these should all be in the main article. At present, the lede contains much information that is not mentioned elsewhere. It therefore fails as a summary. The new information needs to be moved to an appropriate section of the article along with its references, and the lede reduced until it is a summary of the article content. I B Wright (talk) 13:46, 9 January 2013 (UTC)

The lede doesn't have to contain citations but it can, there's nothing wrong with it (WP:LEADCITE). The lede contains most of the citations because the existing article was so poorly cited. We can add citations to the article, but there is no reason to remove them from the lede. Similarly, most of the points from the lede are discussed in the article, and the few that aren't can be added. Each of the points in the lede is a notable part of the audion story, and at minimum the article should cover them. If it doesn't it is because it is so unevenly written. I'd be glad to add them. I thought you wanted to improve the article? --Chetvorno^TALK 18:25, 9 January 2013 (UTC)

Some writers compose an opening statement before writing the main article text. I think that the encyclopedia article, following Wikipedia's house style guideline at WP:LEAD pretty much recommends first writing the article body, then going back and summarizing the body in the lead section. Expanding the lead section first is not wrong, specifically, and it should not be reverted, but a strong effort should be made to improve the article body such that the lead section is a summary of the body. Along the way, I expect that emphasis on this or that point in the article body may indicate some changes to the lead section, to reflect the emphasis. Binksternet (talk) 19:04, 9 January 2013 (UTC)

The previous lede was very incomplete and had a few blatantly wrong statements, so I felt there was an urgent need to rewrite it. While I was at it I thought I might as well make it a good summary of the topic. The article does at least mention most of the points in the lede, but I'll work on improving it. --Chetvorno^TALK 19:35, 9 January 2013 (UTC)

The audion modus operandi

You are probably aware that I am working on a section for this article describing how the audion worked and how it differed from the vacuum triode. My efforts are at approaching a point where it can be added, but I have a few problems with which other editors may be able to assist.

The first is that the interactions in the device are sufficiently complex that any description of operation is by necessity quite long. I have managed to distil it down to around four pages (I can describe the MO of the vacuum triode in one paragraph). Any shorter requires leaving important information out and would only result in readers adding and tags throughout it. I am not entirely convinced that they are not going to do that anyway. My problem here is: do I add this as a rather long new section to the existing article or do I create a new article for those sufficiently interested to want to know how it works?

My second problem is the ever present problem of references. The original material from which the précis has been made was produced around 30 years ago. It was produced in a professional capacity and was drawn from a number of reliable sources including much provided by scientists of the United Kingdom Atomic Energy Authority and the now defunct Marconi-Osram valve company. Unfortunately, all those sources are no longer available to me, so I cannot quote chapter and verse. The strict rules of citation require citations for any material likely to be challenged. I could let my modesty prevail and assume that few other editors have a sufficient understanding (or indeed any understanding) of the technology to challenge anything. The difficulty here is that the audion has been unknown to mainstream electronic engineers for well over a century, and with that knowledge gap, most engineers do now assume that it is just a common or garden triode (which it certainly is not).

I do not like to see information on obsolete technologies disappear so that no one can find out how something worked. Wikipedia is the obvious vehicle for carrying such articles. My personal view is that I would like to see a tag that can be placed at the head of such articles identifying that the article describes obsolete technology for which no contemporary citations can be found. The process of peer review and improvement (vandals permitting) should ensure a good article. There are quite a few such articles on Wikipedia, but they occasionally fall prey to editors who insist on citations for every full stop. One good example is Pentagrid converter. This article has been around for some years, but inevitably some editor has, in recent times, added the {{refimprove}} tag at the top even though contemporary and easily verifiable references are unlikely to be easily forthcoming.

How does everyone feel about this? I B Wright (talk) 12:50, 10 January 2013 (UTC)

First: a very long section may be too much for this encyclopedia article, or it may be appropriate. I cannot tell without seeing it. Certainly if the article was shepherded to Featured Article status, it would require a strong section discussing how the device worked.

Second: Adding text without cites is asking for trouble. If you can find even the article titles but without the page numbers you'll have a better chance. Another possibility is for you to write what you know to be true and then look for the most likely facts to be challenged and verify them from other sources. When I'm writing about a subject new to me I write the text while I have the sources open in front of me. On the other hand, when I'm writing about topics in my area of professional expertise (audio engineering) I do what I recommended to you—I write what I know and then search for support. It is an iterative operation, too, with article text modified by the references that are found. Binksternet (talk) 16:20, 10 January 2013 (UTC)

The material was word-of-mouth from these scientists, I B Wright? Are you sure it isn't available from a published source? The audion was a pretty important device, and plasma physics and gas-filled tubes like thyratrons have been an active area of research since that time; I would think the original gas-filled audion would be pretty well-analyzed by now. I assume you've searched Google Books; it has interesting historical stuff like John Ambrose Fleming (1919) The thermionic valve and its developments in radio-telegraphy and telephony, Hendrik van der Bijl (1920) The thermionic vacuum tube and its applications, Owen Richardson (1916) The emission of electricity from hot bodies. Along the lines of what Binksternet said, I often write the article, then google phrases from sentences I need support for, to try to find sources.

As I'm sure you're aware, there is an enormous amount of technical information that is available but hard to get: articles in journals that are available online, but require a subscription; journals that have ceased publication and are only available in paper form, in-house technical organs of companies like Osram, PhD theses, out of print books. This is kind of drastic, but if you live near a university, for a nominal fee you can often get privileges at their technical library. They will have online subscriptions to most major journals and through their online system you can search for historical articles and download them. The librarian can get copies of paper journals they don't have through interlibrary loans. I'm not suggesting you do this, but I'd bet money the audion analysis is out there. I think your explanation is going to be a valuable, much needed addition to the article, but like Binksternet I think that without sources it's problematic. I'll look around and see if I can find any sources that might be of use to you. --Chetvorno^TALK 20:09, 10 January 2013 (UTC)

Thanks for the feedback guys. Also the list of references below here. Unfortunately, I don't live particualrly near a university so access to their material would be problematic. I'm not going to be around again until sometime next week. What I have decided to do is to upload the magnus opus to my sandbox (hopefully, I can fit this in later today). Once there, you can have a look and see what you think. Discussion on it is probably best carried out here. I B Wright (talk) 12:55, 11 January 2013 (UTC)

Possible references

I am moving to this talk page a list of "External links" added by Glrx:

• Okamura, Sōgo, ed. (1994), History of the Electron Tube, Tokyo: Ohmsha, Ltd., pp. 97–103, ISBN 90-5199-145-2; p. 98 terms audion as "soft-vacuum tube"; operation is unstable.
• Hawkins, Laurence Ashley (August 1955), "Research and Dr. Langmuir", American Heritage, 6 (5), American Heritage, The current belief at the time of Alexanderson's visit to Langmuir was that the residual gas had something to do with the emission of electrons from the hot filament. But Langmuir made this study and found that that was not so, that electrons came out from the hot filament just as freely in the highest possible vacuum as when there was some residual gas. The reason for the audion's stopping to function if the vacuum became too good was that in high vacuum, the first electrons coming out from a filament, being negative charges, built up a negative field around the filament and repelled other electrons that were trying to leave the filament.; also, "in X-ray parlance, a soft tube".
• "Irving Langmuir–Creator of the Super-Tube", Popular Science: 63, October, 1922 {{citation}}: Check date values in: |date= (help)
• Brittain, James E. (May 1984), "Langmuir on the theory and application of the vacuum tube", Proceedings of the IEEE, 72 (5): 612; introduction by James E. Brittain of Irving Lanmuir's 1915 IRE paper.
• Langmuir, Irving (May 1984) [1915], "The pure electron discharge and its applications in radio telegraphy and telephony", Proceedings of the IEEE, 72 (5): 613–624, doi:10.1109/PROC.1984.12900, ISSN 0018-9219
• Langmuir, Irving (September 1915), "The pure electron discharge and its applications in radio telegraphy and telephony", Proceedings of the Institute of Radio Engineers, 3 (3), Institute of Radio Engineers: 261–293
• Langmuir, Irving (1915), "The pure electron discharge and its applications in radio telegraphy and telephony", General Electric Review, 18, General Electric: 327–339

"However, the operation of the audion is in many ways quite different from that of the pure electron device operation in the way I have described above."

"In the audion, as well as in the Lieben-Reisz relay, the amplifying action appears to be largely dependent on gas ionization, even when the device operates well below the point at which the blue glow appears. The action is probably somewhat as follows: there is normally present a small amount of gas ionization, due to the passage of the electrons between cathode and anode. The presentce of the positive ions partly neutralizes the space charge which limits the current flowing between cathode and anode. The presence of the positive ions partly neutralizes the space charge which limits the current flowing between the electrodes. If a small positive potential is applied to the grid the velocity of the electrons passing by it is somewhat increased and they therefore produce more ions in the gas. Besides this, as the potential on the grid is increased, the number of electrons passing the grid is increased, and this again tends to increase the amount of ionization. A very slight increase in the amount of ionization brought about in this way very greatly reduces the space charge and therefore largely increases the current that can flow between the electrodes. Thus, with a given construction of grid, filament, and plate, the relaying action may be very greatly increased beyond that which would occur if no gas was present."

• http://www.computermuseum.li/Testpage/Evolution-of-Vacuum-Tubes.htm
• DeForest Radio Co. v. General Electric Co., 283 U.S. 664 (May 25, 1931). "Of critical importance in the present controversy is the effect of the presence of gas within the tube." "In consequence, the low vacuum tube is more sensitive both as a detector and as an amplifier than a tube of high vacuum." "August 20, 1912, the earliest date claimed for Langmuir, was rejected rightly, we think, by the District Court, which held that Langmuir was anticipated by Arnold in November, 1912. But before the earlier date, De Forest sought and obtained a high vacuum in the audions used as amplifiers, and observed that when the vacuum was too low the blue glow effect occurred at from 15 to 20 volts. In order to secure higher voltages from the audions used as amplifiers and to procure the requisite high vacuum, he had some of the bulbs re-exhausted while superheated. By August 1912, the Telegraph Company used De Forest amplifying audions at 54 volts, and by November, they were used by another at 67 1/2 volts. This was possible only because the tubes had thus been exhausted of gas, which would otherwise have ionized with blue glow at from 20 to 30 volts."

These are not really external links but possible references. Some of the primary sources should only be used for a specific fact, if needed. Later studies will be preferable. Binksternet (talk) 22:45, 10 January 2013 (UTC)

How the audion works

My attempt at a new article section or even an article in itself is available for preview here. I suggest discussion takes place here. I B Wright (talk) 13:19, 11 January 2013 (UTC)

That's fascinating. I see what you meant about grid current; with gas in the tube the grid will draw current due to positive ions when it has a negative potential. Here's some questions I had:

• You say "The audion was a medium impedance device (impedances were of the order of a few kilohms). A radio detector of even that medium impedance was unknown prior to the invention of the audion." Didn't crystal detectors have AC impedances of at least a few kilohms? And the Fleming diode: it didn't have that much impedance at these low signal levels?
• I think something should be said about the "avalanche breakdown" or "ionization" mode, since sources say it was a problem in the audion, reducing the dynamic range over which it could be used as an amplifier. In that connection it also might be mentioned that gas-filled triodes like the thyratron are a descendant of the audion, although they are mostly used as switches and not in the amplifying mode. BTW, that article also mentions there were triodes in the 1920s that continued using the audion technique, incorporating a small amount of gas to increase their sensitivity as a radio detector.
• The 1915 Irving Langmuir article , "The pure electron discharge and its applications in radio telegraphy and telephony" that Glrx found has a short description of the audion operation (p.334) that you might use as a source for some of it. I'm sure he published a more extensive analysis somewhere. By the way, Langmuir mentions the effect of positive ions in canceling some of the space charge due to the electrons, he seemed to think that this is responsible for a good deal of the increased plate current. I didn't see anything about that in your article. Was Langmuir wrong about that, or was I just not paying attention?

--Chetvorno^TALK 20:48, 11 January 2013 (UTC)

First, thanks for posting the draft. Writing takes a lot of work.

The article has a strong WP:POV that tries to prove that an audion is not a vacuum tube. WP does not allow editors to make value judgments; WP would need to quote a source for that statement.

In addition, the proof has a poor foundation. For example, starting with the assumption that an audion operates without any bias on its grid, therefore it must have a positive ion current to discharge the grid, therefore it must have residual gas, therefore it cannot be a vacuum tube. Evidence negates the assumption. Armstrong's paper discusses using an audion with grid bias. An audion is the device, and the device might be used several ways. The device can also be imperfect: leakage, blue-glow of death, and low input impedance. In addition, there is the US Supreme court decision that De Forest made audions with a high vacuum around 1912 -- before Langmuir. Unfortunately, Armstrong tells us what the grid voltage is but not the anode voltage.

MIT (1943, p. 155) discusses "Characteristics of high-vacuum thermionic diodes". The text has the residual gas having a negligible effect. The text states that "audion" and "kenotron" are used for the diode. (De Forest 1906 2-electrode audion patent.) It also states that "The names audion and valve are also applied to thermionic tubes with grids." The discussion was in the high-vacuum context.

The court decision also addresses baking the envelopes to attain the high vacuum. Outgassing is problem in vacuum work. The article has the glass working as a getter rather than a source. There needs to be a reference.

There are many references that mention blue glow and soft rather than hard vacuum. That does not mean that all audions were soft vacuum.

In effect, the article criticizes using the audion in a circuit without grid bias. The circuit works if there is a soft vacuum, but it is unstable in the long run. (IIRC, the US Supreme Court decision said the unstable mode is avoided.) The article then goes on to explain how the later hard vacuum tubes could exploit their residual gas. Certainly the 6BS7 is a vacuum tube; operating it a particular residual-vacuum mode doesn't kick the 6BS7 out of the vacuum tube class. To put it another way, Armstrong's paper suggests a 1915 audion could be used as either a vacuum tube or a residual gas tube. It's very possible that De Forest did not understand grid bias or how the audion actually worked, but that doesn't mean his invention was not a vacuum tube.

De Forest's 1908 patent describes tubes "comprising in their construction a gaseous medium", but claims 1 through 21 describe an "evacuated vessel"; only claim 7 includes "a sensitive conducting gaseous medium". The patent does not teach about precise pressure levels or explain that too hard a vacuum causes the device to fail. (Langmuir's paper shows that many thought residual gas was needed.) The patent declines to describe any theory of operation, but it explains that isolating the grid with a capacitor increased the sensitivity. (line 102) If nothing else, De Forest accidentally discovered a better biasing method for the tubes he had at hand. (Yes, patent figure 1 may fail if the tube has a hard vacuum and the cap C′′ has low leakage.)

I'm suspicious of the anode explanation. With 22 volts, there will be a tough time kicking electrons free (electrons with more than (2X) the ionization potential). If the vacuum is poor, the mean free path will be small, and few electrons will have the energy to ionize. If the vacuum is good (long MFP), then some electrons will have the energy, but it is unlikely they will hit any molecules before reaching the anode. Consequently, the positive ion current component will be small and arise from the tail of the distribution. There seems to be little opportunity for Townsend avalanche gains (freed electron freeing a second electron). Small positive ion currents have a huge effect on the space charge. MIT (1943, pp. 205–208) ("Effect of gas in a thermionic diode") states that "the number of positive ions produced by the electrons is not more than a few percent of the the number of the electrons". Page 208 states:

If a high-vacuum tube is desired in which gas shall play no appreciable part in the over-all behavior, the pressure must be reduced until the number of electrons that make ionizing collisions in crossing the inter-electrode space is a very small fraction of the total number of electrons that cross. Because of the very great effect of positive ions upon the space-charge limitation of current, the fractional increase in the space-charge-limited current caused by ionization is much larger than the fraction of the electrons that ionize. To reduce the effect of the gas on the space-charge-limited current to, say, one per cent, it is therefore necessary to reduce the gas pressure until far fewer than one per cent of the electrons ionize gas particles. The pressure must often be reduced until only one electron in 10,000 or more ionizes; in practical high-vacuum tubes the pressure is frequently of the order of one billionth of an atmosphere.

The name audion coming from audible ions needs a source. I found this book on Google with a limited page view that mentions "audible ions" (previous page not visible), but then gives the fragment De Forest "called his first tube the 'Audion' precisely because it enable [sic] us to hear what was otherwise undetectable and invisible -- radio waves --" (next page not visible). The statement may be true, but I'd expect it to be easy to find sources.

I disagree with some of the differences stated, but that's for another time.

• MIT (1943), Applied Electronics: A First Course in Electronics, Electron Tubes, and Associated Circuits, John Wiley, ASIN B000L3A7EK

Glrx (talk) 18:47, 12 January 2013 (UTC)

I agree that the article has a WP:POV problem in that it seems to be trying to prove that the audion was not a "vacuum tube". The first paragraph that says that the audion should not be called a triode is also a clear WP:POV which is not supported by WP:RSs, as I've mentioned above. The categories into which modern engineering taxonomy has put this device are not ours to change on WP, and are not important anyway. --Chetvorno^TALK 19:44, 12 January 2013 (UTC)

Here's a source for the name "audion" = "audible" + "ions" Hijiya, Lee De Forest, p.77]; it was named by De Forest's assistant. --Chetvorno^TALK 19:44, 12 January 2013 (UTC)

Thank you gentlemen for your feedback and comments. I think that you are both guilty of reading far more into my draft article than is there and certainly making assumptions based on today's somewhat limited knowledge.

My draft was not an attempt at proving anything. It was a factual statement of the modus operandi of the De Forest audion précised from a much larger work drawn from many available sources at the time that that was compiled. It is certainly not my point of view on anything. That the audion contianed low pressure gas was not considered in doubt. First, De Forest's patent on the device specifically stated that it contained low pressure gas. Second, none of De Forest's circuits for the device made any provision for grid bias. It is well established that a true vacuum triode will not work without grid bias (except at anode current too small to be of anything but very limited use), but a device with low pressure gas will.

As far as Armstrong's paper was concerned, it should be remembered that this paper was not a paper on the finer points of the audion operation to any scientific standard. It is also clear from that paper that Armstrong himself did not fully appreciate exactly how it operated (as noted Armstong's theory required grid bias, yet it is known that the original audions had none). Indeed, Armstrong seems to discuss both types of device within the same paper. Armstrong did not originally produce his paper as a scientific discussion, but as part of his evidence trail to satisfy a court that he invented the regeneration technique to counter De Forest's claim that he did not. The aim of the paper was to convince the court that Armstrong understood how the audion worked and De Forest did not (and in all honesty, Armstrong could have said anything as it was unlikely that the judge would notice). The US supreme court's decision that De Forest was making high vacuum tubes was obtained by De Forest purely on a legal technicality. There is no other evidence that he actually did so prior to its emergence elsewhere. It should be remembered that De Forest was extremely well versed in exploiting vested interests and how to fight patent disputes. By contrast, Armstrong most certainly was not, and it would seem that Langmuir was not either.

You express dount about the anode votage of 20 to 25 volts. All De Forests circuits for the audion show an anode voltage supply obtained from a 22.5 volt battery.

My first paragraph does not state that the audion should not be called a triode. Indeed the very first sentence stated, "The audion is a triode device in that it featured three electrodes". What the piece says is that the audion was not called a triode by De Forest, but an audion. There are no contemporary advertisements for a De Forest triode audion, or a De Forest audion triode. This advert from 1916 does not have the word 'triode' in it anywhere (note: also the hollow warning against infingers of his patent - it also fails to reference the triode by name). Even the vacuum triode was not described as a vacuum triode at the time but simply as a 'vacuum tube' - at least in America. It was the rest of the English speaking world that adopted 'triode valve' to distinguish it from Fleming's 'valve' now redesignated 'diode valve'. The term triode only really came to the fore in America once devices with two grids were invented (the tetrode and the French bi-grille - two completely different devices).

The audion was (at the time) considered a high impedance detector, though by today's standards is medium impedance. Earlier crystal detectors were low impedance devices, even at the low forward voltages encountered in the radio detector mode. Such a detector was required to have (what at the time was described as) high resistance headphones in series with it to ensure that the impedance presented to the tuned circuit was not low enough that it would lower the 'Q' of the circuit too much. One work that I have states that the headphone resistance for use wiith a crystal set should be "around 5000 ohms". This was a book published in 1920. It does not say how that resistance was arrived at, but it does not take a genius to work out that such a figure is a trade off between having enough current for audible sound and not lowering the 'Q' of the tuned circuit appreciably. I B Wright (talk) 15:58, 15 January 2013 (UTC)

And I meant to add: that the description was never intended to be a technically comprehensive description of the MO. It would be a lot longer if it was. It is a (vastly) simplified explanation intended to make any subsequent article accesible to the Wikipedia readership. It is a great shame that so much of the knowledge has been lost to mainstream engineers over the century or so since its development, that the operation has become sufficient of a mystery, that it is almost having to be reinvented. I B Wright (talk) —Preceding undated comment added 17:55, 15 January 2013 (UTC)

I'd like to get away from arguing about the names of these tubes to the more substantive issue of how "soft vacuum" tubes differ from "hard vacuum" tubes in operation. The fact is the issue of gas in the tube was not understood by most of the engineers of the time (excluding De Forest, Armstrong, and Langmuir), and no distinction was made back then. As far as I can tell, in the 1913-1923 era some radio books referred to all three-electrode amplifying tubes as "audions", including the hard vacuum "pliotrons" from GE, and some referred to them as "triodes", including De Forest's audion. In addition, as you have both admitted, De Forest's "soft vacuum" audion could "harden" by gas adsorption during use. Nowadays the audion is considered the first triode, and is sometimes called a vacuum tube. We can simply explain this in the article, without taking a position. We can explain that, for the purpose of the article, the term "audion" refers to a soft vacuum tube, and "vacuum triode" refers to a hard vacuum tube. --Chetvorno^TALK 20:53, 15 January 2013 (UTC)

Knowing quite a bit about vacuum tube technology, I found the description of how the audion worked interesting (even if I did have to read it two or three times to fully understand it). The article really does need a description of how the device worked. There seems to be some discussion above as to whether De Forest really produced a soft (low pressure gas filled) tube or really produced a vacuum tube. This is easily resolved because De Forest could not have produced a high vacuum triode even if had wanted to simply because the technology to produce the high vacuum required did not exist in 1906. It is now known that operation of the gas filled tube would slowly convert it into a high vacuum tube, but that was unknown to De Forest at the time that he conceived the device. Fleming's diode 'vacuum tube' would also have been a low pressure gas filled device for the same reason, but in the case of that device, I assume that there was no visible external difference as both a gas filled and vacuum diode will rectify. DieSwartzPunkt (talk) 17:28, 2 February 2013 (UTC)

Is there a reference for a poor vacuum improving over time? My impression is that a hard vacuum gets worse over time unless particular steps are taken: the vessel must be baked (encourage outgassing now rather than over time); modern vacuum tubes also employ a getter to both improve the initial vacuum and maintain the vacuum over time. Glrx (talk) 00:13, 15 February 2013 (UTC)

It has long been known that gassy devices develop a higher vacuum during operation. The gas is adsorbed by the electrodes and the glass envelope. By contrast, high vacuum devices will become gassy due to leakage through the matal to glass seals. This was a particular problem in early cold cathode X-ray tubes with their large electrodes and glass surface area. Users of the lower power and cheaper tubes of the type you could buy from your local radio or even toy store^* had to suffer the reduction in output as the tube aged, much like today's users of DVD and Blu-ray drives put up with a short life due to the laser(s) output reducing as they age.

Larger X-ray tubes had a regassing device as part of the glass work. It was a small side module containing mica. As the tube hardened, the rising anode voltage caused a flashover in the mica module automatically restoring gas pressure. A photograph of such an auto regulating tube can be found here (about half way down). Without a getter, or any gas control device, any tube whether low pressure or vacuum will eventually settle on an internal pressure where the adsorption balances the leakage but it will not be at a high vacuum.

^* I'm not kidding! - I even have a book published in 1920 that shows a schoolboy how to build an X-ray machine for experimentation but if you wanted the easy approach, more than one toy manufacturer made X-ray kits with spare tubes being easily available. I B Wright (talk) 17:50, 24 February 2013 (UTC)

Edit 4 Mar 2013

I'm not sure what it is with some people, but this is another fairly brief Wikipedia article which is completely dwarfed by the size of its talk page :-). As with so many other important technological developments whose potential was vastly underestimated at the time, in the light of the unprecedented research facility afforded by the Internet, the accepted "history" has been exposed as "mostly bunk", based on uneducated guesswork, and "house-of-mirrors" pieces by lazy journalists.

Until recently Google News used to have an "Archives" section, where you could look at newspapers dating back the the 19th century. A lot of them were "Pay-per-view", but there were quite a few free ones as well, and in the 1920s most newspapers had "Wireless" sections of one sort or another. It is sometimes quite staggering the difference between "accepted" reality, and reports written at the time.

I found the real story of Kenotrons and Pliotrons in a 1919 newspaper, and I was intending to link to it on this page, but unfortunately, Google no longer supports that function.

This article is also starting to suffer from "Wiki-rot" where endless rounds of pointless editing have completely erased the original meaning of some sentences. Not replaced them with something different but still valid, but replaced them with gibberish in some cases. Elekas (talk) 02:03, 4 March 2013 (UTC)

Lede

I've corrected the first paragtraph (again!) Somebody appears to have taken it upon themselves to summarize the rest of the page, without actually reading it! The Triode Audion was not invented in 1906, and it never produced any linear amplification until around 1912.Elekas (talk) 03:35, 12 July 2013 (UTC)

I'd certainly agree that the body of the article needs an overhaul, but I'm concerned about the removal of the thoroughly sourced statement that the Audion was the first triode from the introduction. I think this is important to help nontechnical people understand the continuity of development of vacuum tubes. What is your rationale for removing it? --Chetvorno^TALK 03:54, 12 July 2013 (UTC)

Reverted above unsourced changes. Corrected the statement claiming the Audion was the first amplifying tube; the negative resistance mercury lamp amplifiers invented by Ernst Ruhmer and Adolf Pieper preceded it. --Chetvorno^TALK 16:04, 13 July 2013 (UTC)

"but I'm concerned about the removal of the thoroughly sourced statement that the Audion was the first triode from the introduction."

Thoroughly sourced or not, most of the "history" of this subject cosnsits of de Forrest's propaganda, all dating AFTER the fact. The Pliotron blew all his Poulsen Arc radiotelephone systems clean out of the water, and he was clearly desperate to hang onto his Navy contracts.

The main points:

1. The 1906 Audion was not a Triode

2. The vacuum triode works on a completely different principle from the "triode" Audion

3. If you read Langmuir's original paper, you will see that it would have been impossible for de Forrest to have ever produced a working linear amplifier.

4. Langmuir wasn't even trying to produce a linear amplifier, he was merely trying to stabilize the action of the Audion

— Preceding unsigned comment added by Elekas (talk • contribs) 02:09, 22 October 2013 (UTC)

1. Reading contemporary sources, I don't see that the term "triode" was ever used exclusively to distinguish "high-vacuum" three-element amplifying tubes, as you are using it. Some of the early sources used the term "Audion" generically to refer to all three-element tubes, including the high-vacuum GE "Pilotrons", while some used the term "triode". When they wanted to draw a distinction between gas-filled and high-vacuum tubes, the terms "soft" tube and "hard" tube were usually used. The modern descendant of the Audion, the thyratron, is referred to as a "gas triode".

2. From a modern POV, the basic principle of operation of the Audion is used in the triode and all modern amplifying tubes. The crucial point is that the voltage on the grid controlled the filament-plate current, enabling it to amplify. From a history-of-technology POV the invention of amplification is what is important; Langmuir making the tube linear was a refinement.

3. Regardless, the article needs to follow current usage.

I absolutely agree with you that De Forest, a sleazeball self-promoter, was not a reliable source on his own invention. By the way, you can sign your posts by typing ~~~~. Cheers. Chetvorno^TALK 10:09, 22 October 2013 (UTC)

I don't see the sources as all De Forest propaganda. I see a heavy bias against De Forest that includes misstatements (e.g., "The problem was that (possibly to distance his invention from the Fleming valve) De Forest's original patents specified that low-pressure gas inside the Audion was essential to its operation..."; De Forest's triode patent was agnostic. The patent mentions "gaseous medium" (a term that probably is not precise in 1908 when discussing conduction; space charge is an electron gas), but the patent refers to evacuated vessels in the claims. De Forest's 1908 patent is agnostic on how it works: "inasmuch as the explanation of this phenomenon is exceedingly complex and at best would be merely tentative, do not deem it necessary herein to enter into a detailed statement of what I believe to be the probable explanation."). I think it is well settled that De Forest was devious and didn't understand the details, but he did make a working triode.

1. De Forest used "audion" for all his vacuum devices: the 1906 Flemming diode detector ripoff (MIT above), the diode tube with external control solenoid (picture at NYPL?), the triode with the grid on the wrong side of the cathode (1908 patent), the triode with the grid in the right place (1908 patent), and the Federal triode audion with high vacuum (no blue glow at high plate voltage) that defeated the pliotron patent (court decision). I don't know, but I'd guess the ambiguous term audion fell out of favor (MIT still used it for diode/triode in 1940) and the term pliotron never really caught on. Certainly when screen and suppressor grids were added, it probably became easier to distinguish the tubes systematically: diode, triode, tetrode, pentode. The conventional meaning of audion is now a vacuum triode because that was the important discovery. I agree with most of Chetvorno's comments, but disagree with the thyratron comment.

2. Focuses on how De Forest used the tube rather than how it could be used. That's what Armstrong's paper teaches: a better way to use the audion. Armstrong is not clear about his plate voltages.

3. I don't recall that sense from Langmuir's paper and other sources. Langmuir resolved the academic debate whether residual gas was needed for the tube to operate. Some thought residual gas was essential, and others thought not. Langmuir had better vacuum technology than De Forest had in 1908, but that does not mean De Forest and Federal were not pulling the best vacuum that they had available.

Glrx (talk) 22:00, 25 October 2013 (UTC)

Glrx, you seem to have done a lot of research in this area; do you know any source that gives the different models of Audions that De Forest's companies (and his licensees) produced over the years? And particularly which were "soft" and which "hard" tubes? It'd be nice to include a list in the article. I added a bunch of images of early De Forest gear to the article, but as Elekas rightly pointed out, some of them (maybe all) used "hard" tubes. BTW, there's a source that says some Audions had two filaments, I don't know if they had different temp. ratings or if it was so when one burned out there would be a "spare". --Chetvorno^TALK 23:22, 25 October 2013 (UTC)

I don't know of such a source. I'm also leery of the "soft" and "hard" distinction because it doesn't seem to have had such a major impact. The early tubes were soft and had the blue glow issue, but I've not come across contemporary references that really address the issue. The MIT reference describes the residual gas problem and claims that the vacuum should be a billionth of an atmosphere, but it does not trace the issue of improving the vacuum. A vacuum technology website had a nice timeline of vacuum technology, but I haven't looked at it recently. My gut tells me that blue glow was viewed as bad, and higher vacuum solved blue glow without ruining the tube characteristics. De Forest could have improved his vacuum and his audion without understanding the space charge issue. Armstrong's paper also taught everybody to use DC biasing and look at transfer curves.

I have a dim memory of dual filament audions, but do not recall the purpose. I have some residual gas analyzer tubes that have two filaments, and the purpose is to use one until it burns out and then switch to the other. Given the expense of an audion, a dual filament might be reasonable. But I see other reasons, too. Filaments were dicey.

Glrx (talk) 00:47, 27 October 2013 (UTC)

Many thanks. What do you mean when you say the "soft" and "hard" distinction didn't have a major impact? --Chetvorno^TALK 19:31, 27 October 2013 (UTC)

There are a couple modern sources that try to make a big deal of soft v hard to discredit De Forest, but I don't see that emphasis in contemporary sources. How about fairness: is Fleming's valve not a vacuum tube because it didn't have a hard vacuum? NBS 1922(?) uses audion/pliotron for hard vacuum; recounts Armstrong's biasing paper; relegates soft vacuum to a footnote. Morse 1925 (surveying invention/patents), an author with pro-De Forest and pro-Fleming/Marconi ties, did not address the soft/hard issue; he does mentions a dispute about the invention of grid leak biasing between De Forest and Langmuir. (Other sources seem to resolve the issue as if your caps are too good, then add a leak resistor.) Morse also credits De Forest with the B battery for the diode and somebody besides Armstrong for the C battery. It may have been Morse who recounts that it took a long time for audions to catch on, so when they did catch on, they were hard vacuum. Morse also credits De Forest with the metallic filament (tantalum). De Forest is doing all these improvements and he wasn't looking to improve the vacuum? The courts gave De Forest credit for high vacuum. Glrx (talk) 00:13, 30 October 2013 (UTC)

I agree. I can sort of see their point of view about Langmuir; he (and I assume Arnold at Western Electric) was really first to understand what went on in the Audion and transformed vacuum tube technology from art to science, and he didn't use De Forest's patents. But it seems to me what makes a triode is the addition of the grid and the ability to amplify. I don't think the original 1906 Lieben-Reisz tube had a grid. It's ironic that Langmuir, more responsible than anyone for the "hard" triode, doesn't get credit for hard vacuum. --Chetvorno^TALK 05:21, 30 October 2013 (UTC)

There's a lot of confusion. Is Langmuir responsible for hard vacuum tubes? IIRC, the courts determined that de Forest licensee Federal was making hard vacuum tubes before Langmuir and gave de Forest priority. As I understand it, Federal's goal was to avoid the blue glow of death at high plate potentials. The solution would be pretty obvious to an engineer at the time. Langmuir clarifed the theory. Langmuir's paper described competing theories, he did experiments, and he said that one of the theories was right. Langmuir explained the space charge issue, but I don't know if other researchers were already thinking of that. I'd have to look up Child and Richardson. In any event, Langmuir was contributing to an understanding of how the device worked. The thinking at the time was clouded by experience with cold cathode X-ray tubes. Also, in a sense, the hard vacuum is minor improvement. The early tubes had poor emission, poor geometry, and poor field strengths. Armstrong's biasing would also work with a soft vacuum tube. I don't have the reference handy, but I think a de Forest employee wrote a QST article that explained de Forest sold both soft and hard audions. The soft tubes were used for sensitive detectors. Glrx (talk) 19:31, 7 August 2014 (UTC)

Prototype audion image caption

Glrx I think we need to say something more about this image than that it was "an Audion". My original caption is supported by the image source. De Forest says this was the "first tube with the control element in the form of a grid" in the tube, ergo the prototype for his Jan 29, 1907 patent and all triode vacuum tubes that followed. That's why I put it in the article. --Chetvorno^TALK 23:55, 3 August 2014 (UTC)

But it does not read like an intended qualified statement. De Forest played with external control, so there were other "triodes". I don't have time to chase this now.

http://www.oddmix.com/tech/audion_de_forest_triode.html

Glrx (talk) 19:46, 7 August 2014 (UTC)

That's why the caption says "The first prototype Audion with the grid between the filament and plate." BTW, in the same source is a photo of another of his prototype Audions, the "two-plate" tube which appears in the same Jan 29, 1907 patent, in which the third electrode is a second plate on the other side of the filament. That will be good in the article too. --Chetvorno^TALK 22:28, 7 August 2014 (UTC)

My complaint is the restriction can be misread. Glrx (talk) 21:49, 9 August 2014 (UTC)

In what way? Do you mean the word "prototype" might make readers think there were no Audions before this one? --Chetvorno^TALK 01:13, 10 August 2014 (UTC)

Sources

There is a long series of articles in Radio News magazine, available online, by Gerald F. J. Tyne, "The Saga of the Vacuum Tube", which contains a detailed history of the development of the Audion, fully sourced. The articles also include a number of historical pictures not available elsewhere. The copyright has lapsed so the images are public domain. Here are my unedited notes on the contents:

• Part 3, Radio News, May 1943, p. 26 Discovery of the Edison Effect. Contributions of Huff, Bequerel, Guthrie, Edison, Elster & Geitel, Hittorf, Fleming, Howell

• Part 4, Radio News, July 1943, p. 30 The Fleming Valve

• Part 5, Radio News, August 1943, p. 26 The patent dispute between De Forest and Fleming over the Fleming valve, later use of the valve,, Wehnelt rectifier tube, Weagant tube

• Part 6, Radio News, September 1943, p. 26 Description of De Forest's process of inventing the Audion. Assistant Babcock named it “Audion” after Latin audire “to hear” and Greek suffix -ion from ienai “to go”; thus a device to hear electricity in motion. De Forest made reference to Fleming's work but said the Audion worked quite differently; as a “relay”. Shows many of his patent drawings and early audion.

• Part 7, Radio News, October 1943, p. 26 The first Audions sold. Grid type Audion disclosed 3/14/1907. In 1909 to increase conductivity, double grid and plate Audions, called “double Audions” were sold at higher price. In 1913 filament was changed to “Hudson X” design, with tantalum wire wrapped around tungsten. Audions first offered for sale in Sept 1909 issue of “Modern Electrics”, fig 47 is copy of ad. First bulbs were spherical, w/ 2 horseshoe filaments, single grid and plate. The 2nd filament wire was not connected to base but brought out in cotton sleeve. When 1st burned out, 2nd was attached to base w/ rubber band. Avg filament life was 35-100 hrs. Regular “S” grade had tantalum filament. “X” grade cost more. Audions were only sold as part of detector units, containing rheostat & switches. The cheapest, RJ4, cost $18, a fortune for experimenters. Experimenters wanted the bulb, they could build their own filament & plate rheostat circuits. When bulb burned out, owner had to send back a grid and plate to get another one. Since later bulbs had two grids & plates, owners of burned out bulbs would send back one set to DeForest and sell the other through ads in magazines to another experimenter to get a bulb. First article on Audion by John V. L. Hogan Jr. De Forest assistant, in October 1908 “Modern Electrics”. He said at that time there were 6 varieties of Audion: (1) flame Audion (2) arc Audion, (3) 2 element U-wing type (4) external electrostatic control type (5) external magnetic control type (6) grid type. Description: 6 v incandescent tantalum filament, 10x15 mm platinum plate 3mm from fil, grid of No. 22 wire halfway between.

De Forest and assts. Tried back then to use it for audio amplification, but were unsuccessful, perhaps because of impedance mismatch between Audion and earphones. Although its sensitivity as detector depended on amplification, it was not used as an amplifier by anyone until 1912, 5 years later. Reasons for limited use of Audion during first years: (1) Reputation of wireless was damaged by stock-jobbing schemes (2) most radio experimenters were boys who couldn't afford the Audion (3) Audion was erratic and little more sensitive than crystal detectors, (4) threat of patent litigation deterred the marine radio industry from using them. In 1912 De Forest was working for Federal, who needed an amplifier for their Poulsen arc receivers. De Forest & assistent H. B. Van Ettin, tried to get it to amplify again, this time using AF transformers to couple earphones to tube. They succeeded July – Aug 1912 in getting real amplification out of a “double” Audion. This was when De Forest observed howling feedback. Through John Stone Stone, he got meeting w/ ATT and sold it to them as repeater.

• Part 8, Radio News, November 1943, p. 26 Development of the Lieben-Reisz tube. Pictures and patent drawings.

• Part 9, Radio News, December 1943, p. 30 De Forest's sale of the rights to Western electric. The Western Electric search for a repeater. Photos of the mercury vapor negative resistance repeater, developed by Arnold in 1912. (31) Pictures of 10 successive prototype Audions show Arnold's progress at improving it. Fig 66 is first high vacuum tube.

• Part 9, Radio News January 1944, p. 38, Part 10, Radio News March 1944 p. 50, Part 11, Radio News April 1944 p. 54] Development of the Western Electric "hard" triodes. Type 101 The life of the tube was increased from 400 hrs to 40,000 hours.

• Part 12, Radio News June 1944 p. 52 De Forest's development of the Audion 1912 – 1919 De Forest's stock fraud trial. Prosecutor Stephenson's remark that the De Forest Co's only assets were patents on the Audion a strange device that had proven to be worthless. Acquitted 12/31/1913. With $90,000 from ATT he began to develop “Oscillions” for xmtrs, also called “singers” on account of oscillations. Type OJ3 Oscillion Telephone copy of aircraft xmtr sold to amateurs 1917. Type S used in electronic musical intrument demonstrated 1915. “Shaw base”. Higher power oscillions: 1500V 250W A competing tubular soft triode detector, the “Audiotron” was sold, which unlike the Audion could be purchased alone. In response a tubular Audion, type T, w/ 1 filament, was brought out at same price as Audiotron. During WW1 all civilian radio was banned, and De Forest made “VT-21 for Signal Corps, Vp = 20V, If = 1.1A, amplification factor 10-12, plate resistance 60k. CF-185 for Navy, first coated filament, tungsten wire grid, tungsten plate, If = 0.85A, life 5000hour. 1914 Marconi sued for infringement of US Fleming patent, De Forest countersued Marconi. Marconi confessed infringement and was enjoined from making Audions. De Forest suit went to trial 9/20/1916, Marconi won and was upheld on appeal, and an injunction restraining De Forest from making Audions was issued. Thus a stalemate: neither co. could produce receiving Audions. De Forest was guaranteed immunity for Audions made for gov't during WW1. Marconi sued US gov't for these, which Supreme Ct decided 1943 that Fleming patent was void.

Stalemate was broken May 1919 when De Forest and Marconi got together and decided O B Moorhead could mfg Audions for De Forest and Marconi would be distributor. First tubes were Moorhead Electron Relay (unbased soft tube for rcvrs) and Moorhead VT amplifier (hard amplifying tube). Relay went to based design using standard “Shaw” base. VT life was claimed to be 1500 hrs. These were only rcving tubes legally available until founding of RCA. Electron relay was replaced type 20, hard tube. De Forest began to produce higher power oscillions, legal since the stalemate only covered rcving tubes. 250W and 500W tubes appeared on market 1920. Fleming patent expired 1922, leaving De Forest free to make rcving tubes. De Forest co. was reorganized. Type H for amateurs operated at 150W input and Vp of 500 - 3000V

• Part 13, Radio News September 1944 p. 46 Langmuir's early development of the Audion at GE. The reason GE was interested was that it needed a modulator for the Alexanderson alternator. By 1913 Alexanderson had made alternators that produced several kW at up to 200kHz. They tried a generator with its field excited by the microphone, magnetic amplifier, and a 3 electrode mercury arc tube, without success. In 1912 Alexanderson heard about the Audion and obtained one and showed it to Langmuir and Coolidge. Although it was low power and they needed a high power amplifier, langmuir though he could improve it. Others had noticed the blackening that accompanied the “blue glow” discharge. Langmuir researched it and decided it was caused by normal evaporation of the filament. Langmuir experimented with incandescent lamp filaments, made tubes w/2 filaments, heated one and measured current to the other. He found that current obeyed Richardson's law at low fil temps, but leveled off at higher temps. He deduced the presence of space charge of electrons which above a given temp, limited the current. By pushing evacuation to the limit he made bulbs which could stand 250V to 500v and found that space current varied as 1.5 power of voltage.

In 1913 he began working on Audion. He made tubes with both grid and plate made of wire srceens, so he could pass a current through the wire during evacuation to heat them and bake out occluded air. He developed a tube that could stand 250V. In connection with this work he suggested to coolidge, who was working on x-ray tubes, that he use a heated cathode to produce electrons. That was the genesis of the Coolidge x-ray tube.

• Part 14, Radio News November 1944 p 46 The invention at GE of the Kenotron, Pliotron, Dynatron, and Magnetron by Langmuir, Dushman, Hull, 1913 – 1921

• Part 16, Radio News March 1945, p. 52 The Moorhead “Electron Relay”, compromise between De Forest and Marconi

That's most of the Audion material. The series goes on to describe development of the the first British and French triodes during WWI by Round and Ferrie. I've read up to Part 21 so far. The series does not appear in consecutive issues of the mag but often skips several issues. --Chetvorno^TALK 21:41, 7 August 2014 (UTC)

"Marconi sued US gov't for these, which Supreme Ct decided 1943 that Fleming patent was void."

The US Supreme Court determined that the Fleming US (American) patent was void. The Supreme Court couldn't void Fleming's extensive British Empire patents - which laid claim to the triode as nothing more than a natural extension of the diode (this supported by de Forest's inability to explain the action of the triode). What is the significance of the Supreme Court's 1943 ruling? Not a lot, I'd say - particularly as solid state appeared not long after.2A01:4B00:AE0E:6200:DC7D:D977:5E79:92BB (talk) 23:27, 17 March 2024 (UTC)

Another source that may have some unique information: Lloyd Espenscheid (July 1959) "Discussion of 'A History of Some Foundations of Radio-Electronic Technology'" Proceedings of the IRE, Vol. 7, No. 7, p. 1253. A lot of detail about the invention of the triode amplifier. Says Fritz Lowenstein built the first working Audion amplifier in 1911 by removing the grid capacitor (grid leak) used in De Forest's grid detector circuit, and adding negative grid bias (the "C" battery) which made the tube operate class A, and applied for a patent, which was delayed. Meanwhile, in 1912 De Forest got the tube operating as an amplifier and AT&T bought the rights in 1913. In 1917 Lowenstein's patent was granted. Because of the existence of the German Lieben-Reisz tube he wasn't granted rights to the amplifier circuit, only the "C" battery. But AT&T's Audion amplifiers used "C" bias, so AT&T was forced to buy Lowenstein's patent for $150,000. Says Lowenstein also invented the first Audion oscillator in 1911. --Chetvorno^TALK 06:46, 24 October 2015 (UTC)

History

The history can be greatly improved by new facts that have come to light in Steven Johnson's book Where good ideas come from, Chapter 11, Riverhead books, 2011. — Preceding unsigned comment added by Garfield Garfield (talk • contribs) 22:19, 12 March 2015

External links modified

Hello fellow Wikipedians,

I have just modified one external link on Audion. 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/20140511182508/http://www.privateline.com:80/TelephoneHistory3/empireoftheair.html to http://www.privateline.com/TelephoneHistory3/empireoftheair.html

When you have finished reviewing my changes, please set the checked parameter below to true or failed to let others know (documentation at {{Sourcecheck}}).

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) 08:10, 21 October 2016 (UTC)

Diode or Triode

@Glrx: It would be useful if you could expand on your reinstatement of the dubious tag which says, "Audion refers to diode and triode".

The first Audions were built with two electrodes (filament and anode) in the glass structure (in 1906). However, they had a wire wrapped around the glass to influence the electron (and ion) stream inside. This wire constitutes a third electrode (mostly) capacitively coupled into the device (in the same way that the trigger electrode is external to a xenon flash tube). The device was therefore a triode before the grid was physically incorporated into the glass structure in around 1908. Is this what you are referring to as the 'diode audion' or are you aware of some true diode audion of which I am not? DocFergus (talk) 13:59, 21 March 2018 (UTC)

No. He copied Fleming's valve and called it an Audion. See Hilya at 75. https://books.google.com/books?id=JYylHhmoNZ4C&pg=PA75&lpg=PA75&source=bl&f=false And yes, he tried a lot of other things such as external electromagnets (which do not supply ions and are not electrodes) and unusual placements of other electrodes within the tube. Think Audion is the class of devices rather than a triode. Glrx (talk) 17:28, 21 March 2018 (UTC)

I disagree. I added 5 references to the lead which say the Audion was the first triode. If we parse the lead to omit the fact that the Audion was an amplifier and the first triode, we omit the most notable features of the invention. This is a common error on WP; expanding the lead definition to cover marginal cases until it is incomprehensible or no longer a meaningful definition. The common meaning of the term "Audion" both then and today is the three electrode grid Audion, with the grid between the filament and plate. It was the only Audion which had any significant use, and the only one which became a commercial product. The two electrode and external electrode versions were just experiments. We should include them in the article, of course, but they do not define what the finished product was. --Chetvorno^TALK 18:16, 21 March 2018 (UTC)

Copying somebody else's device and applying a different name is not a development worthy of inclusion. The device, in both its 1906 guise and the latter 1908(ish?) development of adding an internal third electrode makes all the recognisable developments, triodes (but not a vacuum triode). As Chetvorno states, the version with the grid betwixt the filament and the anode was the only version that was commercially useful. De Forest would have produced a substantial number of experimental devices along the way because, as he famously admitted, he had no idea how the audion worked. The characteristics that are recorded in various patents and papers of the time betray that the device was operationally very different to a true hard vacuum triode. Although I have no idea how the wretched thing worked, I could probably theorise an MO from the known material.

Edwin Armstrong's paper on the device is a start but not very useful. It was not written to a scientific standard but for the benefit of a judge in a vaguely related patent dispute where Armstrong was trying to convince said judge that De Forest did not have the necessary technical knowledge to have invented the disputed technology (and was probably right given that he hadn't). He failed because, De Forest had deeper pockets to pay good lawyers and he was well acquainted with the commercial exploitation of technology (at which Armstrong was relatively naïve) even if he (De Forest) didn't understand it.

And for the benefit of Wtshymanski who keeps on insisting that the audion is a vacuum tube: all the contemporary material of the era, including De Forest's patent and Armstrong's paper clearly show the device to have residual gas inside it. Indeed: the primary failure mode was the gas being adsorbed by the glasswork, grid and anode as it functioned, ironically slowly turning it into a vacuum triode (if only De Forest had the technical knowledge to appreciate what was going on!). Without the residual gas, the audion would not work because the grid had no independent negative bias on it which is essential in a true vacuum tube (None of De Forest's circuits using audions show any bias), which means that as the gas pressure decreased, the grid voltage would become more and more negative eventually cutting the device off. I am aware that more recent references state that the audion was the first triode vacuum tube, but that can be put down to the modern day lack of knowledge as to what the audion really was coupled with the fact that no one today really fully understands how it worked (if, indeed, anyone ever did). DocFergus (talk) 18:04, 22 March 2018 (UTC)

DeForest's patents insist that the envelope is "evacuated". That's a vacuum - his tubes wouldn't have worked at all at atmospheric pressure, the evacuation was critical to making them work. --Wtshymanski (talk) 19:38, 22 March 2018 (UTC)

I restored Wtshymanski's vacuum tube edit. MIT's (1941?) Applied Electronics textbook goes into detail about required vacuum levels and the effects of residual gas. All tubes generate ions from the residual gas; the issue is how much it affects the operation of the tube. DeForest used the typical high vacuum of the time (the same that Edison used in his light bulbs and Fleming in his valve; is Fleming's valve not a vacuum tube?). DeForest's Space Tel patent disclaims an understanding of how Audions work, and at that time there was debate in the physics community about whether or not gas ions were needed for conduction. DeForest did not need to understand how it worked to invent a vacuum tube triode. DeForest tried a lot of weird arrangements, but he got one that worked. There is a debate about hard versus soft vacuum, and GE/Langmuir lost that debate in court; harder vacuums were used in Audions to avoid the blue glow before Langmuir published his results. Glrx (talk) 20:49, 22 March 2018 (UTC)

The available information suggests that the level of vacuum that De Forest achieved was limited by the available vacuum pumping technology. Fleming's diode almost certainly would have had residual gas inside it, but as it had no functionality beyond passing current one way only, no one is likely to have noticed or cared. I have attempted to calculate the ideal gas pressure for best operation of an audion. Assuming that the residual gas was air (and why wouldn't it be?) whose ions would be approximately 3000 times heavier than electrons, maximum anode current occurs at about 660 Pascals (more or less 150th atmospheric pressure). Ideal audion operation would be at around half that value (because you need the current to rise and fall). My gut feeling is that that figure looks to be a little high, but I can't argue with the physics. Just found a reference which records the ideal gas pressure for maximum sensitivity (i.e. current) for a gas filled photocell as 1100 Pascals for Argon gas. As its atomic mass is roughly double that of air, my figure is in the right ball park (if not necessarily entirely accurate).

You say harder vacuums were used to avoid the blue glow. All De Forest's audion circuits that I have access to show that he operated his audions from an HT (B+) supply of just 22.5 volts. This is far below the 100 volts or so that is required for a blue glow to occur. Having said that, of course, it is not impossible that he did try higher voltages in the course of his trials. DocFergus (talk) 14:08, 23 March 2018 (UTC)

My understanding is DeForest had to operate early Audions at low plate voltages to avoid significant ionization. 20 eV is a tipping point where a collision has a good chance of ionization rather than just excitation. IIRC, blue glow was showing up at 65 V; visible excitation suggests ionization is happening, too. 65 V also suggests avalanche multiplication (3 stages of 20 eV). Figure if the ion current was OK at 22.5 V, then it would be 8 times higher at 65 V. The court case implies engineers knew higher plate voltages were better, but when the voltage got too high, the tube performance fell and there was a blue glow. The glow indicated a gas discharge to those skilled in the art, and the solution was to remove the gas. Early Audions also did not have the advantage of improved biasing. The amplifier applications would not use ion biasing and would therefore not be as finicky. Glrx (talk) 23:02, 23 March 2018 (UTC)

Here's an actual Audion's characteristics: p. 264, fig. 8. According to the researcher the blue glow onset is between 33 and 37 V. ----Chetvorno^TALK 13:06, 24 March 2018 (UTC)

I have had two goes at answering the points above, and twice I have been hit by an edit conflict losing my entire contribution in the process. I am not having a third go. DocFergus (talk) 13:21, 24 March 2018 (UTC)

I can live with the current (Wtshymanski) version of the intro. I agree with Glrx's arguments that it should be called a vacuum tube. @DocFergus: modern sources understand how gas tubes work. I think one reason modern sources call it a "vacuum tube" is because with hindsight we now see it was the ancestor of every amplifying vacuum tube and started the "vacuum tube" era which dominated electronics for 40 years. Modern engineering taxonomy places it in the "vacuum tube" family tree, and we should respect that. The fact that the Audion had more gas than other "vacuum tubes" is in the intro, so I don't think readers will be misled; not everything has to be in the lead.--Chetvorno^TALK 01:26, 23 March 2018 (UTC)

I hope someday one of us can get around to rewriting the body of the article, rather than just the intro; it needs it. I am working on other articles now, but if no one else does I will probably get desperate enough to take a whack at it at some point. One good source is Gerald Tyne's 1943 Saga of the Vacuum Tube], which is available online serialized in Radio News magazine (links in Talk:Audion#Sources), it has a detailed footnoted history of the development of the Audion. --Chetvorno^TALK 01:49, 23 March 2018 (UTC)

As a compromise: can we not describe it as a 'soft vacuum tube'. At least that wouldn't contradict the next sentence. DocFergus (talk) 14:08, 23 March 2018 (UTC)