Talk:MOSFET/Archive 3

External links modified
Hello fellow Wikipedians,

I have just added archive links to 4 one external links on MOSFET. Please take a moment to review my edit. If necessary, add after the link to keep me from modifying it. Alternatively, you can add to keep me off the page altogether. I made the following changes:
 * Added archive http://web.archive.org/web/20130610140024/http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=1049902 to http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=1049902
 * Added archive http://web.archive.org/web/20130903030232/http://www.powerguru.org/wordpress/wp-content/uploads/2012/12/SEMIKRON_application_manual_power_semiconductors.pdf to http://www.powerguru.org/wordpress/wp-content/uploads/2012/12/SEMIKRON_application_manual_power_semiconductors.pdf
 * Added archive http://web.archive.org/web/20160101234641/http://www.automotivedesignline.com/showArticle.jhtml;?articleID=191900470 to http://www.automotivedesignline.com/showArticle.jhtml;?articleID=191900470
 * Added archive http://web.archive.org/web/20160102202735/http://www.semikron.com/skcompub/en/application_manual-193.htm to http://www.semikron.com/skcompub/en/application_manual-193.htm

When you have finished reviewing my changes, please set the checked parameter below to true to let others know.

Cheers.—cyberbot II  Talk to my owner :Online 16:19, 27 February 2016 (UTC)


 * In typical bot framed idiocy the first link is to a paywall where reading the article will cost me a week's food.

Weatherlawyer (talk) 15:48, 6 October 2016 (UTC)

FET as name for MOSFET
I just undid an edit where someone added "FET" as an abbreviation for MOSFET. I feel like this is not valid, as it implies that MOSFET's are the only type of field-effect transistors, which is clearly not the case. On that same note, I realise that most devices we call MOSFET are actually not really MOS, but Poly-oxide-silicon, so it's kinda an iffy situation. Opinions? TheUnnamedNewbie (talk) 10:23, 11 January 2017 (UTC)


 * Along the same line of thought, could someone maybe add the phonetic pronounciation of MOSFET? (so, moss fett) TheUnnamedNewbie (talk) 10:26, 11 January 2017 (UTC)


 * I suppose my thought would be to allow it. Yes there are others, but MOSFET (or, as you note, SOSFET) are so much more common. It is usual to give the more common form the simpler name, and require the appropriate qualification for the less common form.  Any idea what the ratio of JFET to MOSFET production is?  Include every transistor in every integrated circuit and discrete device produced? Gah4 (talk) 10:49, 11 January 2017 (UTC)


 * I don't know any numbers, but I am aware that JFET's are far less common, by many orders of magnitude. However, I don't think that means that FET is a acceptable name for a MOSFET, especially since a FET is more a conceptual thing (steering current flow with an electric field) and a MOSFET is a practical implementation. Perhaps a compromise would be to mention that in industry lingo, "FET" is often used to refer to a MOSFET, but not all field-effect transistors are MOSFETs? Perhaps something like this (feel free to improve the wording, just trying to get the concept):

The metal–oxide–semiconductor field-effect transistor (MOSFET, MOS-FET, or MOS FET) is a type of [field-effect transistor] (FET) used for amplifying or switching electronic signals which has an insulated gate whose voltage determines the conductivity of the device. Although FET is sometimes used when refering to MOSFET devices, other types of field-effect transistors also exist.


 * TheUnnamedNewbie (talk) 12:36, 11 January 2017 (UTC)


 * Close enough for me. A quick search found http://www.ti.com/lit/ds/symlink/opa655.pdf, which, as I finally found from the simplified internal schematic, is actually a JFET. I suspect FET is used often enough referring to MOSFET, though.  One that I find funny is CMOSFET, which doesn't make sense to me. Gah4 (talk) 13:36, 11 January 2017 (UTC)


 * I changed the introduction. TheUnnamedNewbie (talk) 11:58, 12 January 2017 (UTC)


 * Oppose. The JFET isn't dead yet. Andy Dingley (talk) 11:31, 11 January 2017 (UTC)

Naming
It was at that time the Bell Labs version was given the name bipolar junction transistor, or simply junction transistor, and Lilienfeld's design took the name field effect transistor. OK, for one, the name transistor came from the contrast with transconductance, the description for vacuum tubes as voltage controlled current sources. (That is, dI/dV). When (year) and who named the Lilienfeld device, which as I understand it didn't really work, a field effect transistor? That should satisfy the. Gah4 (talk) 15:40, 18 May 2017 (UTC)

External links modified
Hello fellow Wikipedians,

I have just modified 2 external links on MOSFET. 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/20140727084407/http://www-device.eecs.berkeley.edu/bsim/Files/BSIMCMG/BSIMCMG106.0.0/BSIMCMG106.0.0_TechnicalManual_20120313.pdf to http://www-device.eecs.berkeley.edu/bsim/Files/BSIMCMG/BSIMCMG106.0.0/BSIMCMG106.0.0_TechnicalManual_20120313.pdf
 * Added archive https://web.archive.org/web/20151228041321/http://www.itrs.net/ to http://www.itrs.net/

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

Cheers.— InternetArchiveBot  (Report bug) 22:23, 26 July 2017 (UTC)

External links modified
Hello fellow Wikipedians,

I have just modified one external link on MOSFET. 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/20141110225738/http://equars.com/~marco/poli/phd/node20.html to http://equars.com/~marco/poli/phd/node20.html

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

Cheers.— InternetArchiveBot  (Report bug) 18:55, 10 January 2018 (UTC)

Can you prove (or cite) a negative?
In the history section appears the words, "Bell Labs was able to work out an agreement with Lilienfeld, who was still alive at that time (it is not known if they paid him money or not).[citation needed]"

Ok. Wait. I change my mind. Maybe nothing is amiss here.

When I first read this (very late/early) my brain thought the part that needed the citation was whether or not he (Lilienfeld) had been paid any money. My issue with this is that if he received no money you will most likely never be able to prove conclusively And cite it because very few people documenting such a history would note that he didn't get paid. But that doesn't mean conclusively that he didn't.

But... now that my brain has re-read this section for theb4-billionth time as I write this I now understand that the issue at hand is citing whether or not Bell Labs was able to work out an agreement with Lilienfeld.

Sorry. But I posted this anyone because maybe someone will read it and realize that all of us can work to be clearer and we make mistakes in interpretation - especially at such late/early hours.

Thank you all. 166.142.172.242 (talk) 07:05, 9 September 2017 (UTC) David


 * The following was recently removed, though I believe that it belongs:


 * Twenty five years later, when Bell Telephone attempted to patent the junction transistor, they found Lilienfeld already holding a patent, worded in a way that would include all types of transistors. Bell Labs was able to work out an agreement with Lilienfeld, who was still alive at that time (it is not known if they paid him money or not). It was at that time the Bell Labs version was given the name bipolar junction transistor, or simply junction transistor, and Lilienfeld's design took the name field effect transistor.


 * I heard something like this when first learning about ohmic contacts. The story was that the Lilienfeld device didn't work because he didn't know about making ohmic contacts. While the Lilienfeld patent would have expired, Bell would have to show that it wasn't prior art for the junction transistor. Having Lilienfeld agree would have helped. I didn't look for any source, though. Gah4 (talk) 17:06, 14 May 2018 (UTC)


 * OK, some of it is here: Patent Battles. One problem is that what thought they were making, what Shockley thought up, was closer to the Lilienfeld FET. It was only after they had the point contact junction transistor working, that they started to figure out why it worked. Gah4 (talk) 17:16, 14 May 2018 (UTC)

Confusing structure in Applications : CMOS : Analog
The Applications section has Analog under CMOS circuits - not clear from the content if that is the intended structure. Would it be better to have as the top level structure : Digital logic, Signal switching, Power switching, Other analog applications ? Also seems strange to have NMOS logic and Power MOSFET under Other types as they have been discussed above (eg under Applications). - Rod57 (talk) 09:19, 11 June 2018 (UTC)

Proposed talk page edit to delete External links modified sections
Per ongoing discussion at Village_pump_(proposals) and Template_talk:Sourcecheck I'd like to delete the above External links modified section(s). Any objections ? - Rod57 (talk) 09:27, 11 June 2018 (UTC)

Merge MISFET
I suggest that MISFET is merged here. The article is only a single short paragraph and MOSFET already says nearly all of what is there. As I understand it, the difference is only a question of the materials used, so it is unlikely that MISFET would ever be substantially different from this article even if it were fully expanded. SpinningSpark 18:52, 8 March 2016 (UTC)
 * strongly object. These are different concepts. Wikipedia is not paper to squeeze everything into one page. That the page is short means semi engineers don't give a dime for wikipedia, but it does not mean there is nothing to write. Staszek Lem (talk) 16:26, 29 April 2016 (UTC)


 * Maybe. If the MISFET article can be expanded a bit to talk about the materials that are alternatives to oxide, and if transistors of that sort are not already described in MOSFET, then it can stay separate.  Dicklyon (talk) 16:48, 29 April 2016 (UTC)
 * Exactly; there is not a bit but lots. However it is a highly specialized area, and things are experimental, and experts in materials science are too busy to waste their time here. Staszek Lem (talk) 17:01, 29 April 2016 (UTC)


 * Agree. (and then redirect).  MESFET should stay separate, but I don't see much reason to keep MISFET separate. For many years, silicon (and so not metal) gates have been used, without changing the acronym. Insulators other than SiO2 are used, but most often still oxides. Even if an oxygen free insulator is used, the physics isn't so different. Gah4 (talk) 11:32, 11 January 2017 (UTC)

dum — Preceding unsigned comment added by 66.254.239.203 (talk) 15:13, 25 March 2019 (UTC)

Software used to create File:MOSFET Structure.png?
Hi! Anyone know what software used to create ? (This is the first diagram shown in this article.) I need to be able to draw my microelectronic substrates in 3D.... Thanks! --Blue.painting (talk) 18:29, 22 June 2019 (UTC)

CMOS logic consumes over 7 times less power than NMOS logic, and about 100,000 times less power than bipolar
The article says: CMOS logic consumes over 7 times less power than NMOS logic,[9] and about 100,000 times less power than bipolar. Because of the way different logic families scale with clock speed, these numbers are not very useful. Fast switching CMOS uses a lot of power, not switching at all, nearly zero. Which one do you use for the comparison? TTL and NMOS have only a small dependence on clock rate. Gah4 (talk) 21:18, 27 November 2019 (UTC)


 * The sources are comparing them during the 1970s-1980s. Not sure what the difference is now. I've clarified in the article that the comparison is for the 1970s-1980s. Maestro2016 (talk) 01:31, 28 November 2019 (UTC)


 * OK, completely different. As above, CMOS current is pretty much proportional to clock speed, and speeds have increased from MHz to GHz. Transistors get smaller (less current) but more of them (more current). There are CMOS processor now that run at 100W. Gah4 (talk) 02:23, 28 November 2019 (UTC)

scalng
The article makes some comparisons between BJT and MOSFETs that I am not sure are, in general, true. The scaling laws are different for the two. In the early days of CMOS, it was much slower than TTL. However, CMOS gets much faster when it scales down, while TTL doesn't, so that after not so long, CMOS was faster. (There were complications in CMOS manufacturing that had to be overcome, though.) As for high power, are high power MOSFETs so much more common than high power BJTs? Also, I believe that much of analog IC work is still BJT, or a mix of the two. Gah4 (talk) 21:59, 15 January 2020 (UTC)


 * I've now added a sentence to the article mentioning that MOSFETs were initially slower than BJTs. As for power MOSFETs, they account for the majority of the power electronic market as of 2010, with a 53% share, followed by IGBTs with 27% share and then BJTs with just 9% share. As for analog ICs, this this 2002 source says that MOSFETs account for at least 50% or more of the analog IC market. Maestro2016 (talk) 15:43, 22 January 2020 (UTC)

Totally Useless Article
The perhaps half-a-dosen academics worldwide who research FETs this article wouldn't say anything new. For the other perhaps 250 million who just wants to know how to design a circuit with a FET, the article says NOTHING. — Preceding unsigned comment added by 115.70.177.64 (talk) 06:46, 16 November 2015 (UTC)

• Agree. OTOH, the pages on Common Source, Common Drain etc. are incredibly useful, but not linked to anywhere in the article. The Applications section needs a subsection on amplifiers, outlining the three main types of MOSFET amplifier, links to their pages and a diagram of at least one type of amplifier. Might also be able to concatenate all the small signal characteristic tables into one. 2A02:C7D:76BC:A500:DFC3:2B:250A:2F91 (talk) 00:05, 30 May 2018 (UTC)
 * I suspect most articles about things, don't say how to use them, but just what they are. For one, there is WP:NOTHOWTO, but in many cases the use should go somewhere else. AC_power_plugs_and_sockets tells everything you might want to know about them, except how to use one. Gah4 (talk) 19:10, 31 March 2020 (UTC)
 * I suspect most articles about things, don't say how to use them, but just what they are. For one, there is WP:NOTHOWTO, but in many cases the use should go somewhere else. AC_power_plugs_and_sockets tells everything you might want to know about them, except how to use one. Gah4 (talk) 19:10, 31 March 2020 (UTC)

Sockpuppet investigation opened
Hi, I was asked to review the history of Maestro2016's edits to this page to see if I thought Maestro2016 was a sockpuppet of User:Jagged_85. Jagged_85 did huge damage to the encyclopedia a few years ago by adding thousands of edits of apparently properly referenced text where, if you actually checked the reference, you found it did not in fact support what Jagged_85 had added to the article. Jagged_85 particularly specialised in the promotion of Islamic technology and Islamic 'great men'. After spending some time reviewing Maestro2016's edits on various articles I think they are indeed a sockpuppet of Jagged_85. See Sockpuppet_investigations/Jagged_85. Regardless of whether Maestro2016 is a Jagged_85 sockpuppet or not, what my investigation highlighted was that Maestro2016 has made hundreds of dubious edits adding thousands of characters to this article which are not really supported by the references. For example making it sound like the Islamic 'great man' Mohammad Atalla single-handedly invented MOSFET, which in turn created all of the modern electronic world. While Atalla and the invention of MOSFET are clearly important, if you follow the references you invariably find that Maestro2016's claims are exaggerated compared to the source, or slanted to ignore factors other than Atalla, or cherry picked out of context to make it sound like Atalla should have been given a Nobel (when he was mentioned in passing in an article on someone else), etc. etc. I found references which were plagiarised from engineering course pdfs; taken out of context from workshop abstracts; referencing the wrong technology on the wrong calculator. I would personally consider any edit added by Maestro2016 where the references have not been checked and cleared by another editor as suspect. There may therefore be a case to going back to the state of the article before Maestro2016 got involved, and then adding back in any reasonable material as references are checked. The relevant edit I believe would be this: The diff to the current page is this:. Merlinme (talk) 13:23, 18 May 2020 (UTC)
 * Sorry, I missed a few. The edit before Maestro2016 got involved is actually this: The diff to the current page is this:  Merlinme (talk) 13:37, 18 May 2020 (UTC)


 * Couldn't you at least wait for the sockpuppet case to actually come to a conclusion before posting these claims here? And are you seriously suggesting we go back to the problematic old version of the article? I've already explained in the sockpuppet case page why the old version is problematic: You're misrepresenting my edits. Before I began editing the article, the History section of MOSFET consisted of just a single sentence on Lilenfield followed by a paragraph on Atalla and Kahng, and that's it. There was no mention of Heil, Shockley, Bardeen, Brattain, Sah, Wanlass, Deal, Grove, Sarace, Klein or Faggin anywhere. Who added their contributions to the article? That's right, me! Your claim that I ignored their contributions is nonsensical when I was the one who added their contributions to the article in the first place. You're making it sound as if I dedicated the History section almost entirely to just Atalla, when he actually only got two paragraphs (one shared with Kahng) out of the eight paragraphs I wrote in the History section. You're focusing on those two paragraphs and ignoring the six other paragraphs where I described the contributions of others. And for the record, I had previously already added the contributions of Sarace, Klein, Faggin and others to the MOS IC section." You're essentially suggesting we erase the contributions of Heil, Shockley, Bardeen, Brattain, Sah, Wanlass, Deal, Grove, Sarace, Klein, Faggin, etc. simply because you want to downplay Atalla, who just so happens to be Middle-Eastern (there's no evidence he was "Islamic"). Your bias is clearly showing. Maestro2016 (talk) 14:06, 18 May 2020 (UTC)
 * The main reason I'm suggesting going back to the previous version is because of the problematic sourcing. I found significant problems with most of the sources I checked, and I don't have time to check all of them. Are you going to justify plagiarising an engineering course pdf, for example? You've made corrections to some of the problems I've highlighted, but really, given the extent of the problems, I think it would be better to go back to the version before the addition of so many problematic sources. If the material is correctly referenced it can be added back in. Merlinme (talk) 14:41, 18 May 2020 (UTC)
 * I've withdrawn the sock accusation. I still have big problems with the use of sources. Copying this here from the sock investigation:
 * "In 2018, the Royal Swedish Academy of Sciences which awards the science Nobel Prizes acknowledged that the invention of the MOSFET by Atalla and Kahng was one of the most important inventions in microelectronics and in information and communications technology (ICT)". Saying that the Nobel awarding academy 'acknowledged' something sounds like it won a special prize of some description. In fact the Academy issued a press release about the actual winner, who won for the integrated circuit, which the Academy describes as 'the driving force' of microelectronics. After that, in one sentence at the end of the section, they mention MOSFET and the microprocessor as also important. If this reference is needed at all, (there are surely lots of other references which could say similar things more succinctly), it needs to be put in context: "In its award of half the Nobel prize for Physics to Jack Kilby in 2018 for his part in the invention of the Integrated circuit the Royal Swedish Academy of Sciences specifically mentioned MOSFET and the microprocessor as other important inventions in the evolution of microelectronics." That would at least make it clear that it was mentioned in passing as part of an award that was not being given to Atalla and Kahng. And where does it say 'one of the most important'? The sentence starts "Other important inventions include", and they mention a couple, but the implication is surely that there are other important inventions that for whatever reason they're not going to mention. Adding "along with the integrated circuit and the microprocessor", as you have recently done since I made my initial criticisms, is only a marginal improvement, mainly because anything that fails to make it clear that the Academy was giving a prize for the integrated circuit, not MOSFET, is giving a misleading impression of the source.
 * Or this as a source: Is an engineering course pdf really the best source for the claim "The MOSFET is by far the most widely used transistor in both digital circuits and analog circuits, and it is the backbone of modern electronics"? Is this a WP:Reliable Source for an WP:Exceptional claim? Is this a WP:Reliable Source for any claim at all? Quite apart from the fact the original edit was almost certainly a copyright violation, the PDF has no author; what do we know about the fact checking process that went into making that statement?
 * The problem I have is that practically every source I checked I found some issues with. The calculator references have now been fixed, but that suggests a worrying lack of care in adding them in the first place. And I've only checked a handful of the references which were added. I still think there would be a case for returning the page to how it was before the addition of so much problematic material. Bad references are a lot worse than no references at all, because it gives an illusion of authority. Alternatively, I guess you could go through the article, and for each reference say: 1) Is this a Reliable Source? 2) Is it reported accurately and used in context? 3) Is it adding something to the article? Merlinme (talk) 22:38, 18 May 2020 (UTC)
 * Fair enough. I think the last option is probably the best option. I'll check each of the sources to see if there's any such errors or discrepancies, and make changes where necessary, and then you can comment on those changes. Though it's going to take quite a bit of time to get through them. Maestro2016 (talk) 00:38, 19 May 2020 (UTC)
 * Consider also how things fit in with the rest of the article. This is the most general article on MOSFETs, so some details are not needed, or could go in another article. Gah4 (talk) 01:24, 19 May 2020 (UTC)
 * Consider also how things fit in with the rest of the article. This is the most general article on MOSFETs, so some details are not needed, or could go in another article. Gah4 (talk) 01:24, 19 May 2020 (UTC)
 * Consider also how things fit in with the rest of the article. This is the most general article on MOSFETs, so some details are not needed, or could go in another article. Gah4 (talk) 01:24, 19 May 2020 (UTC)

sign and mode
As well as I know, for enhancement mode, PMOS (p-channel) is in n-type Si, and NMOS (n-channel) in p-type Si. That is, in either substrate of that type, or in a well of that type. The other way around for depletion mode. Gah4 (talk) 01:21, 19 May 2020 (UTC)
 * Not necessarily the other way around for depletion mode; just different channel doping. I.e., in depletion-load nMOS process, both the enhancement-mode and depletion-mode devices are made in the same p-type substrate. Dicklyon (talk) 03:44, 19 May 2020 (UTC)
 * The article says: P-type MOS (PMOS) logic uses p-channel MOSFETs to implement logic gates and other digital circuits. N-type MOS (NMOS) logic uses n-channel MOSFETs to implement logic gates and other digital circuits. But PMOS uses p-channel FETs in n-type Si, and NMOS uses n-channel FETs in p-type Si. Gah4 (talk) 04:37, 19 May 2020 (UTC)
 * The article says: P-type MOS (PMOS) logic uses p-channel MOSFETs to implement logic gates and other digital circuits. N-type MOS (NMOS) logic uses n-channel MOSFETs to implement logic gates and other digital circuits. But PMOS uses p-channel FETs in n-type Si, and NMOS uses n-channel FETs in p-type Si. Gah4 (talk) 04:37, 19 May 2020 (UTC)