Talk:Voltage doubler/GA1

Latest comment: 12 years ago by Malleus Fatuorum in topic GA Review

GA Review edit

Article (edit | visual edit | history) · Article talk (edit | history) · Watch

Reviewer: Malleus Fatuorum 23:47, 30 June 2011 (UTC)Reply

Lead
  • "The term is usually applied to circuits consisting of rectifying diodes and capacitors only; other means of doubling voltages are not included." Why are they not included? Surely either this article has the wrong title or they ought to be included? I'm putting this review on hold immediately pending a satisfactory explanation of the article's scope.
  • Voltage doubler is a stock term for this kind of circuit which is rarely, if ever, applied to other circuits formats. Neither broadening the scope (which would result in a collection of more or less unrelated items) nor changing the title (to something less commonly used) would be appropriate here. Another point here is that these circuits are ac input and dc output. Other means of raising voltage are transformer (ac/ac) and chopper circuits (dc/dc) for instance. A gbooks search supports my assertion that the term commonly means diode/capacitor circuits and this book explicitly gives that as a definition. At the very most, a disambiguation hatnote would be appropriate if there were other articles on Wikipedia which specifically covered voltage doubling by other means, but I do not know of any. SpinningSpark 01:38, 1 July 2011 (UTC)Reply
  • "Rarely, if ever" isn't good enough. Is it or isn't it? If you're restricting the scope of this article to ac–dc voltage doublers then that needs to be made clear, preferably with a brief explanation or link to other forms of voltage doubler. Ought it, for instance, to be called a voltage-doubling circuit, to distinguish it from a transformer? Malleus Fatuorum 01:52, 1 July 2011 (UTC)Reply
If you are not prepared to accept the evidence of the sources and prefer your own opinion I am not really sure what I can say here. The first sentence of the article directly and unambiguously makes clear we are talking about ac to dc circuits. That voltage doubler is a stock phrase is made clear by the many thousands of books that so use it. This is a clear case of WP:UCN. That these must nearly all be talking about the circuits described here is shown by the near identical number of results from the searches "voltage doubler" and "voltage doubler" diode. I can also confirm from my own extensive experience as a professional electronics engineer that this is what the phrase would normally be taken to mean.
Voltage-doubling circuit is not so common (gbook search) and is not a stock phrase, it is a descriptive title. Arguably, transformers should be included in an article so named, nevertheless, the gbook hits are still pretty exclusively coming up with the same circuits.
While I am sure that some examples of other circuits described as voltage doublers could be found, this would be descriptive rather than a stock phrase. The term car is normally taken to mean an automobile and so redirects. it does not redirect to a disambiguation page or article that includes the uses as car of a train, carriage or chariot, or a car of a fairground ride. This is because that is what car unqualified invariably means unless specified otherwise. SpinningSpark 09:35, 1 July 2011 (UTC)Reply
You need to resolve the opening to clearly establish the scope of the article: "The term is usually applied to circuits consisting of rectifying diodes and capacitors only; other means of doubling voltages are not included." What you appear to be saying with that is that there are several different kinds of voltage doublers, but you're going to ignore all of those that don't consist only of rectifying diodes and capacitors for some reason. You seem in addition to be restricting the scope to ac-dc voltage doublers, which is perhaps what the article ought to be called. Malleus Fatuorum 14:05, 1 July 2011 (UTC)Reply
I am not gratuitously excluding anything. Look at this from the other side starting with what the article is actually about. The article is about a particular circuit. That circuit is universally called voltage doubler. Any suggestion to change the article title to something other than that used in over 8,000 textbooks is just plain wrong. The rest of it is merely to clarify what we are talking about for the benefit of readers who do not already know the terminology. It is not essential to the article to have this clarification, but it is highly useful to the uninformed reader. There should be no disambiguation in the title, this article is overwhelmingly what is meant by voltage doubler. Any disambiguation, if it is ever needed, would rightly be in the title of other articles. SpinningSpark 16:58, 1 July 2011 (UTC)Reply
In which case you need to remove the gratuitously teasing "... usually applied to circuits consisting of rectifying diodes and capacitors only; other means of doubling voltages are not included". Why so coy? Why can't you just say plainly in the lead something like "Voltage doublers are circuits composed only of rectifying diodes and capacitors, designed to deliver a dc output voltage twice the peak value of the ac input voltage"? If the "other means of doubling voltages", such as transformers would not be called "voltage doublers", as you seem to be saying, then why would a reader need to be told that the article on voltage doublers doesn't deal with them? Malleus Fatuorum 18:12, 1 July 2011 (UTC)Reply
  • The opening sentence is a good enough definition, IMHO. Like Malleus, I do have a slight issue with the last sentence though, specifically "consisting of rectifying diodes and capacitors only". Doublers can also made with switched capacitors, for example, so we should keep things as general as possible. Maybe word it like "basic doublers are made with diodes/capacitors", leaving out "only" (the definition already excludes transformers). -Roger (talk) 19:58, 1 July 2011 (UTC)Reply
Thanks for looking Roger, I've changed it in line with your suggestion. Switched capacitors, though, are a dc device and would not fall under the current definition. SpinningSpark 22:19, 1 July 2011 (UTC)Reply
OK, I'm much happier with that, so now we can get on with the rest of the review. Malleus Fatuorum 22:47, 1 July 2011 (UTC)Reply
True, overlooked that. We could consider taking out the "AC" from the first sentence, and adding a new sentence linking to the charge pump article (as an example of a diode-less doubler). -Roger (talk) 01:28, 2 July 2011 (UTC)Reply
If you do that, this would no longer be a GA quality article and the GAN would have to be withdrawn, at least while the article was worked on. At the moment the article is about something very specific and the scope is well-defined. If going down that road, I can't see how we could help but expand into an overview article of all means of increasing voltage in which case the material we have here now ought to be somewhere else. SpinningSpark 09:57, 2 July 2011 (UTC)Reply
I'm not too familiar with the GA rules, wouldn't that just count as a minor change? The rest of the article seems fine, it just might be worth noting that doublers generally consist of capacitors and some sort of switch (diodes, transistors, etc.) -Roger (talk) 21:00, 2 July 2011 (UTC)Reply
That takes us right back to the heart of my original objection to this article I'm afraid, which is its intended scope. "Doublers generally consist of ..." implies that there's a class of doubler that this article is ignoring. Malleus Fatuorum 21:11, 2 July 2011 (UTC)Reply
Right up to the point Roger mentioned switched capacitors I was prepared to defend to the death the scope of this article. A better definition might be that a voltage multiplier multiplies the voltage by an integer amount n, and in the case of the voltage doubler n = 2. They do not multiply by 3.1415926 for instance. Transformers obviously do not fit into this class, if they were included why not amplifiers? - they have voltage gain after all - and to continue down that line ends up with a monster, pointless article spanning all of electronics. The same argument holds for the class of dc-dc converters in general. However, switching capacitors are an exception; these circuits do multiply by n integer, or at least a rational number. So I am inclined to withdraw the GAN, or at least delay it, until this can be included and while we are about it we could do something on silicon chip implementations which have their own peculiar restrictions. Malleus, you are still wrong and are wilfully misreading what Roger has said. Roger, this is your fault and I think emotional blackmail should be applied to get you to do some work on this article :). SpinningSpark 00:00, 3 July 2011 (UTC)Reply
I am misreading nothing. Do you wish to withdraw this nomination? Malleus Fatuorum 00:23, 3 July 2011 (UTC)Reply
Yes, withdraw, but before you close it I would like to bottom out how we are going to constrain the scope of the article. I believe you have misread Roger's statement: I don't want to put words into his mouth (and I am sure he will correct me if I have got this wrong) but my paraphrasing of his statement would be that the general class of voltage doublers are composed of capacitors and switching elements and the specific sub-class has diodes as the switching element whereas your interpretation seems to be that the majority of the class of voltage doubler are composed of capacitors and switching elements with a minority composed of something else.
The problem with any attempt to put a rigid definition to the scope is going to be sourcing for the statement. I don't believe that such a firm definition will be found in the sources because the industry has no pressing need to create such a definition. The approach of textbooks is to present their examples of the circuits under a section headed "voltage doubler". There is no definition supplied; voltage doublers are defined by the examples presented. So although those skilled in the art are going to have a fairly clear idea of what is meant it is going to be hard to justify any definition. Both Roger's capacitor+switch and my n=2 definition come close this understanding but both are essentially OR in a Wikipedia sense. Although here is a technical dictionary definition that pretty much defines as the article is now and would exclude Roger's switched capacitors - but I don't think it would be wise to take that too literally. SpinningSpark 01:35, 3 July 2011 (UTC)Reply
I think that scoping discussion is best held elsewhere, probably on the article's talk page. But if you want to keep this review page open for a bit longer then that's no problem. Malleus Fatuorum 01:40, 3 July 2011 (UTC)Reply
I was hoping to get a view from you as the reviewer on what you would find acceptable. SpinningSpark 01:50, 3 July 2011 (UTC)Reply
My fundamental problem remains the fuzziness of the definition of a "voltage doubler" in conjunction with the generally/usually composed of capacitors and rectifying diodes, but sometimes other switching elements as well/instead. The starting point has to be a clear and authoritative statement of what a voltage doubler actually is – OK, I know it's a circuit that provides a dc output voltage of twice the peak ac input voltage, but why do we keep getting caught up in the types of circuit elements it's composed of? Either a circuit doubles the voltage or it doesn't; why the distinction between those composed only of diodes and capacitors and others? Is there a historical reason? Malleus Fatuorum 02:18, 3 July 2011 (UTC)Reply
I can give you (at least I can now we have had this discussion) a clean and precise definition that is free from all qualifying phrases and free of all mention of specific components (at least till the point the article starts discussing specific or "most common" circuits) and would be recognised as valid by others skilled in the art. The problem is that I do not believe that I could back up that definition with sources - although I will do a more thorough search now that we have hit this problem. The article is fuzzy because the industry itself is fuzzy, as I said above it has no particular need to be precise and I suspect that a similar problem would be hit on the definition of many other circuit classes if one were to push it too far. It is quite likely that this discussion has only ever occured on Wikipedia - the breadth of our scope means we have a clear need for disambiguation that is absent elsewhere. So if the industry remains obstinately fuzzy do we follow it with a fuzzy definition or implant our own OR definition for the sake of clarity? SpinningSpark 10:55, 3 July 2011 (UTC)Reply
I'm very much afraid that in the absence of a reliable source for a generally accepted definition of voltage doubler that corresponds to the contents of this article (i.e., circuits composed only of rectifying diodes and capacitors) I can see no option other than to extend the scope of this article to include those voltage doublers that do not meet that narrow definition. Malleus Fatuorum 13:30, 3 July 2011 (UTC)Reply
As far as sources go I have so far presented two on this page which actually bother to give a definition at all, and both would support the article exactly as it is now without any expansion or amendment. You are surely not suggesting that the lack of definitions should mean we should include transformer, for instance, when out of 8,000 textbooks which cover voltage doubler not one has been found which even mentions transformer as a voltage doubler, no engineer would call transformer a voltage doubler and an uninvolved expert editor when asked for an opinion, although suggesting a different addition, did not include transformer either. We are supposed to write from the sources on Wikipedia, if the sources think a term has limited scope, even if they indicate that passively by what they cover rather than explicitly with a definition, why should Wikipedia choose to define the term as a broader scope? I have changed my mind about withdrawing the nomination. I do not want to do a lot of work on the article only to have this same debate again on resubmission. I would now prefer to let you fail the nomination and appeal the decision at GAR as it is clear that you cannot be convinced. SpinningSpark 16:10, 3 July 2011 (UTC)Reply
Fair enough, although it is in fact you who have failed to convince me. Good luck at GAR. Malleus Fatuorum 16:33, 3 July 2011 (UTC)Reply
Notes
  • What makes this a reliable source?
  • Because it comes from a physicist who has written two books on high-voltage electricity and the page has actually been cited in scholarly papers and patents. It was originally included because it very clearly distinguished between the Villard and the Greinacher circuits. Many other sources have them horribly confused, even treating them as synonyms. However, it appears to now only be referencing the full-wave bridge circuit (not quite sure how that happened). This is a well-known circuit for which book sources are easily found. SpinningSpark 00:00, 3 July 2011 (UTC)Reply
  • A physicist whose hobby is high-voltage electricity it appears. Malleus Fatuorum 01:05, 3 July 2011 (UTC)Reply
  • Ok it is a hobby page but the man publishes in the field of particle physics - just the place where they build high-voltage voltage multipliers, the original Cockcroft-Walton generator was developed and built specifically for particle physics. The citations to his "hobby" page in published works are a far better indication of scholarly reliability than merely having got a book past a publisher (which he has also achieved - twice - in his "hobby" field). I really think reliability is established far better for this source than the majority of random gbook search results, many of which are highly suspect for accuracy. SpinningSpark 10:55, 3 July 2011 (UTC)Reply
The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.