Talk:Rectifier/Archive 1

more stuff
whats the ripple frequency for a half wave rectifier plugged into a line voltage?

does the filte rcapacitor act as a low pass or high pass filter?

does the filter capacitor reduce ripple voltage or ripple frequency?

if a diode fails in a ful wave, what happens to ripple voltage and frequency?

IF THESE INCREASE     THIS WILL MAKE RIPPLE VOLTAGE resistance load current

if a filter capacitor opens, ripple voltage equals? what if it shorts?

just some interesting questions you may want to include in the article!

Peak loss
Maybe it should be pointed out that the peak loss for full wave rectification is 2 times 0.7, as two diodes are involved? Done that - JJ

Music
I first encountered rectification when researching how to build myself a guitar effects pedal. The output frequency of a full rectifier is effectively double the input frequency, and is therefore convenient in producing an "octave" effect.

This is really about the extent of what I know about the topic, and if someone more knowledgeable could insert it into the article in a better articulated fashion, I think it would be a good addition. If you think I articulated the idea well enough, feel free to just cut and paste it into the article. Shaggorama 00:04, 6 March 2006 (UTC)

"Rectifier efficiency"
The eight / pi^2 figure does not refer to the ratio of output power to input power; it only refers to vdc^2/vac^2. The two are not at all the same. 19% of the input power is not dissipated by the rectifier under ideal conditions. That part needs to be rewritten.


 * Remember the rectifier effiency is the power of the output DC component over the AC input power. Not the total output power over total input power (which would be 100% for an ideal fullwave rectifier).


 * A quick google search for "rectifier efficiency" does not show this to be a common definition for those words. I can't imagine why it would be; why would that number (average Vout divided by peak Vin, all squared) ever be meaningful? I have therefore removed the definition, and the rest of the section was more or less completely incoherent, so I removed that as well.

I have no idea, either, where it came from. However, if you look at the math (integrate), the ratio of peak DC voltage to mean DC voltage (not RMS, but true mean) for a rectified wave, is VpeakDC/VmeanDC = pi/2. So your voltage when you filter it, will actually increase by that much. And power by the square of that. But these give "efficiencies" of greater than 1, basically because the smoothed DC output is at AC peak voltage, not AC (RMS). If the AC(peak) is AC(RMS)*SQRT2 and we forget diode drop and set AC(peak) = DC(peak), then the ratio of AC(RMS) to DC(mean) voltage is:

VAC(RMS)/VDC(mean) = [V(peakDC)/(SQRT2)]/V(meanDC) = pi/(2SQRT2) = 1.11

The inverse of this is about 90%.

So I don't know where this figure used comes from. The only thing that gives 4/pi^2 is the square of VDC(mean)/VDC(peak), or VDC(mean)/VAC(peak). But I can't see the physical meaning of either ratio. S B Harris 03:05, 20 August 2006 (UTC)

Change article title
I suggest a change of title. This article is called rectifier (after the device) then goes on to talk mostly about rectification (the process). I suggest that John Dalton 21:10, 15 March 2006 (UTC)
 * This rectifer article be renamed to rectification.
 * rectifier be redirected to diode


 * I would go along with the rectification change, although when done you know somebody's just going to carp that the article goes on to talk about various kinds of rectifiers, and should be called THAT. You want verb or noun? I object to the change to diode because many of the old high-power systems (before the fancy silicon high power diodes) like arc rectifiers and AC-powered DC-generators, were certainly not diodes in the sense that we know them. They need to be added into the article as has been noted, but nobody has gotten around to it yet. Eventually. S  B Harris 09:07, 4 September 2006 (UTC)

anyone fancy
drawing the input/output graphs for the 3 phase bridge? ;) Plugwash 22:32, 3 September 2006 (UTC)

Notice the diagram for the full rectification. Shouldn't there be a flat portion between each hump? Until each diodes passes its specified voltage (~.6-.7V for Si diodes) then there will be no current across the load?
 * Yes, real diodes will introduce a small flat portion, i wonder though if introducing this into the diagrams will confuse more than it will help though. Plugwash 18:02, 23 October 2006 (UTC)

Rectifier Relationships
I'm doing my lab report on rectifier circuits and I really need rectifier relationships both Vdc and Vrms values produced by all types of rectifiers. Well current might be added as well. I have some, might add them, when I finish, cause it is due soon. Can anyone please add them? —The preceding unsigned comment was added by Dmitrij.ledkov (talk • contribs) 21:51, 12 February 2007 (UTC).


 * The relatonships depend on the design of the rectifier, the properties of the components, and the load. We might be able to provide approximate relationships; how accurate do the relationships need to be? --Gerry Ashton 22:44, 12 February 2007 (UTC)

"Rectifier Diode" image
This is not a diode this is an SCR (notice the gate connection). —The preceding unsigned comment was added by 203.166.229.52 (talk) 14:36, 19 February 2007 (UTC).

3-phase rectifier
Is there a significance in which order the 3 live wires are fed into it, or can the connection be arbitrary? --194.251.240.114 23:45, 27 March 2007 (UTC)
 * The answer to that question should be obvious from the schematic. Plugwash 20:53, 30 March 2007 (UTC)

What is "RMS"?
what is a "rms"?


 * RMS stands for root mean square. If you have alternating voltage applied to a resistor, and you would like to know how much DC voltage, applied to the same resistor, would produce the same power, you can take the voltage at each instant, square it, and then find the average squared voltage over a suitable period of time (one cycle of a sine wave, for example). Then take the square root of the mean squared voltage.


 * The same procedure works for current too. --Gerry Ashton 16:52, 27 July 2007 (UTC)

Other uses of rectifier
There are other definitions of the word Rectifier that have been ignored, such as:

rectifier - a person who corrects or sets right; "a rectifier of prejudices"

"rectifier." WordNet® 3.0. Princeton University. 05 Sep. 2007. .

&

rectifier - A worker who blends or dilutes whiskey or other alcoholic beverages. "rectifier."

The American Heritage® Dictionary of the English Language, Fourth Edition. Houghton Mifflin Company, 2004. 05 Sep. 2007. .

I think its misleading to simply have the alternating to direct current transformer definition rather than include all the definitions of rectifier.

Just throwing it out there. —Preceding unsigned comment added by 128.210.38.14 (talk) 12:28, 5 September 2007 (UTC)

Now fixed with a dab (disambiguation) page.

Untitled
In an alternator with an eight-diode rectifying unit, what are the last two diodes for? I can't find any information on this.

Some alternators are four-phase rather than three-phase. In the schematics I'm familiar with, these are the units that require eight diodes. (two diodes for each phase) —Preceding unsigned comment added by 67.161.213.4 (talk) 01:29, 30 November 2007 (UTC)

Images are slightly too big. I increase the text font and that shoves the image to right.

a rectifier is a circuit which converts bidirectional current(ac) to unidirectional(dc). earlier rectifier was a single diodeconnected to the output of a transformer.but there was some ac still present in the output which is measured by a factor called ripple factor.therefore a low ripple is desired in output of rectifier the above arrangement is known as half wave rectifier, then to remove the ac a full wave rectifier was used this arrangement was of two diodes.

This page should discuss the older technologies that have been used for high-capacity DC power systems, such as motor-generator sets and mercury arc rectifiers. -- Anonymous, 05:25 June 11 (UTC)

Have been trying to find out what a half-wave rectifier is, and there's nothing on the page itself about this.

This page should also discuss controlled rectifiers

Don't cut and paste
Don't cut and paste content from "Allaboutcircuits.com", it's an infringement of copyright, which takes only seconds to detect. If you do add content from another source, don't drop it hodge-podge into an article; make some attempt to merge it sensibly with the organization of the article. --Wtshymanski (talk) 19:06, 12 February 2008 (UTC)

Three-Phase Bridge Rectifier
It is "Three-Phase Half Bridge Rectifier", but not "Three-Phase Bridge Rectifier". For "Three-Phase Bridge Rectifier" is need 12 diodes (Three parallel Gretz Bridges).195.208.208.111 (talk) 15:01, 15 May 2008 (UTC)
 * No, a full-wave 3-phase bridge needs only 6 diodes. --Wtshymanski (talk) 17:59, 15 May 2008 (UTC)

It is not a full-wave 3-phase rectifier. It is not Full Bridge rectifier. It is "3-phase Half Bridge rectifier. 3-phase Full Bridge rectifier needs 12 diodes.195.208.208.111 (talk) 09:13, 16 May 2008 (UTC)
 * You may be thinking of a reversing converter which uses 12 SCRs to obtain operation in all four quadrants, or possibly a 12-pulse rectifier which is fed by both star and delta windings (30 degree phase shift, 6 phases). However, a full-wave three-phase bridge rectifier only needs six diodes - I'm looking at Table 13-5, page 13-24 in the 11th ed. of Standard Handbook for Electrical Engineers which shows a full-wave 6-pulse rectifier, using both halves of the input AC waveform. --Wtshymanski (talk) 14:21, 16 May 2008 (UTC)

It is 3-phase 3-Half Bridge rectifier, known as "star-Larionov" rectifier on 6 diodes. 3-phase 3-Full Bridge rectifier consist of 3 parallel Gratz Bridges. Every Gratz Bridge consist of 4 diodes. All 3-Full Bridge rectifier consist of 12 diodes. These rectifier is new and litle known. Yow can see russian page (article) "Rectifier" ("Выпрямитель") in russian section of Wikipedia.195.208.208.111 (talk) 15:00, 16 May 2008 (UTC)
 * You have the advantage over me of reading and writing both languages, but could you look at http://ru.wikipedia.org/wiki/%D0%98%D0%B7%D0%BE%D0%B1%D1%80%D0%B0%D0%B6%D0%B5%D0%BD%D0%B8%D0%B5:Full-wave_rectifier3.png

and explain why you want to use 12 diodes when 6 suffice? "Gratz bridge" and "star Larionov" are terms unknown to me. All my other references say 6 diodes are enough for a full-wave bridge., --Wtshymanski (talk) 00:38, 17 May 2008 (UTC) If you have one phase voltage you can rectify it by one Full Bridge rectifier on 4 diodes (Gratz Bridge). If you have 3-phase voltage you can rectify it by 3 Full Bridge rectifiers on 4 diods ich (3 Gratz Bridge) (12 diods). After that you can connect 3 Full Bridges parallel or secventionaly and get 2 new 3-phase rectifiers: "3 parallel Full Briges" and "3 secventional Full Bridges". After that you can see, what known 3-phase rectifier "star-Larionov" or "delta-Larionov" on 6 diodes is Half Bridge rectifiers. Rectifier "3 parallel Full Bridges" have litle internal activ resistance than "delta-Larionov" and litle temperature of cuprum of coil and litle lost of energy (fuel) on heating of cuprum of coil. From thise rectifiers only rectifier "3 secventional Full Bridge" is full-wave rectifier, anower are partitional-wave rectifiers.195.208.208.111 (talk) 09:23, 19 May 2008 (UTC)
 * I think you mean copper when you say cuprum. I suggest you draw out the schematic diagram of a 12-diode bridge and I think you'll see that you have several pairs of diodes effectively in parallel. Given that, you can replace the pair of diodes with a single diode and still have a full-wave bridge rectifier with only 6 diodes. Even the Russian Wikipedia page you referenced shows a 6-diode bridge and no 12-diode, 3-phase bridge at all. I believe you are mistaken. --Wtshymanski (talk) 14:03, 19 May 2008 (UTC)
 * Yes, I mean "copper" when I say "cuprum". Yes, if you have schematic diagram of 12-diode bridge, you can replace the pair of diods with a single diode and still have a not full-wave not bridge rectifier, but particaly-wave half-bridge rectifier with only 6 diods, known as "delta(triangl)-Larionov", thise rectifier have another internal activ eqvivalent resistance, than 12-dide bridge. Yes, Russian Wikipedia page has not schematic diagram of 12-diode bridge, but have text with 12-diode bridge. I not mistaken.195.208.208.111 (talk) 09:04, 20 May 2008 (UTC)
 * Sorry, I believe you are incorrect as I can find no reference that says 12 diodes are necessary for a 3-phase full-wave bridge. And I've been on the General Electric DC drives course - I'm sure they would have mentioned this! --Wtshymanski (talk) 17:57, 20 May 2008 (UTC)
 * Full-wave rectifiers are the 1-phase full-bridge rectifier (Gretz bridge)(4 diods), 3-phase 3 secventional full-bridge rectifier (12 diods) and the same. The 3-phase 3 half-bridge rectifier (Larionov) (6 diodes), 3-phase 3 parallel full-bridge rectifier (12 diodes) are not full-wave. In these rectifiers small parts of sinus wave are not used. You no find reference becorse thise rectifiers are new and small known. I`m sorry.195.208.208.111 (talk) 06:50, 21 May 2008 (UTC)

I'm doing to write the rfp for rectifier, precision air, ups and grounding. I try to get more information, but not yet. And next year i'll propose new data center, that's include all facility. Howerver if who's can help me. Please contact me by e-mail:praewpat@dtac.co.th. —Preceding unsigned comment added by 202.91.23.4 (talk) 12:32, 26 August 2008 (UTC)
 * See HVDC for more information on the use of 12 diodes. Biscuittin (talk) 19:47, 8 October 2009 (UTC)

Article layout
Solid state diodes are mentioned in the lead section but not under "Rectification technologies". Is this intentional? Biscuittin (talk) 19:51, 8 October 2009 (UTC)


 * You seem to have spotted a big hole in the article, although the section on "Selenium and copper oxide rectifiers" is, in fact, describing solid state devices and does manage to mention silicon in passing. The whole assumption of the section seems to be that silicon is taken for granted as the technology of choice and it only covers other technologies in detail.  It should also mention germanium, which at one time had widespread applications, at least in the small-signal electronics world, now overtaken by silicon, and historically probably more important than selenium whose application was mostly limited to automotive.  Sp in ni  ng  Spark  22:00, 8 October 2009 (UTC)
 * OK, I have added a brief section on solid state diodes with a link to the main article Diode. Biscuittin (talk) 22:41, 8 October 2009 (UTC)

Synchronous rectifier
The "Synchronous rectifier" section needs attention. I have never heard of an electro-mechanical device of this type being used on a locomotive, and both the links describe solid-state devices. Biscuittin (talk) 09:15, 9 October 2009 (UTC)


 * Google books gets a lot of hits for both mechanical synchronous rectifier and locomotive mechanical rectifier, many of them relevant. The statement is certainly referenceable, but whether or not it is current practice to build locomotives that way I couldn't say.  Sp in ni  ng  Spark  10:32, 9 October 2009 (UTC)
 * I suspect there is confusion between a synchronous mechanical rectifier (where a synchronous motor opens and closes contacts) and a synchronous motor-generator set (where a synchronous motor drives a dynamo). The latter system was used in locomotives in the early 20th century but I doubt if any are still in service. Biscuittin (talk) 12:16, 9 October 2009 (UTC)
 * I have added a "citation" tag and moved the irrelevant links to "See also". Biscuittin (talk) 12:28, 9 October 2009 (UTC)

Fourier Transform of a rectified signal
This article is useful and interesting, but surely some mathematical theory would fit in well. Namely, I would like to see a rectified signal described in terms of the signum function (see here), or $$\text{rect}(x)(t)=|x(t)|=x(t)*\sgn(x(t)) $$

Additionally, it might be worthwhile to note that the Fourier Transform of a rectified sinusoid, say for

$$x(t)=a(t)\sin(\omega_0 t),\quad y(t)=\text{rect}(x)(t) $$

The Fourier Transform is given by

$$\mathcal{F}(y(t))= Y(\omega)= -{2 \over \pi}\sum_{n=-\infty}^\infty {1 \over 4n^2-1}A(\omega-2n\omega_0) $$

Does this fit into this article? Is this already somewhere else I should be looking? Thanks Eccomi (talk) 03:28, 19 January 2010 (UTC)

220.225.122.254 (talk) 15:46, 30 January 2010 (UTC)what is transformer utilisation factor(TUF) in rectifiers

Useless formulas
This edit restored formulas which purport to give invormation about DC and RMS output voltages of rectifiers. I contend they are useless because useful rectifiers have filters, and there is no obvious relationship between the values given by these formulas and the design criteria for a filtered DC rectifier. Also, since filtered DC rectifiers are the norm, people are constantly driving by and "fixing" the formulas, perhaps mentally inserting a filter capacitor into the schematic even though it isn't there.

Also I challenge the formulas and am removing them now as unreferenced material. Jc3s5h (talk) 02:09, 18 December 2010 (UTC)


 * The reintroduction of formulas places new demands on the diagram. Without formulas, the diagrams could be thought of as qualitative. Now that the formulas are present, they must be quantitatively correct. I submit that the formula are either incorrect or impossible to evaluate with the present diagrams. In particular, how can a bridge rectifier and half wave rectifier have the same RMS voltage without a filter? Jc3s5h (talk) 16:41, 19 December 2010 (UTC)
 * Well, we could keep edit warring, or *someone* could actually fix the formulas to match the diagrams, etc. I'd go do it, but both my arms and legs are broken and my eyes have been pecked out by vultures. --Wtshymanski (talk) 21:03, 19 December 2010 (UTC)

Funny picture
the picture in the article http://upload.wikimedia.org/wikipedia/commons/thumb/2/2a/VacRect2E.png/220px-VacRect2E.png

I really like the picture, I dont think its inappropriate but the depressed smiley is very funny... is this really coincidence? not only is the mouth sad, also its eyebrows... — Preceding unsigned comment added by 157.193.12.135 (talk) 22:20, 15 August 2011 (UTC)

Full-wave_rectification: square root of 3 term in DC voltage calculation should be clarified
http://en.wikipedia.org/wiki/Rectifier#Full-wave_rectification

The source of the square root of 3 term in the DC voltage calculation is unclear. In summary, I'd like to propose that the square root of 3 term in DC voltage calculation be explained, perhaps with reference to another article if one exists or with an additional diagram here. --Poushag (talk) 14:00, 20 September 2011 (UTC)


 * See the discussion about the diagram above. It is incorrect and needs redoing to show that point.  Spinning  Spark  14:47, 22 September 2011 (UTC)

3-phase rectifier diagram
I have modified the 3-phase rectifier diagram here and shown alongside avoiding the clutter in the original image and keeping all the axes in same scale. --ElectroKid 19:55, 17 June 2011 (UTC) — Preceding unsigned comment added by Krishnavedala (talk • contribs)


 * That's much better, but still have some objections. The time scale (3 s) is very unlikely, more realisticaly it would be 20 ms or 16.7 ms depending on country.  A better idea would be to present a relative time scale in terms of T, ie 0, 0.1T...T.  I would also remove the Vavg and Vrms guide lines.  These are not very helpful and potentially confusing - the article talks about average and rms voltages of the output waveform whereas these are of the input waveform.  A very minor point; the suffixes should not be italicised as the suffix does not represent a quantity.  Spinning  Spark  23:05, 18 June 2011 (UTC)


 * Updated eith the suggestions. Also, the source code in the description has been rewritten to reflect the changes. --ElectroKid (talk • contribs) 19:31, 30 June 2011 (UTC)


 * Yes, I like that one. You still have the suffixes italicised, it should be Vpeak, not Vpeak.  The word "peak" is not referring to a quantity, although the text of the article needs correcting for this error as well.  Spinning  Spark  21:09, 30 June 2011 (UTC)
 * Now fixed  Spinning Spark  22:39, 30 June 2011 (UTC)

Both the original (suckindiesel's) png and your svg regarding 3-phase full-wave rectification are at odds with the graph in "Power Electronics: Converters, Applications, and Design" by Mohan, Undeland & Robbins, Wiley, 3e. Specifically, please look at Figure 5-32, which is reproduced here (http://www.4thintegrationconference.com/downloads/Basic%20Power%20Electronics%20Concepts_Ozipineci_ORNL.pdf) on pdf page 18. You'll notice that the peaks of the full-wave rectified signal are not simultaneous with the peaks of any of the three input phase lines. Additionally, the peak amplitude of the full-wave rectified signal appears greater than half the peak-to-peak amplitude of any input phase. I have not compared your half-wave rectified graph. I have also not looked at your MATLAB code, but if I were to guess, I'd imagine it's been extended from a single phase rectification scenario. I'll have a look as time permits, thx. — Preceding unsigned comment added by 70.239.5.174 (talk) 18:21, 30 August 2011 (UTC)


 * Agreed on both points. The reason is that the output is related to the line-to-line voltage whereas the input is related to the phase voltage.  The relation between them is $$\scriptstyle V_\mathrm{LL}=\sqrt 3 V_\mathrm p$$ in amplitude and 60° phase difference.  Spinning  Spark  20:30, 30 August 2011 (UTC)


 * Hmm, I'm familiar with the line-line and line-neutral relationship, but are you saying that Wikipedia's graph and the one in "Power Electronics" agree? Also, I just noticed that the time scale increments are not uniform.  Nonetheless, would you agree that the blue and green signals (above) reach their peak difference (line-line) at 2/3 the distance from the "0.5T" to "0.8T" label, yet the full-wave rectified output shows a minimum at that point? (orig:70.239.5.174) 70.239.12.78 (talk) 19:15, 31 August 2011 (UTC)
 * No, I am agreeing that Power Electronics is right and the Wikipedia diagrams are wrong. The scale on the diagram is uniform, it is just incorrectly marked.  I think the editor has rounded 0.25 > 0.2 and 0.75 > 0.8 etc.  The output should show peaks at 0T, 1/6T, 1/3T, 1/2T, 2/3T and 5/6T which I think agrees with your rather more complicated way of expressing it if the rounding error to 0.8 is removed.  Spinning  Spark  20:21, 31 August 2011 (UTC)


 * I think I figured out what went wrong. I did plotted the individual diode potentials whereas the actual signal is the difference between the two diode potentials. I plotted them and they turned out to match the $$\scriptstyle\sqrt{3}V_{\mathrm{peak}}$$ potential (line# 106 in the source code). The time scales were indeed rounded, now they are good. Please verify. Thanks for the inputs. --ElectroKid (talk • contribs) 17:25, 22 September 2011 (UTC)


 * I would question what the dotted plots of rectified phase voltage in the third diagram now mean. Those waveforms do not exist anywhere in the 3-phase rectifier circuit.  I would suggest you substitute the unrectified phase waveforms which will be both more meaningful (diagram will then show input and output voltages) and would be consistent with the half-wave diagram.  Spinning  Spark  19:59, 22 September 2011 (UTC)
 * Done.--ElectroKid (talk • contribs) 15:58, 23 September 2011 (UTC)

Flame rectifier redirects here - but no section on it
Flame rectification redirects here, and this article refers to flame rectification in several places. We should really have either a section to describe flame rectification, or a page dedicated to it.

If someone looks for a page on "flame rectification" they almost certainly know what a rectifier is - so pointing them at this page doesn't make sense if we don't have any explanation of what _flame_ rectification is. — Preceding unsigned comment added by 38.113.0.254 (talk) 20:46, 21 December 2011 (UTC)


 * There is no reason to believe that readers looking for flame rectification will know what rectification means. In fact, the obvious literal meaning of flame rectification has no connection with the electronic device.  It will probably come as a surprise to many electrical engineers that flame has a rectifying property.  Anyway, you are welcome to add something to the article on this.  Here's a source to help you write it.  Spinning  Spark  21:02, 24 December 2011 (UTC)

Questions
1. "Unlike the rectifier above, only aluminium electrodes were used, and used on A.C., there was no polarization and thus no rectifier action, but the chemistry was similar."

This sentence is strange. Also, why is it "A.C." and not just "AC"?

2. "The electrodes were shaped such that the reverse breakdown voltage was much higher than the forward breakdown voltage. Once the breakdown voltage was exceeded, the 0Z4 switched to a low-resistance state with a forward voltage drop of about 24 V."

Is it forward or reverse breakdown voltage?

ICE77 (talk) 05:16, 27 December 2011 (UTC)


 * 1. is indeed a confused sentence. The point that the device could be used in an AC circuit seems to be unrelated to the use of the device as a lightning arrestor.  The points should probably be made separately.
 * 2. Presumably the device was not intended to break down in the reverse direction. When in the forward direction it did break down (at some unspecified voltage) after which the voltage dropped to 24 V until the next half cycle put it into reverse bias again.  Spinning  Spark  12:09, 27 December 2011 (UTC)

Full wave centre-tap rectifier
I have reverted the text of this description. An editor did not seem to like the back-to-back description although this is perfectly correct and its meaning is explained in brackets. Possibly the misunderstanding was caused because the layout of the drawing does not make clear the back-to-back topology. The replacement text though, was very confusing talking of a diode on each outer pole. The circuit requires both diodes to be connected to the same pole of the output. While the editor undoubtedly meant to say the poles of the transformer it has ended up with something the reader has no chance of understanding unless they know already.

By the way, the statement about the voltage being half a full wave rectifier is not really correct, it depends entirely on the turns ratio of the transformer.  Sp in ni ng  Spark  16:18, 19 March 2009 (UTC)

it should be pointed out that the full wave center tap rectifier is two half wave rectifiers back to back and maximum current cannot be obtained until more diodes are added - in a simular arrangement to the three phase bridge as a matter of fact. Charlieb000 (talk) 10:37, 17 March 2012 (UTC)

Error in equation
There appears to be an error in the equation for the average DC output voltage of a 3-phase rectifier, in the section on full wave rectification. There is a factor of 2 missing in the divisor. Thus the voltage caculated from this equation is too high by a factor of 2. — Preceding unsigned comment added by 175.145.64.170 (talk) 09:32, 6 July 2012 (UTC)


 * What makes you think that?  Spinning Spark  23:18, 6 July 2012 (UTC)

If you apply the equation as it stands, to, for example, the direct rectifiaction of 400V 3-phase mains, the answer is 935Vdc. This is definitely not so with a 3-phase full-wave bridge rectifier. I am an electrical engineer who has designed and applied rectifiers of many types for over 30 years in my career. I have also tutored students in power electronics at a technical college. Follow this link for an extract from a textbook on the subject: http://web.ing.puc.cl/~power/paperspdf/dixon/21.pdf Ian.a.mckenzie (talk) 10:10, 8 July 2012 (UTC)


 * I think the input voltage in the equation is meant to be taken as the phase voltage, not the line-to-line voltage, so in your example one would start with 230 V, not 400 V. I agree this is not very clear from the text, but the phases of the waveforms in the accompanying diagrams show this to be so.  There is some discussion of this in the threads above concerning the diagrams.  I have confirmed your account so you should now be able to edit the article if you wish.  Spinning  Spark  12:32, 8 July 2012 (UTC)

Sqrt(3) factor doesn't belong. — Preceding unsigned comment added by 207.67.13.206 (talk) 21:03, 21 September 2012 (UTC)
 * Did you even read any of the above discussions?  Spinning Spark  21:54, 21 September 2012 (UTC)

Half-wave rectifiers are never used?
I have removed this from the article,


 * A half wave rectifier is rarely used in practice. It is never preferred as the power supply of an audio circuit because of the very high ripple factor. High ripple factor will result in noises in input audio signal, which in turn will affect audio quality. Advantage of a half wave rectifier is only that its cheap, simple and easy to construct. It is cheap because of the low number of components involved. Simple because of the straight forwardness in circuit design. Apart from this, a half wave rectifier has more number of disadvantages than advantages!

First of all it is factually incorrect, half wave rectifiers are the norm for peak detectors and envelope followers. Claiming it has high ripple requires some further explanation, the ripple at the output of a rectifier with nothing connected to the load is exactly the same for a full-wave and half-wave rectifier and equal to the peak voltage. The actual ripple in practice depends on the load and the size of the reservoir capacitor. The same ripple can be achieved in a half-wave rectifier by using a larger capacitor compared to a full-wave. The real difference is that doing so will result in larger current pulses in the half-wave case. Thus, a higher current rating device is required (as well as a larger capacitor) and the result may well not be a cheaper circuit at all. On the other hand, vacuum tube circuits have better current handling and the cost of the devices was significant, unlike silicon diodes for a low-power rectifier. I would prefer to see something here cited to a reliable source. At the very least the last sentence (or two sentences) needs deleting because it is repetitive and uses the unencyclopaedic exclamation mark.  Spinning Spark  19:14, 6 November 2013 (UTC)


 * Another factor is that the full-wave rectifier ripple is twice the frequency of the half-wave making it easier to filter.  Spinning Spark  19:17, 6 November 2013 (UTC)

Addition to full wave section
The circuit description just added to the full wave section is a bit problematic. For one thing the description says the current delivered is lower on the negative half cycle so this is not really full-wave. Perhaps 3/4 wave is more accurate. More importantly, this is such an obscure design that I don't think it really belongs here, even if it does work as claimed, especially as it is unreferenced and is not covered in numerous electronics texts. I also don't understand how this is supposed to be used as an inverter: the output is dc, not ac.  Spinning Spark  21:51, 11 April 2014 (UTC)

Author of adittion: 1) I have no idea how to reference chrestomatic things I had used at least my last 30 years. One of possible references are http://www.danyk.cz/reg60v.html in circuit``s most rightest side. Yet I was finding this circuit at russian `Radio` journal of 70~ties, and picked it up to my arsenal. Never I met it mentioned at simple educative literature. 2) Sorry about pictures used via dropbox, simply wikipedia engine not allows me to add anything into picture format. If You are permitted and able, please help to convert given links to Wiki standard pictures. Thanks in advance. 3) About full wave or half wave. My understanding of full wave is when current in the load flows all the time, but for half wave rectifier flows only 50% of time. Here is 100%, therefore it is undoubtly full wave model. However, if we define the full wave by side of input, then indeed this is less than 100%, but I see obvious any definitions from side of result - who between users is interested how to get result except us designers. 4) To have a place or have not into wikipedia. I see it obvious, if something exist over the Planet, it MUST have a certain mention-place at wikipedia. If sweep it out completely, simply good and very effective thing will be annihilated of collective memory of civilization. If You argue that this is not the best location - okay, but where to put it then? I am bit skeptic to build own term - 3/4 wave rectifier. It is nice and logic term, but never never heard. I think the wikipeadia ought not to be originator of new terms. 5) Inverter or converter. Any device using half-bridge or full bridge to feed the ferrite transformer must be labeled inverter, in no respect do it consumes DC or AC, and do it gives on the load the DC or AC. In the circuit I referenced, input is AC and exit DC, but it can be in few minutes resoldered to have DC input and AC output or AC-->AC or DC-->DC. It is just universal inverter. As I told, inverter engineers are competiting for efficiency of device over 90%. If diode has at least 0,8V U(F), then at say 100 A current the each diode losses alone will be 80W or 2%. If to use 4 diodes or 2 diodes, that is big difference there. Thus, probably this is the best rectifier circuit over the history, You said You want to kill. — Preceding unsigned comment added by 109.229.204.220 (talk) 16:10, 12 April 2014 (UTC)
 * Firstly, the issue of pictures is easily fixed. I am an administrator here and can give you the file upload permission.  But first you must open an account (and then tell me the account name).  It is even easier to upload to  where you are not required to have your account confirmed and has the additional advantage that the images can then be used on all Wikimedia projects, not just English Wikipedia.
 * I won't argue with you any more on what the circuit should be called or where it belongs, it is not important. However, you are wrong that "if something exist over the Planet, it MUST have a certain mention-place at wikipedia".  We have criteria for inclusion—one of the most important is verifiability, which means that this thing that "exist over the Planet" must be found in reliable sources.  I cannot read the source you linked to but Google translate seems to tell me that it is a private website describing a home project construction.  It is not an acceptable source.  Information that cannot be verified does not belong here.
 * What we need is a source, or sources, from a textbook or a description in a scholarly journal. I have searched through quite a few books now, including Advanced DC/AC Inverters and can't find anything that looks remotely like it.  Sources do not have to be in English, a reference to a Russian language source would be fine if you have one.  Spinning  Spark  18:44, 12 April 2014 (UTC)

I have removed the following text from the article for now pending discussion and provision of sources. At the very least it needs cleaning up and that cannot be done properly without a good source to refer to.  Spinning Spark  15:47, 14 April 2014 (UTC)

External links modified
Hello fellow Wikipedians,

I have just added archive links to 1 one external link on Rectifier. 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 https://web.archive.org/20140105172551/http://digitalcontentproducer.com:80/mag/avinstall_ac_dc to http://digitalcontentproducer.com/mag/avinstall_ac_dc/

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 16:39, 28 August 2015 (UTC)

The DC and RMS voltage equations are totally f***ed
They basically imply that the voltage frequency changes based on the peak voltage. Otherwise how would the ratio between a curve and a line be equivalent? I'm not going to extensively cover the whole article being generally wrong, since everything here about voltage is highly plausible. Germanium diodes dump black off of unbalanced nothing, and silicon diodes dump silver off of unbalanced nothing. They've got bands. Get it? "Ban"ds? That's how diodes work. My book was wrong. Seems like somebody modified this article since I wrote it.
 * Which section? I can't see the bit you're complaining of. Andy Dingley (talk) 09:13, 24 October 2015 (UTC)
 * Nothing you have written makes any sense, except the hint that you might be banned. SpinningSpark 10:39, 24 October 2015 (UTC)

"Half-wave rectification" or "Half wave rectification"?
Should it be half-wave rectification or half wave rectification?

Sample: subsection Half-wave rectification uses both.

--Mortense (talk) 20:46, 2 February 2016 (UTC)


 * The discussion on the language desk suggests that the hyphenated form is preferred, so I'll make that change.   D b f i r s   10:14, 7 February 2016 (UTC)

Dubious circuit, probably dangerous.
In the section headed "Rectifier Output Smoothing", there is a diagram with a caption commencing with "Rc filter... ", which depicts a single phase bridge rectifier with the negative side of the DC grounded. I doubt that this is actually possible unless you want to drag large currents to and from your earth/ground connection.`Lathamibird (talk) 03:10, 27 March 2017 (UTC)
 * This arrangement is very common and there are no problems provided that V1 is isolated from mains ground (i.e. a transformer secondary). Also SPICE simulations need a "ground" node 0 to work properly, even if it is not connected to a real ground/earth electrode. SV1XV (talk) 03:41, 27 March 2017 (UTC)

Rectifier output smoothing
This section reads like a summary of a college paper. There's no formulas or math. At a minimum, for each elementary filter, we should give the voltage output, ripple voltage output, and ripple current output. We don't discuss frequency domain ripple of filter types (i.e. non-Butterworth, etc), noise filtering (SMPS), resonance effects, voltage harmonics, current harmonics, effect of power factor and voltage regulation of the transformer, impact of source voltage and load variation. On the output side, we don't discuss voltage regulation and current regulation of the filter choices and rectifier choices (particularly, the voltage multiplier types). The companion articles Capacitor input filter, Choke input filter and Series resonant choke either aren't created or they're stubs.<\brain dump> Sbalfour (talk) 00:22, 28 October 2017 (UTC)

The diagram here shows a rectifier and capacitor input filter; however, the resistor's role is ambiguous: in many such diagrams, the load is represented by a resistor, but it could also be construed as part of a parallel RC shunt filter, or it could be a bleeder resistor. I preemptively relabeled the diagram as if the resistor were part of the filter, but I hardly ever see such a filter - it beggars an explanation. Why only an example of a cap input filter? How about a choke input filter; and where might a voltage regulator appear in the circuit? A nice set of diagrams is what is needed here. Sbalfour (talk) 17:14, 28 October 2017 (UTC)

The section title is pretentious; the filter following a rectifier has to do a whole lot more things than reduce ripple, and they're equally vital:
 * adjust voltage to that required by the circuit
 * ensure smooth current delivery
 * do voltage regulation
 * supply reactive power to compensate for imbalances between VA of load and source
 * filter 'inherited' noise like RFI and EMI, and harmonics of the line frequency
 * filter generated noise like SMPS switching noise, diode switching noise, frequency domain ripple
 * damp resonances caused by stray capacitance, leakage inductance, etc
 * protect against pathologies like component shorts and opens, inductor flyback current, inrush current, voltage spikes, etc

Sbalfour (talk) 19:31, 28 October 2017 (UTC)
 * How do filters do half of these things, such as voltage regulation?  Now these things are usually done, but it's not a filter that's doing them. Andy Dingley (talk) 23:14, 28 October 2017 (UTC)
 * A bit of literary license: of course, a voltage regulator if part of the circuit, is not a filter, and neither are other things like flyback diodes, bypass capacitors, leakage resistors, etc. Components between the rectifier and circuit need to do these things, and are part of the consideration when employing rectification.  Sometimes we don't need any of them, like battery chargers.  I do think we should cover the basics including regulators in this section, but maybe gloss over "how to" issues like flyback diodes, etc. Cheers, Sbalfour (talk) 16:24, 29 October 2017 (UTC)
 * I think you need to bear in mind that Wikipedia articles are not meant to be a design manual nor a student textbook. There is a place for formulae on Wikipedia, but it is neither necessary nor desirable to cover every detail of design.  Many of the functions you mention are not intrinsically welded to rectification.  Those circuit blocks can be considered (more or less) separately.  The primary things that do need to be considered together with the rectifier are the resevoir capacitor and the load impedance. Subsequent filtering is (more or less) a linear circuit consideration and can be treated separately. SpinningSpark 17:02, 29 October 2017 (UTC)
 * I'm a bottle head, and see as many choke inputs as capacitor inputs ;-) so I feel obligated to scrawl about them. And in such circuits, I never see an SS regulator,  of course.  A rectifier only does part of the job of rectification, <<- and there isn't a separate article for that.  Where do we discuss the rest?  We have to carefully consider superposition of AC & DC current, voltage, and power there, and the math as well as engineering aspects are intricate.  I've written in a topical style, which you seem happy with, so I'll try to stick to that for now.  We rather passively mention load current in connection with input capacitor sizing; what do you want to see there?  If the load has highly variable impedance, the filter will have to do some impedance matching...  Sbalfour (talk) 17:36, 30 October 2017 (UTC)

2-phase rectification
In the United States, most homes are supplied with 3-wire "two-phase" power: 2 hot wires and one neutral, with the hot wires being each at potential 120V, but 180 degrees out of phase. The hot wires are therefore at potential 240V with respect to each other, and 240V can be supplied for such things as air conditioners, electric ovens and clothes dryers, electric heating, arc welders, electric and hybrid vehicle battery chargers, and heavy duty power tools. Some of those uses would be DC power, i.e. rectified. Do we need to cover "2-phase" rectification in sections like Rectifier Circuits and Conversion ratio? Sbalfour (talk) 14:51, 31 October 2017 (UTC)
 * It's not two phases, it's split phase. I'm not saying no-one has ever hooked diodes to the two legs of a split-phase system to make DC, but I've never seen it - and it would be a full-wave rectifier as we already discuss. --Wtshymanski (talk) 21:30, 31 October 2017 (UTC)

Rectifier efficiency
I'm dumbfounded by what this section could mean. There is an earlier discussion of this above, in a similarly named talk page section. Three editors wrote that they don't know where it came from, and two that it isn't meaningful; one deleted the entire section, and someone later restored it verbatim. The primary source from which the section seems to have been copied (altered just enough to escape copyvio) is a set of slides by a post-grad student teaching a college class to earn his PhD. They're posted on the class website. Grrrrrrr... Would a college student's essay be a good enough source? How about a high schooler's? Where do we draw the line? We draw it here: the source must be a professional peer-reviewed journal with a reputation for accuracy. So I'm deleting the source. That leaves the section unsourced, except for the last sentence.

I was so shocked to read that rectifier efficiency is only 81%, that I immediately got out my breadboard and checked it, because I'd never seen that kind of loss in a circuit. I dropped a resistor across the input, then across the output of the rectifier, and measured the available power. The ratio was marginally less than 100%; there was a small amount of dissipation in the diodes. Then I conducted a second experiment by placing a large capacitor at the output of the rectifier, and repeated what I did in the first experiment. Again, the ratio of available power on input and output was nearly 100%, minus a very small additional loss through the ESR of the capacitor. Only real power can heat a resistor, and a capacitor cannot add any real power to the circuit - all real power comes from the source. So, a capacitor cannot change the "efficiency" of rectification. Whew!

I do see in the literature, that this computation is sometimes called "conversion ratio", or "rectification ratio", rarely to never, "rectifier efficiency"(that's why a hokey college paper got cited). So I've renamed the section. But the prior objections that it is meaningless remain; just because we can calculate something, and do calculate something, doesn't mean we should calculate something. Rectifier output is part of a process of rectification, completed by a filter, and almost never used without a filter, so it's darn useless to specify anything about it.

Now, the section is mostly two numbers, possibly not even correct (who says?), preceded by some trivial arithmetic. I can do the arithmetic mentally, it doesn't need to be there - it's filler text - so I'm deleting it. Now it comes down to what really DOES need to be there - where these magic ratios of $${V_\mathrm{peak} \over 2}$$, $${I_\mathrm{peak} \over 2}$$, $${V_\mathrm{peak} \over \pi}$$, $${I_\mathrm{peak} \over \pi}$$, also possibly not correct, come from (PROVE IT, and when you've done so, place it into the encyclopedia, with reliable source). Ahhhhhhhh..., now that's a tough nut, because you need the calculus to get there, and probably the editor who copied the section wasn't aware of that, and even if he was, probably couldn't do it.

Then, like all circuits, it should be verified in the lab. Stick a rectifier on the breadboard, and measure the AC and DC power components of the output. And how do you do that? I don't mean with a DMM that conveniently splits the voltage into an AC voltage with DC offset - how do we know that's reliable? Pretend you're tasked with showing the DMM is right. I think you might find the experiment quite challenging.

I humbly submit that this section is junk; don't we have something more relevant to say?

Sbalfour (talk) 20:50, 29 October 2017 (UTC)

What this section really should be, is a set of descriptions of various measures of rectifier circuit function, relating various measures to each other (including conversion ratio, as one of them). I've found TUF:transformer utilization factor; RF:ripple factor; FF:form factor, peak factor.

Sbalfour (talk) 21:54, 14 November 2017 (UTC)

At least one of the formulas given for the three-phase full-wave rectifier must be wrong, because dividing them gives a conversion ratio much greater than 100%. 209.179.92.189 (talk) 04:56, 28 October 2019 (UTC)