Talk:Leakage inductance

Validity of sources II

Revision as of 19:35,27 November 2017 provided maintenance tags for Refined inductive leakage factor section including in terms of single source tag template.

Attributing a single source tag for Refined inductive leakage factor section, which relies on Hameyer 2001 source is not justified because the accompanying Refined inductive leakage factor derivation infobox shows how using multiple sources other than Hameyer 2001 source the two pats of the same section derivations arrive at the same end result. That is, the single-source attribution is in fact a multiple source section proving that the section as a whole stands together in a robust manner. The onus is on other editors to disprove the two streams of derivation of the section.

In DNA analysis this proof of ancient ancestor's origin based on living DNA testing is called triangulation. This is what happened with Refined inductive leakage factor section - Multiple-sourcing by triangulation.Cblambert (talk) 05:26, 8 August 2018 (UTC)

Its been awhile. Isn't the Hameyer 2001 a citation to course notes? Constant314 (talk) 05:36, 8 August 2018 (UTC)

Yes. I was busy on heavy-duty project which pre-occupied me. Hameyer 2001 was an extensive unpublished book-quality document targeting a student audience available online, which document has been cited in a number of papers based on a superficial online search. I have provided Hameyer's academic & research credentials in Talk section immediately above.Cblambert (talk) 05:55, 8 August 2018 (UTC)

If it is available on-line, that would be great. Can you provide the link. It is not obvious. Constant314 (talk) 15:53, 8 August 2018 (UTC)

As mentioned in previous Talk section, it is unfortunately no longer available online. I will call Hameyer to see if I can get a copy of the pdf that used to be posted online. Be that is it may Hameyer 2001 is a valid source because there must be millions of sources on Wikipedia that were once available but though no longer readily available are still considered to be valid sources.Cblambert (talk) 20:16, 8 August 2018 (UTC)

In any case, Hameyer 2001 is shown to be valid by triangulation reflected in Refined inductive leakage factor derivation infobox.Cblambert (talk) 20:26, 8 August 2018 (UTC)

I have retrieved the Transformer article change in which the link to pdf http://materialy.itc.pw.edu.pl/zpnis/electric_machines_I/ForStudents/Script_EMIHanneberger.pdf isremoved from Biblio reference: 06:11, 17 March 2013 (diff | hist) . . (+270) . . m Transformer (Clean up Hameyer reference)Cblambert (talk) 21:25, 8 August 2018 (UTC)

It appears to be a dead link. Constant314 (talk) 22:12, 8 August 2018 (UTC)

Yes. It is dead link, which is why it was removed from the Transformer article change mentioned in my last comment here.

Also, making allowance for fact that WP articles must not contain original research, the write.com link at Data Triangulation: How the Triangulation of Data Strengthens Your Research provides interesting insights into data triangulation applicable for certain WP articles.Cblambert (talk) 22:24, 8 August 2018 (UTC)

See also Old url for Hameyer section above signed Cblambert (talk) 12:44, 17 January 2017 (UTC) - - - Sign up for this comment: Cblambert (talk) 23:00, 8 August 2018 (UTC)

Data triangulation seems to be about data credibility, not reference credibility, although there is no reason to argue that Hameyer himself is not an esteemed author.

The long and short of it is that Hameyer 2001 was significantly cited in connection with Transformer GA article as well as for Induction motor and Leakage inductance articles using old url, the pdf of which is no longer available online.Cblambert (talk) 23:25, 8 August 2018 (UTC)

If it is no longer available anywhere, then it doesn't serve the purpose of a reliable source as it it cannot be checked by other editors.Constant314 (talk) 01:52, 9 August 2018 (UTC)

Can you access Knowlton 1949? I can but the vast majority of WP editors can't? Yet the source is reliable. You can access Brenner & Javid but the vast majority of WP editors can't? Yet Brenner & Javid is a reliable source. You have not reacted to my triangulation argument about Refined inductive leakage factor derivation infobox, which by definition is valid. I any case I will phone Hameyer.Cblambert (talk) 03:25, 9 August 2018 (UTC)

Wikipedia says:

• "The term "published" is most commonly associated with text materials, either in traditional printed format or online. However, audio, video, and multimedia materials that have been recorded then broadcast, distributed, or archived by a reputable party may also meet the necessary criteria to be considered reliable sources. Like text sources, media sources must be produced by a reliable third party and be properly cited. Additionally, an archived copy of the media must exist. It is convenient, but by no means necessary, for the archived copy to be accessible via the Internet." Hameyer 2001 is available from reliable 3rd party -- RWTH Aachen University.Cblambert (talk) 03:36, 9 August 2018 (UTC)

Until this issue is clarified to other WP editors' satisfaction, I have just a few minutes earlier:

• Removed Refined leakage factor section
• Retained Refined leakage factor infobox
• Deleted Item f & Eq. 3.7e from Refined leakage factor infobox

The ball is in my court.Cblambert (talk) 04:03, 9 August 2018 (UTC)

Do you have access to Hameyer 2001?Constant314 (talk) 19:17, 9 August 2018 (UTC)

I unfortunately never downloaded a copy of the pdf. I have called Hameyer with no answer. Will try again but earlier in the day but he may be on vacation . . .Cblambert (talk) 23:04, 9 August 2018 (UTC)

'Well! I'll be darned!' It turns out that the Hameyer 2001 (cum Hameyer 2004) document is actually available at the link https://web.archive.org/web/20130210003139/http://materialy.itc.pw.edu.pl/zpnis/electric_machines_I/ForStudents/Script_EMIHanneberger.pdf. I came across in reviewing the Hameyer 2001 reference cited in the Induction motor article. I have duly downloaded a copy of the errant pdf link and would encourage others to do the same before the link disappears again. This is really good news but what a do so and waste of time and keystrokes. I will soon be restoring the Refined leakage factor section & Retained Refined leakage factor sidebox to the way it was yesterday. There is surely a lesson to be learned by Wikipedia from this experience including in terms of a pressing need to create of a new Data/DNA/etc./etc. triangulation article. 'Hooray!'Cblambert (talk) 16:07, 10 August 2018 (UTC)

Note that the single source issue invoked so emphatically in this Leakage inductance article clearly seems to be a red herring if one considers Wikipedia's treatment in Articles with a single source. That is, there can be no question that the 'Magnetizing and leakage flux in a magnetic circuit' diagram and associated equations can somehow be derived from a source other then Hameyer 2004.Cblambert (talk) 18:20, 10 August 2018 (UTC)

We're still not out of woods. I notice that this new link dated 2004 shows improperly formatted math symbols throughout the document. I have sent an e-mail to IEM RWTH Aachen University's Petra Jonas-Astor & Kay Hameyer asking them to provide a properly formatted document, which is known to have existed in the 2001 version of the document.Cblambert (talk) 21:44, 10 August 2018 (UTC)

Yes, I noticed that also. Constant314 (talk) 23:14, 10 August 2018 (UTC)

In meantime, I have added Erickson & Maksimovic 2001 and Kim 1963 references in bibliography and associated footnoted citations to Refined inductive leakage factor section's Fig. 6 caption and 1st sentence to support Hayemer footnoted citationsCblambert (talk) 21:13, 12 August 2018 (UTC)

Latest changes to Refined inductive leakage factor section & Refined inductive leakage factor derivation sidebox make it clear there is enough triangulation of sources and equations to make it clear that Hameyer 2001 is not actually needed to prove Eq. 3.7a to Eq. 3.7e.Cblambert (talk) 00:22, 14 August 2018 (UTC)

I think equation 2.2 where several definitions of "a" are given, needs to be addressed.

$a=N_{P}/N_{S}=v_{P}/v_{S}=i_{S}/i_{P}={\sqrt {L_{P}/L_{S}}}$ ------ (Eq. 2.2)

These are only approximately equal and it is the last expression that must be used to make the math work out exactly. The other expressions are exact, I think, only when the leakage inductance is zero.

$a={\sqrt {L_{P}/L_{S}}}\simeq N_{P}/N_{S}\simeq v_{P}/v_{S}\simeq i_{S}/i_{P}$

The expression for turns ratio under the transformer in fig. 4 sort of implies this, but suggests turns ratio determines inductance ratio whereas it is actually inductance ratio that determines $N_{P}/N_{S}$. :

If you want to keep using :$a=N_{P}/N_{S}$ Then explain that $N_{P}/N_{S}$ is not the literal turns ratio, but is the actually an adjusted turn ratio required so that $a={\sqrt {L_{P}/L_{S}}}$.

In fact, I would replace $N_{P}/N_{S}$ everywhere with $a$ except equation 2.2. That would include fig.3 and 4. The expression for turns ratio under the transformer in fig. 4 should be replaced with $N_{P}/N_{S}{\overset {\underset {\mathrm {def} }{}}{=}}{\sqrt {L_{P}/L_{S}}}$ which, if I got the right symbol says that the turns ratio is defined as the square root of the inductance ratio. I can redraw the figures if needed and address equation 2.2 if you would prefer.Constant314 (talk) 12:23, 14 August 2018 (UTC)

I have changed Eq. 2.2 to read:

$a={\sqrt {L_{P}/L_{S}}}\approx N_{P}/N_{S}\approx v_{P}/v_{S}\approx i_{S}/i_{P}=$ ------ (Eq. 2.2).

The footnoted citation for Eq. 2.2 continues to read:

Brenner & Javid 1959, §18-6 The Ideal Transformer, pp. 597-600: Eq. 2.2 holds exactly for an ideal transformer where, at the limit, as self-inductances approach an infinite value ( $L_{P}$ → ∞ & $L_{S}$ → ∞ ), the ratio $L_{P}/L_{S}$ approaches a finite value.

Quoting from Brenner & Javid, I have further added footnoted citation for Fig. 4 that reads:

"Fig. 18-18 In this equivalent circuit of a (nonideal) transformer the elements are physically realizable and the isolationg property of the transformer has been retained."

There is in my view no need to change nonideal equivalent circuit in Fig. 4 with the addition of this quoted footnoted citation exactly as taken from Brennan & Javid's Fig. 18-18. Cblambert (talk) 15:59, 14 August 2018 (UTC)

Come to think of it, strictly speaking, for Wikipeida audience consumption, a turn ratio should be called the 'ratio of the primary to secondary turns'. Erickson & Maksimovic 2001 may do well to refer on p. 33 to effective turns ratio as $n_{e}={\sqrt {L_{P}/L_{S}}}$. Trouble is Erickson & Maksimovic 2001 seem to be the only source using this turns ratio distinction.Cblambert (talk) 17:01, 14 August 2018 (UTC)

Also, as mentioned in the next section below: 'When the adjusted turns ratio N is equal to the ratio of the system-rated voltages, the ratio is called nominal, and the transformer is omitted from the single-line diagram in a per-unit system. When the adjusted turns ratio N is not equal to the ratio of the system rated voltages, it is said the transformer has an off-nominal turns ratio.' This is a not insignificant turns ratio distinction.Cblambert (talk) 17:49, 14 August 2018 (UTC)

I propose Eq. 2.2 to read:

$a={\sqrt {L_{P}/L_{S}}}=N_{P}/N_{S}\approx v_{P}/v_{S}\approx i_{S}/i_{P}=$ ------ (Eq. 2.2).

That's a good step. I think that there needs to be note regarding figure 3 that the circuit is a correct model no matter what value is used for a, so long as the transformer in the circuit is an ideal transformer. And there needs to be a note that fig 4 is derived from fig 3 by assuming $a={\sqrt {L_{P}/L_{S}}}$. Constant314 (talk) 17:26, 16 August 2018 (UTC)

Fig.3 & Fig.4 are exactly the same as Brenner & Javid show it in their figures. There is a difference between 'equivalent for all values of the "number" a' and 'no matter what value is used for a '. The note that Fig. 4 is derived from Fig. 3 by assuming $a={\sqrt {L_{P}/L_{S}}}$ IS shown from Eq. 2.12:

$a={\sqrt {L_{P}/L_{S}}}$ ------ (Eq. 2.12).

It remains that in Fig. 3 $a=N_{P}/N_{S}$ and in Fig. 4 ${\sqrt {L_{P}/L_{S}}}=N_{P}/N_{S}$, $a={\sqrt {L_{P}/L_{S}}}$ being simply used to derive Fig. 4 in the same terms as are used in Fig. 1. It also remains that Brenner & Javid say "The most popular value chosen for the "number" a is the turns ratio which the actual transformer has." Cblambert (talk) 05:28, 17 August 2018 (UTC)

It should be clear from these last few comments that the equivalent circuits in Fig. 3, Fig. 4 & Fig. 5 are all equivalent to each other, with:

• the ideal transformer in Fig. 3 being captioned $a=N_{P}/N_{S}$ because the other elements of the circuit are expressed in terms of $a$.
• the ideal transformer in Fig. 4 being captioned ${\sqrt {L_{P}/L_{S}}}=N_{P}/N_{S}$ because the other elements of the circuit are expressed in the same terms as are used in Fig. 1.
• no ideal transformer in Fig. 5 is order to depict the circuit in its simplest possible terms.

But the ideal transformer in Fig. 3 & Fig. 4 could just as well be captioned $a=N_{P}/N_{S}={\sqrt {L_{P}/L_{S}}}$, this captioning being implicitly understood in Fig. 5 as well.Cblambert (talk) 15:24, 23 August 2018 (UTC)

Redundant symbols

We have $L_{oc}^{pri}=L_{P}$ and $L_{oc}^{sec}=L_{S}$ and since $L_{oc}^{pri}$ and $L_{oc}^{sec}$ do not appear in any figures, it looks like we could eliminate $L_{oc}^{pri}$ and $L_{oc}^{sec}$ and have two fewer symbols. It might improve the readability. Constant314 (talk) 22:13, 22 August 2018 (UTC)

I don't agree. $L_{oc}^{pri}$ and $L_{oc}^{sec}$ are measurements that are equal to the respective inductances. This is significant. Maybe it should be expla'ned but deleting terms to improve readability is not a good enough reason.Cblambert (talk) 15:09, 23 August 2018 (UTC)

I propose simplifying by combining to $L_{oc}^{pri}=L_{P}$ is primary self-inductance & $L_{oc}^{sec}=L_{S}$ is secondary self-inductance as now shown in article.Cblambert (talk)

A factor at play in Fig. 1 is that the image is used in WP article of other languages.

The use of the term "Magnetizing inductance" is misleading

In figure 1 the term $L_{M}$ is defined as "magnetizing inductance", I think a better term would be "coupling inductance". The present definition suggests that the magnetizing inductance changes if the coupling coefficient $k$ changes. In real transformers, when voltage is applied to the primary and the secondary is left open circuit, there is a current that flows from the voltage source through the transformer primary. This current is commonly called the magnetizing current. The magnitude of the magnetizing current depends upon the voltage applied and the Primary inductance, $L_{M}+L_{P}^{\sigma }$, not just $L_{M}$. If the coupling coefficient went to 0 (remove the secondary) it implies the inductance responsible for magnetizing also goes to zero and so the magnetizing current should be infinite for any applied voltage. Clearly this is not the case. In fact, the coupling coefficient has no impact on magnetizing current.

Instead of calling $L_{M}$ "magnetizing inductance", if it were called "coupling inductance", I think this would avoid the potential confusion. I think it is also more descriptive. Leakage inductance is the fraction of winding inductance that does not couple to other windings, coupling inductance is the fraction that does, and the coupling coefficient defines the portion.

This is not to be confused with mutual inductance. The two terms are related but not equal.

If we replaced "magnetizing" with "coupling" everywhere in the article I think it will make the article more clear.

--IflyHG (talk) 18:49, 4 November 2022 (UTC)

I double checked. It looks correct. By "coupling inductance" did you mean "mutual inductance" which has the symbol $M$?

I think that you might be referring to equation 1.1c which I have copied here

$L_{M}=L_{P}\cdot {k}$ ------ (Eq. 1.1c)

You would expect the magnetizing inductance to be constant and not a function of k. That is more or less correct. What the equation tells you is that decrease k, you must increase $L_{p}$, the inductance measured at the terminals. Constant314 (talk) 02:27, 5 November 2022 (UTC)