Talk:Network analysis (electrical circuits)

Y-Δ article
See Y-delta transform for material moved to an existing page.

Charles Matthews 21:24, 18 Jun 2004 (UTC)

branch current
does anyone know why this method isn't mentioned on the page? Plugwash 15:13, 8 Dec 2004 (UTC)

Opening paragraph
The opening paragraph needs to be rewritten. First of all, it ought to use the phrase "analysis of resistive circuits", or the page should be moved. Secondly, I'm pretty sure the definition of when two circuits are equivalent is wrong. It currently says:
 * Two electric circuits are said to be equivalent with respect to a pair of terminals if the voltages across the terminals and currents through the terminals are identical for both networks.

I think it should be something like:
 * Two electric circuits are equivalent with respect to a pair of terminals if equal voltages across the terminals produce equal currents.

But I'm not an electrical engineer, so I'd prefer if someone else actually changes it. Dbenbenn 22:07, 24 Dec 2004 (UTC)

DC circuit analysis
Not realizing that this article existed, I created the DC circuit analysis article. Accordingly, I have added a link to this article instead of adding any additional detail to mine. Alfred Centauri 19:52, 23 August 2005 (UTC)
 * thats what redirects are for ;) Plugwash 21:42, 23 August 2005 (UTC)
 * How nice - did you bother to read my article to see if any part of it could be merged with yours? Oh, never mind.  I'll do it myself... Alfred Centauri 22:56, 24 August 2005 (UTC)
 * Done. I have planned to write two articles entitled AC circuit and AC circuit analysis.  These seem like appropriate and conventional titles to me.  Are you aware of similar articles with different titles? Alfred Centauri 23:09, 24 August 2005 (UTC)
 * There is already a note on this page on how to extend the methods to AC cuircuits which i have just expanded a bit. I don't know of any articles specifically about AC analysys though. Plugwash 23:43, 24 August 2005 (UTC)
 * Since this article is about resistive circuits, it seems reasonable to me to limit the scope to just that. The mention that these techniques can be generalized to AC circuits seems like a good place to put a link to AC circuit analysis.  The new article would not need to rehash all the methods covered here except to point out some new effects such as resonance and complex power.  My goal with these articles is not to rewrite what has already been written on Wikipedia.  Instead, I want to give a relatively brief summary of definitions, concepts, new results, and then point to other articles for additional information. Alfred Centauri 00:23, 25 August 2005 (UTC)

Article can be generalized
I propose this article be renamed "Analysis of passive circuits" or something since its not specific to resistive circuits. Roger 18:58, 27 June 2007 (UTC)

Contents of old Circuit Analysis article
The contents of this article have been pasted here as it is now a redirect. I don't think there is much that needs to be incorporated into the article page, but for what it's worth, here it is.  Sp in ni ng  Spark  19:39, 5 April 2008 (UTC)

Circuit analysis normally refers to finding all node voltages and all branch currents for an electrical circuit.

The term circuit analysis is often used to mean linear circuit analysis. However non-linear circuits may also be analyzed. Resistive networks normally consisting of a single independent source and resistors may also be analyzed using simpler techniques, however the term resistive network analysis should be used for this. Unfortunately, some have taken to using the term circuit analysis to describe resistive network analysis, which is misleading at best.

Linear DC circuits include independent voltage and current sources, dependent current and voltage sources, and linear resistors. Linear AC circuits also include at least one linear differential component (capacitors and inductors) along with at least one AC source. If a circuit does not include capacitors and inductors DC circuit analysis may be used. If a circuit includes one or more linear differential component and an AC source AC circuit analysis must be used.

DC linear circuit analysis techniques
There are several methods of linear DC circuit analysis.

(1) Nodal analysis ("node") (2) Mesh analysis ("mesh") - Does not work for complex 3D cases (3) Superposition - normally done in conjunction with node or mesh if the circuit has dependent sources (4) Source transformations - a limited technique (5) Equivalent circuits - normally done in conjunction with node or mesh

AC linear circuit analysis techniques
The methods for AC circuit analysis are generally the same as for DC circuit analysis. However the inclusion of capacitors and inductors (linear differential elements) means that complex math, or phasor must be used.

The effective resistance, or impedance, for such components are
 * $$ Z(C) = \frac{1}{j \omega C} $$ and $$ Z(L) = j \omega L \,$$,

where $$ j = \sqrt{-1}$$, $$ \omega = 2 \pi f $$, $$f$$ = the frequency of the AC source, C = the capacitance, and L = the inductance. In short the introduction of $$j$$ means that the math is much, much more onerous.

Notes on merger
This article used to be called Analysis of resistive circuits. If anyone feels they need to write specifically on resistors, please note we still have the article resistive circuit which could badly use some more material. The article Circuit analysis has been merged. If anyone feels that they need to write specifically on circuits rather than networks, please note we still have the article circuit theory.  Sp in ni ng  Spark  20:04, 5 April 2008 (UTC)
 * Why the name change to Network analysis (electronics)? The concepts of the analysis are just as valid in, for example, the analysis of high-voltage transmission networks, or even household wiring, without any involvement of active electronic components. &mdash; BillC talk 22:03, 5 April 2008 (UTC)
 * I take your point, I have no strong view on this. I was merely following the precedent set by several other analysis articles (which I am way too tired to find right now) which also required a disambiguation in the title.  Also, the article (original as well as this one) is in Category:Electronic circuits, which is kind of self defining.  Sp in ni  ng  Spark  01:40, 6 April 2008 (UTC)
 * As the article correctly states, the analysis that follows is only applicable to passive, linear circuits, which eliminates electronic components. For this reason, having the term 'electronics' in the title is most inappropriate. I suggest Network analysis (electric circuits). &mdash; BillC talk 15:24, 6 April 2008 (UTC)
 * Well I cannot find a version of the article that actually says that. It certainly does not say that now and neither did the version prior to the move.  True, it has had some extensive edits so it might be in an intermediate version, but I have looked at quite a few and could not find it.  I do not agree with your categorisation of "non-linear" as "electronics".  Much of electronics deals with purely linear designs, eg filter designs.  I take the physical sciences view of electronics - the science of free electrons - thus including all power distribution in the subject rather than the narrow view of its got to have chips in it to be electronics.  On the other side, non-linear analysis is certainly required in high voltage distribution, transformer hysterisis, for example, is an issue.  You might also note that there is now a large amount of information on non-linear analysis in the article.
 * As I said in reply to your first post, I really have no axe to grind with the name. I will not object to you changing the name.  However, please check "What links here" and amend as appropriate as a second move is certain to cause double-redirects and broken links.  Also, you might want to get an opinion from Rogerbrent first as he has been editing the article as well.  Sp in ni  ng  Spark  16:27, 6 April 2008 (UTC)
 * Looks like the article has already been renamed, but recent edits added stuff on non-linear and active networks so BillC's point is no longer valid. Also, even the electronics article characterizes passive, linear components as electronic. -Roger (talk) 18:00, 6 April 2008 (UTC)

3-dimensional mesh analysis
I have removed the stuff about mesh analysis not being possible in 3D networks. I say it is possible, eg a cube of resistors can be analysed quite satisfactorily with mesh analysis. Mesh analysis is based on Kirchoffs laws which are universally applicable so it must be true. If you think I am wrong please provide a reference that says so or a counter example.  Sp in ni ng  Spark  20:18, 5 April 2008 (UTC)

Never mind, I have found a reference. I am right that a cube of resistors can be analysed with mesh, but strangely it turns out to be planar if you draw it the right way. I will put this back in.  Sp in ni ng  Spark  08:48, 6 April 2008 (UTC)

New external link?
I have an online circuit analyzer program, which runs in every JAVA-enabled browser. It includes theoretical explanation as well.

I think it could be a good external link, but I don't want to add it because of the conflict of interest gudelines. Could an editor please check it, and add it to this page, if he/she thinks that it's relevant enough? The page:

http://www.petertoth.eu/circuitanalyzer.html

Peter —Preceding unsigned comment added by 84.206.41.132 (talk) 06:10, 12 June 2009 (UTC)

Wye-delta claim in article
Where it says "It can be shown that this is sufficient to find the minimal network for any arbitrary network with successive applications of series, parallel, Y-Δ and Δ-Y; no more complex transformations are required." Where is a proof of that?

In fact, I don't think so. Consider a tesseract with edges made up of resistors of various resistances. The problem is to find the equivalent resistance across two opposite vertices. Each vertex has four resistors connected to it. Nowhere is there a triangle in the network. Therefore we cannot apply neither a delta, nor a wye transformation. Neither can we find any parallel or series resistors. So we are stuck, if we are only restricted to series, parallel, Y-Δ and Δ-Y transformations. QED.

Is there an error in the above, or in the article's claim? — Preceding unsigned comment added by 128.250.80.15 (talk) 06:32, 9 June 2011 (UTC)


 * You are right, this needs to be taken out.  Spinning Spark  13:57, 9 June 2011 (UTC)

contradiction with node article
Just noticed this page says that a node is "a point at which terminals of more than two components are joined," while the node article says that "node refers to any point on a circuit where two or more circuit elements meet." So I'm not sure if nodes connect more than 2 or at least 2 components, but I thought I should point this out. 201.171.99.70 (talk) 21:12, 1 December 2013 (UTC)
 * Yes, a node can be a junction of two elements. However, in analysis it is usual to treat series connected elements as all belonging to the same branch of the graph, hence such nodes do not occur.  Spinning  Spark  09:37, 2 December 2013 (UTC)

Y-Δ Button
The Y-Δ button appears twice in the menu at the side, under impedance transforms. I know only basic editing, so please fix and remove this comment.
 * There are two separate buttons for Y-Δ and Δ-Y. They were meant to link to the relevant sections of the article, but the section links had become broken. Now fixed. SpinningSpark 14:11, 9 June 2016 (UTC)

Mesh and nodal analysis only valid for linear circuits?
In the Superposition section of this page, it is said that "Mesh analysis and node analysis also implicitly use superposition [...]" Which book said this? Nodal analysis is based on applying Kirchhoff's laws in a particular manner, but that doesn't require linear elements. Kirchhoff's laws are based on charge and energy conservation principles, which are indepedent of the linearity/non-linearity of the network (although yes, I'm aware these two laws aren't valid for circuits with magnetism unless we consider Faraday's law, and are neither valid for distributed-parameter circuits such as transmission lines.) Alej27 (talk) 05:44, 29 December 2019 (UTC)

I've added a citation: Once we have nonlinear elements, we cannot write the equations in matrix form. KVL/KCL does not require network elements to be linear and mesh/node analysis requires the network to obey KVL/KCL. But it does not follow from that that a network obeying KVL/KCL will obey superposition. Yes, the network can still be solved by algebraically manipulating the equations, but it cannot be solved by any linear analysis method. SpinningSpark 16:24, 20 January 2020 (UTC)


 * I think the idea here should be reexpressed as "... are present. Matrix methods used for mesh analysis and node analysis are also only applicable to linear circuits [Keep reference] but mesh analysis and node analysis methods may still be used to derive a set of n equations for n variables. Superposition of ..."
 * Such a reformulation would solve 2 minor problems with the current statement:
 * The current text states that the analyses may only be used with linear circuits, but the reference only mentions the ability to write equations in matrix form. The link between the statement from the reference to the statement in this text is missing.
 * The current text says that mesh analysis and node analysis are not applicable to non-linear circuits. But these analyses are in fact very useful for non-linear circuits. When dealing with a non-linear circuit one still has to derive a set of non-linear equations somehow.
 * Also note that, after using mesh analysis or node analysis methods to get equations, one does not necessarily need to use matrix methods. Even for linear cases, one may choose to solve the equations by methods such as substitution or numerical methods etc.
 * I didn't want to change the text right away, without discussing it here, since I am not a regular in this article. Kcceao (talk) 18:30, 27 April 2021 (UTC)
 * I removed the incorrect statement about mesh and nodal analysis not being used for non-linear circuits. I believe that is where this discussion is heading.  I left the citation because I did not know what it was supporting.  Constant314 (talk) 20:41, 27 April 2021 (UTC)

"ECA---Electrical Circuit Analysis" listed at Redirects for discussion
An editor has identified a potential problem with the redirect ECA---Electrical Circuit Analysis and has thus listed it for discussion. This discussion will occur at Redirects for discussion/Log/2022 May 4 until a consensus is reached, and readers of this page are welcome to contribute to the discussion. Steel1943 (talk) 19:04, 4 May 2022 (UTC)

Antimetric electrical network
Please, mention Antimetric electrical network in this article. Thanks! — Preceding unsigned comment added by Alexander Davronov (talk • contribs) 19:23, 11 May 2022 (UTC)
 * I don't think its important enough to be included here (and I'm the person who wrote that article). WP:PROPORTIONAL. SpinningSpark 19:00, 12 May 2022 (UTC)