EEVblog® Electronics Community Forum

Electronics => Beginners => Topic started by: little_carlos on April 13, 2016, 11:54:18 pm

Title: are kirchoff laws useless?
Post by: little_carlos on April 13, 2016, 11:54:18 pm
in the time ive been in electronics, never in my life ive seen someone using kirchoff laws analysis to design a circuit or repairing something.
are they useless? by this i mean they have no application at the time of designing something, or have you used them?
Title: Re: are kirchoff laws useless?
Post by: mrpackethead on April 14, 2016, 12:04:22 am
As a professional Design engineer, its second nature, and i dont' even think about it.      If we did'nt know that all of the current flowing in, flowed out,  designing things woudl be interesting to say the least. 

When you are designing something as simple as  Fuse board, you are applying kirtoffs laws, and you might not even be releasing it.

Useless No. So simple that you might not even release it.   So complex at times, that it might give you a head ache.

There is a well known problem that floats around universitys, where they make a "cube" of  1ohm Resistors an they apply 1 Volt across the diagnoal corners, and ask you to calculate the current in one of the resitors..   
Title: Re: are kirchoff laws useless?
Post by: Len on April 14, 2016, 01:13:27 am
in the time ive been in electronics, never in my life ive seen someone using kirchoff laws analysis to design a circuit or repairing something.

Then you've never seen someone design a circuit. You can't even put a resistor in series with an LED without using Kirchoff's voltage law.
Title: Re: are kirchoff laws useless?
Post by: retrolefty on April 14, 2016, 01:34:15 am
I believe that Kirctoff's laws greatest value is in helping learning the proper fundamentals of basic electronics theory. After that, as already stated, it just becomes part of ones understanding and one can fall back on it when presented with a specific situation in designing or troubleshooting. 
Title: Re: are kirchoff laws useless?
Post by: BobsURuncle on April 14, 2016, 04:33:28 am
Blasphemy!
Title: Re: are kirchoff laws useless?
Post by: rx8pilot on April 14, 2016, 04:51:57 am
As a professional Design engineer, its second nature, and i dont' even think about it.      If we did'nt know that all of the current flowing in, flowed out,  designing things woudl be interesting to say the least. 

When you are designing something as simple as  Fuse board, you are applying kirtoffs laws, and you might not even be releasing it.

Useless No. So simple that you might not even release it.   So complex at times, that it might give you a head ache.

There is a well known problem that floats around universitys, where they make a "cube" of  1ohm Resistors an they apply 1 Volt across the diagnoal corners, and ask you to calculate the current in one of the resitors..

That is a rather good answer - in some cases it is so intuitive that your mind does not even make the connection that it KVL/KCL. The crazy resistor exercises are not only harder to solve, but harder to mentally connect with a real life circuit. They do, however play a daily role with any designer.
Title: Re: are kirchoff laws useless?
Post by: jitter on April 14, 2016, 05:17:49 am
in the time ive been in electronics, never in my life ive seen someone using kirchoff laws analysis to design a circuit or repairing something.
are they useless? by this i mean they have no application at the time of designing something, or have you used them?

I'm pretty sure that Ohm's Laws are used with a lot more awareness than Kirchoff's Laws, but that doesn't mean they're useless.

If you're repairing something in a circuit, you must always be aware of the fact that things can be in parallel and that currents divide and add up in nodes, but their sum will always be equal to 0 in each node. What goes in must come out. So, this can even be a helpful rather than useless!
Title: Re: are kirchoff laws useless?
Post by: DimitriP on April 14, 2016, 05:21:25 am
in the time ive been in electronics, never in my life ive seen someone using kirchoff laws analysis to design a circuit or repairing something.
are they useless? by this i mean they have no application at the time of designing something, or have you used them?

If you have ever driven  accross a bridge with moderate traffic in a hurry, you've experienced both of  them ...and so did the bridge. 
Title: Re: are kirchoff laws useless?
Post by: rsjsouza on April 14, 2016, 09:28:35 am
As others suggested, Kirchoff's laws become second nature. My example where  they are imperative is to calculate the behaviour of transistor-based circuits: both DC bias and AC low noise response. Oh well, perhaps people don't do these calculations by hand anymore... :)
Title: Re: are kirchoff laws useless?
Post by: Karlo_Moharic on April 14, 2016, 10:23:37 am
The original question makes no sense because circuit analysis as such is based on Kirchhoffs laws.
They are to electronics what Newton laws are to mechanics or First ans Second law of thermodynamics are to thermodynamics.
Title: Re: are kirchoff laws useless?
Post by: tszaboo on April 14, 2016, 12:19:31 pm
1.) NO.
2.) If anyone disagrees, see point 1.)
Title: Re: are kirchoff laws useless?
Post by: Cerebus on April 14, 2016, 12:49:40 pm
The original question makes no sense because circuit analysis as such is based on Kirchhoffs laws.
They are to electronics what Newton laws are to mechanics or First ans Second law of thermodynamics are to thermodynamics.

I'd say that closest parallels in general physics are that the voltage law is cognate with the law of conservation of energy and current law with the conservation of mass.

Of course Einstein came along and now we have the conservation of mass-energy instead. I'll let Maxwell take the blame for where Kirchoff's laws break down.
Title: Re: are kirchoff laws useless?
Post by: rrinker on April 14, 2016, 12:52:36 pm
 Absolutely not. It's true, it does just become something you do without thinking about it. The example of a resistor and LED is spot on, most of my electronics over the past many years has been as part of my model railroad hobby, and I am forever explaining the WHY of calculating the value for a series resistor for LEDs (often used instead of incandescent bulbs for headlights and such). Sure, you calculate the resistor using Ohm's law, but why does that even result in a proper limiter for the LED? Kirchoff's laws, of course. Why can you put multiple LEDs in series and then use a single, smaller dropping resistor? Kirchoff again.

Title: Re: are kirchoff laws useless?
Post by: Brumby on April 14, 2016, 02:36:39 pm
This:


Useless No. So simple that you might not even release it.


The fundamental simplicity of the laws belies their value.  They are pretty much a 'common sense' thing.  The application of them in less than simple circuits is where it gets interesting.
Title: Re: are kirchoff laws useless?
Post by: mrpackethead on April 14, 2016, 02:50:28 pm
This:


Useless No. So simple that you might not even release it.


The fundamental simplicity of the laws belies their value.  They are pretty much a 'common sense' thing.  The application of them in less than simple circuits is where it gets interesting.


Did i accidently say somethign profound.
Title: Re: are kirchoff laws useless?
Post by: Brumby on April 14, 2016, 02:52:54 pm
Some might say so....
Title: Re: are kirchoff laws useless?
Post by: Iwanushka on April 14, 2016, 05:45:58 pm
These things are quite useful...
Title: Re: are kirchoff laws useless?
Post by: jpb on April 14, 2016, 05:59:07 pm
Kirchoff's law is the basis of linear circuit simulators and also non-linear ones (eg SPICE) so if you use a simulator you're using Kirchoff's law.

Basically such simulators work by adding all the conductance terms in a large matrix which gives you an matrix equation of the form G.V = I.

Kirchoff's law gives you the I and by inverting G (actually by using LU factoring in SPICE I think) you get the Vs.

SPICE works by linearising the circuit at each time step before doing the solution.

Title: Re: are kirchoff laws useless?
Post by: bson on April 14, 2016, 06:09:39 pm
Most of the basic circuit questions asked here can be answered with some very basic understanding of how to do network analysis.  So yes, KCL and KVL are useful.  Also understanding their applicability to the complex plane is essential.  (And of course the analytical understanding that complex numbers form a closed field just like reals, with the only exception that complex numbers are unordered.)  DC (real) network analysis is first-year educational coverage in any EE program, at least it was for me in the early 80s, and for good reason.
Title: Re: are kirchoff laws useless?
Post by: bson on April 14, 2016, 06:17:09 pm
Not everything is easily simulated, and sometimes creating test scenarios for simulation requires clever circuit design to create specific sources and load behavior.
Title: Re: are kirchoff laws useless?
Post by: IconicPCB on April 14, 2016, 09:54:47 pm
The point is they are not laws that Mr Kirchoff formulated.. they are observation of real world behavior of circuits.

Rejecting them would mean a need for new reality.
Title: Re: are kirchoff laws useless?
Post by: Karlo_Moharic on April 15, 2016, 10:46:38 am

 .... I'll let Maxwell take the blame for where Kirchoff's laws break down.
[/quote]

What do you mean by  that ?
Title: Re: are kirchoff laws useless?
Post by: Brumby on April 15, 2016, 11:50:25 am
James Clerk Maxwell
Title: Re: are kirchoff laws useless?
Post by: rstofer on April 15, 2016, 03:09:07 pm
James Clerk Maxwell

Electromagnetics!  That was a tough course!  I'm not sure I ever did understand curl or divergence as mathematical operations.  I still have the text, 40+ years later, but I don't spend any quality time with it.

Try to analyze an op amp integrator without KVL or KCL!  We certainly can't have current piling up at the summing junction so whatever current flows through the timing resistor must flow through the timing capacitor because there is no current flow into the ideal op amp.  KCL tells us that current can't pile up at a node.  What goes in, goes out!

https://en.wikipedia.org/wiki/Op_amp_integrator


Title: Re: are kirchoff laws useless?
Post by: Cerebus on April 15, 2016, 03:39:18 pm

 .... I'll let Maxwell take the blame for where Kirchoff's laws break down.


What do you mean by  that ?

Kirchoff's laws break down as soon as there's any external electromagnetic coupling into the circuit. The obvious case is where current flows into an antenna and disappears out of the end as radio waves. Maxwell is, as someone else pointed out, James Clark Maxwell of the famous eponymous Maxwell's Equations that describe electromagnetism, also quite well known as a farmer of daemons which he leased out for impossible tasks (https://en.wikipedia.org/wiki/Maxwell's_demon (https://en.wikipedia.org/wiki/Maxwell's_demon)).
Title: Re: are kirchoff laws useless?
Post by: IanB on April 15, 2016, 05:01:18 pm
The obvious case is where current flows into an antenna and disappears out of the end as radio waves.

It certainly doesn't!  :o
Title: Re: are kirchoff laws useless?
Post by: Cerebus on April 15, 2016, 05:39:44 pm
The obvious case is where current flows into an antenna and disappears out of the end as radio waves.

It certainly doesn't!  :o

Well of course it doesn't actually disappear, but in the naïve view where there is only KCL and KVL at least some of current into an antenna (aka a wire that just ends and doesn't close a loop) just goes 'somewhere else'. It only makes sense once you include a equivalent circuit for an antenna that implicitly hauls the world of electromagnetism into the nodes and vertices of the KCL/KVL world.

And I think a naïve view is appropriate here when the OP asks "are kirchoff laws useless?"(sic). If you want to try explaining exactly what's really going on to the OP such that he properly understands it and uses no simplistic or naïve examples, I will gladly observe with, depending on how well you do, awe or amusement.
Title: Re: are kirchoff laws useless?
Post by: IanB on April 15, 2016, 06:24:42 pm
Well of course it doesn't actually disappear, but in the naïve view where there is only KCL and KVL at least some of current into an antenna (aka a wire that just ends and doesn't close a loop) just goes 'somewhere else'. It only makes sense once you include a equivalent circuit for an antenna that implicitly hauls the world of electromagnetism into the nodes and vertices of the KCL/KVL world.

This is just not correct. KCL is observed perfectly in an antenna, it has to be. The current law is an expression of the conservation of charge, and that is a fundamental and inviolable law of physics. Unless charge leaks out of the antenna by corona discharge or some other mechanism, current is conserved.

If you were to pick a point on the wire feeding the antenna and then integrate the current flow across that boundary over time, then the total would approach zero over a sufficiently long time interval.
Title: Re: are kirchoff laws useless?
Post by: bson on April 15, 2016, 06:26:29 pm
It only makes sense once you include a equivalent circuit for an antenna
All circuits are equivalent circuits.
Title: Re: are kirchoff laws useless?
Post by: photon on April 15, 2016, 06:33:11 pm
Well of course it doesn't actually disappear, but in the naïve view where there is only KCL and KVL at least some of current into an antenna (aka a wire that just ends and doesn't close a loop) just goes 'somewhere else'. It only makes sense once you include a equivalent circuit for an antenna that implicitly hauls the world of electromagnetism into the nodes and vertices of the KCL/KVL world.

This is just not correct. KCL is observed perfectly in an antenna, it has to be. The current law is an expression of the conservation of charge, and that is a fundamental and inviolable law of physics. Unless charge leaks out of the antenna by corona discharge or some other mechanism, current is conserved.

If you were to pick a point on the wire feeding the antenna and then integrate the current flow across that boundary over time, then the total would approach zero over a sufficiently long time interval.
What is lost is not the current (I) but the energy (QV). Since charge (Q) is conserved, the energy loss is from heat dissipated from the battery and the other matter in the antenna.
Title: Re: are kirchoff laws useless?
Post by: IanB on April 15, 2016, 06:37:41 pm
What is lost is not the current (I) but the energy (QV). Since charge (Q) is conserved, the energy loss is from heat dissipated from the battery and the other matter in the antenna.

The energy loss from heat dissipation is a parasitic loss, which ideally would be as little as possible. The energy dissipation of value is of course by electromagnetic radiation.
Title: Re: are kirchoff laws useless?
Post by: Cerebus on April 15, 2016, 07:19:20 pm
Well of course it doesn't actually disappear, but in the naïve view where there is only KCL and KVL at least some of current into an antenna (aka a wire that just ends and doesn't close a loop) just goes 'somewhere else'. It only makes sense once you include a equivalent circuit for an antenna that implicitly hauls the world of electromagnetism into the nodes and vertices of the KCL/KVL world.

This is just not correct. KCL is observed perfectly in an antenna, it has to be. The current law is an expression of the conservation of charge, and that is a fundamental and inviolable law of physics. Unless charge leaks out of the antenna by corona discharge or some other mechanism, current is conserved.

If you were to pick a point on the wire feeding the antenna and then integrate the current flow across that boundary over time, then the total would approach zero over a sufficiently long time interval.

I think you're missing the point here and also being unnecessarily pedantic - you've passed the test, you're an engineer. The whole sub-thread here is the result of a throwaway explanation of a throwaway reference to Maxwell. The point, if there is any, is that KCL and KVL do not fully account for electromagnetics. You can worry away at the details of a throwaway example for as long as you like but it won't alter the fact that (1) in context, it doesn't really matter, (2) I don't really care if I was pedantically, exactly correct in the precise technical detail of a throwaway remark that was meant to merely illustrate something.
Title: Re: are kirchoff laws useless?
Post by: albert22 on April 15, 2016, 08:55:30 pm
For an EE Kirchoff laws are useful and unavoidable. Provided that can be applied;
quote from wikipedia

Quote
Both of Kirchhoff's laws can be understood as corollaries of the Maxwell equations in the low-frequency limit. They are accurate for DC circuits, and for AC circuits at frequencies where the wavelengths of electromagnetic radiation are very large compared to the circuits.
...
Limitations

KCL and KVL both depend on the lumped element model being applicable to the circuit in question. When the model is not applicable, the laws do not apply.
...

For a carpenter they are useless.
Title: Re: are kirchoff laws useless?
Post by: IanB on April 15, 2016, 08:59:12 pm
The point, if there is any, is that KCL and KVL do not fully account for electromagnetics.

I'm disagreeing with you because I think they do fully account for electromagnetics.

I think KCL expresses a general rule about conservation of charge, and KVL expresses a general rule about the conservative nature of electric fields. As such they are (when accumulation terms and propagation delays are neglected) axiomatically true. They are true in transformers, in antennas and anywhere else that you can measure currents and voltages without disturbing the circuit.

Since we seem to be disagreeing about this point it appears we need another expert person to arbitrate.
Title: Re: are kirchoff laws useless?
Post by: photon on April 15, 2016, 11:11:09 pm
What is lost is not the current (I) but the energy (QV). Since charge (Q) is conserved, the energy loss is from heat dissipated from the battery and the other matter in the antenna.

The energy loss from heat dissipation is a parasitic loss, which ideally would be as little as possible. The energy dissipation of value is of course by electromagnetic radiation.
Energy_in = power supply. Energy_out = EM waves + losses. Losses = heat. No loss of charge or current.
Title: Re: are kirchoff laws useless?
Post by: IanB on April 15, 2016, 11:45:00 pm
Energy_in = power supply. Energy_out = EM waves + losses. Losses = heat. No loss of charge or current.

Of course. We are in complete agreement on that. Maybe I misunderstood your earlier comment when you seemed to imply that the energy losses were only to heat.
Title: Re: are kirchoff laws useless?
Post by: mrpackethead on April 16, 2016, 12:33:41 am
beetroot
Title: Re: are kirchoff laws useless?
Post by: rfeecs on April 16, 2016, 12:43:22 am
How about just reading the Wikipedia article:
https://en.wikipedia.org/wiki/Kirchhoff%27s_circuit_laws#Kirchhoff.27s_voltage_law_.28KVL.29 (https://en.wikipedia.org/wiki/Kirchhoff%27s_circuit_laws#Kirchhoff.27s_voltage_law_.28KVL.29)

Quote
Both of Kirchhoff's laws can be understood as corollaries of the Maxwell equations in the low-frequency limit. They are accurate for DC circuits, and for AC circuits at frequencies where the wavelengths of electromagnetic radiation are very large compared to the circuits.

Quote
Limitations
KCL and KVL both depend on the lumped element model being applicable to the circuit in question. When the model is not applicable, the laws do not apply.
KCL, in its usual form, is dependent on the assumption that current flows only in conductors, and that whenever current flows into one end of a conductor it immediately flows out the other end. This is not a safe assumption for high-frequency AC circuits, where the lumped element model is no longer applicable.[2] It is often possible to improve the applicability of KCL by considering "parasitic capacitances" distributed along the conductors.[2] Significant violations of KCL can occur[3] even at 60Hz, which is not a very high frequency.
In other words, KCL is valid only if the total electric charge, \scriptstyle Q , remains constant in the region being considered. In practical cases this is always so when KCL is applied at a geometric point. When investigating a finite region, however, it is possible that the charge density within the region may change. Since charge is conserved, this can only come about by a flow of charge across the region boundary. This flow represents a net current, and KCL is violated.
KVL is based on the assumption that there is no fluctuating magnetic field linking the closed loop. This is not a safe assumption for high-frequency (short-wavelength) AC circuits.[2] In the presence of a changing magnetic field the electric field is not a conservative vector field. Therefore the electric field cannot be the gradient of any potential. That is to say, the line integral of the electric field around the loop is not zero, directly contradicting KVL.
It is often possible to improve the applicability of KVL by considering "parasitic inductances" (including mutual inductances) distributed along the conductors.[2] These are treated as imaginary circuit elements that produce a voltage drop equal to the rate-of-change of the flux.
Title: Re: are kirchoff laws useless?
Post by: Macbeth on April 16, 2016, 12:44:33 am
in the time ive been in electronics, never in my life ive seen someone using kirchoff laws analysis to design a circuit or repairing something.
are they useless? by this i mean they have no application at the time of designing something, or have you used them?
]https://www.youtube.com/watch?v=WBfAEeEzDlg] (https://www.youtube.com/watch?v=WBfAEeEzDlg)

Daves EEVBlog #819 had over 2000 thumbs up "if you found this useful"

and

]https://www.youtube.com/watch?v=8f-2yXiYmRI] (https://www.youtube.com/watch?v=8f-2yXiYmRI)

EEVBlog #820 had nearly 2000 thumbs up. So I would say there are only a handful of naysayers say they are useless.(Seriously, who thumb downed these videos?)
Title: Re: are kirchoff laws useless?
Post by: uncle_bob on April 16, 2016, 01:05:34 am
Hi

Ok, so here we all have gone for darn near two pages of agreeing with each other. Not a single peep I can see from the original poster of the thread. Makes one sort of wonder about the original post and it's context. Either we have scared him away (quite possible), or the post was in jest (also possible), or something else altogether (like he's in jail :) ....Who knows ...

Bob
Title: Re: are kirchoff laws useless?
Post by: Karlo_Moharic on April 16, 2016, 11:49:28 am
James Clerk Maxwell

I know who Maxwell is , but I don't understand what he meant by saying that Maxwell is to blame for Kirchhoff laws breaking down ?
Title: Re: are kirchoff laws useless?
Post by: uncle_bob on April 16, 2016, 01:34:49 pm
James Clerk Maxwell

I know who Maxwell is , but I don't understand what he meant by saying that Maxwell is to blame for Kirchhoff laws breaking down ?

Hi

Fundamentally the basic Kirchoff laws do not consider currents in anything other than "conductors". Current is nicely contained and thus can be analyzed on that basis. Conductors may be wires, or volumes of materiel, they still contain the current.

As soon as you have propagating waves in free space generated by currents (or currents induced by waves), they no longer are nicely constrained. If you break down the constraint, you break down the applicability of that particular law. The filed of electromagnetic waves is pretty much all based on Maxwell's famous equations.

Yes, there are other "AC" things that you need to consider in circuits. Energy storage can take place in a number of ways. It does need to be accounted for. The fundamental basis for all this still is mesh and node equations. Without that, your analysis stops dead. Without Kirchoff, no mesh and node ....

Bob
Title: Re: are kirchoff laws useless?
Post by: rstofer on April 16, 2016, 07:09:43 pm
in the time ive been in electronics, never in my life ive seen someone using kirchoff laws analysis to design a circuit or repairing something.
are they useless? by this i mean they have no application at the time of designing something, or have you used them?

You can't even power an LED without both KVL and KCL.
The LED spec: 20 mA If at 2.0V
The battery spec:  12 VDC 0 Ohms internal impedance

KVL gets us the fact that the resistor needs to drop 10V => 12V at battery - 2V at LED => 10V across the resistor
KCL gets us the fact that the current through the resistor is 20 mA => the current in the loop is the same everywhere so the resistor current is exactly equal to the LED current
Ohm's Law gets us the fact that the resistor should be 500 Ohms - R = E / I = 10V / 0.02A = 500 Ohms

The thing about these Laws is that they are frequently applied, like in the case of the LED, without even thinking about the laws themselves.  It just becomes second nature.  But the fact that the are intuitively applied doesn't make them useless.  They aren't simply suggestions, they are LAWS!


Title: Re: are kirchoff laws useless?
Post by: onlooker on April 16, 2016, 08:07:01 pm
I  guess the OP can ask the question better with a more defined scope. For example, one may ask: does anyone still explicitly use Kirchhoff's laws to solve circuitry problems involving  more than, say, 5 or 10 variables in the design work?

In other words, I guess, the proper question is about how often Kirchhoff's laws are consciously used; It is not about the correctness of Kirchhoff laws in a particular type of EE work.

There are many alternatives to explicitly using Kirchhoff's laws. Examples are simulation software, relying on experiences, using application notes or similar design, or simply by trial and error.

The often mentioned example of LED serial with a resistor is a good case in point. I believe most people knows how the current can be calculated long before aware of the name, Kirchhoff and his laws. The point is, for simpler situations, there are simplified rules. They also likely existed before Kirchhoff's laws even if they now can be regarded as a special application of Kirchhoff' laws.
Title: Re: are kirchoff laws useless?
Post by: uncle_bob on April 16, 2016, 08:08:53 pm
Hi

About the only person who has *not* posted to this thread more than once is the original poster ...

Bob
Title: Re: are kirchoff laws useless?
Post by: T3sl4co1l on April 16, 2016, 11:34:01 pm
Well of course it doesn't actually disappear, but in the naïve view where there is only KCL and KVL at least some of current into an antenna (aka a wire that just ends and doesn't close a loop) just goes 'somewhere else'. It only makes sense once you include a equivalent circuit for an antenna that implicitly hauls the world of electromagnetism into the nodes and vertices of the KCL/KVL world.

It's naive to think KCL applies everywhere, globally, at all instants.

Realize that a schematic is an abstraction of the real world.  There are no true capacitors, nor inductors.  An RLC network responds instantaneously to an input, something no real circuit can do.

KCL works because, in an abstract circuit, there is no concept of a speed of light.  Therefore KCL is true globally at all times.

When we include EM, KCL doesn't cease being useful, but it is necessarily restricted to infinitesimal points, or point-like areas for the purposes of analysis.

An example of a "point like area" is a transmission line port: for frequencies below the higher modes of the TL, it is true that current into one terminal of a port equals current out the complementary terminal of that port.  (A TL is a four terminal, two port component.  A port is simply a pair of terminals at one end of the line: +/- for twisted pair, signal and ground for coax.  We ignore the common mode or shield current for this purpose, as well.)

For an antenna, the current into the base is balanced by displacement current into the EM field.  Just as it is for any resonant circuit, except the "capacitor" happens to be a propagating EM wave, instead of a bunch of E bottled up in a lot of dielectric. :)

Tim
Title: Re: are kirchoff laws useless?
Post by: munkeyman1985 on April 16, 2016, 11:39:06 pm
They get used all of the time. You might not be thinking about it, but any one building, testing, or troubleshooting electronics must be using them.
Title: Re: are kirchoff laws useless?
Post by: T3sl4co1l on April 17, 2016, 12:06:50 am
Fundamentally the basic Kirchoff laws do not consider currents in anything other than "conductors". Current is nicely contained and thus can be analyzed on that basis. Conductors may be wires, or volumes of materiel, they still contain the current.

This is, at best, a misinterpretation of nature, and at worst a gross corruption of the underlying laws.

Consider:
1. If KCL applies only within conductors, then transmission lines wouldn't be possible.  The current would leave the far port, the instant it arrives at the near port.
2. TLs exist.
3. Therefore KCL doesn't apply within conductors.

Further consider:
1. If KCL doesn't apply to fields, then there need be no conservation of displacement currents.
2. Displacement currents are necessarily conserved.  (Faraday's law and electric induction.)
3. Therefore KCL does apply.

As I said before, you can't apply KCL over an arbitrary volume, because the currents will in general be unequal over time or space.  But it is necessarily locally true, because the fields are not discontinuous.

Tim
Title: Re: are kirchoff laws useless?
Post by: uncle_bob on April 17, 2016, 02:11:52 am
Fundamentally the basic Kirchoff laws do not consider currents in anything other than "conductors". Current is nicely contained and thus can be analyzed on that basis. Conductors may be wires, or volumes of materiel, they still contain the current.

This is, at best, a misinterpretation of nature, and at worst a gross corruption of the underlying laws.

Consider:
1. If KCL applies only within conductors, then transmission lines wouldn't be possible.  The current would leave the far port, the instant it arrives at the near port.
2. TLs exist.
3. Therefore KCL doesn't apply within conductors.

Further consider:
1. If KCL doesn't apply to fields, then there need be no conservation of displacement currents.
2. Displacement currents are necessarily conserved.  (Faraday's law and electric induction.)
3. Therefore KCL does apply.

As I said before, you can't apply KCL over an arbitrary volume, because the currents will in general be unequal over time or space.  But it is necessarily locally true, because the fields are not discontinuous.

Tim

Hi

Any analysis of a transmission line that is based *only* on KCL and *not* on Maxwell will be a failure. The most fundamental thing on the first day of any transmission line course is Maxwell. To say that currents flow through vacuum (without Maxwell) simply is not in any way correct.

Bob
Title: Re: are kirchoff laws useless?
Post by: EEVblog on April 17, 2016, 02:21:33 am
I've seen Dave talk about the principle when he did the OpAmp video. IIRC it was in relation to input bias current. He may not have mentioned it by name.

Correct. Kirchoff's laws are so fundamental and second nature, you don't even realise you are using them.
It's "obvious" that current into a node equals the total current out the node, but that's Kichoff's laws for you. It's the 2nd thing you usually learn in DC circuit theory after ohms law.
Hardly anyone goes around saying "because of Kirchoff's Law, the current into the opamp junction must flow this way", because there is simply no need to repeat it because it's so obvious. Just like I don't always say "because of ohms law..."
Title: Re: are kirchoff laws useless?
Post by: EEVblog on April 17, 2016, 02:25:34 am
There are many alternatives to explicitly using Kirchhoff's laws. Examples are simulation software

Which ironically make extensive use of Kirchoff's laws to solve the circuit.
Title: Re: are kirchoff laws useless?
Post by: T3sl4co1l on April 17, 2016, 05:58:49 am
Any analysis of a transmission line that is based *only* on KCL and *not* on Maxwell will be a failure. The most fundamental thing on the first day of any transmission line course is Maxwell. To say that currents flow through vacuum (without Maxwell) simply is not in any way correct.

It would be a gross corruption of my words, to think I implied Maxwell's equations do not apply.

Tim
Title: Re: are kirchoff laws useless?
Post by: mrpackethead on April 17, 2016, 03:53:57 pm
Trolling? I would be like to hear any circuit, no matter how simple, defies KXL or can be analyzed without KXL in any forms.

I fail.
Title: Re: are kirchoff laws useless?
Post by: zapta on April 17, 2016, 04:58:32 pm
...It's the 2nd thing you usually learn in DC circuit theory after ohms law.

I think that Ohm's law is also based on Kirchoff. If you can't assume that the current in both ends of the resistor are the same, then 'resistor current' is not well defined.
Title: Re: are kirchoff laws useless?
Post by: Kalvin on April 17, 2016, 05:39:30 pm
...It's the 2nd thing you usually learn in DC circuit theory after ohms law.

I think that Ohm's law is also based on Kirchoff. If you can't assume that the current in both ends of the resistor are the same, then 'resistor current' is not well defined.

This is nice video "M.I.T.-Walter Lewin- Complete Breakdown of Intuition - Part1":
https://www.youtube.com/watch?v=eqjl-qRy71w (https://www.youtube.com/watch?v=eqjl-qRy71w)
Title: Re: are kirchoff laws useless?
Post by: IanB on April 17, 2016, 07:01:15 pm
This is nice video "M.I.T.-Walter Lewin- Complete Breakdown of Intuition - Part1":

I think he did a sneaky trick there. He failed to discuss some important things that would affect the observations.
Title: Re: are kirchoff laws useless?
Post by: Kalvin on April 17, 2016, 07:43:59 pm
This is nice video "M.I.T.-Walter Lewin- Complete Breakdown of Intuition - Part1":

I think he did a sneaky trick there. He failed to discuss some important things that would affect the observations.

I think it was a psychological trick, not as much a physical trick. But anyway, nice demonstration and makes one think how that is possible.
Title: Re: are kirchoff laws useless?
Post by: zapta on April 17, 2016, 08:00:34 pm
...It's the 2nd thing you usually learn in DC circuit theory after ohms law.

I think that Ohm's law is also based on Kirchoff. If you can't assume that the current in both ends of the resistor are the same, then 'resistor current' is not well defined.

This is nice video "M.I.T.-Walter Lewin- Complete Breakdown of Intuition - Part1":
https://www.youtube.com/watch?v=eqjl-qRy71w (https://www.youtube.com/watch?v=eqjl-qRy71w)

Does this really break Kirchoff?  I would think that once you have magnetic field that induce current in the wires, the wires not become current or voltage sources and should be modeled accordingly for Kirchoff analysis.
Title: Re: are kirchoff laws useless?
Post by: onlooker on April 17, 2016, 09:03:37 pm
Yes, it is a real breakdown of KVL, since the potential difference between point D and point C is no longer well defined or unique (or call it path dependent);

This, as expected,  will show real effects when one use a voltage meter to measure the voltage between D and C. Depending on whether the meter's probe cable laying on the left side or right side of the solenoid one will have different readings and I think this is the part that can't be modeled by distributed battery sources.
Title: Re: are kirchoff laws useless?
Post by: T3sl4co1l on April 17, 2016, 09:40:04 pm
He cheats because he's applying fields to a diagram, one which, by its definition, is agnostic of fields.

As I said before, the circuit diagram has no representation of the speed of light, nor the magnetic or electric fields that construct the EM field.

You can place two wires on a schematic, as close together as possible, and they will still have zero picofarads of capacitance, and zero nanohenries of mutual inductance!  It is only when you draw that "===" symbol between two wires, that you symbolize a capacitor, within which you get electric field, and electric field only.  Or that curly symbol to indicate inductance, wherein you get magnetic field, and magnetic field only.

The correct schematic representation requires a transformer to be drawn somewhere; if it is drawn where the battery used to be, everything is perfectly fine again.

If the schematic is not abstract, but meant as a crude mechanical drawing of a real system, then one will find the voltage measurement depends upon, not where the voltmeter probes are connected, but upon what paths the two voltmeter leads take away from those points.  The voltage drop across each resistor will of course be sensible, but the voltmeter loop itself will see a more significant amount.

There is a second fiction present, also: no real solenoid is infinite in length, and even an infinite solenoid has magnetic potential around it (if little-to-no flux density outside of it).  Performing this experiment with a finite length solenoid will subject all those paths to varying amounts of fringing flux, thus making the voltages dependent as I just said.

Tim
Title: Re: are kirchoff laws useless?
Post by: bson on April 17, 2016, 10:02:43 pm
I think that Ohm's law is also based on Kirchoff. If you can't assume that the current in both ends of the resistor are the same, then 'resistor current' is not well defined.
It's not.  Ohm's law defines the abstract property of resistance as U/I.  It's abstract because nothing ever actually exists that you can point to and say "that right there is resistance".  Wherever there is voltage and current there is by definition resistance.  The current can be limited by a huge number of reasons; Ohm's law simply bundles these into the concept of resistance.

Not to be a confused with a resistor, which ideally has constant resistance.
Title: Re: are kirchoff laws useless?
Post by: zapta on April 17, 2016, 10:10:36 pm
I think that Ohm's law is also based on Kirchoff. If you can't assume that the current in both ends of the resistor are the same, then 'resistor current' is not well defined.
It's not.  Ohm's law defines the abstract property of resistance as U/I.  It's abstract because nothing ever actually exists that you can point to and say "that right there is resistance".  Wherever there is voltage and current there is by definition resistance.  The current can be limited by a huge number of reasons; Ohm's law simply bundles these into the concept of resistance.

Not to be a confused with a resistor, which ideally has constant resistance.


Let's see how the abstract Ohm's law works without assuming kirchoff, let's say that we have a black box with two  leads connected in a circuit. The voltage across the box is 1V, the entering current at lead #1 is 1A and the existing current at lead #2 is 2A. Which of the two currents would you use to compute the resistance of the box using Ohms law?

Ohm's law assumes that there is a single current through the box, hence the dependency on kirchoff.
Title: Re: are kirchoff laws useless?
Post by: T3sl4co1l on April 17, 2016, 10:35:28 pm
If it helps, Ohm's law for fields is E = rho * J.

J is the current density (a vector field), which necessarily has to obey KCL because of charge conservation.

This also says E and J are in the same angle and phase, which is correct.  For reactive conditions (say, permeable or dielectric material), this need not be the case.  (A metal can be represented as a dielectric with a large imaginary component to e_r.)

Tim
Title: Re: are kirchoff laws useless?
Post by: onlooker on April 17, 2016, 10:40:03 pm
Quote
Ohm's law assumes that there is a single current through the box, hence the dependency on kirchoff.

1st, Ohm's law was discovered 20 years before Kirchhoff's laws.  The best one can say is that  Kirchhoff's laws extended some aspects of Ohm's law.

Yes, "Ohm's law assumes that there is a single current through the box", or the "quasi-stationary current" condition. But, I do not think it is directly related to KCL.  KCL by itself is just about charge conservation.  The "quasi-stationary current" condition is an additional requirement to make KCL helpful/useful in circuit analysis. 

An example is the already mentioned transmission line.  The charge conservation law stays, but the "quasi-stationary current" condition is no longer valid. That is, one can still say KCL is valid (for any "node" if  a "node" means a cross-section of the line). But, KCL (or charge conservation) alone is not much helpful for analyzing the circuit.
Title: Re: are kirchoff laws useless?
Post by: TimFox on April 17, 2016, 10:53:34 pm
Ohm's law is approximate.  All resistors show some non-linearity.
Kirchoff's laws are exact (when you pay attention to the conditions).
Title: Re: are kirchoff laws useless?
Post by: filssavi on April 17, 2016, 11:08:27 pm
The point, if there is any, is that KCL and KVL do not fully account for electromagnetics.

I'm disagreeing with you because I think they do fully account for electromagnetics.

I think KCL expresses a general rule about conservation of charge, and KVL expresses a general rule about the conservative nature of electric fields. As such they are (when accumulation terms and propagation delays are neglected) axiomatically true. They are true in transformers, in antennas and anywhere else that you can measure currents and voltages without disturbing the circuit.

Since we seem to be disagreeing about this point it appears we need anothere expert person to arbitrate.

Since kvl and kcl for you explain all Electromagnetism why are we bothering to kill half of our brain trying to get around math formalism of Maxwell equations
Since you clearly are smarter than me Can you explain how fiber optics work (in depth not just current goes in the laser diode and gets out the detector) modes, dispersion and all with just kvl and kcl
Title: Re: are kirchoff laws useless?
Post by: IanB on April 17, 2016, 11:30:29 pm
Since kvl and kcl for you explain all Electromagnetism why are we bothering to kill half of our brain trying to get around math formalism of Maxwell equations
Since you clearly are smarter than me Can you explain how fiber optics work (in depth not just current goes in the laser diode and gets out the detector) modes, dispersion and all with just kvl and kcl

I have not said they explain electromagnetism. I have simply said that they are not invalidated by electromagnetism.
Title: Re: are kirchoff laws useless?
Post by: IconicPCB on April 18, 2016, 12:05:31 am
We should ignore Kirchoff.. and analyse everything in terms of Maxwell's work. ( my contribution to the war)

Title: Re: are kirchoff laws useless?
Post by: orolo on April 18, 2016, 01:42:51 am
...It's the 2nd thing you usually learn in DC circuit theory after ohms law.

I think that Ohm's law is also based on Kirchoff. If you can't assume that the current in both ends of the resistor are the same, then 'resistor current' is not well defined.

This is nice video "M.I.T.-Walter Lewin- Complete Breakdown of Intuition - Part1":
https://www.youtube.com/watch?v=eqjl-qRy71w (https://www.youtube.com/watch?v=eqjl-qRy71w)

Does this really break Kirchoff?  I would think that once you have magnetic field that induce current in the wires, the wires not become current or voltage sources and should be modeled accordingly for Kirchoff analysis.
It's a trick, of course. It is implicit in the circuit that the two resistors are on the secondary side of a transformer: you don't have R1 in series with R2 alone, in a complete lumped description you have L2 (the secondary, one loop inductor around the magnetic field) in series with R1 and R2. KVL applies to these three elements, as usual. Tricky old professor.
Title: Re: are kirchoff laws useless?
Post by: EEVblog on April 18, 2016, 02:10:39 am
Yes, it is a real breakdown of KVL, since the potential difference between point D and point C is no longer well defined or unique (or call it path dependent);
This, as expected,  will show real effects when one use a voltage meter to measure the voltage between D and C. Depending on whether the meter's probe cable laying on the left side or right side of the solenoid one will have different readings and I think this is the part that can't be modeled by distributed battery sources.

And cue the overunity crowd!
Title: Re: are kirchoff laws useless?
Post by: EEVblog on April 18, 2016, 02:11:43 am
We should ignore Kirchoff.. and analyse everything in terms of Maxwell's work. ( my contribution to the war)

Basic circuit design would get pretty tedious very quickly.
Title: Re: are kirchoff laws useless?
Post by: Helix70 on April 18, 2016, 02:52:44 am
Ohm's law is approximate.  All resistors show some non-linearity.

I don't think so. Care to explain what you mean by this?
Title: Re: are kirchoff laws useless?
Post by: T3sl4co1l on April 18, 2016, 06:05:35 am
Ohm's law is approximate.  All resistors show some non-linearity.

I don't think so. Care to explain what you mean by this?

Ohm's law really isn't, not to nearly the same degree as KxL anyway.

The only thing that can be said to be "ohmic" is space itself, but that's tricky and kind of useless.  I'll explain in a moment.

There are myriad examples of non-ohmic devices.  Diodes are excellent.  Even among regular metals, depending on what time and voltage scales you're talking about, you can experience:
- Long time scales: self heating (e.g. light bulb filament)
- High current density: electromigration (and similar effects like memristance, in ionic conductors)
- Extremely high current density, high voltage: breakdown, ionization, plasma (e.g., The Z Machine)

It's somewhat miraculous, and rather handy, that so many ordinary metals and compounds are ohmic at all!

In bulk materials that aren't perfectly pure metals (there's a suitable physics definition for this), you can get diode junctions between crystal grains and mating surfaces.  For example, nickel plated connectors are undesirable for sensitive RF, due to possible mixing/modulation/distortion, due to Ni-NiO-Ni junctions on the connector surfaces.

Possibly the best (read: most linear over the largest dynamic range) ohmic configuration is no material at all: ohms defined by the EM field itself.  The impedance of free space is the ratio between electric and magnetic field strengths: just as the ratio of voltage to current gets you a resistance, or the ratio of inductance to capacitance gets you a resistance [squared].  It's kind of useless, because you don't get a resistor with two terminals; a wideband antenna is the closest representation of this, but contains metal.  And anyway, the side effect is beaming EM radiation off to infinity, which might not be desirable (also, any reflections received by the antenna will change the terminal impedance).

Even this will break down at some energy density, because nothing is forever; I'm not sure what the actual intensities required are, but a particle-physics description will go something like, photon up-conversion using virtual particles as reaction mass; eventually resulting in pair production of electrons and protons (and other assorted things, at ever-higher energy levels).  The ultimate mass-energy density limit of course being a black hole, but that's truly beyond astronomical: even light-speed quasar jets aren't quite *that* intense.

In any case, EM is truly and wholly fundamental to all of physics: it gave rise to relativity, is a crucial ingredient in QED (perhaps the most accurately proven theory in history), and extends all the way to the bottom of the Standard Model (where classical fields-as-we-know-them give way to particle-like descriptions of interactions, and numerous other charges become as or more important on the smallest scales, such as quark flavor charge).

Indeed, since KxL is just another conservation law, we could say something more general: that laws of that nature are much more widespread, so that whether you're talking loop voltage, or node current, or atomic quark color, or anything else that's conserved, it's the same mathematical symmetry in the system.  Of course, we don't call all those symmetries as "Kirchoff": that's limited only to the ones in electronics.

Tim
Title: Re: are kirchoff laws useless?
Post by: filssavi on April 18, 2016, 06:14:51 am
Since kvl and kcl for you explain all Electromagnetism why are we bothering to kill half of our brain trying to get around math formalism of Maxwell equations
Since you clearly are smarter than me Can you explain how fiber optics work (in depth not just current goes in the laser diode and gets out the detector) modes, dispersion and all with just kvl and kcl

I have not said they explain electromagnetism. I have simply said that they are not invalidated by electromagnetism.


Well about that, it's not true either, KVL and KCL are used in lumped model circuit analisys so if you are low frequency (or better said the considered wavelenght are longer than your circuit lenght) they are indeed valid

If you start going upin frequency (or in circuit length think distribuzione grid wise) you will start seing that wires are not ideal any more they will delay and distort your waveform

These are trasmissione line effects and here regular KVL and KCL does not apply any more, what is done usually is to considerazione a lumped model equivalente of the whole line and then you can indeed stil use kirchoff's laws bug that is only an approximation its  not the real deal and as such it has some limitations

So if Kirchoff's law are of any good depends whether  you can just treat the TL as a black box or not if you can then you can measure it's properties (or more likely S parameters) and merrily use KVL/KCL if however this is not the case and you need to know exactly what is going on in the middle of the line then you cannot use them you need to step up your game and start consider full Maxwell's equations

P.S. it might be semantics but since optics fiber are Electromagnetism and KCL/KCL are not valid in this domain then they are invalidated by Electromagnetism (EM for short) in general, now if by EM you really mean the subset of EM we all generally work in so geometrically small low frequency circuits than yes they are and not invalidated but that is only a mall subset of EM field theory not the whole thing
Title: Re: are kirchoff laws useless?
Post by: photon on April 18, 2016, 09:29:14 am
I think that Lewin is doing a magic trick in this video, pulling a rabbit out of a hat just for show. It is sleight of hand and does not stand up to closer inspection.

KCL is a statement of the conservation of charge and KVL is a statement of the conservation of energy. These conservation laws are true. If Lewin thinks he has an example circuit where KVL does not apply, then he has a circuit where the conservation of energy does not apply. But the conservation of energy always applies if you consider all sources of energy.

Lewin's example:
http://videolectures.net/site/normal_dl/tag=28248/non-conservative_fields-do_not_trust_your_intuition.pdf (http://videolectures.net/site/normal_dl/tag=28248/non-conservative_fields-do_not_trust_your_intuition.pdf)
is magic. The sleight of hand is that he introduces an energy source (a solenoid) from outside of his example circuit. He adds energy to the circuit without adding a circuit element to account for it.

You could find an example where Maxwell's equations do not apply in the same way. Just write the equations and omit a current or magnetic source.

I say bullshit.
Title: Re: are kirchoff laws useless?
Post by: filssavi on April 18, 2016, 11:13:43 am
No he is not doing tricks he is considering a physical circuit made of copper and carbon (for resistors) where you have a magnetic field coupling in a closed loop of conductor in other words the middle loop is an antenna that picks up magnetic field from the outside world this induces a voltage in the circuit and hey presto.

That said you could model this effect with a the varing voltage source and only then KCL/KVL could be applied
but that is the electronic equivalent of hiding dirt under the carpet

KVL and KCL are not the laws of conservation!, they are statements derived by laws of conservazione when applied to a specifico set of assumptions (specifically that a lumped element equivalente circuit can be used and that there are no external time varying H or E field in which your physical circuit is immersed)

Conservati in of energy is valid only globally locally can be made not to be true by cleverly picking the boundary ofthe local part in this example if you ignore the energy carried from the magnetic field than yes conservation is broken because you neglected a source of energy
Title: Re: are kirchoff laws useless?
Post by: orolo on April 18, 2016, 01:04:04 pm
If one looks for a complete breakdown of KCL, it surely must happen at the quantum level, for example, in molecular scale circuits with tiny currents. If a single electron travels along a nanowire, and it reaches a bifurcation, it can propagate along the different wires simultaneously, with different amplitudes, including being reflected back. The electron might tunnel from node to node against the potential, or diffuse along the whole nanocircuit, if it doesn't find anything to interact with. I'm not sure if Ohm's or Kirchoff's laws can be easily reformulated at the operator level, but I have a feeling that KCL at least might, being an statement of conservation (eg. in the example before, the probability current in a node, times charge, must be conserved).