[alancalverd]

   Now, Alancalverd seems to be denying E and B fields and their importance in delivering power in electrical circuits, that's something I might be able to change with some discussion.

It  is entirely reasonable that a static  E field can be detected where there is a potential gradient, and obvious that there is a static B field around a conductor carrying a steady current, but varsigma was talking about an electromagnetic field, which I take to mean a time-varying and self-propagating field generated by accelerating charges.

Mea culpa,  possibly.

[varsigma (OP)]

It  is entirely reasonable that a static  E field can be detected where there is a potential gradient, and obvious that there is a static B field around a conductor carrying a steady current, but varsigma was talking about an electromagnetic field, which I take to mean a time-varying and self-propagating field generated by accelerating charges.

Actually I was talking about the difference between an open circuit with no current flowing, and a circuit with a DC current. When you apply the current (resp. the voltage), the changes in the circuit (inside and outside) propagate through it, at c.
That has to hold as well for AC. It's the changes in the E and B fields that propagate.

If instead you attach an open wire to a battery terminal, it has the same potential as the terminal for the same reason.

[Eternal Student]

Hi.

   There's a bit of a problem with your phrasing @varsigma,   which could easily leave people muddled or confused.

 First line implies some differences exist:

Actually I was talking about the difference between an open circuit with no current flowing, and a circuit with a DC current.

Later line doesn't:

If instead you attach an open wire to a battery terminal, it has the same potential as the terminal for the same reason.

When you apply the current (resp. the voltage), the changes in the circuit (inside and outside) propagate through it, at c.

    There needs to be a careful use and understanding of the word "changes".   It would be helpful to make a distinction between the changing of the potential and the (final) change in potential that is achieved.
    Your sentence reads as if the final state of the circuit is reached much like something is transmitted at the speed c.   That doesn't happen.   If you close a switch near the battery, it can take a while for the region away from the battery to reach its final steady state.  Indeed it can take a while for the region immediately next to the battery (and switch) to reach its final steady state.    However, the changing of the potential can propagate at the speed c:   If the potential of a piece of wire near the battery starts to rise, then the potential of a piece of wire far away from the battery will then start to rise with a delay time between those two which is precisely as if something is travelling at about c.
     The total time taken to reach something approximating the steady state is short anyway because the electrons don't have to move far in the wires (or components) to start establishing the final E fields.   Specifically, the electrons can move sideways (e.g. from the centre of the wire to the outer surface) which is a distance of a few millimetres (width of the wire) to start establishing the surface charge distributions and this will get you close to the conditions of the final steady state .   None-the-less that is some distance that the electrons have to move, so no part of the circuit is going to jump to the final steady state conditions instantly.

   You can consider circuits like this:

circuit.jpg (27.95 kB . 1056x345 - viewed 993 times)
   Once the switch is closed,   the bulb starts to glow within   1 (metre) / c    seconds,     as if power has passed in a straight line through the open space between the switch and the bulb  -  it has clearly not travelled all along the length of the wires.   However, this is only the first or early glow of the bulb.   The final steady state will not be reached for a while longer and the bulb will progressively get brighter.
    Presenting diagrams and experiments like the above is a slightly softer introduction to the idea that E and B fields are important.   Most of the useful ideas you may want to present can be seen or exhibited and many school level notions of how electrical power is delivered can be dismissed (e.g. pushing a train of charges through the wire, which should only start getting power to the bulb at ~ speed of sound in the wire  and demands the entire length of the wire is travelled).

Best Wishes.

[paul cotter]

That's spot-on from a physics perspective. An engineering interpretation would be that as any conductor has inherent inductance there would be an emf induced of value -Ldi/dt which slows the rise in current. A rigorous analysis would require transmission line treatment with the parameters of inductance and resistance of the conductor and interconductor capacitance and admittance( in air 0 ).

[alancalverd]

power has passed in a straight line through the open space between the switch and the bulb  -  it has clearly not travelled all along the length of the wires. 

AAAGH! (to quote the bad guy getting his comeuppance in all the best comics).

Power is the rate of transfer of energy. This is a science forum, not a school for bad journalism!

What this thread is all about is the difference between phase velocity (v) and group velocity (u) of electrons (and holes). In the absence of inductive effects, v = c, and u is the drift velocity we calculated a few pages ago.

The observed delay in a filament lamp reaching full brightness is due to thermal inertia. The current flow is virtually instantaneous to maximum and the power actually decreases as the filament reaches white heat. If we replace it with a LED or a spark gap, there is no delay - every electron that moves, generates a photon.

And before the pedantic sharks attack, yes, the power dissipated in a LED or spark gap does indeed increase with time but again this is a thermal effect (tungsten has a positive temperature coefficient of resistance, semiconductors and plasmas generally negative) not a consequence of the finite value of v or u in the wires!

[paul cotter]

Pedantic shark reply: one can have zero resistance as in a superconductor but inductance is always present and hence there will be a finite rise time to the current. I like rigorous analyses although I often miss some factor. A rigorous analysis settles a question, in my opinion.

[alancalverd]

We used to make precision resistors for lowish frequencies with a bi-filar winding: make a hairpin of the length of wire you need, then wind it around a suitable core. As the current is flowing in the opposite direction in adjacent parts of the conductor, there is no net inductance! IIRC the Vishay precision foil resistor uses the same principle in a flat format and behaves as pure R up to a few MHz.

At the other end of the scale, I've used an analog delay line to make a very stable sinusoidal oscillator with no harmonics. We had the material in stock: it looked like a coaxial cable  but instead of a braid, the "shield"  was a single continuous winding with a ?ferrite core. You just asked the storekeeper for  "30 microseconds" or whatever, and he cut the appropriate length off the reel. 

Sadly, knowledgeable chainsmoking storekeepers and sweaty analog electronics seem to have been abolished. Or was it the other way around?

[paul cotter]

Yes Alan, i'm quite familiar with these low inductance resistors, popular for meter shunts and anywhere transient induced ringing is to be minimised. Any finite length of conductor will have inductance and thus a finite rise time for the current- it's the interconnecting conductors i'm concerned with. Off topic but I bet a thick film resistor would beat any contrawound wire resistor in terms of low inductance.

[varsigma (OP)]

And before the pedantic sharks attack, yes, the power dissipated in a LED or spark gap does indeed increase with time but again this is a thermal effect (tungsten has a positive temperature coefficient of resistance, semiconductors and plasmas generally negative) not a consequence of the finite value of v or u in the wires!

LEDs and transistors have a smooth v-i response, except at the beginning or end of a switch from off to on (or on to off).
In the initial and final parts of a switching (or equilibrium change), the response is chaotic, briefly. It's also more interesting than the smooth response which is just boring old linear stuff.

[Eternal Student]

Hi.

A rigorous analysis would require transmission line treatment with the parameters of inductance and resistance of the conductor and interconductor capacitance and admittance( in air 0 ).

    Yes.   I've seen seen transmission lines modelled like this:



    I'm not an electrical engineer but I had always assumed that is just a model.   There are no capacitors or inductors between the power lines  (or along the lines) but none-the-less the thing acts as if there is.  This model is useful for analysing what happens mainly because it allows one to use more conventional ideas or notions in electronics to determine what is happening.    However, the capacitance is not found in discrete little capacitor components spaced every few metres,  it is much more of a continuous thing spread out over all the length of the wire.   Similarly the inductors aren't found as discrete components.   The model is just a way of emulating the way E and B fields will be established in each wire and how they would propagating through the space between the wires etc.

    I don't think there's a perfect model based on how the E and B fields would spread or a microscopic explanation of what is happening to charges in the wires either.   A lot of those models assume simple geometry (like round wires rather then cheese wedge shapes in cross section  along with the wires usually being quite straight instead of curved wiggly wires).   More generally the first moments of current flow (i.e. at the moment the switch is closed) is often explained with an awkward hybrid of mechanisms - some of it is a bit like a train of charges being pushed along the wire, some of it is more like charges moving to the surface of the wire etc.   Additionally a real world circuit never has the behaviour we would want for the E and B fields spreading through space:    (1) Air is a mixture of stuff and fields don't move through it exactly like a vacuum,    (2) The battery, wires and load are NEVER the only things in the universe.    There's always some other ferromagnetic material and/or fields sourced from other things in the region.   For very long power transmission lines there's no reason to assume that the power station in Nottingham is the place where all the power being consumed in my light bulb is coming from,  even if the wiring was only connected to that Nottingham power station.  My light bulb can have power delivered to it from whatever E and B fields are in the local vicinity to it.   To say that another way,  the conservation of energy only requires the Nottingham power station delivers energy to the fields that exist in space equal to the amount of energy my light bulb can pull out of space.  However the  path of the energy flow may be from Nottingham to  Rodney Smith's light bulb  and not my light bulb,  meanwhile whatever power station and power transmission lines that Rodney Smith's house is wired to may be where my energy for my light blub is coming from.

{Overlap with @varsigma  who finished writing before I did.    Yes... (I think) the early moments of current flow is usually explained with an awkward hybrid of mechanisms,  i.e. we don't really have an exact model or understanding for that - just some ideas we can patch together well enough.   I'm no expert, that's just my opinion.}

Best Wishes.

[alancalverd]

Now you (varsigma) are talking like an engineer! Yes, in any real junction device there will be some effective junction capacitance. Not sure about "chaos" but in a real circuit you can expect some ringing after an impulse.

[paul cotter]

In a simple circuit consisting of a dc supply, two wires and a load resistor the current never reaches the predicted v/r value: due to the inductance of the leads the current approaches an asymptote of v/r.( I accept I am being a bit extreme here )

[alancalverd]

So in an ac circuit the current never reaches its theoretical peak value, then reverses and falls a bit short of its negative maximum, and so on, until it diminishes to zero. The moral is, leave all your electrical equipment switched on and save money!

[varsigma (OP)]

This might seem like I'm just bouncing ideas around, but I reassure you this is relevant to the topic at hand (what was that again?).

I'd like to canvas an opinion or two about the likeness between two well-understood things, namely energy and information.

I'll list some of the coincidental things:

Energy is conserved, so is information. Both depend on a local context, i.e. both depend on locality.
Energy can be transported from place to place, so can information. Both can be lost along the way, via dissipative processes.

Information can be written or stored. Energy can also be stored.
Information can be erased, but this only changes a pattern, so it's agreed that the pattern is "blank". Energy can be converted which in a sense erases some energy, although conversion of energy is itself a process that uses energy.

[Eternal Student]

Hi.

I'd like to canvas an opinion or two about the likeness between two well-understood things, namely energy and information.

   You could start a new thread and put a Poll on it.  It appears right at the top of the thread and is automatically updated.  People can easly select a response and a bar chart of results is automatically generated (or hidden until some deadline is reached).  Experiment with the options and you'll see for yourself.

See  https://www.thenakedscientists.com/forum/index.php?topic=85028.msg681214#msg681214   for an example of a poll and some screenshots showing you exactly which buttons to push to create one.

Best Wishes.

[alancalverd]

For a start, energy is not well-understood. Indeed most of the rest of varsigma's post shows that it isn't!

[paul cotter]

With one tap on a keyboard I can eliminate any arbitrarily large amount of information, in principle. Energy on the other hand cannot be lost. Yes it can be dissipated through lossy mechanisms but a thorough energy audit will reveal 100% with an increase in entropy.

[varsigma (OP)]

With one tap on a keyboard I can eliminate any arbitrarily large amount of information, in principle.

Can you explain what you mean by eliminate?

Do you mean something like, make indistinguishable from a background?
Is energy well-understood? Does the meaning of well-understood need a review? It's a thing that is certainly well-understood in a mathematical sense. It is not however, a thing that can be said to be patterned, except in a rather abstract sense. That sense is an increase or decrease in entropy. Entropy is system-dependent.

[paul cotter]

On a simple computer without cloud connection a low level format instruction, activated by the return key, can be used to wipe any amount of information, in principle.

[alancalverd]

Not sure whether it is still the case, but the National Curriculum used to require primary school teachers to establish that "Energy is the 'go' of things", thus guaranteeing that at least one generation had absolutely no idea of what physics, chemistry, life, and the entire universe, is about.

Any journalist will tell you that power is the ability to make things happen and is synonymous with force, strength and energy, and resilience is the ability to deform and dissipate energy.