Re: Inertial-dampening systems

From: Gregory L. Hansen (glhansen_at_steel.ucs.indiana.edu)
Date: 02/08/05 

Date: Tue, 8 Feb 2005 02:36:51 +0000 (UTC)

In article <1107814751.492551.78350@f14g2000cwb.googlegroups.com>,
 <msadkins04@yahoo.com> wrote:
>
>Gregory L. Hansen wrote:
>> In article <1107743305.741695.255450@c13g2000cwb.googlegroups.com>,
>> <msadkins04@yahoo.com> wrote:
>
><snip>
>
>> >
>> >(2) Take your silly "one-turn secondary" comment literally, > >for a
>moment; take that wire, disconnect it from the
>> >transformer circuit, and suspend it in mid-air by some
>> >well-insulated means, but well within the same distance
>> >of the magnetic field generated by the primary. Now send
>> >high-voltage *DC* through the primary; then quickly
>> >increase that voltage to a much higher DC level. Does
>> >the "varying magnetic field" generated by it cause gross
>> >current to flow through the body of that disconnected
>> >secondary wire?
>>
>> I've just done that. A spool of wire (I don't know how many > turns)
>connected to a function generator (Stanford Research
>> Systems DS345) with a 10 ohm resistor in series, and a
>> scope (Tektronics 2445B) across the resistor to monitor the
>> current. And a one turn secondary with a 10 ohm series
>> resistor, and a second channel on the scope across that
>> resistor. I put a square wave on the primary, and damned
>> if the secondary didn't spike every time the primary changed
>> state. I put a sawtooth wave on the primary, and found a
>> square wave on the secondary with an amplitude that
>> increases with the frequency of the sawtooth on the primary.
>
>Of course you did. Do you know why? Because you created a completed
>circuit consisting of: the first part of the wire, the resistor in
>series with it, the lead of your scope coming off the far side of the
>resistor, the scope innards, and the other scope lead connected to the
>near side of the resistor. You created a grounded, completed circuit,
>and lo and behold, the magnetic pulse induced an electric current. But
>only through that circuit. There was no current flow through the rest
>of that wire, of course.

You're halfway through freshman physics, right?

The scope, and voltmeters in general, have an input impedence at least in
the megaohms. FET inputs are easily in the gigaohms. It's true that some
current was flowing into the scope, but a realistic figure is that about
a millionth of the current flowed through the scope while the rest flowed
through the resistor. The scope has approximately no effect on the
circuit, which is exactly the intent of its design.

On the other hand, I put a resistor in the loop specifically so that there
would be a voltage drop that could be measured, and selected ten ohms
because it's large compared to the resistance of the copper wire but small
compared to the input impedence of the scope.

>And if there had been no resistor and no
>scope, but just a piece of wire, why on earth would you claim to think
>that current would flow down its length? Electric current requires two
>things: (1) a path; (2) an electromotive potential difference. A
>straight length of isolated wire provides neither. A closed loop of
>isolated wire provides neither. There is no current flow there.
>(Another possibility is direct induction on your scope, but I tend to
>discount that.)

You haven't gotten to Maxwell's equations yet, have you? It's true that a
potential that drives a current around in a loop of wire cannot be created
by a collection of static charges. I never said otherwise. Now I'm going
to quote that equation again, are you ready for it?

  curl E = -dB/dt

What is the definition of the curl, and how does that relate to currents
going around in a circle? You know what dB/dt means, right?

>
>A straight wire is not a circle and gross current cannot flow along the
>conductor circularly. A closed loop of wire, though a circle, provides
>no path because there is no potential difference. EMF at the "start of
>the circle" (wherever you wish to define that) is the same as at the
>end, because the end is also the start. It's a circle. Superconductor
>ring currents are different, because there it is not externally applied
>EMF which keeps the electrons going *once they have been accelerated*,
>but rather their own momentum and the absence of electrical resistance.
>
><snip>
>
>> >
>> >(3) You can't have it both ways. If a magnetic field of
>> >16 T, propagating, washing over a frog in an ambient
>> >magnetic field of 0, is "quasi-stationary", then so is a
>> >propagated change in that field from 16 T to 32 T. The
>> > difference in both cases is 16T.
>>
>> Quasi-stationary (I suppose quasistatic is a more common
>> term) means the field changes slowly enough in the region
>> of interest that propagation effects can be ignored.
>> That doesn't mean the magnetic field won't change in the
>> region of interest. The field versus time will be
>> determined by the current versus time in the magnet.
>> But in the quasistatic approximation, if you change the
>> current at point A, you don't worry about how long it
>> takes that to affect the field at point B.
>
>First, a field propagating at c does not cause changes slowly. A
>faster influence cannot be postulated in contemporary physical models.

What we have here is a failure to communicate.

Suppose you have a magnet that happens to put one tesla per amp into a
test region, it starts with zero current, and you have your finger on the
current adjust.

Over the course of one second, you turn the knob so that the current goes
from zero to one amp. At the end of one second plus L/c where L is the
distance from the coil to the test region (e.g. for L=10 cm, L/c=3e-10
seconds) the magnetic field is one tesla. Yes?

You keep turning up the current adjust so that at two seconds there is two
amps through the magnet, which means at 2.0000000003 seconds there is two
teslas in the test region. Are you still following me?

Keep turning the knob so that at 3 seconds there's 3 amps in the magnet,
and at 3.0000000003 seconds there is three teslas in the test region.

The quasistatic approximation is where we say "This is stupid, any wave
effects there are too weak to do anything observable, my fingers don't
have picosecond resolution, and my timer can't measure picoseconds. I'm
going to stop writing all those sig-figs."

But suppose Mark Adkins thinks we still have to include the propagation
time. That's fine. Calculate the change in magnetic field over some time
interval, e.g.

  (3 tesla - 2 tesla) / (3.0000000003 sec - 2.0000000003 sec)

  = 1 tesla/sec

Propagation delay drops out anyway, whoopee.

>By influence, I don't mean the time it takes the machine to warm up, or
>the time it takes to travel across the chamber, I mean the time it
>takes to travel the length of the frog's body.
>
>And you can't ignore it, because that change in magnetic field strength
>is what causes magnetization. The subsequent, continuing, field of the
>same strength merely maintains it, and does work if it moves the frog.
>Consider how a cyclotron works. The magnets cannot impart additional
>rotational energy to an electron whirling around in it: they can only
>maintain its orbit.

STATIC magnetic fields cannot do work on a particle. A dB/dt is not
static, it produces a -curl E, and an electric field can do work.

>
>In order to magnetize that frog, you have to increase atomic
>spin-energy in each of its atoms. That's what that 2 amps of
>non-dissipative atomic microcurrent represents (and by the way, these
>are ubiquitous, not surface, currents, because every molecule in its
>body is magnetized. This magnetization is negative relative to the

Maybe you're thinking of a flux. A current is a net flow. I had assumed
at some point you'd seen in your text a diagram with a bunch of little
boxes with arrows drawn inside an object that shows how a segment of
current about one point is cancelled out by the segment of current about
an adjacent point.

    -->-- -->--
   | | | |
   ^ v ^ v
   | | | |
    --<-- --<--
    -->-- -->--
   | | | |
   ^ v ^ v
   | | | |
    --<-- --<--

>field, because M = kB and k is negative here. You can't increase the
>spin energy of the atomic electrons directly by means of a magnetic
>field; the magnetic field must induce an electric field to do that.
>That occurs when the ambient magnetic field strength increases from 0
>to 16 T as the field propagates over that frog when it is first
>exposed. The continuing magnetic field maintains the magnetic
>polarization that is necessary to keep those microcurrents stable,
>instead of entropically decaying via spontaneous emission of radiation.

Yes, thank you. 

-- 
"'No user-serviceable parts inside.'  I'll be the judge of that!"

Relevant Pages

  • Re: any value to "NORET_TYPE" macro?
    ... Robert P. J. Day wrote: ... so it looks like a thoroughly obsolete macro which can be tossed. ... So the way to "test" your patch would be to see if there is any increase / ... decrease in the number of *warnings* blurted out by gcc during kernel build ...
    (Linux-Kernel)
  • Re: [RFC] enhancing the kernels graphics subsystem
    ... A simple place to start is OOPS display while in graphics mode. ... driver and left the GPU messed up. ... This implies that the kernel driver needs to track the dimensions and ...
    (Linux-Kernel)
  • Re: OT Question for Iggy
    ... superconducting magnet with a homogenous cylindrical volume of about 2" ... It's the magnet for a Fourier transform ... For whatever reason Oxford favored small diameter wire and lots of turns, ... which has the effect of making the inductance of the coil very large. ...
    (rec.crafts.metalworking)
  • Re: Coulomb and Lorentz Gauge
    ... mainly between the deflection coils and the beam current, ... and wires that the magnetic force will causes the outer tires to move inward just as wire wrapped on a nail will bunch-up when we ... The beam current is perpedicular to the wire in a focusing magnet and parallel to the wire in a deflection magnet. ... some of the oil droplets fall through a small hole cut ...
    (sci.physics.relativity)
  • Re: More Help on RPM Calculations?
    ... Then fix a magnet to your wheel somewhere. ... Wrap some wire around a chopstick and hold it near the wheel. ... I have a 4-channel digital scope sitting on my desk. ...
    (rec.motorcycles)