Next Meeting: September 2: Social gathering Next Installfest: TBD Latest News: Aug. 18: Discounts to "Velocity" in NY; come to tonight's "Photography" talk Page last updated: 2002 Mar 27 00:12
[vox] physics time (was: Smoothing out the powerline signal?)

# [vox] physics time (was: Smoothing out the powerline signal?)

```hey there,

i've been really busy lately, so i haven't been following this thread.
my favorite brand of mathematics is differential geometry.  but my *2nd*
favorite brand of math is vector calculus.  the the study of E+M IS the
study of vector calculus.  ergo, i really like E+M.  :-)

begin ME <dugan@passwall.com>
> On Tue, 26 Mar 2002, Mark K. Kim wrote:
> > On Tue, 26 Mar 2002, ME wrote:
> > > How close is your fridge to your computer and other devices?
> >
> > Umm... it's a mini fridge... doubly serving as my printer stand :)
> >
> > I can't believe an EM from the fridge can disturb a huge device like an
> > electron gun inside the CRT, especially when the monitor is shielded.

i think that depends on the current draw, since the magnetic field from
a wire is, roughly:

B = I*L / distance from wire     (see my comment on biot-savart below)

> > Plus, the power line is fluctuating all the time (60Hz) but that's not
> > very apparent at all; why would it become so apparent only when the fridge
> > turns on/off?  Can you convince me it's at least plausible?

when a "fridge turns on", what actually happens is that a compressor
turns on.  it compresses a gas to above atmosphere pressure, puts it in
the vicinity of the place you want to cool and then lets the gas rapidly

the first law of thermodynamics says:

change in energy = change in heat energy - work done on the system

delta U          = dQ + dW
delta U          = C delta T + P delta V

C is the heat capacity (constant), P is pressure, V is volume, T is
temperature.  in the case of a fridge, the change in energy = 0 (assume
an ideal fridge).  this gives:

C delta T = -P delta V

so if you let a gas expand rapidly, delta V goes up.  this makes delta T
negative.  the gas cools.

anyway, compressors are notoriously *very* current intensive.  without
the compressor going, a fridge should draw almost no current at all.  i
think it really does pass the plausible test.

> Now, the amount of EMI generated by a powered device is (for the most
> part) rather steady. (Yes, I know it is not entirely true, but just for
> simplicity let us assume it is.)

i'm not sure that's right.  it's all based on current draw, and a fridge
draws a huge current when the compressor is on, and almost none when the
compressor is off.

> Now, in the case of a device shifting on, there is (assuming no other
> devices, and ignoring the earth's own field,

at 10^-4 tesslas, the earth's B field ain't gonna do much of anything.
:)

> It is possible to generate electron flow with magnets (magneto, or think
> turbine with hydro-electric power),

modification: it's possible to generate electron flow with magnets
*moving* relative to a conductor.  the key here is moving.  no moving
magnets, no induction.

> but it is also possible to create a
> strong magnet with electron flow. They are actually linked.

the curl of E = - time derivative of B

so a changing magnetic field is a source for a non-conservative electric
field.  helmholtz theorem states that any vector field A can be written
down as a sum of a curl source and a divergence source:

A = curl(B) + div(f)

for some vector potential B and scalar potential f.  so the electric
field has a conservative divergence source (stationary charges)  and
non-conservative curl source (changing magnetic field).

EMF = -time derivative of magnetic flux

magnetic flux is is "how much B field goes through a surface".  this
illustrates that moving magnets are the key.  if the magnet is
stationary, flux is not a function of time.  so it's derivative is 0.
if the derivative is 0, EMF is 0.

> Magnetic fields, can move electrons. (Ever made your
> own electromagnet by wrapping wire in a cyll like a spring around a
> screwdriver, and apply a DC current through the wire?)

changing magnetic fields produce a non-conservative electric field,
which can accelerate both stationary and non-stationary charges.

static magnetic field can accelerate *only* electrons which are moving
in a direction that's non-parallel to the magnetic field.

> Once you can convince yourself of the impact of EMI on a CRT with your own
> magnets and see how distance plays a very strong hand in it wit the source
> and destination, then you may be able to accept other EM interference as a
> possible problem. I do not recall all of the equations or constants
> (physics was many years ago), but I seem to recall it being something like
>   q(1)q(2)
> K---------- = F
>     r^2

that's the force that one stationary charge has on another stationary
charge.  i think the relationship here might be:

F = q(E = v cross B)

which gives the force on a moving electron due to E and B fields in SI
units.

the distance relationship you're looking for is actually in the
calculation of B, via the biot-savart law:

B = integral  I dl cross r
------------
r^3

so B itself has a 1/r^2 relationship itself.  when you compute this for
a line charge, it turns out that B is a function of 1/r.  that is, B
drops off with 1/distance from the wire causing the B field.

pete

ps- at one time in my life, i really did do more physics than computers.
:)
_______________________________________________
vox mailing list
vox@lists.lugod.org
http://lists.lugod.org/mailman/listinfo/vox

```

• Follow-Ups: