Why are the 'Fixed Stars' so FIXED?

"Jerry" wrote in message
oups.com...
On Jul 16, 4:15 am, George Dishman wrote:
On 15 Jul, 10:23, Jerry wrote:

On Jul 15, 1:25 am, HW@....(Henri Wilson) wrote:
On Sun, 15 Jul 2007 00:46:58 +0100, "George Dishman"
wrote:
"Henri Wilson" HW@.... wrote in
messagenews:vjmi935pingk5bt6apah6d0saailmiun87@4a x.com...
On Sat, 14 Jul 2007 18:19:02 +0100, "George Dishman"

http://www.georgedishman.f2s.com/Henri/fit_lum.png

That page wont come up. Do you have the correct address?

I just clicked the link and it worked fine.

It still wont come up...... Don't know why.

Perhaps it is your pop-up blocker(s)?

It's not Henry's fault, the link doesn't work for
me from work though it did from home. I'll
have to investigate.

The links were working for me when I wrote that post.
Then they disappeared. Most likely I was viewing cached
pages???

Yes, the server went down and they restored
from a backup that was 48 hours old. The
graphics are available again now.

George

Sorry this has taken a while.

"Henri Wilson" HW@.... wrote in message
...
On Sun, 8 Jul 2007 23:16:50 +0100, "George Dishman"
wrote:
"Henri Wilson" HW@.... wrote in message
. ..
On Sun, 8 Jul 2007 09:49:59 +0100, "George Dishman"

Henry's snip of the context restored

If you differentiate the measured radius of the star,
you get the radial velocity. Differentiate the radial
velocity and you get the radial acceleration. The
shape of the luminosity curve matches the velocity,
not the acceleration.


George, you are definitely going backwards now. What you say is true
only if
light speed is constant.

Nope, velocity is the derivative of the radius in
_all_ theories. Schoolboy calculus as you said Henry.

that's not in dispute.

It is precisely what you disputed above.

We are discussing a theory that say it isn't.

There is no observed component
of the acceleration curve in the luminosity.

BUNCHING George!... BUNCHING!!!!

Yep, that is the effect we agreed to call ADoppler
and there is NONE detectable for L Car.

How would you know from that miserable curve?

Because it matches the integral of the velocity
curve, not the second integral, including allowing
for the variable speed distortion.

It depends on emission delay ...

Yep, "VDoppler".

and c+v at emission...

More accurately the rate of change of v at
the time of emission, that gives the ADoppler term.


as welll as distance of
course.

To a lesser extent as we are always well beyond
the equalisation distance.

You are missing the main point again,George.
It is 'photon density' that matters. Photon density is mainly ADoppler
dependent.

You are missing the point entirely Henry, the
curves are of "photon density", well strictly
relative photon arrival rate.

I have a whole page of matched
cepheids now.

None has a radius curve, they do not distinguish between
VDoppler and ADoppler.



You don't seem to understand, since they don't
differentiate between the possibilities, they
are worthless.

You seem to only recognise VDoppler....

Without a radius curve, you can tweak your parameters
to match the luminsity with either. If you want to
show me evidence that it is ADoppler and not VDoppler
you need something more. For LCar we have that, the
radius curve suffices, but L Car turns out to have all
VDoppler and no ADoppler, so where is your evidence?

How can you tell if it is A or VDoppler, George ?

Because I can do schoolboy differentiation. I fit
the velocity curve like this

http://www.georgedishman.f2s.com/Henri/fit_vel.png

To do that I used these _real_ parameters:

http://www.georgedishman.f2s.com/Henri/sine_wrong.png

and then adjusted to take account of the variable
speed effect to the the observed curves:

http://www.georgedishman.f2s.com/Henri/sine_right.png

Then I checked the radius

http://www.georgedishman.f2s.com/Henri/fit_rad.png

Then I look at the luminosity

http://www.georgedishman.f2s.com/Henri/fit_lum.png

The ADoppler curve is the time distorted version
of the derivative of the velocity curve and it is
nothing like the luminosity.

VDoppler certainly can't produce that much brightness change.

Agreed, ballistic theory cannot match L Car.

Of course I knew it was a bloody exponential...or at least something
like
an exponential.
Actually, if you think about it, it might not even be exponential.

Partial dv by ds is proportional to v-c/n.

No. You cannot even assume that condition is true.

To be physical, the function must be "well behaved"
in the mathematical sense and you say speeds above
c/n get reduced while speeds below get increased so
as v-c/n - 0, it must reduce to first order. Come
on Henry, you claim to know maths so I shouldn't
have to help you out like this.

Solve the
schoolboy integral Henry, it is an exponential.
Obviously if n varies with s, you get other solutions
too.

George, think about this:


A


B


C



1__2___3____4_____5______6____-v,a D


An accelerating source emits pulses of light at equal time intervals
at
the
points shown. The speed of each pulse is c wrt its source.
You can easily imagine how the pulses bunch together as they
approach
the
three
points A, B, C and D.


A line length problem. They should be as I have now drawn them above.

Just consider point D, we are only
interested in the radial components and any transverse
component of the speed affects the radial only via
Pythagoras as a second order speed term so changes the
result by less than one part per million.

You can see that the pulse density distribution at A, B, C and D can
be
manipulated by curving the path of the source.

They are still in the wrong places. A, B and C should be in line
directly
above
D.

OK, that makes a bit more sense. Now draw a line from
point 1 to each of points A-D. We are always ignoring
the transverse component because it is second order
hence negligible, so project the velocity and
acceleration at each of points 2-6 onto the 1-D line
and you get "radial velocity" and "radial acceleration"
for each. Those are the 'v' and 'a' terms we are using.


Nothing you are saying is relevant to this process.

Think again. Consider points 1 through 6 as the surface
of the star as it expands away from the centre. The
photon rates you will calculate will depend on the speed
and acceleration of the surface, both of which can be
found by differentiating the radius.

So where do you include observer distance in your approach George?


Since speed equalisation happens over a distance
that is much smaller than real observer distances
(one light minute compared to 4000 light years for
J1909-3744) the overall distance doesn't affect
the result. The change in distance affects arrival
time by the diameter of the orbit - 3.8s for the
pulsar. I take that into account but you said you
didn't bother.

Here we go again...changing the subject to get out of a tight spot....

Not really, maybe I gave to much detail. Let me
repeat, you ask "where do you include observer
distance in your approach" and I answered that
it is included as the VDoppler term, OK?

Where?

..... you haven't included it in the important calculation.

Sorry Henry, I slightly jumped topic there. We were
first talking about the bunching due to motion of the
source towards the observer between pulses. That is
taken into account by the (1+v/c) term in the photon
arrival rate equation.

You also asked about the total observer distance.
Without speed equalisation, it appears as "d" in
the term 1/(1-da/c^2).

With speed equalisation, once the speeds are matched
any farther travel doesn't affect the apparent relative
photon rate which becomes asymptotic to a value given
by 1/(1-Ra/c^2) where R is the characteristic distance
of the speed equalisation.

I hope that's clearer.

OK, now apply that to the carrier and sideband
frequencies independently. Then inverse transform
the three to get the received waveform. What speed
does the modulation travel at? Show your working ;-)

I don't see the point.
What are you getting at.

Yet again Henry, consideration of sidebands allows
you to calculate the Doppler shift directly from
the speed of modulating pulses hence a speed of
c+v determines the shift.

You don't have a model for individual photons.

I don't need a model, I just need to know that
they deflect by the same angle as the classical
wave on hitting a grating which is proved by
the photomultiplier experiment. Again this is
something I have pointed out dozens of times.
When are you going to stop trying to change the
subject and address the proof?

My photon model conforms with grating diffraction behavior and
includes
corrections for light speed variations.
...
Still trying to change the subject Henry? Your
model must also tell you that the deflection
angle tells you the wavelength of the reflected
photon hence my argument remaains valid.

I don't know which argument that its...

It got snipped, I have restored the context above.

And I have restored it again.

The argument I presented was that single photon
observation of grating deflection shows that the
photons mostly land where the classical analysis
says the peak intensity should occur hance my use
of classical analysis is valid.

That doesn't worry me. The experiment has never been done with a fast
moving
source.

It doesn't need to be, photons carry the energy so
where they land is where the maximum intensity
occurs. Let me be clear Henry, it doesn't matter
what angle they get deflected through, it only
matters that the classical intensity is a maximum
at the same place as where the photons land. That
means the methods of prediction are equivalent.

...but my theory explains why an HST
grating can detect its own orbit movement whereas YOUR theory says it
should
not.

Wrong, SR predicts the deflection correctly, but
you are just changing the subject again.

No, this is important.

This has nothing to do with the topic in hand,
you are just trying to weasle out of the subject.

YOUR theory says the HST should NOT detect its own movement because its
grating
is wavelength sensitive only and there is no way the movement of the
grating
can affect the absolute wavelength of the incoming light.

Sorry Henry, wavelength _is_ affected by Doppler.

Go back and read your posts again.

Just use some imagination George.

No, I'll just point out you are now telling me
the opposite of what you said before.

You probably misinterpreted what I said before.

Whatever, I said you needed to make the sphere
at the bottom of your animation static relative
to the barycentre of the binary system. You now
seem to be saying the same.

Use some imagination George.

No Henry, it's your job to say what you mean. You
can't tell me one thing one day and the opposite
the next and then say "Use some imagination" when
your contradictions are pointed out.

I didn't say you were wrong. The middle diagram shows just that. ..so
how
could
you possibly tell me somethingi already knew?

If you look back at the conversation, I said the top
and middle diagrams were fine, it is the bottom one
that applies and there should be a single merged
sphere which is at rest relative to the barycentre of
the binary system.

It wont be at rest.

Then the pulsar should have shown distortion due
to the ADoppler from the edge of the sphere.

It will move...but by less than it would without the second
star.


The stars are of different size. .....the sphere wil move...or at least
have
moving 'bumps' on it.

Because the stars are different, the larger moves
less than the smaller. The combination of the small
motion of a large source and vice versa gives a
merged sphere which is static wrt the barycentre.

It wont be static. It will wobble....by an amount depending on relative
star
sizes and 'sphere' diameters.

See above.

Nope, it's less than a millionth of the gravitational
bend (from memory, Craig Markwardt posted the details
about a year or more ago in reply to Sean). You can
easily separate the effects since the optical is
frequency dependent while gravitational is not.

The 'Wilson sphere' around a star bends the light.

As I said, optical effects are eliminated by
frequency dependence.

Wilsonian spheres are not frequency selective.

"Wilsonian spheres" are nothing more than the solar
plasma which is most definitely frequency dependent.

They are a lot more than that. I think they are probably closely linked to
gravity fields.

ROFL, Henry you are a card. You do realise that you
just explained that ballistic theory predicts half
the bending by gravity because it doesn't take account
of ... gravity :-)

The variation in the distance from source to observer
is what produces the "VDoppler". If the source is
moving towards you at constant speed v then photons
emitted at times separated by dt will travel v*dt less
distance so they will take v*dt/c less time to arrive
than t. That reduces the period so increases the
frequency by the ratio we call VDoppler.

That effect is generally negligible. Forget it...

Henry, you have been telling me I have got it all
wrong because I didn't take this into account. Now
you realise I have been all along, you tell me to
"forget it". Make your mind up.

George, you can generally assume the pulses all move the same distance.

No, I assume each pulse starts from a position which
is closer to the observer by v/f where v is the speed
and f is the frequency (photon rate). I also assume
the pulses travel at speeds that differ by a/f where
a is the acceleration. Both those factors affect the
arrival rate.

It is the time they take to get there that matters. The time differences
in
crossing the orbit are very small compared with the total time taken.

That affects the arrival _time_, not the rate.

But as you know, we need to include it in the case of pulsars and contact
binaries which have very short extinction distances.

You always need to take it into account, but we were
talking about the photon arrival rate equations I
posted, not the arrival time equation.

Well it is certainly tough for you. It makes your argument look like a
joke...

It means your claim to have matched the curves
is nonsense by your own admission.

BaTh produces curves of the right shapes and magnitude changes.
That's pretty impressive don't you think?

http://www.georgedishman.f2s.com/Henri/fit_lum.png

I don't that fit is impressive at all, in fact I
don't think it fits at all.

other
variables...

Sorry Henry, the shape is wrong for ADoppler and you
just said (correctly for a change): "VDoppler doesn't
produce anything like the observed brightness
variations.", so no, you haven't matched anything.

The shape is NOT WRONG.

http://www.georgedishman.f2s.com/Henri/fit_lum.png

The ADoppler shape is hopelessly wrong.

Yes Henry, and you _still_ can't do it after how many
weeks? Here are the plots again, just apply schoolboy
calculus to the bottom one and you get those above.

The dispute is not over the calculus. It is about the shapes of typical
curves.

http://www.georgedishman.f2s.com/Hen...lustrative.png

The bottom one is similar in shape to the measured values

It isn't.

Here is the shape:

http://tinyurl.com/239mw6

They match.

That's nothing like your radius curve

Really? Have another look:

http://www.georgedishman.f2s.com/Henri/fit_rad.png

It is as good as anything youhave produced.

and any fool can see it is nothing like a
best fit to the points anyway.

A fool is often deceived which is why scientists
uses mathematical methods.

The authors obviously knew the answer they wanted and proceeded to draw a
curve
to suit. .

ROFL, Henry you really are clueless. Nobody "draws
a curve" in a real paper, they calculate the values
and plot the result.

Exactly, and the curves you are producing do _not_ match
the top plot, they match the middle plot which is VDoppler,
not ADoppler.

Can you not get it into yopur head that ADoppler produces a similar
curve.

Are you incapable of doing schoolboy calculus (your
own words). Take the smoth curve he

http://tinyurl.com/239mw6

Differentiate once to get velocity then a second
time to get acceleration and compare them. Get it
into _your_ head that the curves are NOT similar
at all.

George, those curves are not similar. ..but the resulting photon density
curves
ARE.

Henry, for the umpteenth time, those ARE photon
density curves.

You are performing the wrong calculation.
For god's sake George, forget your equations and let the computer do the
work.

The computer _did_ calculate them for me, I just
cut and paste over the observed curves and rescaled
to fit.

Have a look at RT Aur.
It's curve is nothing like yours.

Of course not you idiot, you have a VELOCITY curve
for RT Aur and you are comparing it againgst my
RADIUS curve for L Car.

well george, please try to produce 'photon density' curves vs distance, in
future. These are what we want.

No, what you want is relative photon arrival rate
(number of photons per second arriving divided by
the number transmitted) versus the phase, and guess
what Henry, that's what those curves show.

George, the fact that most cepheid velocity and brightness curves are
roughly
in phase tells me that any companion must be fairly distant...

The fact that the period is of the order of days
tells you it isn't.

that's quite far enough. Our moon orbit every 28 days and rarely
eclipses
earth.

Rarely? How many times a century on average does it
eclipse Regulus for example? Now do the statistics
for the number of known Cepheids and see how often
we should detect an eclipser.

....but YOU claimed most cepheids don't have companion objects ....

No, be careful. Many Cepheids will be part of binary
systems but the period of that binary is unrelated to
the variability of the star. What you were suggesting
was that the Cepheid itself was a pair, a visible star
and a dark companion. What I am saying is that that is
not the case, they are single "huff-puff" stars as you
put it. However, you still expect the ballistic model
to produce dynamic effects as a result of the variable
radius, that's what I am discussing. Do you follow that?

Not without a companion.

Every star in orbit has a companion or two....

Nope, only about half.

How and why would a star be in orbit if it didn't have something else to
orbit, George.

About half are in binaries, the rest are simply
in galactic orbits. Our Sun for example is not
part of a binary system.

Yes it is.

No it isn't Henry Don't be silly.

Everything is in some kind of orbit around a mass centre.
Think about it.

Sure, the sun is gravitationally bound to the galaxy.

Of course the period might be huge and the orbit very unstable.

The period is a bit over 200 milllion years IIRC, and
the Sun oscillates through the plane with a period of
about 80 million years, but it isn't bound to any other
star so it is not part of a binary system.

Right so the cause is VDoppler, not ADoppler. Nice
of you to finally admit it.

It is also similar if the shift is 100% ADoppler.
You ar still missing the point.

No Henry, you are missing the point, if it was
ADoppler, the radius curve would not match.

Forget the radius curve George.
Produce the pulse density curves at the observer distance. That's what we
need.

I'll even give you a hint.

Determine the density distribution over a distance equal to 1
'light.period'
then move the observer along that 'distance' at c.

I'll give you a hint Henry, that's what the existing
curves are.

The variation in the distance from source to observer
is what produces the "VDoppler".

That's wrong for a start.

Go and think about what causes VDoppler.

George, all I can suggest is that you try to write yourt own BaTh program
so
you will get to know what is really involved.

For binaries, it is easy. For a Cepheid, you get all
the complexity of edge darkening, optical depth,
variable composition temperature and pressure
variations and so on. The ballistic theory part is
completely trivial in comparison.

...this is ridiculous. I'm not going to explain again.
The distance across the orbit is generally negligible compared with the
LYs of travel.

Read again what I said, the distance across the orbit
is NOT what we are talking about.

Rubbish.

You have simply lost the plot Henry.

You want to add that distance to the distance traveled by each
pulse...and of course that is required for pulsars.

Sure, but it is small in comparison to the effect of
variable speed. I don't mind if you exclude it from
your program as you previously said you did or if you
include it but I have already accounted for it in the
photon _rate_ calculation and I believe so have you.

I have it, George. Differentiating the radius twice SHOULD match
OBSERVED
velocity curves.

Differentiating radius _twice_ gives radial velocity?
Boy you really have lost the plot!

I said OBSERVED velocity curves.

Why, you keep telling me the velocity curve isn't
affected by ADoppler which is why the luminosity
can vary be sevreal magnitudes while the OBSERVED
velocity curves are still only a few km/s.

Yes, that's how it turns out George.

OK, so stick with that and stop contradicting yourself.

I haven't lost the plot at all George.

Yes you have Henry, you are now relying on something
you told me doesn't happen, and you invented the K
factor to justify.

There is good reason to believe it exists.

It is impossible for it to exist Henry, but you need
to understand some physics to follow the proof I gave
you.

Photons can hardly shrink forever

Of course not, "superposition" applies to EM so
when the back of the photon catches up with the
front, it passes it unaffected. At least that's
what ballistic theory requires.

whereas there is no restriction on inter-photon movement.

I wouldn't accept any of this.

Of course not, you always deny reality when it conflicts
with your fantasies.

It's a silly curve anyway. ..only half the points are present.

The Earth rotates Henry, L Car isn't always above
the horizon, and sometimes there are clouds. Trust
me, living in England this year I have experienced
lots of clouds!

...more excuses....it's the clouds now....

Welcome to reality.

You invented your K number to remove ADoppler from
photons which means it _is_ the true velocity. Your
own program assumes the velocity has the same peak
to peak value for measured and true, or have you
changed that?

George, I give up. You are getting this completely wrong...

I'm just repeating what you told me. Go back and
look at your own posts and then try to clarify
what you are saying.

Write a program that will produce photon bunching vs distance just for
circular
orbits.
You will learn a lot in the process.

Been there, done that.

The fact is that, as you can see on that diagram,
the measured radius curve is an excellent match for
the integrated velocity curve. The only variable is
the distance to the star used to convert the angular
diameter which also matches other methods of measuring
the distance.

That is what we have been talking about for a couple of
weeks now so if you have finally grasped the plot, maybe
you can say something sensible about it this time.

Yes, it's rubbish...
Nobody has actually seen a cepheid radius changing.

Denial again Henry? ESO has measured it.

They used doppler shift.

Nope. Try again.

They use spectral line doppler shift.

Nope. Try again.

George, velocity curves calculated from grating angles with VDoppler
equations should be similar to brightness curves.

Why would you use a Doppler equation to calculate a grating
angle? You've lost the plot again.

Read again what I said...

OK, I think I see what you mean but you claim
luminosity is mostly due to ADoppler because
VDoppler gives insufficient variation. There are
two possibilities - the spectral shift might be
due only to VDoppler as you claimed before or it
might also include ADoppler as I say it must.

In the first case, the velocity is a true measure
and that fits the match of L Car to the derivative
of the radius.

No no. That's only your assumption.

No, it is what you keep telling me, K 1 therefore
there is a negligible ADoppler contribution. You can't
have it both ways Henry, you know VDoppler cannot
produce the magnitude of the luminosity variation so
you have to separate them using your K factor. That
means that the spectral shift is only VDoppler while
the luminosity is both VDoppler and ADoppler.

The shape of the luminosity curve
would be quite different from that of the velocity
curve since it would be mainly due to the
acceleration and that is a very bad fit for L Car.

This is not true.

See for yourself:

http://www.georgedishman.f2s.com/Henri/fit_lum.png

The ADoppler curve is nothing like the luminosity
in either band.

In the latter case you are right, the shapes would
be similar but you are back to the problem that the
observed velocity is should be far greater, the
observed velocity curve would only produce a few
milli-mag variation.

I think you need to sort out exactly what you are
claiming Henry, you are contradicting yourself all
over the place and none of the combinations that I
can draw from it works at all.

I am not contradicting myself. You are simply not getting the message.

You contradict yourself when you first claim K is
very small so that you can explain the luminosity
variation and then also say the velocity curve
isn't just VDoppler. Make your mind up.

George

On Mon, 16 Jul 2007 05:39:48 -0700, George Dishman
wrote:

On 14 Jul, 23:53, HW@....(Henri Wilson) wrote:
On Sat, 14 Jul 2007 09:47:52 +0100, "George Dishman" wrote:
"Henri Wilson" HW@.... wrote in message
.. .

Note that, if you have a short speed equalisation
distance as has been the case with all the examples
so far, the speeds become the same and photon (or
pulse) spacing remains constant even in ballistic
theory.

Yes...but don't get confused between a sphere that moves WITH the star and one
that does not.

In the examples so far, we saw no ADoppler so if
spheres existed, they had to be motionless.

I wish you wouldn't repeat this nonsense over and over, George. Are you trying
to convince yourself that it's true.
ADoppler is the main cause of brightness variation. Variation of 3 or more
magnitudes are easily obtainable with ADoppler.
The VDoppler contribution is generally negligible in all but pulsars and
contact binaries. My 'sphere' demo explains why.



George your equations are a joke.
Where is distance, where is time, where is eccentricity, where are the orbit
properties?

Oh dear, Henry I am assuming you can do algebra at
secondary school level. Time is not shown explicitly
because you are supposed to realise that velocity v,
and acceleration are both functions of time. I could
write v(t) instead of just v but you should be aware
of that convention anyway.

No George, where is brightness vs TIME.

Distance does appear in the equations. From below:

f'/f = c(c+v)/(c^2-da)

That would be the 'frequency' shift of a single photon on arrival, if classical
wave theory applied....and it doesn't. Photons are particles.

where d is the distance to the observer.

Your reading disability is showing.

Ecccentricity etc. don't appear because the equations
are universal, they apply to any source motion
regardless of how it is produced, so orbital motion,
surface expansion of a Cepheid and your 'laser in a
wheel' are all covered. In each case you need to
work out the functions v(t) and a(t) from the
dynamics.

What do you think a computer is for.

Let me give you another example, you could
attach a mirror to a loudspeaker cone and shine a
laser on it. The shift of the reflected light would
also be described by my equations, but you would need
to look at the waveform of audio to work out the
velocity and accelerations involved.

George, you are missing the point.
You are not relating your equations to observed brightness variation vs time.

Yes it is Henry, it is the part that occurs if the
source is moving towards you at constant speed and
mathematically depends only on the velocity.

..and it is responsible for a very small brightness variation due solely to the
energy E variation, = h(c+v)/lambda

Nope, I don't calculate that factor at all, my
equations relate _only_ to the 'photon bunching'
due to the velocity and acceleration.

....source acceleration only. Not velocity.


It is the same as the arrival frequency that you just attributed to
VDoppler.

No, we have been calling this effect ADoppler and
it varies with distance from the source, the farther
away the observer is, the more time the pulses have
to bunch up. VDoppler discussed earlier applies to
constant speed for the source and remains a fixed
factor no matter how far away the observer is.

Yes you are getting the picture George.

Don't try it on Henry, I am not "getting" anything,
we agreed those definitions _weeks_ ago and you lost
the plot. At least you are showing some signs of
remembering them at last.

George, when and if you ever try to turn your equations into predicted
brightness curves, involving eleven variables, you will start to realise why I
have spent ten years training my computer to do the work.


The one thing I can't understand is how you can still
be struggling with such simple equations after talking
about this model for over ten years. I'm sure you were
posting about it in sci.physics.relativity when I was
reading it in 1997. It is hardly surprising that nobody
believes you when you claim to have a science degree yet
cannot work out the equations of your own theory when
they are trivial schoolboy algebra.

George, you are quite free to do all the calculations with equations.

That is what any competent person would do.

However by the time you include all the variables and plot the arrival rate vs
time, you will be a very old man.

You've been working on your program for many
years, I worked out the equations in a few
minutes. You challenged nme to give you a
shortcut for calculating the arrival time
because you thought it was impossible so
I timed how long it took, just under four
minutes to give you an analytic function.

You are yet to produce ONE brightness curve.


I prefer to train my computer to do the work in seconds.

You were talking about this 10 years ago Henry
and you still don't produce the right numbers.

I can match just about any curve George.
What is more, star curves nearly ALL seem to fit into the very narrow range
predicted by the BaTh. Aren't you impressed by that?

George



www.users.bigpond.com/hewn/index.htm

The difference between a preacher and a used car salesman is that the latter at least has a product to sell.

On Mon, 16 Jul 2007 00:20:59 -0700, George Dishman
wrote:

On 16 Jul, 00:16, HW@....(Henri Wilson) wrote:
On Sun, 15 Jul 2007 20:05:22 +0100, "George Dishman" wrote:
"Henri Wilson" HW@.... wrote in message
.. .
Henry, this is the curve I posted some weeks ago
and which formed the topic in this part of the
thread throughout:

http://tinyurl.com/239mw6

ok

If you want to talk about huff puff stars then the radius curve is
unlikely to
have such a sharp turn at the bottom.

However, that's what it does. The sharp turn at
the bottom is where the "light valve" described
in the overview of the Cepheid mechanism I gave
you changes state. Of course the interpretation
with ballistic theory might differ and the set
of curves I used to match the observations were
based on the radius varying as a simple sine wave.

A sharp turnaround like that is similar to the radial velocity of a star in
orbit with e= 0.4
I match cepheids with e typiclly 0.18-0.25

Yes but the star is swelling and shrinking, it
isn't actually in an orbit. You need to think
of that aspect of your program as just a
convenient way of producing an arbitrary motion.

You have far too rapid an acceleration at the bottom.


The star swells and shrinks. While it is swelling
the surface moves towards us at some speed and
after it reaches maximum radius and starts to fall
back it is moving away from us. Speeds are of the
order of 10 to 20km/s in each direction which is
comparable to orbital speeds and should produce
ballistic "photon bunching" effects for the same
reason.

Yes I pointed that out to Jerry some time ago.

So why ask "What does that have to do with ..."?

However the surface doesn't move like a flat piston. One must integrate across
the face.

Yes, that will have the effect of broadening lines
but the same is true in both SR and ballistic theory.

That may be true but it will also affect the BaTh predicted brightness curve
shape slightly. I have to include this refinement in my program.


What that means is that if you choose some arbitrary
function for radius versus time, all the other curves
follow from it. The differ depending on whether you
use SR or ballistic theory but either way you can
adjust the modelled curve to fit and work out what
the true motion was.

I chose to start with ballistic theory by fitting
the velocity:

http://www.georgedishman.f2s.com/Henri/fit_vel.png

I still can't get that website.

Neither can I from work yet this one is still OK:

http://www.georgedishman.f2s.com/Hen...lustrative.png

Both should be in the same directory and both
are PNG format graphics and both work from home.
I'll have to look into it.

OK I can get them all now.
That is a typical Keplerian velocity curve.

Then I found that I could get a rough fit for the
radius:

http://www.georgedishman.f2s.com/Henri/fit_rad.png

Why the sharp point at the bottom? That's not right.


The fit is terrible but that is what you need to
address.

You need to include elliptical orbits and yaw angle settings.

No, it isn't an orbit. You need to allow for the
radius to be an arbitrary function of time.

.....and like I said, cepheid velocity curves could indeed be those of a
pulsating sphere that does not move in S.H.M....but they are also typical
Keplerian orbit curves.
You will have to explain WHY the surface should move as your curves require.
Why such a large acceleration at the bottom compared with the top?


Since the radius and integrated velocity curves match,
the implication is that this is such a situation.

..but in general they don't, George.

They always do Henry, if you disagree show me
one that doesn't.

You took my comment out of context. I meant 'the observed curves do not'.


So apply ballistic theory to the measurements and see
what it predicts for the true values. I am saying it
shows a match in _shape_ of luminosity to VDoppler
although the amplitude is wrong, but the that it says
the shape is entirely wrong for ADoppler.

VDoppler cannot produce anywhere near the observed brightness changes.

Right but it matches the velocity curve and the
radius is a fairly good match too which is all
you need. The SR curve is a better fit implying
the speed is equalised rapidly.

But ADoppler can and will produce the same OBSERVED velocity curves.

ADoppler can.....as well as the right shaped curves..

I'll need to find out why you can't see the
graphics. I always check by clicking the link
in my own posts after the appear on the server
and they are OK from home but these ones don't
seem to work from elsewhere.

When I sort that you will see that the shape
of the ADoppler curve is hopelessly wrong.

George, there are no sharp corners in real curves.

Anyway, my website problem doesn't stop you
using your own program to match the curves from
the paper in the meantime, match the velocity
then compare the radius and you should get the
same as mine.

You are trying to match the Willusion of the true source velocity by assuming
there is no willusion.

George



www.users.bigpond.com/hewn/index.htm

The difference between a preacher and a used car salesman is that the latter at least has a product to sell.

On Mon, 16 Jul 2007 18:45:47 +0100, "George Dishman"
wrote:


"Jerry" wrote in message
roups.com...
On Jul 16, 4:15 am, George Dishman wrote:
On 15 Jul, 10:23, Jerry wrote:

On Jul 15, 1:25 am, HW@....(Henri Wilson) wrote:
On Sun, 15 Jul 2007 00:46:58 +0100, "George Dishman"
wrote:
"Henri Wilson" HW@.... wrote in
messagenews:vjmi935pingk5bt6apah6d0saailmiun87@4a x.com...
On Sat, 14 Jul 2007 18:19:02 +0100, "George Dishman"

http://www.georgedishman.f2s.com/Henri/fit_lum.png

That page wont come up. Do you have the correct address?

I just clicked the link and it worked fine.

It still wont come up...... Don't know why.

Perhaps it is your pop-up blocker(s)?

It's not Henry's fault, the link doesn't work for
me from work though it did from home. I'll
have to investigate.

The links were working for me when I wrote that post.
Then they disappeared. Most likely I was viewing cached
pages???

Yes, the server went down and they restored
from a backup that was 48 hours old. The
graphics are available again now.

It's all OK now.
....pity you wasted so much time on something as irrelevant as those curves...

George




www.users.bigpond.com/hewn/index.htm

The difference between a preacher and a used car salesman is that the latter at least has a product to sell.

On 17 Jul, 00:58, HW@....(Henri Wilson) wrote:
On Mon, 16 Jul 2007 00:20:59 -0700, George Dishman wrote:
On 16 Jul, 00:16, HW@....(Henri Wilson) wrote:
On Sun, 15 Jul 2007 20:05:22 +0100, "George Dishman" wrote:
"Henri Wilson" HW@.... wrote in message
.. .
Henry, this is the curve I posted some weeks ago
and which formed the topic in this part of the
thread throughout:

http://tinyurl.com/239mw6

ok

If you want to talk about huff puff stars then the radius curve is
unlikely to
have such a sharp turn at the bottom.

However, that's what it does. The sharp turn at
the bottom is where the "light valve" described
in the overview of the Cepheid mechanism I gave
you changes state. Of course the interpretation
with ballistic theory might differ and the set
of curves I used to match the observations were
based on the radius varying as a simple sine wave.

A sharp turnaround like that is similar to the radial velocity of a star in
orbit with e= 0.4
I match cepheids with e typiclly 0.18-0.25

Yes but the star is swelling and shrinking, it
isn't actually in an orbit. You need to think
of that aspect of your program as just a
convenient way of producing an arbitrary motion.

You have far too rapid an acceleration at the bottom.

That is what is _observed_.

The star swells and shrinks. While it is swelling
the surface moves towards us at some speed and
after it reaches maximum radius and starts to fall
back it is moving away from us. Speeds are of the
order of 10 to 20km/s in each direction which is
comparable to orbital speeds and should produce
ballistic "photon bunching" effects for the same
reason.

Yes I pointed that out to Jerry some time ago.

So why ask "What does that have to do with ..."?

However the surface doesn't move like a flat piston. One must integrate across
the face.

Yes, that will have the effect of broadening lines
but the same is true in both SR and ballistic theory.

That may be true but it will also affect the BaTh predicted brightness curve
shape slightly. I have to include this refinement in my program.

Sure.

What that means is that if you choose some arbitrary
function for radius versus time, all the other curves
follow from it. The differ depending on whether you
use SR or ballistic theory but either way you can
adjust the modelled curve to fit and work out what
the true motion was.

I chose to start with ballistic theory by fitting
the velocity:

http://www.georgedishman.f2s.com/Henri/fit_vel.png

I still can't get that website.

Neither can I from work yet this one is still OK:

http://www.georgedishman.f2s.com/Hen...lustrative.png

Both should be in the same directory and both
are PNG format graphics and both work from home.
I'll have to look into it.

OK I can get them all now.
That is a typical Keplerian velocity curve.

It is similar which is why you should be able
to use your existing program fairly successfully
though you need the extra curve.

Then I found that I could get a rough fit for the
radius:

http://www.georgedishman.f2s.com/Henri/fit_rad.png

Why the sharp point at the bottom? That's not right.

I explained a few posts back - see at the top:

.. The sharp turn at
the bottom is where the "light valve" described
in the overview of the Cepheid mechanism I gave
you changes state. ..



The fit is terrible but that is what you need to
address.

You need to include elliptical orbits and yaw angle settings.

No, it isn't an orbit. You need to allow for the
radius to be an arbitrary function of time.

....and like I said, cepheid velocity curves could indeed be those of a
pulsating sphere that does not move in S.H.M....

Actually, in ballistic theory it isn't far from
SHM. The illustrative curves are just a perfect
sine wave with variable speed from the source to
the observer.

but they are also typical
Keplerian orbit curves.

That's because radial distance for a circular
edge-on orbit also happens to be a sine wave.
They would also be a typical pendulum curve
or a wave or sound or anything else that is
a sine curve.

You will have to explain WHY the surface should move as your curves require.
Why such a large acceleration at the bottom compared with the top?

Again:
.. The sharp turn at
the bottom is where the "light valve" described
in the overview of the Cepheid mechanism I gave
you changes state. ..


Since the radius and integrated velocity curves match,
the implication is that this is such a situation.

..but in general they don't, George.

They always do Henry, if you disagree show me
one that doesn't.

You took my comment out of context. I meant 'the observed curves do not'.

So did I, I am saying that the observed radius curve
always matches the integral of the observed velocity
curve and challenging you to show me one that doesn't.

So apply ballistic theory to the measurements and see
what it predicts for the true values. I am saying it
shows a match in _shape_ of luminosity to VDoppler
although the amplitude is wrong, but the that it says
the shape is entirely wrong for ADoppler.

VDoppler cannot produce anywhere near the observed brightness changes.

Right but it matches the velocity curve and the
radius is a fairly good match too which is all
you need. The SR curve is a better fit implying
the speed is equalised rapidly.

But ADoppler can and will produce the same OBSERVED velocity curves.

Sure, but then you have to intgrate _twice_ to
get the radius curve instead of once. Since the
true veloicity curve (rather than observed) is
close to a sine, the extra integral would give a
90 degree phase shift so the radius curve would
no longer match. Try it if you don't believe me.

ADoppler can.....as well as the right shaped curves..

I'll need to find out why you can't see the
graphics. I always check by clicking the link
in my own posts after the appear on the server
and they are OK from home but these ones don't
seem to work from elsewhere.

When I sort that you will see that the shape
of the ADoppler curve is hopelessly wrong.

George, there are no sharp corners in real curves.

Yes there are in the _observed_ curve.

Anyway, my website problem doesn't stop you
using your own program to match the curves from
the paper in the meantime, match the velocity
then compare the radius and you should get the
same as mine.

You are trying to match the Willusion of the true source velocity by assuming
there is no willusion.

No, these are the true curves without the illusion:

http://www.georgedishman.f2s.com/Henri/sine_wrong.png

These are the curves after application of the variable
speed:

http://www.georgedishman.f2s.com/Henri/sine_right.png

and this is the match:

http://www.georgedishman.f2s.com/Henri/fit_vel.png

Note that I use the curve _including_ the illusion
for matching.

Henry, please try to think through your comments more
carefully, we are wasting a lot of time with simple
facts you check for yourself. It's easy to see I used
the distorted curve and not the pure sine wave in
matching the velocity. Earlier you said there was no
sharp acceleration in the observed curves but you
can easily see that the curve in the paper does have
that characteristic, and you asked me to explain the
cause of that when it is already the first quote in
the post. I know I have better things to do with my
time and I'm sure you have too.

George

Google said this posted successfully last week
but it appears to have vanished.

On 8 Jul, 12:44, HW@....(Henri Wilson) wrote:
On Sat, 7 Jul 2007 10:50:58 +0000 (UTC), bz wrote:
"George Dishman" wrote in news:f6njt0$g7k$1
:

"bz" wrote in message
.198.139...

Kind of 'lost in the noise', right?

Exactly. It's surprising that people like Sekerin
who pushed this years ago didn't see that problem.

Lack of understanding of the relationship between information theory,
intellegence and noise.

Ever since I saw loopy graphs in that paper on ballistic light[can't remember
the title/author at the moment], I suspected that there was something loopy
about the theory itself.

That was Sekerin's paper. He made the mistake of going beyond the critical
distance. Curves like those never happen.

They don't but they should, we are almost always
well beyond that distance. That was the first hole
in the theory.

Blows holes in his concrete boat.

Not concrete, it has enough holes to be chicken wire.

Either way, it doesn't stay afloat.

BaTh is very much alive and well.
Nothng any of you has said has even dented it.

Henry, when I design circuits with resistors,
I use Ohms Law. If I put a voltage across a
resistor, it tells me what current will flow.
I use it because it is always correct.

The opposite goes for ballistic theory. If I
try to apply it to an iFog or laser gyro, it
says the device will show no fringe shifts,
but they do shift. If I use to to calculate
the Shapiro effect from a satellite on the far
side of the Sun, it says the signal will arrive
early but it arrives late. Every time you try
to use ballistic theory, it gets it wrong. The
theory has no value whatsoever.

The way he looks at it, you would need to accelerate
the whole laser to get bunching. An example might
be a natural maser in a stellar atmosphere of it was
part of a binary system.

He had the rubber cars scrunchin due to bunchin. Plenty of bunchin in lasers
but no sign of scrunchin.

Try spinning a laser in a fast flywheel with is axis pointing in one direction.

Now that given me a great idea.

Move a laser in orbit on some kind of wheel whilst keeping it aligned with the
mirror on the moon.
If its light is pulsed at say 10000 hz, it should be possible to spin it at
such a rate that many pulses (emitted at c+v) arrive back at Earth at the same
instant. This would mean that the reflected light would arrive in bright pulses
with a frequency equal to that of the wheel rather than that of the laser.

Brilliant eh?

Dumb really. You would need to keep the laser aligned
while spinning it. First simplification - put a mirror
on the wheel and shine the laser onto it. Second
simplification - the displacement towards the Moon is
a sine wave so put the mirror on a loudspeaker voice
coil and feed it a sine wave. That way it stays pointing
in the same direction and moves along a straight line.

Using an ultrasonic source, you could get very high
accelerations which theoretically should be sensitive
enough to detect ADopppler.

In reality, Ives and Stilwell already did a similar
test but just for the VDoppler part which confirmed
SR and ruled out the ballistic model.

But the emitter will have to be outside the Eatrh's atmosphere.

I need a theory that is valid if I do the test in air
as well. SR does that, remember Fizeau's experiment.
Ballistic theory will get that wrong too.

George

Henry replied to George:

http://www.georgedishman.f2s.com/Henri/fit_rad.png

Why the sharp point at the bottom? That's not right.

Atoms in the atmosphere of a "huff-puff" Cepheid variable
are initially propelled upward at very high speed, but
slow as they rise, like any ballistic object, reach a
maximum altitude of hundreds of thousands or millions of
kilometers, and fall back until they run into the next
wave of atoms on the way up. The atoms do not gradually
slow down as they near the bottom of their fall, and they
do not gradually speed up as they begin rising.

The whole process is similar to a rubber ball bouncing on
a hard floor: The turnaround at the top is gradual, while
the turnaround at the bottom is fairly sudden and sharp.
The radius vs time curve approximates a cycloid.

The atmosphere of a Cepheid is so tenuous that most of the
falling atoms are passed by rising atoms in the next wave,
and don't collide with rising atoms until they are far
below the visible "surface" of the star. Imagine two
groups of performing dolphins; one group jumps high out of
the water, arcs over at the top of the jump, and just as
they re-enter the water, the second group jumps upward,
passing the first group at high speed. So there can be a
very rapid transition from falling dolphins (or atoms) to
rising dolphins (or atoms).

-- Jeff, in Minneapolis

On Mon, 16 Jul 2007 23:07:13 +0100, "George Dishman"
wrote:

Sorry this has taken a while.

"Henri Wilson" HW@.... wrote in message
.. .
On Sun, 8 Jul 2007 23:16:50 +0100, "George Dishman"

BUNCHING George!... BUNCHING!!!!

Yep, that is the effect we agreed to call ADoppler
and there is NONE detectable for L Car.

How would you know from that miserable curve?

Because it matches the integral of the velocity
curve, not the second integral, including allowing
for the variable speed distortion.

Sorry I don't follow.
The integral of the velocity curve is the radius curve.
The L Car curve is nothing at all like YOUR radius curve.


To a lesser extent as we are always well beyond
the equalisation distance.

You are missing the main point again,George.
It is 'photon density' that matters. Photon density is mainly ADoppler
dependent.

You are missing the point entirely Henry, the
curves are of "photon density", well strictly
relative photon arrival rate.

George, it was only when I programmed your methoid that that I found the added
complexity.
As well as the f'/f factor, you still have to consider the position of that
photon wrt other photons.
I suggest you try to produce a brightness curve using your equation and you
will soon find the problem.
Your method is slightly faster than mine but it is a programmer's nightmare.

I have a whole page of matched
cepheids now.

None has a radius curve, they do not distinguish between
VDoppler and ADoppler.



You don't seem to understand, since they don't
differentiate between the possibilities, they
are worthless.

You seem to only recognise VDoppler....

Without a radius curve, you can tweak your parameters
to match the luminsity with either. If you want to
show me evidence that it is ADoppler and not VDoppler
you need something more. For LCar we have that, the
radius curve suffices, but L Car turns out to have all
VDoppler and no ADoppler, so where is your evidence?

How can you tell if it is A or VDoppler, George ?

Because I can do schoolboy differentiation. I fit
the velocity curve like this

http://www.georgedishman.f2s.com/Henri/fit_vel.png

...but I can get that with purely ADopler George.....and I can match the
brightness variation as well.

To do that I used these _real_ parameters:

http://www.georgedishman.f2s.com/Henri/sine_wrong.png

and then adjusted to take account of the variable
speed effect to the the observed curves:

http://www.georgedishman.f2s.com/Henri/sine_right.png

Then I checked the radius

http://www.georgedishman.f2s.com/Henri/fit_rad.png

Who drew that stupid curve? It's nothing like a best fit.

Then I look at the luminosity

http://www.georgedishman.f2s.com/Henri/fit_lum.png

All you have is a curve of similar shape. Its variation is only a small
fraction of the true brightness change.
So what are you trying to tell me?

The ADoppler curve is the time distorted version
of the derivative of the velocity curve and it is
nothing like the luminosity.

VDoppler certainly can't produce that much brightness change.

Agreed, ballistic theory cannot match L Car.

Of course it can. It's straightforward.

Of course I knew it was a bloody exponential...or at least something
like
an exponential.
Actually, if you think about it, it might not even be exponential.

Partial dv by ds is proportional to v-c/n.

No. You cannot even assume that condition is true.

To be physical, the function must be "well behaved"
in the mathematical sense and you say speeds above
c/n get reduced while speeds below get increased so
as v-c/n - 0, it must reduce to first order. Come
on Henry, you claim to know maths so I shouldn't
have to help you out like this.

To be exponential, the fractional rate of change must be constant. I would not
assume it is constant with distance from a star's surface.


Not really, maybe I gave to much detail. Let me
repeat, you ask "where do you include observer
distance in your approach" and I answered that
it is included as the VDoppler term, OK?

Where?

..... you haven't included it in the important calculation.

Sorry Henry, I slightly jumped topic there. We were
first talking about the bunching due to motion of the
source towards the observer between pulses. That is
taken into account by the (1+v/c) term in the photon
arrival rate equation.

You also asked about the total observer distance.
Without speed equalisation, it appears as "d" in
the term 1/(1-da/c^2).

With speed equalisation, once the speeds are matched
any farther travel doesn't affect the apparent relative
photon rate which becomes asymptotic to a value given
by 1/(1-Ra/c^2) where R is the characteristic distance
of the speed equalisation.

I hope that's clearer.

That part is clear...



It got snipped, I have restored the context above.

And I have restored it again.

The argument I presented was that single photon
observation of grating deflection shows that the
photons mostly land where the classical analysis
says the peak intensity should occur hance my use
of classical analysis is valid.

That doesn't worry me. The experiment has never been done with a fast
moving
source.

It doesn't need to be, photons carry the energy so
where they land is where the maximum intensity
occurs. Let me be clear Henry, it doesn't matter
what angle they get deflected through, it only
matters that the classical intensity is a maximum
at the same place as where the photons land. That
means the methods of prediction are equivalent.

...but my theory explains why an HST
grating can detect its own orbit movement whereas YOUR theory says it
should
not.

Wrong, SR predicts the deflection correctly, but
you are just changing the subject again.

No, this is important.

This has nothing to do with the topic in hand,
you are just trying to weasle out of the subject.

YOUR theory says the HST should NOT detect its own movement because its
grating
is wavelength sensitive only and there is no way the movement of the
grating
can affect the absolute wavelength of the incoming light.

Sorry Henry, wavelength _is_ affected by Doppler.

George, photon wavelength...like any length...cannot be affected by the motion
of another object.




If you look back at the conversation, I said the top
and middle diagrams were fine, it is the bottom one
that applies and there should be a single merged
sphere which is at rest relative to the barycentre of
the binary system.

It wont be at rest.

Then the pulsar should have shown distortion due
to the ADoppler from the edge of the sphere.

It is probably there but tooo small to be detected.


"Wilsonian spheres" are nothing more than the solar
plasma which is most definitely frequency dependent.

They are a lot more than that. I think they are probably closely linked to
gravity fields.

ROFL, Henry you are a card. You do realise that you
just explained that ballistic theory predicts half
the bending by gravity because it doesn't take account
of ... gravity :-)

What makes a gravity field George?


Henry, you have been telling me I have got it all
wrong because I didn't take this into account. Now
you realise I have been all along, you tell me to
"forget it". Make your mind up.

George, you can generally assume the pulses all move the same distance.

No, I assume each pulse starts from a position which
is closer to the observer by v/f where v is the speed
and f is the frequency (photon rate).

That is your principal source of error. 'v' is continually changing according
to the orbit parameters.

I also assume
the pulses travel at speeds that differ by a/f where
a is the acceleration. Both those factors affect the
arrival rate.

They do. ...but v is a function of time as well. For a circular orbit, v=
a.sin(kt).

It is the time they take to get there that matters. The time differences
in
crossing the orbit are very small compared with the total time taken.

That affects the arrival _time_, not the rate.

It can be ignored in most cases...but not pulsars or contact binaries.

But as you know, we need to include it in the case of pulsars and contact
binaries which have very short extinction distances.

You always need to take it into account, but we were
talking about the photon arrival rate equations I
posted, not the arrival time equation.

George, I am wasting too much time discussing this.
My computer does all the calculations...and a lot more.
When you produce a proper brightness curve I will discuss it with you.

Well it is certainly tough for you. It makes your argument look like a
joke...

It means your claim to have matched the curves
is nonsense by your own admission.

BaTh produces curves of the right shapes and magnitude changes.
That's pretty impressive don't you think?

http://www.georgedishman.f2s.com/Henri/fit_lum.png

I don't that fit is impressive at all, in fact I
don't think it fits at all.

It's YOUR curve...and no, it doesn't fit very well at all.



The shape is NOT WRONG.

http://www.georgedishman.f2s.com/Henri/fit_lum.png

The ADoppler shape is hopelessly wrong.

Yes Henry, and you _still_ can't do it after how many
weeks? Here are the plots again, just apply schoolboy
calculus to the bottom one and you get those above.

The dispute is not over the calculus. It is about the shapes of typical
curves.

http://www.georgedishman.f2s.com/Hen...lustrative.png

The bottom one is similar in shape to the measured values

It isn't.

Here is the shape:

http://tinyurl.com/239mw6

They match.

That's nothing like your radius curve

Really? Have another look:

http://www.georgedishman.f2s.com/Henri/fit_rad.png

That curve doesn't match the points at all.

It is as good as anything youhave produced.

and any fool can see it is nothing like a
best fit to the points anyway.

A fool is often deceived which is why scientists
uses mathematical methods.

Any intelligent person can see it is nowhere near a best fit.

The authors obviously knew the answer they wanted and proceeded to draw a
curve
to suit. .

ROFL, Henry you really are clueless. Nobody "draws
a curve" in a real paper, they calculate the values
and plot the result.

Well it is obviously not a best fit curve in this case.



Differentiate once to get velocity then a second
time to get acceleration and compare them. Get it
into _your_ head that the curves are NOT similar
at all.

George, those curves are not similar. ..but the resulting photon density
curves
ARE.

Henry, for the umpteenth time, those ARE photon
density curves.

You are performing the wrong calculation.
For god's sake George, forget your equations and let the computer do the
work.

The computer _did_ calculate them for me, I just
cut and paste over the observed curves and rescaled
to fit.

Have a look at RT Aur.
It's curve is nothing like yours.

Of course not you idiot, you have a VELOCITY curve
for RT Aur and you are comparing it againgst my
RADIUS curve for L Car.

well george, please try to produce 'photon density' curves vs distance, in
future. These are what we want.

No, what you want is relative photon arrival rate
(number of photons per second arriving divided by
the number transmitted) versus the phase, and guess
what Henry, that's what those curves show.

....and that's what my computer does...and it produces the right curve for L
Car....WITH A MAGNITUDE CHANGE OF 0.7



Rarely? How many times a century on average does it
eclipse Regulus for example? Now do the statistics
for the number of known Cepheids and see how often
we should detect an eclipser.

....but YOU claimed most cepheids don't have companion objects ....

No, be careful. Many Cepheids will be part of binary
systems but the period of that binary is unrelated to
the variability of the star. What you were suggesting
was that the Cepheid itself was a pair, a visible star
and a dark companion. What I am saying is that that is
not the case, they are single "huff-puff" stars as you
put it. However, you still expect the ballistic model
to produce dynamic effects as a result of the variable
radius, that's what I am discussing. Do you follow that?

Of course I follow it. Just becasue some pairs have orbit frequencies different
from the huff puff frequency doesn't mean they all do.
I would say that most companions are dark.


About half are in binaries, the rest are simply
in galactic orbits. Our Sun for example is not
part of a binary system.

Yes it is.

No it isn't Henry Don't be silly.

Everything is in some kind of orbit around a mass centre.
Think about it.

Sure, the sun is gravitationally bound to the galaxy.

.....and it is free falling in a gravity field IN SOME KIND OF COMPLICATED
ORBIT.

Of course the period might be huge and the orbit very unstable.

The period is a bit over 200 milllion years IIRC, and
the Sun oscillates through the plane with a period of
about 80 million years, but it isn't bound to any other
star so it is not part of a binary system.

Not a binary....it is gravitationally linked to all the stars in the galaxy.
No star has a fixed position. Each one is moving slowly wrt all the others...as
well as rotating with the galaxy.


Go and think about what causes VDoppler.

George, all I can suggest is that you try to write yourt own BaTh program
so
you will get to know what is really involved.

For binaries, it is easy. For a Cepheid, you get all
the complexity of edge darkening, optical depth,
variable composition temperature and pressure
variations and so on. The ballistic theory part is
completely trivial in comparison.

hahaha...


...this is ridiculous. I'm not going to explain again.
The distance across the orbit is generally negligible compared with the
LYs of travel.

Read again what I said, the distance across the orbit
is NOT what we are talking about.

Rubbish.

You have simply lost the plot Henry.

You want to add that distance to the distance traveled by each
pulse...and of course that is required for pulsars.

Sure, but it is small in comparison to the effect of
variable speed. I don't mind if you exclude it from
your program as you previously said you did or if you
include it but I have already accounted for it in the
photon _rate_ calculation and I believe so have you.
yes.

I have it, George. Differentiating the radius twice SHOULD match
OBSERVED
velocity curves.

Differentiating radius _twice_ gives radial velocity?
Boy you really have lost the plot!

I said OBSERVED velocity curves.

Why, you keep telling me the velocity curve isn't
affected by ADoppler which is why the luminosity
can vary be sevreal magnitudes while the OBSERVED
velocity curves are still only a few km/s.

Yes, that's how it turns out George.

OK, so stick with that and stop contradicting yourself.

can't you get it into your head that hte OBSERVED velocity curves are generally
NOT the true ones.

I haven't lost the plot at all George.

Yes you have Henry, you are now relying on something
you told me doesn't happen, and you invented the K
factor to justify.

There is good reason to believe it exists.

It is impossible for it to exist Henry, but you need
to understand some physics to follow the proof I gave
you.

All the physics suggests that a factor K should exists....for a similar reason
to the fact that cars on the highway don't change length when they bunch
together in a slow section. Rubber cars might, too some extent. (K factor)

Photons can hardly shrink forever

Of course not, "superposition" applies to EM so
when the back of the photon catches up with the
front, it passes it unaffected. At least that's
what ballistic theory requires.

Ah no. The K factor is not necessarily constant.



The Earth rotates Henry, L Car isn't always above
the horizon, and sometimes there are clouds. Trust
me, living in England this year I have experienced
lots of clouds!

...more excuses....it's the clouds now....

Welcome to reality.

and the curve is nothing like a best fit.



Write a program that will produce photon bunching vs distance just for
circular
orbits.
You will learn a lot in the process.

Been there, done that.

No you haven't George. You haven't included all relevant factors.
You have a BaTh equation that gives the ADoppler changes in the distance
between two wavecrests of a classical wave at distance D and the additional
'observed doppler shortening' due to initial speed. You have NOT included the
additional bunching of those two wavecrests wrt the other wavecrests.

I think I got that right..


In the first case, the velocity is a true measure
and that fits the match of L Car to the derivative
of the radius.

No no. That's only your assumption.

No, it is what you keep telling me, K 1 therefore
there is a negligible ADoppler contribution. You can't
have it both ways Henry, you know VDoppler cannot
produce the magnitude of the luminosity variation so
you have to separate them using your K factor. That
means that the spectral shift is only VDoppler while
the luminosity is both VDoppler and ADoppler.

For most stars, both velocity and brighness curves result from ADoppler.

The shape of the luminosity curve
would be quite different from that of the velocity
curve since it would be mainly due to the
acceleration and that is a very bad fit for L Car.

This is not true.

See for yourself:

http://www.georgedishman.f2s.com/Henri/fit_lum.png

The ADoppler curve is nothing like the luminosity
in either band.

that's because you haven't produced a real brightness curve.


In the latter case you are right, the shapes would
be similar but you are back to the problem that the
observed velocity is should be far greater, the
observed velocity curve would only produce a few
milli-mag variation.

I think you need to sort out exactly what you are
claiming Henry, you are contradicting yourself all
over the place and none of the combinations that I
can draw from it works at all.

I am not contradicting myself. You are simply not getting the message.

You contradict yourself when you first claim K is
very small so that you can explain the luminosity
variation and then also say the velocity curve
isn't just VDoppler. Make your mind up.

I have.

George




www.users.bigpond.com/hewn/index.htm

The difference between a preacher and a used car salesman is that the latter at least has a product to sell.

On Tue, 17 Jul 2007 01:26:31 -0700, George Dishman
wrote:

Google said this posted successfully last week
but it appears to have vanished.

On 8 Jul, 12:44, HW@....(Henri Wilson) wrote:
On Sat, 7 Jul 2007 10:50:58 +0000 (UTC), bz wrote:
"George Dishman" wrote in news:f6njt0$g7k$1
:

"bz" wrote in message
.198.139...

Kind of 'lost in the noise', right?

Exactly. It's surprising that people like Sekerin
who pushed this years ago didn't see that problem.

Lack of understanding of the relationship between information theory,
intellegence and noise.

Ever since I saw loopy graphs in that paper on ballistic light[can't remember
the title/author at the moment], I suspected that there was something loopy
about the theory itself.

That was Sekerin's paper. He made the mistake of going beyond the critical
distance. Curves like those never happen.

They don't but they should, we are almost always
well beyond that distance. That was the first hole
in the theory.

It was thought to be a 'hole' but it was wrong.

Blows holes in his concrete boat.

Not concrete, it has enough holes to be chicken wire.

Either way, it doesn't stay afloat.

BaTh is very much alive and well.
Nothng any of you has said has even dented it.

Henry, when I design circuits with resistors,
I use Ohms Law. If I put a voltage across a
resistor, it tells me what current will flow.
I use it because it is always correct.

The opposite goes for ballistic theory. If I
try to apply it to an iFog or laser gyro, it
says the device will show no fringe shifts,
but they do shift. If I use to to calculate
the Shapiro effect from a satellite on the far
side of the Sun, it says the signal will arrive
early but it arrives late. Every time you try
to use ballistic theory, it gets it wrong. The
theory has no value whatsoever.

George, the signal actually arrives early due to its average speed being
greater than c.... but not as early as it should. Hence the 'delay'.


Move a laser in orbit on some kind of wheel whilst keeping it aligned with the
mirror on the moon.
If its light is pulsed at say 10000 hz, it should be possible to spin it at
such a rate that many pulses (emitted at c+v) arrive back at Earth at the same
instant. This would mean that the reflected light would arrive in bright pulses
with a frequency equal to that of the wheel rather than that of the laser.

Brilliant eh?

Dumb really. You would need to keep the laser aligned
while spinning it. First simplification - put a mirror
on the wheel and shine the laser onto it. Second
simplification - the displacement towards the Moon is
a sine wave so put the mirror on a loudspeaker voice
coil and feed it a sine wave. That way it stays pointing
in the same direction and moves along a straight line.

Using an ultrasonic source, you could get very high
accelerations which theoretically should be sensitive
enough to detect ADopppler.

No nothing like that is sensitive enough.


In reality, Ives and Stilwell already did a similar
test but just for the VDoppler part which confirmed
SR and ruled out the ballistic model.

Of course George...


But the emitter will have to be outside the Eatrh's atmosphere.

I need a theory that is valid if I do the test in air
as well. SR does that, remember Fizeau's experiment.
Ballistic theory will get that wrong too.

no it wont.
....and the SR explanation is really the old aether one.

George



www.users.bigpond.com/hewn/index.htm

The difference between a preacher and a used car salesman is that the latter at least has a product to sell.