Why are the 'Fixed Stars' so FIXED?

"Henri Wilson" HW@.... wrote in message
...
On 8 Mar 2007 02:10:36 -0800, "George Dishman"
wrote:

On 8 Mar, 07:52, HW@....(Henri Wilson) wrote:
On Wed, 7 Mar 2007 23:00:25 -0000, "George Dishman"
wrote:
"Henri Wilson" HW@.... wrote in message

....that's not being sidetracked.
Halpha light has the same ABSOLUTE distance between 'wavecrests' no
matter how
it is produced.

Actually, that might be of interest, I'll keep the
comment in for later.

I wont make any claims to that effect if the source is accelerating.

I don't think there will be any disagreement other than
when there is an acceleration.

Unfortunately there is no way of using (distance x velocity) - which is
what my
program can simulate - to produce an extinction distance and a velocity.

The first question is what is the velocity if the
extinction is negligible. The distance you use is a
combination of the extinction and actual distances
so you get the zero-extinction result by setting the
program parameter to the actual distance. That has
been measured by parallax as 1.14kpc or 3520 light
years.

George, even with the HST that figure could be out by 1000LY either way.

First let me correct an error, I should have typed 3720,
not 3520. the error is +40/-30 parsec or about +/120 light
years, good enough for our purposes.

However at 3000LYs, the velocity would have to be very low, 0.000001c.

But I can't do this. The published velocity change, using classical
doppler, is
around 0.9 in 10000. My figure for linear brightness variation should be
twice
the same....about 0.00018.

Something about the method doesn't add up here. My selected velocity,
expressed
as a fraction of c, has to produce a linear brightness variation of twice
the
same fraction. (twice because amplitude is only half the swing)

Be careful, the brightness variation should be twice the
_red_ curve velocity, not the blue curve. These speeds
should be markedly different now.

I can do that.....I get a figure for extinction distance of about one
lightday
or less for a peripheral velocity of 0.0009c and a velocity change of
about the
same fraction.

I don't find that unreasonable...but I am not happy about the method.

That's one point in going through this process, you get
an apprecaiation of how the numbers pan out and sometimes
they can be surprising. Still, that's the reason for
checking the software, is there an error or are the results
genuinely counter-intuitive?

One step at a time Henry, us 3520 light years for the
distance and tell me what your orbital parameters are
for a red curve that is a perfect sine wave with an
amplitude of 27983 m/s.

Like I said George, a condition is that my predicted linear brightness
variation must be the same as twice the fractional velocity curve
amplitude .

The red curve, yes, not the blue curve.

Yes, George, I now understand where we are heading. You might have to
accept that the doppler calculated velocity curve is way out.

Of course! The first step should give you a stupidly low speed
which we then address by various methods.

A very heavy neutron star that is orbitted by a dwarf star would
conceivably
have quite a small peripheral velocity. It just wobbles around the
barycentre...which could be only a few diameters away.

Ah but there is an upper limit to the possible mass of
a neutron star.

who said? Einstein?

No, measurements in accelerators that tell us how much
pressure material can withstand before imploding.

George, when the BaTh takes over the whole of astronomy will have to be
rewritten.

Except that we already know it is wrong.

That's why you go one step at a time, first assume no
extinction.

I can't do that and still get the right match. Extinction for this pulsar
occurs in around 1 Lday....but there is more to this.

It sounds as though you are comparing the wrong velocity,
match the red value and the brightness to the observation
and the blue value will be much less.

I told you. I can produce (D * v) but not D or v.

You also told me " Yes, George, I now understand where we
are heading." but you don't seem to have really grasped it,
or maybe you didn't appreciate how the distance parameter
in your program can be used. Set it to the parallax distance
and you get v for the case of no extinction.

parallax distance is of no use at all.

It is the correct value to use for this first test.

I have to match my 'brightness variation' (which is now based on
bunching...and
bunching is used to determine velocity variation) so that linear variation
in
'brightness' = twice the velocity curve amplitude.

I have to work on this a bit more before I'm confident of the outcome.

If you want to see where this is headed, given that we know the
period, what would be the orbital diameter for the known mass
of the stars? Can we use that to rule out this solution and
consider changing the pitch instead? First let's get your
velocity written down - a first stake in the ground.

George, we don't know the mass of the stars. Even that has been derived
wrongly.

Sorry Henry, the rules of thermodynamics still apply
so mass can be obtained from absolute magnitude, radius
and temperature for the dwarf.

Only a rough estimate at that distance.

Certainly.

All we know with reasonable certainty are the orbit period and the
observed
bunching pattern of the pulses.

We know a lot more than those Henry, spectra carry a wealth
of information. Anyway, using just those what do you get
for the velocity?

Like I said, that is impossible.

I think you have momentarily lost track of which speed
you are matching.

Using this method, attributed to yourself, a condition is that the
velocity
expressed as a fraction of c has to equal half the 'brightness
change'...since
this 'brightness change' is really an indication of speed variation, in
this
case about 0.00018.

Yes, but you match the red curve and then tell me the blue
curve value.

George