On 20 Feb 2007 02:18:17 -0800, "George Dishman"
wrote:

On 19 Feb, 23:17, HW@....(Henri Wilson) wrote:
On 18 Feb 2007 19:20:31 -0800, "Leonard Kellogg" wrote:
Henri Wilson wrote:
...
I use symbolic pulses from a star of constant brightness
emitted at equi-temporal points around the orbit. These
travel at varying c+cos(v) speeds towards a distant obsever.
The rate at which they arrive at the observer should then
simulate its brightness curve there.

So apply that to the pulsar.

There is absolutely no point....unless you can provide a reliable curve showing
the variation in arrival rate of the pulses over time.

Henry, I have already done that several times. In
round figures the PRF is 339 Hz and that is varied
by +/- 30.5 mHz. The exact numbers are in my previous
posts.

The velocity curve that would be published is just
a sine wave (near zero eccentricity) with an
amplitude of c * 0.0305 / 339 = 27983 m/s.

That should be the same
as my 'brightness curve'. I can't make sense of the curve published by Jacoby
et al

Also, I cannot adjust the number of pulses I sample per orbit (122 million in
this case) without changing the code a bit. I can do it but it will take a
little time

You don't need to, it is only the frequency ratio
that matters. You do need to fix the velocity curve
calculation though, I'll explain that in reply to
your post giving the details of your current
calculation.

OK
The frequency variation correspondes to a magnitude change of about 0.2.
(CMIIW)

have look at: http://www.users.bigpond.com/hewn/J1909-3744a.jpg

Here the magnitude change is 1 (too high) but I included it because it shows
how the OBSERVED sine velocity curve corresponds to an orbit with ~0.15
eccentricity with Yaw 90. (periastron furthest from observer). Distance = 4 LY
for this mag change.

and: http://www.users.bigpond.com/hewn/J1909-3744b.jpg
mag change ~0.2
Period = 0.0042 years
max velocity=0.0000933c.

To obtain curve b, I have to plug in a distance of less than 1 LY....more like
0.7 LYs.
This order of 'extinction length' is quite consistent with those I have derived
from short period contact binaries.

In curve b, the magnitude change is smaller and a sine-like red velocity curve
corresponds with an e ~ 0.06, yaw -90.
A circular orbit results in a clearly skewed red curve.
So my theory says the orbit is NOT circular at all.

provide me with a good curve of pulse arrival times and i can probably do what
you ask.

Already done, repeatedly.





George