Perihelion shift of S2
Ed Keane III
Sgr A* at the center of the Milky Way and has a
highly elliptical orbit. Will relativistic effects cause
precession of the perihelion of S2 that can be accurately
predicted? Is there any chance of dark matter
having any effect at such small distances?
Sam Wormley
So far we have not much more than one orbit of data.
http://www.mpe.mpg.de/www_ir/ir_recent.php
Ed Keane III
"Sam Wormley" <swor...@mchsi.com> wrote in message
news:401048DA...@mchsi.com...
It has been observed since 1992 and has a 15.2 year
period. If there is a chance that dark matter is not the
reason for anomalies in orbital motion in the outer regions
of galaxies then one (I) might be curious if everything is
moving as expected at the core. Do we know enough
about mass distribution in this area to make such predictions?
http://www.solstation.com/x-objects/s2orb-b.jpg
greywolf42
news:401012e5$0$21829$38ce...@news.westelcom.com...
Why go so far out? We have two excellent examples close by, and have known
about them since 1977. Long before the much ballyhooed PSR1913+16.
They are DI Herculis and AS Camelopardalis.
These two are much simpler systems with none of the massive complications of
PSR1913+16 or even Mercury. No error has ever been found in the data. They
were selected because of their felicitous identification as eclipsing
binaries and spherical stars -- so there is no doubt about orbital
inclination. They were identified specifically to test GR.
But because GR's predictions are massively off observation (by factors of 2
to 4), these stars have sunk into obscurity. (Everybody 'knows' about them,
but no one talks about them.)
--
greywolf42
ubi dubium ibi libertas
{remove planet for return e-mail}
Cesar Sirvent
"greywolf42" <min...@marssim-ss.com> escribió en el mensaje
news:1012sd7...@corp.supernews.com...
Uhm... just what I was looking for. Could you please give some bibliographic
(or web) references on data about them and some of those analysis "off
observation by factors of 2 to 4" ? Thanks in advance.
car...@no-dirac-spam.ucdavis.edu
> "greywolf42" <min...@marssim-ss.com> escribio en el mensaje
> news:1012sd7...@corp.supernews.com...
[...]
>> Why go so far out? We have two excellent examples close by,
>> and have known about them since 1977. Long before the
>> much ballyhooed PSR1913+16.
>> They are DI Herculis and AS Camelopardalis.
>> These two are much simpler systems with none of the massive
>> complications of PSR1913+16 or even Mercury. No error has
>> ever been found in the data. They were selected because of
>> their felicitous identification as eclipsing binaries and spherical
>> stars -- so there is no doubt about orbital inclination. They
>> were identified specifically to test GR.
>> But because GR's predictions are massively off observation
>> (by factors of 2 to 4), these stars have sunk into obscurity.
>> (Everybody 'knows' about them, but no one talks about them.)
> Uhm... just what I was looking for. Could you please give some
> bibliographic (or web) references on data about them
The best place to start is two papers by Claret, Astron. Astrophys.
327 (1997) 11-21 and Astron. Astrophys. 330 (1998) 533-540.
The first of these analyzes ten relativistic eclipsing binaries for
which observations agree well with the GR predictions. The
second paper concentrates on DI Her, along with two other
systems, AS Cam and V541 Cyg, for which there are significant
discrepancies. One of the things that must be explained is why
there is this inconsistency, that is, why many eclipsing binaries
show the apsidal motion predicted by GR but a few do not. The
one significant difference that Claret finds is that the three cases
that show disagreements wth GR have the three longest periods
of apsidal motion.
(I should note also that the discrepancy for V541 Cyg seems to have
gone away with better measurements -- see Volkov and Khaliullin,
Information Bulletin on Variable Stars, 4680, 1.)
Steve Carlip
Cesar Sirvent
<car...@no-dirac-spam.ucdavis.edu> escribió en el mensaje
news:busbi2$ei7$1...@woodrow.ucdavis.edu...
Thanks. However, I understand that for AS Cam and DI Her the disagreements
are not likely due to poor measurements, right?
Ed Keane III
<car...@no-dirac-spam.ucdavis.edu> wrote in message
news:busbi2$ei7$1...@woodrow.ucdavis.edu...
> In sci.astro Cesar Sirvent <8UMU...@spaxmterrmapsa.esspaxm> wrote:
>
> > "greywolf42" <min...@marssim-ss.com> escribio en el mensaje
> > news:1012sd7...@corp.supernews.com...
>
> [...]
> >> Why go so far out? We have two excellent examples close by,
> >> and have known about them since 1977. Long before the
> >> much ballyhooed PSR1913+16.
>
> >> They are DI Herculis and AS Camelopardalis.
> >> But because GR's predictions are massively off observation
> >> (by factors of 2 to 4), these stars have sunk into obscurity.
> >> (Everybody 'knows' about them, but no one talks about them.)
DI Herculis and AS Camelopardalis? 164 hits on Google.
That is less than no one.
> > Uhm... just what I was looking for. Could you please give some
> > bibliographic (or web) references on data about them
>
> The best place to start is two papers by Claret, Astron. Astrophys.
> 327 (1997) 11-21 and Astron. Astrophys. 330 (1998) 533-540.
http://aa.springer.de/papers/7327001/2300011/small.htm
http://aa.springer.de/papers/8330002/2300533/sc3.htm#sc3.10.
And they are in english (words).
Joseph Lazio
EK> S2 is the closest star to the supermassive black hole Sgr A* at
EK> the center of the Milky Way and has a highly elliptical
EK> orbit. Will relativistic effects cause precession of the
EK> perihelion of S2 that can be accurately predicted?
I'm not sure. I think a larger worry would be whether its orbit will
remain stable over any substantial amount of time. There are a lot of
stars in the central cluster. Their individual gravitational tugs can
perturb the orbit of S2 to the point that it may not close for any
substantial amount of time.
EK> Is there any chance of dark matter having any effect at such small
EK> distances?
Well, in a sense, Sgr A* is dark matter. More properly, the
supermassive black hole, around which is a region that emits radio
radiation that we detect as the source Sgr A*, is (baryonic) dark
matter. So, yes.
--
Lt. Lazio, HTML police | e-mail: jla...@patriot.net
No means no, stop rape. | http://patriot.net/%7Ejlazio/
sci.astro FAQ at http://sciastro.astronomy.net/sci.astro.html
greywolf42
>
> > "greywolf42" <min...@marssim-ss.com> escribio en el mensaje
> > news:1012sd7...@corp.supernews.com...
>
> [...]
> >> Why go so far out? We have two excellent examples close by,
> >> and have known about them since 1977. Long before the
> >> much ballyhooed PSR1913+16.
>
> >> They are DI Herculis and AS Camelopardalis.
>
> >> These two are much simpler systems with none of the massive
> >> complications of PSR1913+16 or even Mercury. No error has
> >> ever been found in the data. They were selected because of
> >> their felicitous identification as eclipsing binaries and spherical
> >> stars -- so there is no doubt about orbital inclination. They
> >> were identified specifically to test GR.
>
> >> But because GR's predictions are massively off observation
> >> (by factors of 2 to 4), these stars have sunk into obscurity.
> >> (Everybody 'knows' about them, but no one talks about them.)
>
> > Uhm... just what I was looking for. Could you please give some
> > bibliographic (or web) references on data about them
A general overview is in Astronomy, Nov. 1995, p 54-59, "The Mystery of DI
Herculis".
The original cite: "The apsidal motion of the eccentric eclipsing binary DI
Herculis - an apparent discrepancy with general relativity", Guinan, E. F.;
Maloney, F. P., Astronomical Journal (ISSN 0004-6256), vol. 90, Aug. 1985,
p. 1519-1528.
http://adsabs.harvard.edu/cgi-bin/nph-bib_query?bibcode=1985AJ.....90.1519G&
amp;db_key=AST&high=3e5ffa223d11852
Abstract:
"The apsidal motion of the eccentric eclipsing binary DI Herculis (HD
175227) is determined from an analysis of the available observations and
eclipse timings from 1959 to 1984. Least squares solutions to the primary
and secondary minima extending over an 84-yr interval yielded a small
advance of periastron omega dot of 0.65 deg/100 yr + or - 0.18/100 yr. The
observed advance of the periastron is about one seventh of the theoretical
value of 4.27 deg/100 yr that is expected from the combined relativistic and
classical effects. The discrepancy is about -3.62 deg/100 yr, or a magnitude
of about 20 sigma. Classical mechanisms which explain the discrepancy are
discussed, together with the possibility that there may be problems with
general relativity itself."
> The best place to start is two papers by Claret, Astron. Astrophys.
> 327 (1997) 11-21 and Astron. Astrophys. 330 (1998) 533-540.
> The first of these analyzes ten relativistic eclipsing binaries for
> which observations agree well with the GR predictions.
A horrible place to start. But it does defend the Faith....
Claret's primary purpose is not to test GR, but to examine the interiors of
stars, due to tidal distortions (from primarily Newtonian gravity). The
authors studiously ignore good tests of GR (eclipsing binaries that are not
tidally-distorted). In fact, they state early on:
"In the papers quoted above we have only analysed the systems for which the
relativistic contribution to the total apsidal motion were small."
In short -- GR is examined only when such effects have been reduced to noise
on a stronger signal. The only mention of the classic Di Her case is when
'dissing' a particular theory.
Keep in mind that eclipsing binaries are rare, due to simple statistics.
Stellar orbits are usually quite a bit larger than the diameters of the
component stars. The odds of a given double star system being aligned with
Earth is very low. The larger the radius of the orbit, the lower the
probability. Of the eclipsing binaries that we know, the vast majority will
unavoidably be close binaries -- because the odds of a binary being aligned
with Earth is inversely proportional to the orbital radius.
But close binaries are distorted into non-spherical shapes, due to the
gravitational potentials of close star systems. These non-spherical shapes
cannot be observed directly. But they can be inferred by a suitable theory
of gravity, due to changes in periastron position. However, if you have to
infer the shape (from GR) -- then you have no test of GR.
Guinan's purpose was explicitly to find those few, rare cases where GR could
be tested without circularity. Claret's purpose was to find all the 'muddy'
systems he could. Because he was looking at theories of stellar
interiors -- not at systems that would test GR.
> The
> second paper concentrates on DI Her, along with two other
> systems, AS Cam and V541 Cyg, for which there are significant
> discrepancies. One of the things that must be explained is why
> there is this inconsistency, that is, why many eclipsing binaries
> show the apsidal motion predicted by GR but a few do not.
Even Claret's conclusion is simply that DI Herculis disagrees with GR --
despite 20 years of 'intensive effort' (quote from Claret paper) to make it
behave.....
The reason is that the 'good tests' of GR (those rarer eclipsing binaries
with large orbital separations, and undistorted stars) have no adjustible
parameters. GR fails on these. But if you concentrate on distorted stars
(which you cannot observe directly) you can find a theoretical distortion to
match every case with GR.
The 'explanation' that good tests don't match GR is well known. (At least
according to the Astonomy paper, above.)
> The
> one significant difference that Claret finds is that the three cases
> that show disagreements wth GR have the three longest periods
> of apsidal motion.
Yes. Exactly as noted in the literature a decade prior to Claret. And
explained by Guinen.
1) long period equals large separation distance.
2) large separation distance equals round stars, due to lack of tidal motion
3) lack of tidal distortion eliminates any complexities and free parameters
from the problem
4) GR is disproved whenever you can't hide the discrepacy in free
parameters.
Good tests of GR disprove it. So the faithful write papers on complex
systems that can be tweaked to match GR.
> (I should note also that the discrepancy for V541 Cyg seems to have
> gone away with better measurements -- see Volkov and Khaliullin,
> Information Bulletin on Variable Stars, 4680, 1.)
Going away just like the apsides of Venus. I'll look this up later.....
--
greywolf42
Good links, thanks.
> And they are in english (words).
The author's primary purpose is not to test GR, but to examine the interiors
of stars, due to tidal distortions (from primarily Newtonian gravity). The
authors studiously ignore good tests of GR (eclipsing binaries that are not
tidally-distorted). In fact, they state early on:
"The eclipsing binaries which present apsidal motion are also useful to test
the predictions of the General Relativity - GR - for the periastron advance.
In the papers quoted above we have only analysed the systems for which the
relativistic contribution to the total apsidal motion were small."
The only mention of the classic Di Her case is when 'dissing' a particular
theory.
So much for Steve's reference. He should know better.
--
Craig Markwardt
"greywolf42" <min...@marssim-ss.com> writes:
> Why go so far out? We have two excellent examples close by, and have known
> about them since 1977. Long before the much ballyhooed PSR1913+16.
>
> They are DI Herculis and AS Camelopardalis.
>
> These two are much simpler systems with none of the massive complications of
> PSR1913+16 or even Mercury. No error has ever been found in the data. They
> were selected because of their felicitous identification as eclipsing
> binaries and spherical stars -- so there is no doubt about orbital
> inclination. They were identified specifically to test GR.
Your claim that DI Her and AS Cam are "much simpler systems" than the
binary pulsars is unsubstantiated. In fact, normal stars are much
more extended than neutron stars compared to their respective typical
Roche lobes (10 km vs 700,000 km for stellar mass ZAMS stars).
Therefore there are a lot of tidal-type effects to consider for stars
that are much reduced for radio pulsars: the presence of a
circumstellar material, misaligned rotation axes, incorrect modeling
of tidal effects, and so on (Claret 1997). These effects can be ruled
out (or stringently limited) for binary pulsar systems.
Also, the presence of an as-yet undetected third star cannot be ruled
out yet, and would cause apsidal motion (eg. in SS Lac; Torres &
Stefanik 2000). If any of these effects are significant for DI Her,
and there is some evidence they may be (see refs.), then the orbit of
DI Her is not a good test of GR, at least not until they are better
understood.
CM
References
Claret, A. 1998, A&A, 330, 533
Torres, G. & Stefanik, R.P. 2000, AJ, 119, 1914
greywolf42
news:onvfmyz...@cow.physics.wisc.edu...
>
> "greywolf42" <min...@marssim-ss.com> writes:
> > Why go so far out? We have two excellent examples close by, and have
> > known about them since 1977. Long before the much ballyhooed
> > PSR1913+16.
> >
> > They are DI Herculis and AS Camelopardalis.
> >
> > These two are much simpler systems with none of the massive
> > complications of PSR1913+16 or even Mercury. No error has ever
> > been found in the data. They were selected because of their felicitous
> > identification as eclipsing binaries and spherical stars -- so there is
no
> > doubt about orbital inclination. They were identified specifically to
> > test GR.
>
> Your claim that DI Her and AS Cam are "much simpler systems" than the
> binary pulsars is unsubstantiated.
Of course it's substantiated. See the references posted in the prior
parallel posts. And I note that you don't dispute the point -- merely try
to evade it.
> In fact, normal stars are much
> more extended than neutron stars compared to their respective typical
> Roche lobes (10 km vs 700,000 km for stellar mass ZAMS stars).
True, but irrelevant.
> Therefore there are a lot of tidal-type effects to consider for stars
> that are much reduced for radio pulsars: the presence of a
> circumstellar material, misaligned rotation axes, incorrect modeling
> of tidal effects, and so on (Claret 1997).
Of course. But Claret 1997 wasn't looking for good tests of GR. He
specifically was focusing on distorted stars -- to check their interiors.
On the other hand, DI Herculis was specifically selected because it is far
enough separated to have no distorted stars. (Distorted stars tend to
increase periastron advance -- not reduce it.)
> These effects can be ruled
> out (or stringently limited) for binary pulsar systems.
Just like they can be ruled out for DI Herculis (which is far enough
separated to avoid all of the above). Though AFAIK, we don't have any
eclipsing binary pulsars. So orbital inclinations are unknown for
pulsars -- unlike the eclisping binaries.
> Also, the presence of an as-yet undetected third star cannot be ruled
> out yet, and would cause apsidal motion (eg. in SS Lac; Torres &
> Stefanik 2000).
It was ruled out just as well as 'Vulcan' was ruled out for Mercury. This
was ruled out 'way back in 1985. And why would spherical eclipsing binaries
ALL have this problem? (Those with long periods.) See 1998 Claret.
> If any of these effects are significant for DI Her,
> and there is some evidence they may be (see refs.),
LOL! The 'evidence' they discuss is that GR is not disproven!
All the ref's say that none of these are known. But 'something' might
be there. Sure, we can't PROVE there is nothing there.
> then the orbit of
> DI Her is not a good test of GR, at least not until they are better
> understood.
LOL! But DI Her is an EXCELLENT test of GR. All of those things have been
looked for -- and none of them exist. Despite 20 years of trying.
Including your references, below.
> References
> Claret, A. 1998, A&A, 330, 533
> Torres, G. & Stefanik, R.P. 2000, AJ, 119, 1914
--
greywolf42
> Craig Markwardt <crai...@REMOVEcow.physics.wisc.edu> wrote in message
> news:onvfmyz...@cow.physics.wisc.edu...
> >
> > "greywolf42" <min...@marssim-ss.com> writes:
{snip}
> > Also, the presence of an as-yet undetected third star cannot be ruled
> > out yet, and would cause apsidal motion (eg. in SS Lac; Torres &
> > Stefanik 2000).
>
> It was ruled out just as well as 'Vulcan' was ruled out for Mercury. This
> was ruled out 'way back in 1985. And why would spherical eclipsing
> binaries ALL have this problem? (Those with long periods.)
> See 1998 Claret.
Even better, see 1985 Guinan:
"Although it is possible to create mathematically a third member of DI Her
that can resolve the discrepancy between the observed and the theoretically
expected apsidal motion, there is no observational data to support its
existence...."
DI Her is a pair of B stars (B4 and B5). The postulated companion is
constrained to be between a B9 to A0 star, due to known variations in the
light curves, at extremely high inclination to the eclipsing pair (exceeding
46 degrees) -- an unstable configuration. And a high-inclination A0 star is
really hard to miss.
{snip}
car...@no-dirac-spam.ucdavis.edu
> <car...@no-dirac-spam.ucdavis.edu> wrote in message
> news:busbi2$ei7$1...@woodrow.ucdavis.edu...
[...]
>> The best place to start is two papers by Claret, Astron. Astrophys.
>> 327 (1997) 11-21 and Astron. Astrophys. 330 (1998) 533-540.
>> The first of these analyzes ten relativistic eclipsing binaries for
>> which observations agree well with the GR predictions.
> A horrible place to start. But it does defend the Faith....
> Claret's primary purpose is not to test GR, but to examine the interiors of
> stars, due to tidal distortions (from primarily Newtonian gravity). The
> authors studiously ignore good tests of GR (eclipsing binaries that are not
> tidally-distorted). In fact, they state early on:
> "In the papers quoted above we have only analysed the systems for
> which the relativistic contribution to the total apsidal motion were small."
This is a grotesque distortion. Claret is doing two things -- modeling
stellar interiors and testing GR. To do this, he first looks at systems
in which the relativistic contributions are small, to test the stellar
modeling. This is done in Claret and Gimenez, A&A 277 (1993) 487,
and Claret and Gimenez, in Inside the Stars (IAU Symp. 137), ed.
W.W. Weiss, A. Baglin, PASPC, 469. These are ``the papers quoted
above.'' Having shown that the stellar models work well, he then
applies them to relativistic systems in the paper I cited.
Here is the context for the statement ``greywolf42'' quotes out of
context, in the third paragraph of the paper I cited above:
``The eclipsing binaries which present apsidal motion are also
useful to test the predictionsof General Relativity for the periastron
advance. In the papers quoted above we have only analysed the
systems for which the relativistic contribution to the total apsidal
motion were small. The results of these papers indicated that using
new opacity calculations, core overshooting, rotation, improved
orbital elements and recent apsidal motion rates the theoretical
predictions are in good agreement with observations. The old
problem, that real stars seemed to be more mass concentrated
than predicted by theory, was solved or at least minored. As we
did not know a priori which was the cause for these discrepencies
we have separated the systems presenting high relativistic
contributions in order to avoid these disagreements with the
theory to be attributed to relativistic effects. In this way we
have divided our investigation into two parts: one concerning
the non-relativistic systems (Claret & Gimenez 1993ab) and the
present work probing the relativistic ones.''
In other words, the claim by ``greywolf42'' is precisely the
opposite of what Claret says.
I am appalled.
Steve Carlip
car...@no-dirac-spam.ucdavis.edu
[...]
> However, I understand that for AS Cam and DI Her the disagreements
> are not likely due to poor measurements, right?
It's not clear. I would suggest reading the second paper of Claret
that I cited (A & A 330 (1997) 533), section 3.10, which discusses
the observational issues. Claret quotes Guinan and Milone, for
example, as stating that the difference in the periastron advance
for DI Her obtained using photographic, visual, and photoelectric
measurements and only photometric data can reach 270%. Further,
the photometric observations have covered only .1% of the apsidal
motion period.
If I had to guess, I would put even odds on problems with the
observations and a fairly prosaic physical explanation, such as
a third body, with some radical problem in GR a distant third.
(For the record, I, like most of the people I know who work in the
field, would *love* to find some set of good observations that
disagreed with GR -- it would make life much more interesting.
It's precisely because I recognize this bias, though, that I'm not
willing to leap at claims that aren't well-substantiated and that
have possible boring explanations. If you want to find something
new, a lot of the work goes into not chasing down blind alleys --
there are just too many of them.)
Steve Carlip
greywolf42
news:bv6vhp$f8t$4...@woodrow.ucdavis.edu...
> In sci.astro Cesar Sirvent <8UMU...@spaxmterrmapsa.esspaxm> wrote:
>
> [...]
> > However, I understand that for AS Cam and DI Her the disagreements
> > are not likely due to poor measurements, right?
>
> It's not clear.
It is perfectly clear that 84 years of measurements, reviewed by dozens of
researchers all clearly show that the DI Her measurements are excellent.
> I would suggest reading the second paper of Claret
> that I cited (A & A 330 (1997) 533), section 3.10, which discusses
> the observational issues. Claret quotes Guinan and Milone, for
> example, as stating that the difference in the periastron advance
> for DI Her obtained using photographic, visual, and photoelectric
> measurements and only photometric data can reach 270%.
It's still a 20 sigma effect against GR.
And why would Steve quote Claret, instead of Guinan and Milone about what
Guinan and Milone stated? Unless Steve never bothered to read Guinan and
Malone (available on the web), and read only papers that claim to support
GR?
> Further,
> the photometric observations have covered only .1% of the apsidal
> motion period.
Irrelevant.
> If I had to guess, I would put even odds on problems with the
> observations and a fairly prosaic physical explanation, such as
> a third body, with some radical problem in GR a distant third.
see 1985 Guinan:
"Although it is possible to create mathematically a third member of DI Her
that can resolve the discrepancy between the observed and the theoretically
expected apsidal motion, there is no observational data to support its
existence...."
DI Her is a pair of B stars (B4 and B5). The postulated companion is
constrained to be between a B9 to A0 star, due to known variations in the
light curves, at extremely high inclination to the eclipsing pair (exceeding
46 degrees) -- an unstable configuration. And a high-inclination A0 star is
really hard to miss.
> (For the record, I, like most of the people I know who work in the
> field, would *love* to find some set of good observations that
> disagreed with GR -- it would make life much more interesting.
A transparently untrue statement -- because Steve won't examine well-known
and well-documented 'anomalies' such as DI Her. Steve's knee-jerk response
is to claim the observations are bad -- when the observations of DI Her are
FAR better tests of GR than the cases he uses for 'support' of GR.
> It's precisely because I recognize this bias, though, that I'm not
> willing to leap at claims that aren't well-substantiated and that
> have possible boring explanations. If you want to find something
> new, a lot of the work goes into not chasing down blind alleys --
> there are just too many of them.)
DI Her is enormously well documented. It was first identified as a GR test
in 1959 -- specifically because it was so much better than Mercury (which is
a dynamic nightmare, loaded with subjective pitfalls). There have been 20
years of intensive efforts to explain the motions of DI Her -- starting with
Guinan (who is a GR-supporter, but an honest one). Observational problems
have been ruled out. Third bodies have been ruled out -- to all but the
priesthood.
But Steve pushes complex studies of distorted stars where the GR effect
is mere noise on a signal.
--
greywolf42
news:bv6tmm$f8t$3...@woodrow.ucdavis.edu...
> In sci.astro greywolf42 <min...@marssim-ss.com> wrote:
> > <car...@no-dirac-spam.ucdavis.edu> wrote in message
> > news:busbi2$ei7$1...@woodrow.ucdavis.edu...
>
> [...]
> >> The best place to start is two papers by Claret, Astron. Astrophys.
> >> 327 (1997) 11-21 and Astron. Astrophys. 330 (1998) 533-540.
> >> The first of these analyzes ten relativistic eclipsing binaries for
> >> which observations agree well with the GR predictions.
>
> > A horrible place to start. But it does defend the Faith....
>
> > Claret's primary purpose is not to test GR, but to examine the interiors
> > of stars, due to tidal distortions (from primarily Newtonian gravity).
> > The authors studiously ignore good tests of GR (eclipsing binaries
> > that are not tidally-distorted). In fact, they state early on:
>
> > "In the papers quoted above we have only analysed the systems for
> > which the relativistic contribution to the total apsidal motion were
> > small."
>
> This is a grotesque distortion.
No, it is a direct quote. And an explicit identification of the stellar
systems being analysed in the paper. Non-spherical stars are not good tests
of GR, due to the number of free parameters that result from the tidal
distortions. Such distortions cause periastron advances far greater than GR
effects. To specifically test GR, one should look to systems that don't
have these complications -- such as DI Her.
> Claret is doing two things -- modeling
> stellar interiors and testing GR. To do this, he first looks at systems
> in which the relativistic contributions are small, to test the stellar
> modeling.
Yes. But in the 1997 paper, he is looking at systems with highly-distorted
stars. Which swamps any GR NNPA by Newtonian effects.
> This is done in Claret and Gimenez, A&A 277 (1993) 487,
> and Claret and Gimenez, in Inside the Stars (IAU Symp. 137), ed.
> W.W. Weiss, A. Baglin, PASPC, 469. These are ``the papers quoted
> above.'' Having shown that the stellar models work well, he then
> applies them to relativistic systems in the paper I cited.
In the 1997 paper, Claret does not apply the models to any other system than
to the highly-distorted stars that he used in the first part.
> Here is the context for the statement ``greywolf42'' quotes out of
> context, in the third paragraph of the paper I cited above:
>
> ``The eclipsing binaries which present apsidal motion are also
> useful to test the predictionsof General Relativity for the periastron
> advance. In the papers quoted above we have only analysed the
> systems for which the relativistic contribution to the total apsidal
> motion were small.
Yep. That hasn't changed.
> The results of these papers indicated that using
> new opacity calculations, core overshooting, rotation, improved
> orbital elements and recent apsidal motion rates the theoretical
> predictions are in good agreement with observations. The old
> problem, that real stars seemed to be more mass concentrated
> than predicted by theory, was solved or at least minored.
For these highly warped stars, Claret was able to bury any discrepancies
deep enough in the free parameters that it no longer showed. However, he
still couldn't tame the simple systems -- those 'long period' systems that
don't have all those free parameters to work with.
> As we
> did not know a priori which was the cause for these discrepencies
> we have separated the systems presenting high relativistic
> contributions in order to avoid these disagreements with the
> theory to be attributed to relativistic effects. In this way we
> have divided our investigation into two parts: one concerning
> the non-relativistic systems (Claret & Gimenez 1993ab) and the
> present work probing the relativistic ones.''
>
> In other words, the claim by ``greywolf42'' is precisely the
> opposite of what Claret says.
>
> I am appalled.
I accurately described both the method and the limitations of Claret's
analysis. Why does it appall you that I disagree with Claret's conclusions?
And why are you ignoring Claret's conclusions that GR still has problems
with systems of 'long' periods?