Pioneer 10 test of light speed delay

wrote in message
oups.com...

George Dishman wrote:
wrote in message
oups.com...

George Dishman wrote:
wrote in message
http://windsor.gsfc.nasa.gov/spacecr...tdf/atdf_data/


I thought I had answered this. The 1988 data at this site is
corrupted but the 1987 data sets are not. By the way what is
the link to your program?

I would prefer to use 1987 data. It is strange that you cant
make a list with this data that is similar to your special 1988
data?


The problem isn't anything to do with reading the data,
as I have said many times, the DSN was only receiving
at one site during 1987 and my test compares results
from two receive sites.
Try any other year with non corrupted data.

I haven't found any corrupted files and in particular
Craig has confirmed the 1988 file is uncorrupted. There
is of course some noise in all the measurements but
that didn't stop me reading the files and it doesn't
stop you either.

Why don't you just find the bug in your software that
is stopping you reading the file. To be honest, I am
starting to think you are only claiming you can't read
the data because you have no answer to the proof it
provides.

The fact that you have
not been able to do this suggests that there is something wrong with
your method that you dont want to acknowledge.

No, it reflects that fact that I have done many hours
of work already, have debugged my code so that it
reads the files correctly and have obtained all the
data I need. I see no point in doing any more when the
problem is at your end, I have better things to do
with my time.

I even sent you the Excel sheet with all the data
already extracted so you didn't need to worry about
getting the data from the files yourself. If you want
to use it on other years, just enter the filename
and press the button. You have no excuse.

George

George Dishman wrote:
wrote in message
oups.com...

George Dishman wrote:
wrote in message
oups.com...

George Dishman wrote:
wrote in message

http://windsor.gsfc.nasa.gov/spacecr...tdf/atdf_data/


I thought I had answered this. The 1988 data at this site
is
corrupted but the 1987 data sets are not. By the way what is
the link to your program?

I would prefer to use 1987 data. It is strange that you cant
make a list with this data that is similar to your special 1988
data?


The problem isn't anything to do with reading the data,
as I have said many times, the DSN was only receiving
at one site during 1987 and my test compares results
from two receive sites.
Try any other year with non corrupted data.


I guess the answer is you have and found that this other
data does not support your contention. This is because your contention
ignores the differences in vantage points at the two locations and the
differences in starting and ending times at the two locations etc.


I haven't found any corrupted files and in particular
Craig has confirmed the 1988 file is uncorrupted. There
is of course some noise in all the measurements but
that didn't stop me reading the files and it doesn't
stop you either.

My program reads all of the files except the 88 file. Why?
Why is this the only file whose data you want to use. Why?

Why don't you just find the bug in your software that
is stopping you reading the file.

It has been debugged and it works with all of the other files.

e
To be honest, I am
starting to think you are only claiming you can't read
the data because you have no answer to the proof it
provides.

You are delusional. Your phony inaccessible data ignores what
any high school astronomy student knows, that the vantage points,
horizons etc of the different sites are too different to permit a
comparison like the one you are trying to make.
The proof is that a trajectory assuming the transmission and
reception site motions are the same does not imply an anomalous
acceleration.


The fact that you have
not been able to do this suggests that there is something wrong
with
your method that you dont want to acknowledge.

No, it reflects that fact that I have done many hours
of work already, have debugged my code so that it
reads the files correctly and have obtained all the
data I need. I see no point in doing any more when the
problem is at your end, I have better things to do
with my time.
Evidently not. What for the nth time is the link to your code?

I even sent you the Excel sheet with all the data
already extracted so you didn't need to worry about
getting the data from the files yourself.

How do I know your excel sheet does not contain made up numbers?


If you want
to use it on other years, just enter the filename
and press the button. You have no excuse.

My excuse is that you have not given a direct link to your program
But all of this is a sideshow. You have not explained why the
difference in vantage points is Madrid and LA and Canberra can be
ignored to make the comparison you propose a valid comparison.

wrote:

I guess the answer is you have and found that this other
data does not support your contention.

You have the October 1987 data so instead of guessing,
why don't you look at the data and see if what I said
is true?

This is because your contention
ignores the differences in vantage points at the two locations and
the
differences in starting and ending times at the two locations etc.

The local horizon limits the periods of reception
obviously, but it doesn't affect the received
frequency when the craft can be seen.

I haven't found any corrupted files and in particular
Craig has confirmed the 1988 file is uncorrupted. There
is of course some noise in all the measurements but
that didn't stop me reading the files and it doesn't
stop you either.

My program reads all of the files except the 88 file. Why?

You haven't said what error you get but if it bombs
out and won't read the file at all, just debug the
code to see why. If it gives you values that are not
what you expect, perhaps it is because your theory
is wrong, but tell me how they differ from mine anyway.

Why is this the only file whose data you want to use. Why?

Because we spent time last year working with that data
and you said nothing about having a problem with it.
A year later, when you discover it proves you wrong,
suddenly the data has become corrupt. Why wasn't it
corrupt when we were talking about it? I suspect it
is because you never even looked at it. You just posted
guesses here without doing any work at all.

We are analysing _your_ idea so you should be doing
_all_ the work for others to peer review, I shouldn't
be doing any.

Why don't you just find the bug in your software that
is stopping you reading the file.

It has been debugged and it works with all of the other files.

e
To be honest, I am
starting to think you are only claiming you can't read
the data because you have no answer to the proof it
provides.

You are delusional. Your phony inaccessible data ignores what
any high school astronomy student knows, that the vantage points,
horizons etc of the different sites are too different to permit a
comparison like the one you are trying to make.

If you took the trouble to understand the method, you
would find that only the longitude matters. It only
takes high school calculus as you say, and I gave you
the base equation. Find the second derivative.

The proof is that a trajectory assuming the transmission and
reception site motions are the same does not imply an anomalous
acceleration.

Hilarious. The effect is barely detectable over 8 years
yet you are still asking for my help in checking your
excel sheet that only has 3 points on the same day. You
are about half a dozen orders of magnitude away from
finding a result.

I even sent you the Excel sheet with all the data
already extracted so you didn't need to worry about
getting the data from the files yourself.

How do I know your excel sheet does not contain made up numbers?

a) Because you can erase all the numbers and just
press the button to get the macro to re-read them
from your own raw data files.

b) Because you can change the fiename and get the same
macro to read the 1987 file and the data compare it
with your own.

c) Because you can interrupt the macro and single step
through every line to check that there is no code
present that can manufacture false values.

As I said:

If you want
to use it on other years, just enter the filename
and press the button. You have no excuse.

My excuse is that you have not given a direct link to your program

I first gave you access to the spreadsheet containing
the code at the end of last year. I can't get Google to
show the message but this was your reply:

[You wrote:]
[I wrote:]
The reception on the 6th/7th illustrates this quite
nicely and I've put some graphs he

http://www.georgedishman.f2s.com/Ralph/index.htm

We are way beyond these graphs which are not the least bit helpful.
Look at good instances in your spreadsheets or mine of 87 receiver
frequencies and transmitter frequencies minute by minute etc which
can

You didn't even bother to read to the bottom of the
page where the spreadsheet was linked.

Some time later you asked again so I split the page so
the spreadsheet was at the top and replied:

"George Dishman" wrote in message
...

wrote in message
oups.com...

George Dishman wrote:
wrote in message
oups.com...

I thought I had answered this. The 1988 data at this site is
corrupted but the 1987 data sets are not. By the way what is
the link to your program?

It is the first link on this page:

http://www.georgedishman.f2s.com/Ralph/index.htm

No. This is the excel file produced by a program that reads the
ATDF
file. Where is the program that produces the input to the excel
program?

When you open the file, the first sheet has a button
marked "Import Data". The program is the code attached
to that button. Press Alt-F11 to enter VBA and see the

This time you claimed you couldn't see the code
so I made it available as text:

"George Dishman" wrote in message
...

wrote in message
oups.com...

Couldn't do it. Just copy and paste the code here.

http://www.georgedishman.f2s.com/Ralph/code.txt

The pages are still there.

But all of this is a sideshow. You have not explained why the
difference in vantage points is Madrid and LA and Canberra can be
ignored to make the comparison you propose a valid comparison.

I did that last year before posting the proof. The
diurnal component depends on the location of the site
in two ways, the latitude and altitude influence the
magnitude while the longitude influences the phase.
(Remember the discussion of "point X"? We agreed all
this before I showed the proof because I knew you
would start rejecting basic maths as soon as you
saw the result.)

The method finds the time of the zero crossing point
which means the magnitude has no effect. You even
said at one point "I know where you are going with
this." and gave a reasonable summary of the method
before I had posted it so last year you understood
this, it has only slipped your mind since you found
out it proves you wrong.

George

wrote:
wrote:

I guess the answer is you have and found that this other
data does not support your contention.

You have the October 1987 data so instead of guessing,
why don't you look at the data and see if what I said
is true?


Unlike you, Markwardt,etc, I have posted my program and data and
trajectory analysis which show that the Pioneer 10 Anomalous
Acceleration is due to the fact that the speed of light delay does not
apply to distances over 3(10^5)km.
That is the spacecraft is shown to be moving differently than as
determined by reconciling the received frequencies with those predicted
based on Doppler shifts assuming the transmission earthsite motions are
different than the reception earthsite motions.
When this correction is made, successive positions and velocities of
Pioneer10 are evident that together with the earthsite motions during
transmission assumed to be the same as at reception give the observed
received frequencies without the need for increasing the velocities of
the spacecraft toward the sun!
That is, the anomalous acceleration is removed if the correct model
of light speed delay is acknowledged!

Now as to your strangely unobtainable data that purports to show
received frequencies at two different sites between craft rise and
craft set times at the different sites about eight hours apart.
You say that the time in each case when the earthsite motion in each
case is perpendicular to the plane that contains the craft-earthsite
line at each time, that these times are not when they should be if the
received frequencies in each case were transmitted a minute before.
The spin component of these different earthsite motions at these
times above and below the equator may be similar but necessarily equal
and the orbital motions are even more obviously different and the
angles of these different motions projected onto the different
craft-site lines in each case are obviously going to lead to different
times when the earthsite motions at these sites are perpendicular to
the craft-earthsite lines at these sites.
I disagree and perhaps we have to agree to disagree.

You haven't said what error you get but if it bombs
out

The program does not bomb. The data that is supposed to be in
certain fields is not there. This only happens for 1988.


Hilarious. The effect is barely detectable over 8 years
yet you are still asking for my help in checking your
excel sheet that only has 3 points on the same day. You
are about half a dozen orders of magnitude away from
finding a result.

You are hilarious. Anderson et al do not give the specific data and
reaon for their accuracy statements.
I do and you have merely to apply this analysis to get the result.
You have no excuse.


a) Because you can erase all the numbers and just
press the button to get the macro to re-read them
from your own raw data files.

b) Because you can change the fiename and get the same
macro to read the 1987 file and the data compare it
with your own.

c) Because you can interrupt the macro and single step
through every line to check that there is no code
present that can manufacture false values.

You have no excuse.

My excuse is that you have not given a direct link to your
program
Just copy and paste the damn code and stop playing this shell game.

writes:
....
When this correction is made, successive positions and velocities of
Pioneer10 are evident that together with the earthsite motions during
transmission assumed to be the same as at reception give the observed
received frequencies without the need for increasing the velocities of
the spacecraft toward the sun!
That is, the anomalous acceleration is removed if the correct model
of light speed delay is acknowledged!

Interesting. And how much data have you analyzed, compared to the 7
or more *years* of continuous tracking data other authors have
examined? Are you still limited by the coarseness of your one-minute
HORIZONS samples? (which leads to up to one-minute errors in the
light travel time, which leads to up to 2*Pi*(1min)/(1 day)*0.5 km/s =
2.2 m/s = 16 Hz doppler shift errors)

And how does your "light speed delay" scenario account for the fact
that in many cases the uplink transmitter is switched off at the same
time that the receiver is receiving valid tracking data?


Now as to your strangely unobtainable data that purports to show
received frequencies at two different sites between craft rise and
craft set times at the different sites about eight hours apart.
You say that the time in each case when the earthsite motion in each
case is perpendicular to the plane that contains the craft-earthsite
line at each time, that these times are not when they should be if the
received frequencies in each case were transmitted a minute before.

George's point is a good one. In your "scenario," where the light
travel time is nearly instantaneous, the time when the spacecraft
transits the zenith meridian will be the time of a local minimum
Doppler shift. He has shown it on his pages. It's basic geometry
too. The changes in velocity of the earth and the spacecraft over
this time period are negligible, so they wouldn't contribute in your
"scenario."

However, as George has shown, using your scenario, the same spacecraft
appears to transit at different longitudes. Thus your scenario leads
to (yet another) contradiction. Theories that lead to contradictions
are erroneous.

....
You haven't said what error you get but if it bombs
out

The program does not bomb. The data that is supposed to be in
certain fields is not there. This only happens for 1988.

The data from 1988 are fine. I have checked, and they produce
excellent tracking solutions. You have been "sure" of your programs
in the past, and sure the the rest of the world was wrong, only to
find your own mistakes and bugs later.


Hilarious. The effect is barely detectable over 8 years
yet you are still asking for my help in checking your
excel sheet that only has 3 points on the same day. You
are about half a dozen orders of magnitude away from
finding a result.

You are hilarious. Anderson et al do not give the specific data and
reaon for their accuracy statements.
I do and you have merely to apply this analysis to get the result.
You have no excuse.

Of course this is ridiculous. Anderson et al spend huge parts of
their paper describing the techniques and the error analysis.

I see you are continuing to spout unsubstantiated statements.

Craig

wrote in message
oups.com...

wrote:
wrote:

I guess the answer is you have and found that this other
data does not support your contention.

You have the October 1987 data so instead of guessing,
why don't you look at the data and see if what I said
is true?


Unlike you, Markwardt,etc, I have posted my program and data and
trajectory analysis which show that the Pioneer 10 Anomalous
Acceleration is due to the fact that the speed of light delay does not
apply to distances over 3(10^5)km.
That is the spacecraft is shown to be moving differently than as
determined by reconciling the received frequencies with those predicted
based on Doppler shifts assuming the transmission earthsite motions are
different than the reception earthsite motions.
When this correction is made, successive positions and velocities of
Pioneer10 are evident that together with the earthsite motions during
transmission assumed to be the same as at reception give the observed
received frequencies without the need for increasing the velocities of
the spacecraft toward the sun!
That is, the anomalous acceleration is removed if the correct model
of light speed delay is acknowledged!

I'll look at yours when you look at mine. As long as you
keep just snipping it or making superficial comments, I'm
not going to put in any effort on your stuff. I will do so
as long as you play fair and think seriously about the test
I have shown you.

Now as to your strangely unobtainable data


It is very strange that you cannot read it when everyone
else can, but that can be resolved, my code is available
for you to use.

that purports to show
received frequencies at two different sites between craft rise and
craft set times at the different sites about eight hours apart.
You say that the time in each case when the earthsite motion in each
case is perpendicular to the plane that contains the craft-earthsite
line at each time, that these times are not when they should be if the
received frequencies in each case were transmitted a minute before.

To be exact, I said the plane that contains the craft and
the Earth's _axis_ but they are equivalent as the motion
is perpendicular to the plane.

The spin component of these different earthsite motions at these
times above and below the equator may be similar but necessarily equal

That is correct, the magnitude depends on the latitude
of the respective sites. However, the phase does not
which is why I use the zero crossing. A * sin(theta)
is zero regardless of the value of A if theta is zero.
Please think carefully about that, you cleary haven't
grasped the importance when we discussed it previously.

and the orbital motions are even more obviously different

You are right in principle but I already addressed that
and showed that the maximum error is less than a degree
while the discrepancy between the sites is 26 degrees.

and the
angles of these different motions projected onto the different
craft-site lines in each case are obviously going to lead to different
times when the earthsite motions at these sites are perpendicular to
the craft-earthsite lines at these sites.

Yes, that's the basis of the test. The two times should
differ by an amount equivalent to the difference in the
site longitudes but they miss by 26 degrees.

I disagree and perhaps we have to agree to disagree.

There is no need to disagree when we are discussing
simple geometry. The objections you raised above are
not valid but perhaps we need to look closer at the
second one for you to be convinced. Really it would
be best to use the analytical approach from the
equation I posted some wweeks ago.

You haven't said what error you get but if it bombs
out

The program does not bomb. The data that is supposed to be in
certain fields is not there. This only happens for 1988.

OK, now we are getting somewhere. Which fields?

Hilarious. The effect is barely detectable over 8 years
yet you are still asking for my help in checking your
excel sheet that only has 3 points on the same day. You
are about half a dozen orders of magnitude away from
finding a result.

You are hilarious. Anderson et al do not give the specific data and
reaon for their accuracy statements.

The data is the same files we are using. The accuracy is
the fully discussed in Chater X and detailed in Table II.
I haven't seen your equivalent discussion.

I do and you have merely to apply this analysis to get the result.
You have no excuse.

I don't need an excuse, getting a correct answer proves
nothing in science when there is already another test
which the theory fails.


a) Because you can erase all the numbers and just
press the button to get the macro to re-read them
from your own raw data files.

b) Because you can change the fiename and get the same
macro to read the 1987 file and the data compare it
with your own.

c) Because you can interrupt the macro and single step
through every line to check that there is no code
present that can manufacture false values.

You have no excuse.

My excuse is that you have not given a direct link to your program
Just copy and paste the damn code and stop playing this shell game.

You just snipped the link! Here it is again:

Text version

http://www.georgedishman.f2s.com/Ralph/code.txt

The excel sheet in which it lives (zipped, 600kb)

http://www.georgedishman.f2s.com/Ral...lph_signed.zip

If you can't unzip that, here is the raw xls file (3.6Mb):

http://www.georgedishman.f2s.com/Ral...March_1988.xls

Once you download that, either go to "Tools", "Macro",
"Visual Basic Editor" or press Alt-F11.

If you get a blank editing area, either go to the menu
and select "View", "Project Explorer" or press Control R.

Under the entry marked "Microsoft Excel Objects" you
should find an entry "Sheet 1 (Control)". Double-click
on that to see the code.

Sorry it's so buried but you'll have to take that up
with Microsoft :-(

The code will read the data files into the second sheet
in the file. After that you can use it for anything you
like including your own analysis.

If you can download the file and run the button on
the "Control" sheet, I'll explain more about how you
can use the fields. They should be fairly obvious.

George

Craig Markwardt wrote:
writes:
...
When this correction is made, successive positions and
velocities of
Pioneer10 are evident that together with the earthsite motions
during
transmission assumed to be the same as at reception give the
observed
received frequencies without the need for increasing the velocities
of
the spacecraft toward the sun!
That is, the anomalous acceleration is removed if the correct
model
of light speed delay is acknowledged!

Interesting. And how much data have you analyzed, compared to the 7
or more *years* of continuous tracking data other authors have
examined?


The wrong conventional model and the closer fitting instantaneous
model is immediately obvious in 60 minutes of data and the claim that
though wrong for some data, the average over 7 years is not too bad
does not help.
Especially since it is not accompanied by any demonsration of how
the analysis is done and how the trajectory is obtained. The
unsubstantiated claim is more suspect coming from someone who has lied
about the existence of transmissions from the craft to the earth even
when the craft has not received transmissions from the earth to be
retransmitted back to earth.


Are you still limited by the coarseness of your one-minute
HORIZONS samples? (which leads to up to one-minute errors in the
light travel time, which leads to up to 2*Pi*(1min)/(1 day)*0.5 km/s
=
2.2 m/s = 16 Hz doppler shift errors)

It doesn't lead to such errors for the sixty minute period chosen.
You appear to be referring to the difference in time values associated
with the received frequencies and the time values associated with the
ephemeris earth site motions. This is a minute as I understand it
although it may be slightly less than a minute. So that the frequencies
received at 6:20 and 6:21 should be associated with the earthsite
position at 6:19 and 6:20 or 6:21 and 6:22. And this may be off by 10
seconds for the last earthsite position and motion data.





And how does your "light speed delay" scenario account for the fact
that in many cases the uplink transmitter is switched off at the same
time that the receiver is receiving valid tracking data?

Why do you still persist in lying about the fact that the received
transmissions could be from the craft transmitter which is always
transmitting to the earth even if it is not retransmitting signal
received from the earth? Perhaps you are insinuating that if the
received tracking data is "valid" that it must by retransmitted
oscillations and not oscillations originating in the craft and
transmitted to earth.

Now as to your strangely unobtainable data that purports to show
received frequencies at two different sites between craft rise and
craft set times at the different sites about eight hours apart.
You say that the time in each case when the earthsite motion in
each
case is perpendicular to the plane that contains the
craft-earthsite
line at each time, that these times are not when they should be if
the
received frequencies in each case were transmitted a minute before.

George's point is a good one. In your "scenario," where the light
travel time is nearly instantaneous, the time when the spacecraft
transits the zenith meridian will be the time of a local minimum
Doppler shift. He has shown it on his pages. It's basic geometry
too. The changes in velocity of the earth and the spacecraft over
this time period are negligible, so they wouldn't contribute in your
"scenario."


No these changes when you take into account the changes in the
craft-earthsite line in each instance are not negligible.







...
You haven't said what error yoisu get but if it bombs
out

The program does not bomb. The data that is supposed to be in
certain fields is not there. This only happens for 1988.

The data from 1988 are fine. I have checked, and they produce
excellent tracking solutions. You have been "sure" of your programs
in the past, and sure the the rest of the world was wrong, only to
find your own mistakes and bugs later.

Given your history of distortion and lying, who would believe
you?
There is no problem with the other files so that there is something
different about the 1988 file.



Hilarious. The effect is barely detectable over 8 years
yet you are still asking for my help in checking your
excel sheet that only has 3 points on the same day. You
are about half a dozen orders of magnitude away from
finding a result.

You are hilarious. Anderson et al do not give the specific data
and
reaon for their accuracy statements.
I do and you have merely to apply this analysis to get the result.
You have no excuse.

Of course this is ridiculous. Anderson et al spend huge parts of
their paper describing the techniques and the error analysis.

Only in general terms. No data and no method is given in any
specific way so there is no way in hell anyone else can check their
claims.

I see you arie continuing to spout unsubstantiated statements.

My statements and analyis are at mysite.verizon.net/r9ns
using the data archived raw ATDF data except that of 1988 which does
not appear to be in the same format as the other files.
This transparent analysis indicates that the anomalous acceleration
is much larger than claimed and is due to assuming the wrong earth site
motions in calculating the doppler shift of the transmitted and
retransmitted back frequencies.

writes:
Craig Markwardt wrote:
writes:
...
When this correction is made, successive positions and velocities of
Pioneer10 are evident that together with the earthsite motions during
transmission assumed to be the same as at reception give the observed
received frequencies without the need for increasing the velocities of
the spacecraft toward the sun!
That is, the anomalous acceleration is removed if the correct model
of light speed delay is acknowledged!

Interesting. And how much data have you analyzed, compared to the 7
or more *years* of continuous tracking data other authors have
examined?


The wrong conventional model and the closer fitting instantaneous
model is immediately obvious in 60 minutes of data and the claim that
though wrong for some data, the average over 7 years is not too bad
does not help.

I notice that you diverted away from the question. Your claims could
have been checked by examining more data than a single tracking pass,
but you did not perform this check.

Papers like Anderson et al and Markwardt have fitted seven or more
*years* of data with a highly successful model. The residuals of
these models are a few *milli*Hertz (rms), and are consistent
throughout the fit, not just a single day. You could have examined
either Anderson et al or Markwardt to see this, but you apparently did
not. So it seems like you continue your unsubstantiated claims.


Especially since it is not accompanied by any demonsration of how
the analysis is done and how the trajectory is obtained.

Again, if you would have bothered to read the Anderson et al. or
Markwardt papers, you would have found extensive descriptions of how
the analysis is done, and how the trajectory is obtained, including
the modeling of the observables, the equations of motions, the
integrator, quality control, the fitting process, and error analysis.
Apparently you did not bother to read these articles, and your
unsubstantiated claims are compounded.

... The
unsubstantiated claim is more suspect coming from someone who has lied
about the existence of transmissions from the craft to the earth even
when the craft has not received transmissions from the earth to be
retransmitted back to earth.

Your claims of a "lie" are also unsubstantiated. I never claimed
there were no transmissions from the spacecraft. You could have
attempted to find a quotation to back up your claim, but you did not.



Are you still limited by the coarseness of your one-minute
HORIZONS samples? (which leads to up to one-minute errors in the
light travel time, which leads to up to 2*Pi*(1min)/(1 day)*0.5 km/s
=
2.2 m/s = 16 Hz doppler shift errors)

It doesn't lead to such errors for the sixty minute period chosen.

How can you make such a claim? When in fact, you found that your
"predicted" frequency was off from the measured one by ~10 Hz.

You appear to be referring to the difference in time values associated
with the received frequencies and the time values associated with the
ephemeris earth site motions.

No, I am not referring to the UTC vs. TDB time difference.

What I am referring to is the fact that the light travel time is *not*
a multiple of whole minutes even though you assume that it is. Of
course the true round trip light travel time will depend on the
distance, and may take on any fractional number of minutes. Because
you used the HORIZONS output, which only has one minute sampling, you
are assuming something that is not true (fractional minute vs. whole
minute). Since your light travel time can be off by up to +/- 1
minute, this leads to a one minute error in the rotational phase of
the earth, which is up to 2.2 m/s linear speed, or 16 Hz maximum
Doppler shift. This is very closely in line with the actual error
that you found.

Furthermore, the spacecraft round trip light travel time gradually
changes during the tracking pass, even though you assume that it is
fixed to a certain number of whole minutes (again through your use of
the HORIZONS output). Thus, this leads to a gradually changing error
in your predicted frequency, which in turn leads to the residual
trends you describe.

A diligent scientist will investigate the sources of error and try to
minimize them. You on the other hand seem blithely unconcerned about
these very real sources of error in your analysis, despite repeated
warnings.


And how does your "light speed delay" scenario account for the fact
that in many cases the uplink transmitter is switched off at the same
time that the receiver is receiving valid tracking data?

Why do you still persist in lying about the fact that the received
transmissions could be from the craft transmitter which is always
transmitting to the earth even if it is not retransmitting signal
received from the earth?

Again, you should be able to back up your claims of "lying." In this
case, if you had read my statement, I said that the *uplink*
transmitter was off, and I said nothing about the spacecraft
transmitter. Thus, your claims of "lying" are unsubstantiated.

... Perhaps you are insinuating that if the
received tracking data is "valid" that it must by retransmitted
oscillations and not oscillations originating in the craft and
transmitted to earth.

No, I am not insinuating that. I am saying that if the uplink
transmitter is off, then there is no way that downlink receiver could
be receiving two-way doppler tracking data. Period. "Two-way," needs
both directions, not just one.

Now, it appears that *you* are insinuating that if the spacecraft
transmitter is on all the time, it is possible to receive coherent
(two-way) tracking data on the ground. That is an erroneous
supposition for several reasons. First, if the spacecraft is
receiving no uplink carrier, then the spacecraft transmitter cannot be
transmitting a coherent downlink at the 240/221 turnaround ratio.

So second, the spacecraft would have to be relying on its own internal
oscillator to generate the (one-way) downlink carrier. While usually
of high quality, spacecraft oscillators are nowhere nearly as stable
as can be achieved on the ground with Hydrogen maser frequency
standards and GPS timing. Thus, high precision tracking is usually
not possible in one-way mode. See for example Rebold et al. 1997,
which shows the non-coherent distortion effects in a Voyager tracking
pass.

In the Markwardt analysis, the goodness of fit was excellent,
regardless of whether the uplink transponder was off at the same time.
[ Of course it was not off one light travel time earlier. ] In fact,
it is quite straightforward to see the transition in the data from
non-coherent (one-way) to coherent (two-way) when the spacecraft
starts receiving the uplink carrier. If you would have checked more
data, you could have seen this effect, but you did not. [ In fact, I
did check your scenario over a span of several years, and it fails
utterly. ]

And finally, even though the receiver is receiving tracking data at
the same time that all transmitters are turned off, the tracking data
still contains the Doppler shift of the *uplink* from one light travel
time earlier. Truly, your scenario is not viable.



....
George's point is a good one. In your "scenario," where the light
travel time is nearly instantaneous, the time when the spacecraft
transits the zenith meridian will be the time of a local minimum
Doppler shift. He has shown it on his pages. It's basic geometry
too. The changes in velocity of the earth and the spacecraft over
this time period are negligible, so they wouldn't contribute in your
"scenario."


No these changes when you take into account the changes in the
craft-earthsite line in each instance are not negligible.

Well, that can be checked. In your scenario, all that matters is the
acceleration of the earth during the tracking pass (i.e. you do not
admit light travel time). The circular acceleration of the earth is
approximately (V_earth)^2 / D_earth = (30 km/s)^2/(1 AU) = 0.6 cm/s^2.
During a 3 hour pass, the maximum velocity change is thus 0.06 km/s
and slowly varying. This is small compared to the rapidly varying ~0.5
km/s velocity changes imparted to the stations due to earth rotation.

The accelerations of the Pioneer spacecraft at 65 AU are (G*M_sun)/(65
AU)^2, or 0.0001 cm/s. Also, completely negligible.

Thus, your "not negligible" claim is not substantiated by even the
most basic calculations. You could have performed these calculations,
but you did not.


The data from 1988 are fine. I have checked, and they produce
excellent tracking solutions. You have been "sure" of your programs
in the past, and sure the the rest of the world was wrong, only to
find your own mistakes and bugs later.

Given your history of distortion and lying, who would believe
you?
There is no problem with the other files so that there is something
different about the 1988 file.

Given that your claims of "lying" are unsubstantiated, your criticism
is irrelevant. In fact, the 1988 data produce the same high quality
solution as the other years.

And it is ironic that you are claiming the data is corrupt and your
program is right. In the past, you also claimed that the data were
corrupt and your program was right, because you were getting "all
zeroes" in the uplink-enabled field. But then you found a bug in your
program (ref. 2). Or how about the time you were convinced that your
spreadsheet was correct and that the world was wrong, but it turned
out that you had entered the wrong uplink frequency in the spreadsheet
(ref. 3)?

To make it plain, the irony is that now you are once again claiming
that everybody is incorrect and that your calculations are correct,
when as usual you haven't supplied much substantiation. (For example,
you haven't considered more than a single tracking pass).

George has even showed you how to load multiple data sets into a
spreadsheet using an Excel macro, and yet you can't be bothered.
Simply amazing.


Hilarious. The effect is barely detectable over 8 years
yet you are still asking for my help in checking your
excel sheet that only has 3 points on the same day. You
are about half a dozen orders of magnitude away from
finding a result.

You are hilarious. Anderson et al do not give the specific data and
reaon for their accuracy statements.
I do and you have merely to apply this analysis to get the result.
You have no excuse.

Of course this is ridiculous. Anderson et al spend huge parts of
their paper describing the techniques and the error analysis.

Only in general terms. No data and no method is given in any
specific way so there is no way in hell anyone else can check their
claims.

Again, this is a laughably unsubstantiated claim, as described above.
Data extraction, data selection, tracking analysis, quality and error
analysis are all discussed in both the Anderson et al. and Markwardt
et al papers.

So it seems that your pattern of making reams of unsubstantiated
claims is continuing rapidly!

CM

References
1. Rebold et al. 1997, The Telecommunications and Data Acquisition
Progress Report 42-131
2. Sansbury Usenet article
3. Sansbury Usenet article

In message , Craig Markwardt
writes

writes:
Craig Markwardt wrote:
writes:
...
When this correction is made, successive positions and velocities of
Pioneer10 are evident that together with the earthsite motions during
transmission assumed to be the same as at reception give the observed
received frequencies without the need for increasing the velocities of
the spacecraft toward the sun!
That is, the anomalous acceleration is removed if the correct model
of light speed delay is acknowledged!

Interesting. And how much data have you analyzed, compared to the 7
or more *years* of continuous tracking data other authors have
examined?


The wrong conventional model and the closer fitting instantaneous
model is immediately obvious in 60 minutes of data and the claim that
though wrong for some data, the average over 7 years is not too bad
does not help.

I notice that you diverted away from the question. Your claims could
have been checked by examining more data than a single tracking pass,
but you did not perform this check.

Papers like Anderson et al and Markwardt have fitted seven or more
*years* of data with a highly successful model. The residuals of
these models are a few *milli*Hertz (rms), and are consistent
throughout the fit, not just a single day.

11.5 years in the full data set! BTW, what caused the increased
residuals at about 1800 days in figure 8 of the
http://xxx.lanl.gov/abs/gr-qc/0104064 paper? Just curious.

Jonathan Silverlight writes:
....
11.5 years in the full data set! BTW, what caused the increased
residuals at about 1800 days in figure 8 of the
http://xxx.lanl.gov/abs/gr-qc/0104064 paper? Just curious.

If you mean the increase in the *scatter* in the residuals, it appears
to have been a noisy few months in 1992 (see for example Fig 3 of the
Markwardt paper). I'm not clear why, although it was around solar
maximum, so the solar wind could have been chaotic. Solar
conjunctions are also a time where the Doppler noise increases, as has
been discussed in Anderson et al and Markwardt.

Not that it made a difference. Including all the noisy data changed
the anomaly in the solution by only 6%.

Craig

--
--------------------------------------------------------------------------
Craig B. Markwardt, Ph.D. EMAIL:
Astrophysics, IDL, Finance, Derivatives | Remove "net" for better response
--------------------------------------------------------------------------