Re: Pioneer 10 test of light speed

From: George Dishman (george_at_briar.demon.co.uk)
Date: 01/09/05 

Date: Sun, 9 Jan 2005 17:14:11 -0000

<r9ns@verizon.net> wrote in message
news:1105281297.505946.192950@c13g2000cwb.googlegroups.com...
>
> George Dishman wrote:
>> Hi Ralph, Happy New Year.
>>
>> Have you received my email reply to your message?
>
> No

I'll try again as it has some useful explanations
on how to use the spread***.

> but I did look at your spread*** based on a program like mine that
> extracts relevant items from the radiometric data. Very good but your
> idea that the cause of changes in the received frequency are not sound
> because these changes could have been caused by changes in the freqency
> transmitted from earth due to changes in the earth transmitter which
> items you have omitted in your spread***.

The file data is in column M headed "Tx Freq". This
is field 116 or bits 2125 to 2160 which appears to
be the actual value divided by 9.6 and rounded to
the nearest integer (i.e. the resolution is +/-4.8Hz).
You can get a more accurate value by using other
fields, see Craig's comments in this document:

http://lheawww.gsfc.nasa.gov/users/craigm/atdf/

Look near the bottom under "FURTHER REDUCTION". You
can extract any field just by adding a column on the
"Extracted" *** and putting the bit numbers at the
top. Leave a blank column at the right hand end so
the macro knows where to stop.

The drawback is that you have to interpolate between
the "type 6" records but it shouldn't be too much
trouble. I might do that myself as it may be of use
to me too, but generally I'm not doing any more work
on your theory because I've already proved it wrong.

> The transmitter frequencies
> at different sites are different and the transmitter frequencies at
> routinely readjusted over time at the same site.

There is only a single DCO (digitally controlled
oscillator) which I believe is used as the
reference both for the uplink and the downlink.
Where there is simultaneous transmission and
reception, the same DCO reference is used for both.
If there was no change in column M, there was no
change in the transmit frequency or Doppler
reference. Anyway, you can easily see the periods
of adjustment on the graphs.

> This may also explain
> the pattern you saw in 1988 data.
> Also you do not show clearly that the 1987 data for Oct 6 and 7 has
> the same pattern as that you showed in the inaccessible data for 1988.

First note the 1988 data isn't inaccessible any
more, the *** will a section from an ATDF file
of _any_ length. That was my reason for writing
it!

I'm not sure what you mean by pattern here. In the
March 1988 data each site both transmitted and
received though for varying durations. In October
1987, all the transmissions were from antenna 12
while all the receptions were at 63 so the
scheduling was quite different. As Jonathan said,
simultaneous reception and transmission isn't
always possible.

My test was to check the consistency of RA from
different sites because neglecting the travel
time has the effect of neglecting the amount the
Earth rotates between receptions. For the March
data, I could get a result from both Madrid and
Canberra and show they were inconsistent using
your theory, but the October data only has
reception at a single site so there's no way to
apply my test to those days.

> Dont show the graphs but show the list of received frequencies at the
> two different sites like you did for 1988.(These are apparently
> different now - site 12 LA is mentioned in place of Canberra or Madrid)

I wish you would look at the graphs more carefully,
they already answer most of the questions you are
asking. One important thing is that they make it
obvious where you have good three-way data and where
it is one-way, tx only or the frequency is settling
after a mode change. Use that to guide you in your
selection of specific values to analyse.

The transmit frequency is in column M though you
can get a more accurate version by using other
columns and a bit of calculation as I explained
above. The received frequency is linearly related
to column Q headed "Doppler". That is in fact the
raw counter output so just needs to be scaled and
offset.

> Also, I wish you would look at
> http://mysite.verizon.net/r9ns/trj071087.xls as a way of obtaining the
> true position and velocity of the craft and help derive the formula for
> the subsequent craft velocity and subsequent position.

I have looked at it, that and your previous comment
is why I used those dates in October for the sample
data in the file.

> The problem is to change by trial and error the given coordinates of
> craft position and velocity so as to produce a closer fit between the
> predicted and observed received frequency at that instant. Then to
> produce the subsequent velocity based on this previous changed given
> velocity and the acceleration of the craft toward the sun from the
> previous changed given position so as to maintain this closer fit.

That isn't the problem at all, the problem is to
test your theory, and for the last two years I have
been pointing out that the location derived from
the diurnal component can do that. Perhaps you
should explain how your test is supposed to work
before we worry about how you can calculate the
values you need.

It is always possible to select a trajectory that
matches a single data point so as a minimum you
need to apply the same method to other points in
the same contact period and then see if you can
find values of K and V etc. that work for them all.
Just using the conventional values are you do at
the moment tells us nothing about your theory.

George

Quantcast