Il giorno martedì 25 aprile 2017 19:15:29 UTC+2, Craig Markwardt ha scritto:
On Monday, April 24, 2017 at 7:17:03 AM UTC-4, wrote:
On Thursday, April 20, 2017 at 7:36:10 AM UTC+10, Craig Markwardt wrote:
On Wednesday, April 19, 2017 at 5:50:36 AM UTC-4, wrote:
On Saturday, April 1, 2017 at 7:47:20 AM UTC+11, Craig Markwardt wrote:
---
---
The question is what tolerance was required for the analysis. Anderson
et al's analysis was quite back of the envelope. A few Watts makes a
difference, so it was important to treat the analysis in more detail.
...
Fig.2 from this link
https://arxiv.org/pdf/1204.2507v1
apparently suggests that the Pioneer anomaly may only be the
result of mismodelling of the solar thermal contribution.

Your interpretation of this paper is incorrect. This paper
(Turyshev et al 2002) demonstrates that most if not all of the
"anomalous" acceleration can be attributed to *internal* thermal
emission, generated by the RTGs and internal electronics.

I note no response.

But such an error can't reside in the mismodelling of the solar
reflection/absorption characteristics of the HGA dish because
that was accurately determined prior to launch,

Just a side note, but thermal coatings do degrade over time, due
to exposure to solar UV as well as ionizing radiation (white
coatings become darker). So one can't be so certain about these
properties. That was one of the aspects covered by the thermal
analysis reported in Turyshev et al's 2012 paper.

It's obvious that either Turyshev or Anderson made a fairly
significant error. But it's strange that Anderson is deemed
wrong when a significant number of others also demonstrated the
existence of the anomaly, including yourself (2002)? Why should
this new result override all others when the average result from
all contributors well and truly favors the anomaly's existence?
That's a funny way to do physics.

What would be funny is to ALWAYS assume that the "average" of some
papers is relevant. ...

I note no response.

But actually it's not true to say no one else was considering
thermal. ...

I note no response.

...
For John Anderson's analysis, each data point was obviously
generated using a common logic. ...

Absolutely not. The data points you are referring to in your own
plot are taken from Anderson's paper, Figure 7. If you had read
the Anderson et al 2002 paper more carefully you would have
understood that this figure represented the early analysis, of
analysts working in an operational environment and trying to
navigate the spacecraft. The analysis was not performed in a
totally rigorous way by the same analysts or with the same criteria.

As they pointed out in the paper, when the spacecraft reached about
20 AU, the search for unmodeled accelerations could begin in more
earnest, and more attention to detail was made... but 20 AU is way
past the so-called "onset" around 5AU! At the time of the Anderson
et al 2002 paper, the early data was lost (unavailable) and only
those graphical figures remained, not the original data. Thus, when
it came time to do new, more careful work, their group had to begin
with data starting about 1980, and there was no way to look for a
putative onset around 5 AU which occurred in the early 1970s.
[ Later, after Anderson's work, a small amount of earlier data was
recovered. ]

Just to be clear, during the early to mid 1970s, the data formats,
the software, and the analysis teams were in much more flux. NASA
was in the business of navigating spacecraft (and not high
sensitivity spacecraft physics), and was just learning how powerful
the radiometric techniques could be. You can go back and read all
of the DSN status documents if you want: they are public and online.
But the point is that what we know and do today to navigate
spacecraft, does not and did not apply to what was done in the early
1970s when this was a new field. You cannot say that anything was
done with a "common logic" or "invariant" comparing today to then.

You act as though analysts from the 1970's were using a set of
physical laws that analysts of today are not familiar with. But
that's not the case at all. Even if modifications have been made
to some of the physical laws the analysts of today would still
know exactly how the first data points showing the onset of the
anomaly were derived.

This is unsubstantiated. I am an analyst today of the Pioneer doppler data, and I do not have enough information from that one "onset" chart to understand how it was derived.

Furthermore, let's be clear what was happening in the early 1970s. The Pioneer spacecraft were the first spacecraft to be sent to the outer solar system, with long cruise phases. Earlier spacecraft were kept inside the Earth-Moon system, or directed to the inner solar system. This new direction required new attention to detail and modeling of the physical effects of the space environment. No, the physical laws did not change, but the modeling capability within the software that was used did not capture all of the physics with the same fidelity.

Effects such as spacecraft spin and new relativistic physics were being incorporated at that time. You can read the paper of Wong & Lubeley from 1974 (AIAA Paper No. 74-845) which demonstrates how the orbit determination groups were incorporating new software with better capabilities.

Effects such as spacecraft propellant valve leaks were harder to understand, and navigation analysts had to make do. What physical laws does one apply to a valve that is leaky???

Before encountering Jupiter, both Pioneer spacecraft performed frequent maneuvers (every few weeks), which disturbed the orbit determination process, and also introduced more possibilities for leaks.

Within the first 20 AU, the solar system radiation pressure is significant. In fact, before Jupiter approach, the radiation pressure is dominant over most of the force terms and is many times the "anomalous" acceleration. You pointed out thermal coatings, but let us not forget that the exact geometry of the spacecraft has an effect. Is the antenna treated as parabolic or flat-plate? What about the cut-outs and antenna feeds, how are those treated? These are not simple "physical law" issues. They have to do with how accurately the spacecraft systems can be modeled. Early in the mission, with computing power limited, some short cuts were inevitably taken.

I would also like to point out that the planetary positions, which we know so accurately today, were not known as accurately at the time. In fact, the Pioneer and Voyager encounters were used to refine the positions of the planets based on the fly-by data. But, before the fly-by, the knowledge of the planet positions ... and the corresponding acceleartions ... were less well known. Thus, there is some uncertainty there too.

So, please do not pretend that the state of affairs in the early 1970s can really be compared to the state today.

The figure which you have referred to which shows the "onset" of the acceleration, you have replotted. What you failed to show was the uncertainty range (error bars) of the points. If you had displayed those as well, you would have seen that the bars are very large and that early data is not necessarily as stringent as you say.



... The analysts weren't stupid back then, they noticed
this error and reported it. They couldn't explain it and neither
could anyone else. Your suggestion that the analyst were too busy
trying to navigate the spacecraft to properly analyze the
evidence is absurd. They weren't exactly alone in the world were
they!

Please let me clear. The Pioneer navigators were brilliant for their time. I referred you to one paper. There are several, and it is evident that they were highly skilled at their craft. BUT... they were also not able to fully exploit the capability of the radiometric data they were using. This was evident in their own records which show they were upgrading the software steadily with new capabilities. That early data MAY NOT have been as accurate as desired.

I also want to say that I am in no way criticizing John Anderson or his team of co-authors. His science track record is quite excellent, and it is clear that they are very capable as well. BUT AGAIN... the question is how good does someone have to be. And the Turyshev paper demonstrated that a higher quality thermal analysis can resolve the Pioneer "anomaly."

...
According to you, "when the spacecraft reached about 20 AU, the
search for unmodelled accelerations could begin in more earnest,
and more attention to detail was made". Why do you think a more
precise analysis of the more distant data points will be of any
use? It becomes quickly apparent where the curve is heading. But
so what? The horse has already bolted.

First of all, the quote is actually what Anderson et al. said in their paper, not me. And second of all, what is your point? If the goal is to accurately measure the "onset" of the anomaly, then the early data which would or could have shown that is largely lost forever. If the goal is to get an accurate measure of the anomaly in late days (20 AU) then the Anderson et al 2002 paper is just fine... but it doesn't prove or disprove any of your claims then. Anderson's paper claims the acceleration is consistent with being constant, quite the opposite of your claim!


The unnecessarily huge error margin I'm referring to is of course
the 25% generated by the unknown performance of the RTG coating.
Precise details of the coating are well known, so why hasn't this
question been addressed years ago. In Turyshev's paper it was
stated that "the properties of the RTG paint are, in principle,
measurable by a thermal vacuum chamber test of a hot RTG
analogue". So how can this huge error still exist? After more
than 40 years!

Problem 1. The exact Pioneer thermal coating materials no longer exist, so what do you suggest we do? (even if the formula still exists, there is significant variance from batch to batch, which provides its own uncertainty)
Problem 2. The space environment destroys thermal coatings. Do you have access all of the relevant space environments? (UV light, ionizing radation, extreme temperatures)
Problem 3. Funding. Doing the job properly requires access to facilities that are expensive, and technicians that are expensive. No money is available for this.

You doing a vacuum test with Plutonium is rather laughable, but please do not: Plutonium is toxic.

Why do I get the impression that huge error margins are a good
thing in the case of the Pioneer anomaly?

Thankfully you have the wrong impression.

The Turyshev team have perhaps managed to squeeze out a positive
result within the error bars, but they certainly haven't managed
to remove the problem. They could only shift it down a little at
best.

You can believe what you wish, but the Turyshev result shows that
the thermal and doppler solutions are consistent - within the
tolerances (see their Figure 4).

"believe" is the key word here isn't it.

I don't see you providing an alternate analysis of the thermal or doppler data. And, other researchers working independently have found similar results (see Bertolami's work that I pointed out earlier). Turyshev et al's work demonstrates within the evidence and known tolerances that the Pioneer "anomaly" has largely been resolved.

CM

.... rerephrasing the last question : is Cassini having a Dppler shift like Pioneer , and yearly and dayly ?