Re: Laser ranging to moon begs questions


John C. Polasek <jpolasek@xxxxxxxxxx> writes:

On 18 Oct 2007 00:35:58 -0400, Craig Markwardt
<craigmnet@xxxxxxxxxxxxxxxxxxxxxxxxxx> wrote:


John C. Polasek <jpolasek@xxxxxxxxxx> writes:

On 17 Oct 2007 08:46:53 -0400, Craig Markwardt
<craigmnet@xxxxxxxxxxxxxxxxxxxxxxxxxx> wrote:


John C. Polasek <jpolasek@xxxxxxxxxx> writes:


Ref. http://sunearth.gsfc.nasa.gov
NASA says laser ranging to corner reflectors on the moon has
determined that the moon is receding at 3.8 cm/year. This remark is
puzzling:
"Even under good atmospheric viewing conditions, only one
photon is received every few seconds".

With a signal that bad, how could they determine the range to within,
say, 1.0 cm out of about 4*10^10 cm? It would seem they would try to
time the onset of the return pulse and do lengthy cross-correlation,
but it's hard to say there is even a signal.


You seem to confusing the issues.
Thank you for your assessment, and especially your declension into 5
genera of confusion a-e, but I believe it's misplaced. We were all
working from the same sparse data base: 3.8cm/yr and one photon every
few radar pulses. We tried to work with that.

But, I found a URL that has the equipment specifications to help with
the discussion: http://cddisa.csfc.nasa.gov
Equipment spec is dated 1994

The web site address you linked is incorrect. More recent information
can be found here. http://ilrs.gsfc.nasa.gov/stations/
All I get is a map of the stations round the world from your site.

True. Did you try to click any of the links? Like, say, McDonald
Observatory?

I goofed the URL; it's complicated: good luck.
http://cddias.gsfc.nasa.gov/slr_sys/mlrs_sys.html

It's still out-dated information.


(a) It is possible to measure the arrival time of a photon very
accurately, probably at the picosecond level for "stopwatch" type of
measurements measurements.
Pulse width is 200 ps; power is 1500 mJ @ 10 Hz with .075microradian
beamwidth. They used a spinning disk as the T/R "switch"
Time resolution with a 10 Mhz clock and 4 plus 4 verniers is stated as
50 psec.
1cm of double range is about 66 psec.

(b) The difficulty is not in the relative ranging, so whether it is
1 cm out of 1e10 or 4e10 is not relevant.
?

It was you who questioned, "with a signal that bad, how could they
determined the range to within, say, 1.0 cm out of 4*10^10 cm?"
First, the premise of your question, that the signal was "bad" was
erroneous. Second, the "out of 4*10^10 cm" part of the question is
not really relevant since the accuracy of the experiment depends on
the precision of time tagging (and modeling), and not on the distance
to the moon. For example, the same basic ~tens of picosecond
precisions are achieved for spacecraft laser ranging.


(c) The real issue is whether a single photon arriving from a
distant source is really that source or a contaminating background
photon. The experimenters have carefully tuned the apparatus so
that only photons within a narrow time and wavelength window are
permitted, so the background contamination is exceedingly small.
I.e. the return signal is still whoppingly bright compared to the
background level.
It's hard to believe some of the specifications: 1500mJ must be 1.5J
(not megaJ) because even then, at 200ps pulse width, the peak laser
power would be 7,500 megawatts, but avg power of 15W. If it's
megajoules, avg power is 15,000 megawatts average and that doesn't
seem possible.

What you find hard to believe is not really relevant. The McDonald
printed laser properties are consistent with the properties of other
laser ranging stations.
Craig you keep reading me wrong. It's not about my belief. (Your minor
in psychology will get you nothing but trouble).
I should have been explicit: I didnt know whether they meant
millijoules or megajoules, so I made the two calculations, so as to
rule one out. I'm going with 15W (but that can't be right).

Brief tutorial in SI prefixes:
m = milli = 10^{-3}
M = Mega = 10^{6}
I note that you didn't provide any substantiation why "that can't be
right."


It's hard to believe the .075 microradian beam width, because on
arrival at the moon the spot would span 28 meters. (Another article
mentioned "4 miles wide"-author's embellishment?). The required aiming
accuracy is unbelievable, certainly requiring better than .075
microradians. How accomplish that?
If the 28 meters is right, I think it's easy to show that there will
be lots and lots of photons coming back. I don't have the time firhgt
now to do the calculation.

It's not clear how the 0.075 microradian figure is derived (for
example, whether that is a property of the laser or the telescope
system), or perhaps whether it is a typo. The newer site logs list 0
to 20 arcsec as the divergence, which is about 0 to 96 microrad.
Perhaps the older document was meant to say 75 microrad (in any case,
it is adjustable).

I note that you ignored these facts.


(The time window is of order 400 ns. >While this
is small, the window still allows ~0.1 km of variation in the lunar
range, which is huge. The wavelength window is also small because
they are using a monochromatic laser).

What is a time window? 400 ns is 2000 times their 200 ps pulse width.

The time tagging receiver has an active time interval of 400 nsec for
each pulse. I.e. for a 10 Hz pulse repetition rate, the receiver is
active for 4 usec out of each second, hence the background level is
reduced by a factor 250,000. Furthermore, the receive aperture has a
selectable field of view which can reduce background.

The point is that the background is extremely small -- smaller than
one photon per 400 nsec time interval -- so the detection of a given
photon is very likely to be a "return" photon.

I note your lack of response.


(d) The experimenters do *not* use a single photon arrival time, but
rather the average of many arrival times, since a *train* of pulses
is sent up and then received. This improves the precision.

The article implied that the return signal strength approximated 1
photon every few seconds ("not that time, not that time, oh there's
one"). That is probably nonsense.

What you consider to be nonsense is not really relevant. The return
photons must arrive in a very tight time window, very tight angular
aperture, and very tight wavelength band. All of these cuts serve to
reduce background to such a degree that a very large fraction of
received photons are real ranging photons.

I note your continued lack of response.


It would be hard to test for the presence of the Pioneer effect with
such a poor signal. Using a Doppler radar signal would seem to be a
better idea, because the range to the moon is only about 1/40,000th of
the distance to the Pioneer at max range and the team was able to
detect minute changes in the Doppler velocity. (Of course, Pioneer
re-transmitted its signal at 7 watts).

and

(e) The experiment is lunar *ranging*, not doppler velocity
measurements.

But the significant finding is the detection of radial velocity,
namely 3.8cm/yr. With Doppler and phaselocking you can "sit on the
signal" for hours and get astonishing accuracies typical of the
Pioneer venture.

By all means, proceed with your experiment!


Doppler is accurate to perhaps 0.01 mm/s at its
best, but this is not nearly accurate enough to detect the signals
of ~0.1 mm *PER YEAR* the experimenters are looking for.

The Pioneer team were able to detect the minute acceleration, Ap, of
8e-10mss, which, multiplied by only 1.38 seconds would equal the
target 3.8 cm/yr. From that standpoint, Doppler is grossly
overqualified.

The Doppler method is sensitive to velocity and velocity *change*. It
took approximately a decade of data to detect and confirm the velocity
change of the Pioneer effect (not 1.38 seconds). In practice, I
believe the doppler technique can detect absolute line of sight
velocities of ~0.1 mm/s at best, which is woefully inaccurate for the
lunar ranging effects.


For that matter there aren't any radio re-transponders on the moon's surface
for this purpose.
You don't need retroreflectors for radar; the large area arrays can
(and did) handle small signals for Pioneer.

It's unclear where such arrays would be (do you mean the Earth? the
Moon?) Transporting anything large to the Moon is very difficult and
very expensive. Transporting a retroreflector is only difficult and
expensive. In any case, no such arrays or retransponders exist on the
moon today so your supposition is rather moot.

There's enough arithmetic to keep everybody busy, but I think I've
done my share and have highlighted some monstrous claims that must
have been confected by the authors.

Your description of "monstrous claims" are unsubstantiated, as
outlined above.


The Pioneer measurements do not solve "range" but rather a
trajectory through the solar system which is consistent with the
doppler velocity measurements.
Oh, come now! Take it easy. I'm just learning.

I find that hard to believe. You've been taking whacks at the Pioneer
"anomaly" for several years.
You haven't heard my latest.

OK!

CM
.

Relevant Pages