<html>
<head>
<meta content="text/html; charset=windows-1252"
http-equiv="Content-Type">
</head>
<body bgcolor="#FFFFFF" text="#000000">
Dear Joe,<br>
<br>
I had been hoping that the second run would happen no later than
September. This is because we are headed out of the country for an
extended period on October 23rd. (To trekking east of
Kangchenjunga, no less - the world's third highest peak, at
28,146'. We won't go to the top!) As I am sure you appreciate, I
am greatly interested in participating in the second run, and thus
in having a voice in what we choose to do with the time. However,
there is realistically NO way I could participate for a full week
prior to our departure (or at absolute max, six days). Thus I have
to be counted out from October 16-17 on through our return in late
November. I'm sorry it has come to this, but, "you only go around
once" as we keep reminding ourselves.<br>
<br>
I have to say that I am less than impressed with Bluejeans. There
are at least three problems with it (not that I think there is much
you can do about them, but you should at least know). First, while
our two-person dedicated session worked perfectly in allowing me to
see what you were presenting, it NEVER works at the time of our
conference calls. Every time I try to get in, I receive, after a
very lengthy waiting time, a message that they are sorry, but they
are experiencing technical difficulties, please try again. The
second, considerably worse problem is that there is strong
moderately variable intensity background noise. It sort of sounds
like one is running a variable (but high) power hair dryer all the
time - a loud, variable, persistent whine all the time. It is loud
enough to make some fraction of the spoken word unintelligible. The
third, and worst of all, doesn't happen all the time, but is
terrible when it does, as it did this morning. The sound keeps
dropping out. I'd estimate the "up" duty factor to be no better
than 30% or so. The dropouts are of two kinds - a total loss of
information, or an enormous reduction in volume. The last 12-13
minutes of today's meeting were basically unintelligible to me,
which is why I finally gave up. I could occasionally catch few word
stretches of the conversation, and guess the remainder in the best
of circumstances, and would be totally in the dark most of the
time. I'm sure JLab likes to use "free" software, but at the same
time, they should appreciate that it comes with consequences. Have
you ever thought about "Go To Meeting"? Mary has far more
conference calls than I do, and it seems to work pretty flawlessly
for her groups, which are both bi-costal and involve a number of
participants. I believe it is still no cost software as well. Is
there any chance of switching - we certainly have enough experience
with Bluejeans from my perspective.<br>
<br>
Anyway, on to my comments. I thought that Dan's analysis of the
systematic effects was both valuable and interesting, though I had
already reached the conclusion by the end of the actual beam runs
that the systematic effects were tiny at best. I think he could
take things one step further by doing a few Chi-squared tests. For
example what is the Chi-squared probability that the four data
points at different values of the 0L02 phase are the same value
(high, I suspect)? Similarly, one could simply make a model for
the various beam positions on the target foil (by the way, I thought
we used 1 (or 2?) mm position changes, instead of one "spot
diameter", but this is easily checked) such that the asymmetry is
the central value plus a linear (or more likely quadratic) term for
left-right displacements and a different one for up-down
displacements, and then fit the data. I wouldn't spend a lot of
time on this, but I believe it would show that there is no
meaningful dependence of the asymmetry on beam position on the
target foil for any realistic beam position motion during the run.
Recall that we ran for hours, and NEVER re-steered the beam from run
to run because nothing had changed on the scale of a small fraction
of the spot size, verified by (occasional) observation. Properly
set up, CW SRF accelerators are well known for their high quality
stability. I guess all this is to say that I think a very small
amount of additional work would pretty cleanly establish that the
systematic effects he looked at are tiny compared to our
experimental error bars (which are small), and thus that we don't
really need to spend much time on systematic studies in the October
run (once we reproduce the January results with convincing
precision, though not on all foils).<br>
<br>
Is seven shifts all we can hope for in October (says he who may not
be able to attend if its the latter part of the month)? Or is it
just what you think will solve all remaining issues?? I personally
would like to see a couple different energies run, for several
reasons. Beam energy meaningfully influences the nuclear size
effect on the analyzing power, and may also affect the size of the
radiative corrections - our greatest uncertainty in my opinion.
While one might not choose to run as detailed a set of measurements
at two other energies (meaningfully higher and lower), it would be
nice to be able to compare the measured asymmetry with what we would
expect hopefully again demonstrating that the nuclear size effect
and the radiative corrections (?) can't be vastly larger than what
we presently believe. Perhaps we should ask Xavier what he things
the energy dependence of the radiative corrections might be??<br>
<br>
I also believe there is some merit in rotating the plane of
polarization by 90 degrees, and demonstrating that the up/down and
left/right asymmetries are basically statistically equal with pretty
good precision (and by inference we have built a good instrument).
Eliminating the dump dipole has great intellectual appeal to me, and
a convincing demonstration would, I believe, add to our result (and
to future use of the polarimeter for helping the halls).<br>
<br>
I have done a fair bit of playing with the information that Dan has
sent regarding his many fits, and have reached the conclusion that
the only factor that influences our measured asymmetry is the energy
cut. I think that studies to further illuminate or understand this
are the most important item we could address in October. But, what
exactly would I recommend I cannot exactly say right now. I do
believe that, were funds available like in the old days, the one
possible improvement to the polarimeter would be to go to a
considerably better scintillator (i.e. considerably improved width
of the elastic peak). I don't know if one might hope to scrounge
such scintillators from long gone experimental equipment or not -
there is certainly a lot of stuff out there. I was interested in
Marcy's fit to the asymmetry versus thickness, where she gets a
lambda of 0.316. All of Dan's many results are consistent with
lambda being 0.324. I assume that Marcy's fit is different from
Dan's but this might be worth a few minutes of someone's looking.<br>
<br>
I liked you analysis along the lines of estimating the analyzing
power of the second scattering. I had had some thoughts along these
lines, hoping to do analytical estimates, but your presentation this
morning was quite nice. To me, the difference between the Lebow and
FESEM thicknesses and their effect on the second scattering
analyzing power was very interesting - they can't both be right, and
I thought the FESEM results looked more believable, though that's
just a qualitative impression from this morning. Clearly more work
is needed here, and this might be part of the October work.<br>
<br>
Anyway, if we are to keep the same foil ladder, then the above
encompasses pretty much what I think we should focus on in October.
I personally feel that there is considerable merit in exploring the
effect of changing Z, but if I am the only
advocate................... I certainly DO NOT advocate changing
the foil ladder before October. And, personally, I have zero
interest in making 499 MHz measurements. I would personally much
rather invest in making it "quick and easy" to switch to and from 62
or 31 MHz for measurements for the halls. <br>
<br>
Maybe I'll have some other thoughts later. I gather from what you
said you will be gone for a month. Mary and I are gone in
mid-September for a bit. Among other things we are getting a
special tour of the Hanford reservation (where they made the
plutonium for the bombs). I'm sure you appreciate that I'll have a
hard question or two for our poor tour guide! I'm not sure when all
this activity morphed from a perceived national need to a clearly
criminal activity, but it certainly did that morph in my mind. Now
we sit on ~ 44 metric tons of declared excess plutonium which is
beginning to look like it will cost as much to get rid of as it cost
to make in the first place.............. The "cleanup" has been
going on for ~ 20 years now, and <u><b>nothing</b></u> is yet
cleaned up. Probably because we are only spending about $2B
annually on the effort! See how you become curmudgeonly in old
age??<br>
<br>
Anyway, enjoyed the part of the program I could hear this morning.
More later. I think Dan is doing a fine service to our effort.<br>
<br>
Best,<br>
Charlie<br>
<br>
<br>
<br>
<br>
<br>
<div class="moz-cite-prefix">On 8/12/2015 1:06 PM, Joseph Grames
wrote:<br>
</div>
<blockquote
cite="mid:36706554.6214557.1439409976386.JavaMail.zimbra@jlab.org"
type="cite">
<pre wrap="">If you haven't already then please send me your (un)availability for the period Oct. 4-25, the likely window of opportunity to perform a 2-3 day Mott Run 2.
thanks,
Joe
_______________________________________________
Mott mailing list
<a class="moz-txt-link-abbreviated" href="mailto:Mott@jlab.org">Mott@jlab.org</a>
<a class="moz-txt-link-freetext" href="https://mailman.jlab.org/mailman/listinfo/mott">https://mailman.jlab.org/mailman/listinfo/mott</a>
</pre>
</blockquote>
<br>
</body>
</html>