[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
remarks to the proposal
Dear PrimEx Friends,
I went through the proposal and have as was maybe expected a lot of remarks.
My mail might sound very negative, but I think it
is better to get main issues out of the way before the submission to a
PAC. I think several basic points are missing and need for my feeling be
addressed before the proposal can be presented to the PAC in January.
Before I list specific remarks to each section let me make some global
remarks on what is missing in the proposal.
It is argued very strongly that the FCAL energy and angular resolution is
not good enough to do the measurement.
But I miss actually the key arguments for this, because nowhere in the
proposal there is a section with minimal requirements for the experiment.
Yes good resolution is better but what would be good enough.
Let me give an example what systematics on the measurement would you get
using the FCal. For you systematics is the dominant error as you are
not statistics limited and some systematics is independent of the choice
of the calorimeter.
What makes getting a feeling for the requirements also very difficult, that
there is no section in the proposal which describes the MC simulation.
Which process have been simulated to obtain figure 44. Which
assumptions have been made in the simulation.
An other big missing part is the simulation of electromagnetic background,
like pair-production, bhabha scattering and Compton.
You want to run with the solenoid off, fine but the solenoid confines all this
background produced in the target to a very small angular range.
What is completely missing and essential, how the readout and trigger of the
experiment is planned. There is no way to integrate your Fastbus readout
in the readout of Ha Hall D. In addition I would have a problem believing
that in 10 years from now fastbus is at all an option, no spares no
support for repair and so on and so on.
And obviously one cannot write events to tape with the tagger rate of
7 10^6 x 3.7 = 26 MHz
One other topic missing is the discussion of radiation damage, maybe the
lead tungsten crystals have really zero, but it needs to be at least
mentioned.
I also think not mentioning any contribution to the base equipment is
a major concern, see separate mail by Curtis.
Here are my points section by section.
section 2 motivation:
you ask for 120 days of running, the 120 days are beam time on target,
including an efficiency for CEBAF it will be more. Lately we run 24 weeks
which is 168 days, so you ask basically for a year of running. For this
the motivation is very weak. You give general arguments but you give nowhere
what your measurement will really constrain and what the community will know
on all this in 2015 and how the proposed experiment will compare with this.
I would need more time to make more concrete remarks on
the physics and the impact of the proposed experiment on non-pertubative QCD
and chiral symmetry breaking, but I think the motivation needs
significant strengthening.
section 3:
one part of the motivation is the discrepancy between the Collider results
and the Cornell primakoff measurement it was pointed out to me that the Cornell
result was reanalyzed and the discrepancy is gone. this paper is
even listed in your proposal (17). I think it is a must in a proposal to
describe the experimental situation to date accurately. Such the plot showing
the world data needs an update.
I think it would be good to show how the distance target calorimeter influences the separation for different processes.
Point 4 on the formfactors needs citations.
section 4:
I think variables should be introduced the first time they are used this is
also true for the sections before.
You want to measure g+he^4 ---> eta+he^4, but as you don't detect
the recoiling He^4, actually a plot on the momentum of the recoil would be
good, I have a hard time to see how you ensure your reaction was not
g+he^4 ---> eta+X. i think I miss the point because there is no detailed
description of the analysis technique to extract the events.
How can you normalize to the Cornell data which don't agree with the
Collider data you don't know they are correct.
I think it would be helpful to show a comparison of the pi0 and
eta distributions ala fig 4 to support the point made.
Have any nuclear effect like measured at HERMES and described in
A. Airapetian et al, Nucl. Phys. B 780 (2007) 1-27
http://www-hermes.desy.de/notes/pub/publications/attlong.v10.0.pdf
been taken into account for the incoherent production?
Section 5
Section 5.1
I think as a minimum the www-page to the material provided for the
tagger and beamline review should be given in reference 20.
In general if appropriate citations to the work by the GlueX collaboration
should be given, example pair-spectrometer.
In general I think I would start from the requirements of the electron beam,
than the radiator and than describe the photon beam.
Absolutely nothing is described in the proposal about the electron beam
and which radiator is used to get the photons.
On page 15 top paragraph you say the broad band hodoscope can operate only at
moderate beam intensities. the photon spectrum can be tagged above 9 GeV
for 10^8 photons in the coherent peak on target, I think that is a huge rate.
Section 5.4
What can the photon monitor do the active collimator does not provide.
It is really not clear to me why it is needed in addition.
Section 5.5.
Section 5.5.1.
I mentioned already the electromagnetic background in the target. But in front
of the hycal different to the setup in Hall B will be material from the CDC and
the FDC. The exact radiation length can be provided, but keep in mind
it is different for different radii. Have you done any simulation showing
what the impact is on your veto counters due to early conversions of photons?
How will you do the calibration runs using e+e- pairs in Hall D.
Section 5.5.2
I think the hole size needs to be motivated by a background simulation.
---------------------------------------------------------
Comments from Alex:
p.47 "...we assume 7 10^7 beam photons on target"
I would specify the photon beam energy range here (is the cut off energy
set
to an eta production threshold ?). If you lower the threshold
you get much more beam photons; the shape is 1/E.
------
p 18. Pair Spectrometer
" ... 5 - 10 Hz coincidence rate..."
There are two points here:
1) 20 micron thick diamond and 2.2 microampere electron beam correspond
to high-lumi runs. However, 7.6 10^6 beam photons on a target in the
energy
range between 10.5-11.7 GeV required on page 47 correspond to low-lumi
runs
(the size of the primary collimator is 5 mm in diameter
here)
2) The pair spectrometer rate in the beam energy range of 10.5-11.7 GeV
for
high-lumi runs with a 5x10^-3 converter, and a 3.4 mm collimator is
about
25 Hz. Therefore, for low-lumi runs with 10-3 thick converter, and
5mm collimator the rate is (25 Hz /5/10 )*2 = 1 Hz. In principle such a
thin converter might not be needed to monitor the flux of incoherent
photons; 5x10^-3 thickness might be appropriate.
(updated plots of the ps rate are in
http://www.jlab.org/Hall-D/software/wiki/index.php/Pair_Spectrometer)
-------------------------------------------------------------------------
Section 6:
Section 6.2
I think a number is needed what the contribution of the 30cm long target to the
angular resolution is.
Here of course my argument from above comes in play again which resolution
do you really need.
Section 7:
Is the 0.55^o consistent with your hole size, and the caption of figure 37.
Section 8 & 9:
would need more time to comment on it in a intelligent way
Section 10:
Section 10.1
again a detailed explanation of what has been simulated is missing.
what does 7*10^7 equivalent photons / s mean. Is that all photon energy range
from 0.2 GeV to 11.7 or ?
Beam time is calculated with no efficiency for CEBAF, is this the std for
JLab?
This measurement is systematic limited, I think a table just listing
the contributions without further explanation is not enough, especially
as some numbers have not yet been addressed, like event selection and
background subtraction. Actually how have this numbers been
determined without
background simulation.
This became a very long email, sorry
I hope some of my remarks are helpful.
cheers elke
p.s. here some remarks from Boris Kopeliovich on the theory part
the theoretical part seems to me the main source of systematic errors (if
it is included at all), and I have strong doubts that the formulas are
correct.
There are actually no formulas in the proposal, mostly verbal description,
and I didn't find any reference how nuclear effects are calculated. There
is no mentioning of the Coulomb phase, etc.
Calculations for spin amplitudes, similar to what is needed for Primakoff,
were done in:
B.~Z.~Kopeliovich and T.~L.~Trueman,
%``Polarized proton nucleus scattering,''
Phys.\ Rev.\ D {\bf 64}, 034004 (2001)
[arXiv:hep-ph/0012091].
B.~Z.~Kopeliovich and A.~V.~Tarasov,
%``The Coulomb phase revisited,''
Phys.\ Lett.\ B {\bf 497}, 44 (2001)
[arXiv:hep-ph/0010062].
( `,_' )+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=
) `\ -
/ '. | +
| `, Elke-Caroline Aschenauer =
\,_ `-/ -
,&&&&&V Jefferson Lab +
,&&&&&&&&: HALL-D 12C / F381 121-A Atlantic Avenue =
,&&&&&&&&&&; Suite 8 Hampton, VA 23664 -
| |&&&&&&&;\ 12000 Jefferson Ave +
| | :_) _ Newport News, VA 23606 Tel.: 001-757-224-1216 =
| | ;--' | Mail: elke@jlab.org Mobil: 001-757-256-5224 -
'--' `-.--. | +
\_ | |---' Tel.: 001-757-269-5352 =
`-._\__/ Fax.: 001-757-269-6331 -
+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+=-+