Difference: QuestionsST (1 vs. 14)

Revision 142011-04-22 - PeterWinter

Line: 1 to 1
 
META TOPICPARENT name="MuCapRun8"
Changed:
<
<
<--   
-->

Simple Questions

to Steve

  1. Define units used (ns, cm)?
    I tried to label units whenever stated and on histogram axes.

  2. Also show zoom at cluster size distribution in Fig 3.4, should we check different cluster size cuts? The spark cuts have an amazing effect!
    Spark cuts will be clarified in the next version (already in the working document). The explicit cut on cluster size is responsible for the large effect.

  3. Cathode or should be displayed in Fig 3.7, if that's your favorite.
  4. What don't we understand on the ePC autocorrelation plots?
  5. show 4 fold coinc autocorrelation.
    Added to next draft.

  6. Show autocorrelation of mSC, mSCA and mPC as used in the PP
  7. Show mSC correlations of mSC, mSCA and mPC as used in the PP
  8. Determine drift time as function of TPC x z (lower priority)
  9. Plot 3.24 in liner scale cutting the first bin. That's an important plot conc eDet effects.
    Added Oct. 4.

  10. mu+p scatter: x label for fig.8.1, why does sum fit give smaller lam than individual gondolas in Fig. 8.2
    There are not enough scatter events for the fit range 100 to 24000 ns, leading to a bad fit with higher lambda.

  11. what are the abs(phi1-phi2)<1 tracks in fig 9.2
    I never figured this one out conclusively, if these are secondary particles, misreconstructions or what. Data with no beam (cosmics runs) also show these coincident tracks.

  12. Can one get a good chi on fit T-coinc and not phi-opp e when applying the double counting procedure?
  13. Does the DNL phase effect of the CAEN affect the fit result? (differential non linearity)
    Added lifetime vs. rebinning phase study to next draft. The fits appear consistent with the numerical simulation I did some months ago (referred to in the Numerical Studies chapter).

  14. Why is the chi2 in Table 10.2 eTrack CathOr 1.09 and 0.97 in Tab.10.3?
    (figure labels changed slightly. The table in question is "Lifetime vs. eDetector Treatment") The difference is in the first case there is no impact parameter cut.
  15. In Table 10.5 the overall mu inefficiency should be the extrapolation axis. Is the mPC efficiency ~ overall mu inefficiency? We should play the 0.1, 1% suppression game with the overall PP input (needs discussion).
  16. Are the lifetime vs Mu SC/PC deadtime studies performed for the whole prod 50 dataset?
    Yes, except the 80 unskimmed files (out of 1403) not at NCSA.

  17. Do you have electron veto in capture search?
    Yes.

  18. Fitting of capture spectra?
    I attempted this a bit some months ago and did not have any more success than Tom at getting the literature nitrogen values from CalibN2 data.

  19. Do you have spark cuts on trailing edges like Tom?
    No, not beyond the block cut for >100 trailing edges in the shared module MCaenCompProcessRaw.
  20. Fig 3.23: Can you get better Zgond-Zepc resolution when playing with the light velocity in the gondola?
    This was not studied systematically. We could look at Zgond-Z(epc1*epc2@gond) with different velocities for a few runs.
  21. How large are the per run group fluctuations mentinoed by Tom on p. 52.
    Lifetime vs. Run Group section has been added to my report (Oct. 2).

to Tom

  1. Show autocorrelation of mSC, mSCA and mPC as used in the PP
  2. Show mSC correlations of mSC, mSCA and mPC as used in the PP
  3. Determine drift time as function of TPC x z (lower priority).
  4. I did not understand the processing of muSC router signals. Let's discuss.
  5. mPC1: where does the 200 ns mPC deadtime come from electronically? Time over threshold from Genna's electronics?
  6. page 9: ClusterInterval and MaxGap values missing
    The values can be found in Appendix B, "Berkeley-specific settings," p.78.
  7. for both Tom and Steve: how did you treat the case of 2 muPC tracks coincident with mSC?
    For me, if the two muPC1 hits are distinct they are regarded as separate muon hits, so it would be considered an incidence of pileup.
  8. how does the mPC TPC inpact parameter distribution look like (lower priority)
    See my 3/22/05 worklog entry "Spatial Matching of muPC1 Hits with TPC Tracks,"
    http://weak0.physics.berkeley.edu/weakint/research/muons/private/tbanks_dir/TeleConf/2005Mar22/2005Mar22.html
    I did not deem spatial matching to be feasible, which was later corroborated by Steve's difficulties in fitting muon tracks to straight lines.

  9. Discuss and potentially update Muon Detector Inefficiency note?
    At some point I would like to update the note's quotes for the individual detector inefficiencies, but otherwise the note's conclusions remain valid.
  10. p.20: Steve gets an experimental "light velocity" of 67mm/ns.
    That seems plausible: the gondolas are 900 mm long, and the width of the upstream/downstream time differences in my Figure 12 is about 16 ns. This gives a velocity of v ~ 900mm/16ns ~ 56 mm/ns. I still do consider this resolution is good enough to allow robust eSC z-tracking, espcially given concerns about interpolator nonuniformity.
  11. p. 21: what roughly is the fraction of combinatorical double counting at this stage? Where do you get the "actual efficiencies" from?
    The "actual efficiencies" are calculated from the reduced set of eSC hits: e.g. (ePC1+eSC)_red/eSC. Working on fraction of double-counting...
  12. Define local spark condition, bookended?
    If I cut all data around sparks, I must bookend muSC hits from the edges of those cuts, just as I have to bookend from the edges of the data blocks.
  13. Did we perform an independent COMP analysis, so as to get independent of CAEN errors?
    I did not, although I believe that Steve did. We do, however, compare CAEN and COMP muSC hits in the shared muon analysis modules, and the software cuts the data block if there are too many discrepancies.
  14. Continue Fig 25 above 80 keV, to see whether say 100 keV is excluded. Compare to mu+, because SRIM actually calculates mu+. Efficiency seems to be between 100 and 50%. Needs discussion.
  15. Reference to stability graph of TDC400 drift interval. "sometimes incorrectly", too vague, database should indicate incorrect settings.
    The known time periods when the TPC voltages were improperly set are noted in Francoise's modified shift summary and/or memo on run categorization. My own investigation into the TPC drift stability can be found at
    http://weak0.physics.berkeley.edu/weakint/research/muons/private/tbanks_dir/TeleConf/2005July5/2005July5.html
  16. p.32: z distribution somewhat broader because of larger pitch is not correct, rather due to wider induced cathode distribution. Table with geometry offsets. Monte Carlo results showing the impact par distribution would be nice.
    It is perhaps true that electron hits induce wider clusters in the cathodes than in the anodes 103104 ; nonetheless, the ePC z-resolution is inherently coarser than in x and y because the anodes are more closely spaced than cathode separations. Most of my geometry offsets are available in Appendix B, "Berkeley-specific settings," p.78.
  17. Magnet effect: Bending of muon track for completeness. e and mu bending calculated in MC?
    Only the electron track's curvature is addressed in my MC. Muon track curvature is not really of importance, because we only look at the muon's stopping point, not its trajectory, and therefore do not perform any straight-line extrapolations (see infeasibility of muPC1/TPC track matching in #8 above).
  18. The impurity capture search as well as the properties of the impurity capture events is only sparsely described in 5.7. More information or reference would be very helpful as this is a key UCB analysis contribution.
  19. p.41. Negligible effect on rate, improves chi**2. Quantify.
    See the end of the preceding paragraph.
  20. 6.3 Rebinning: I understand you used rebinning of time differences? Tom and Steve should document their binning/rebiining methods in a picture.
    That is correct: I rebin the T(eSC)-T(muSC) time differences in the lifetime histogram. This seemed to be the least worst approach. I am not sure how to illustrate the rebinning procedure beyond the prose description on p.42.
  21. Where is the e-vector mu-stop pairing described? Are there combinatorical ambiguities remaining, which fraction. What is their lifetime?
    The mustop/e-vector pairing is described in section 5.5 on the impact parameter (p.31). The actual joining of mustop/e-vector pairs is a rather trivial operation, as one looks for temporal coincidence in the interval +- 40 us. I have not looked at the fraction or lifetime of ambiguous matches.
  22. p.44 Where are the results for the different combination of muPC AND/OR, ePC AND/OR.
    http://weak0.physics.berkeley.edu/weakint/research/muons/private/tbanks_dir/TeleConf/2006Jun27/2006Jun27_updated.html
  23. mu diffusion and gondola effect. Include reference to MC simulation and elimination during extrapolation.
    MC simulation -- http://weak0.physics.berkeley.edu/weakint/research/muons/private/tbanks_dir/TeleConf/2006Feb21/2006Feb21.html
    elimination during extrap -- http://weak0.physics.berkeley.edu/weakint/research/muons/private/tbanks_dir/TeleConf/2006Jun6/2006Jun6.html
  24. fig.44 studies are interesting. Event display picture a'la Steve would be helpful to understand the definitions and their variations.
  25. Confirm again cause of mSC problem periods.
  26. Explain Fig. 47 based on mu detector note. Is few x 10-4 safe?
  27. Do we understand gondola distribution Fig.49. Start time fit of this data?
  28. 7.1 Muon scatters. Important section. Errors need to be discussed. Tom uses >=4 pixel, Steve >=8. Sensitivity to gondola effect?
  29. p.60. Variety, vague... Either summarize or refer to impurity report.
    See Bernhard's assessments of the impurity situations for Runs 8, 9, 10:
    http://weak0.physics.berkeley.edu/weakint/research/muons/private/bernhard_analysis.html
  30. p.61 How clean is impurity curve for EHEH in prod-50 data. i.e. BG subtraction.
    This should be qualitatively apparent by comparing the left- and right-hand plots in Figure 50.
  31. p.63. How can we prove that eps^EH_O~eps_N ?

Advanced Questions

  1. Which studies of lambda versus ePC cluster parameter values were performed?
    Do we need more. Cluster size cut, incomplete gondola's?
  2. to Steve: Your ePC1xePC2xeSC pairing difference is a critical difference to the UCB analysis. What are the properties of the multiple ways pairing tracks. What is their fraction, what is their lifetime etc. Could there properties be checked with the MC data?
  3. Justify that extra pixel>8 are ok, compared to > n, where n=1-10. Possibilities are
    lifetime vs pixel cut
    gondola effect vs pixel cut
    delta_y vs pixel
    pixel impact parameter
  4. Bernhard, please summarize in your talk what MC has to say about the gondola effect. In particular, can it reproduce Steve's fig. 7.4?
  5. Do we have a global consistent view of the PP inefficiency over the run?
  6. How to chose lifetime and error from the different possibilities?
  7. Are the results of Steve's lifetime vs MuSC /PC deadtime studies consistend with Tom's MuDet note ( http://ten.npl.uiuc.edu:8085/MuCap+Notes/39).
  8. Wall stops. If we have an electron distribution of n(t)= A (exp(-r0*t) + b exp(-r t), the first order correction to the lifetime is del_r0/r0 = b (r-r0)/r0. Thus if (r-r0)/r0 is order 1, we need to know b (wall stop fraction) to 20 ppm. Do we have any direct constraints on that? What is the effect of double muons within deadtime? Other crazy mechanisms to have 20ppm wall stops.Carbon coat in the future? Any indication of returning capture products from lower Al plate or glass?
  9. We should study the effect of the TPC time resolution on the inconsistent capture fits. Can we improve by selecting single pixel capture events?
  10. Should we analyze 4.8 kV data.
  11. Estimate threshold uncertainty for capture finder between 4.8 and 5.0 data.

Future Games

  1. Let's analyze higher energy muons to get the TPC y resolution, FADC, future.
  2. Let's determine the z resolution by comparing with FADC reconstruction.
  3. Determine TPC energy resolution.
  4. Bernhard, Peter: what is the capture neutron n-p scattering signature in the TPC
  5. m+p scatters: Could Bernhard include those in the Geant and produce a distribution of where and when each muon is stopped.
  6. mu+p scatters: run 10 analysis with higher TPC gain essential
  7. How to proceed with Brendan's diffusion MC?
  8. MC (Bernhard, Brendan) could corroborate the TPC shell dependence of the lifetime.
  9. Does an accidental fit to the negative times gives consistent results?
  10. Design a MC challenge for Steve mp diffusion reconstruction.
>
>
Please go to https://muon.npl.washington.edu/twiki/bin/view/Main/QuestionsST
 \ No newline at end of file

Revision 132006-10-04 - StevenClayton

Line: 1 to 1
 
META TOPICPARENT name="MuCapRun8"

Revision 122006-10-04 - TomBanks

Line: 1 to 1
 
META TOPICPARENT name="MuCapRun8"

Revision 112006-10-04 - StevenClayton

Line: 1 to 1
 
META TOPICPARENT name="MuCapRun8"

Revision 102006-10-03 - TomBanks

Line: 1 to 1
 
META TOPICPARENT name="MuCapRun8"

Revision 92006-10-03 - TomBanks

Line: 1 to 1
 
META TOPICPARENT name="MuCapRun8"

Revision 82006-10-02 - PeterKammel

Line: 1 to 1
 
META TOPICPARENT name="MuCapRun8"

Revision 72006-10-02 - StevenClayton

Line: 1 to 1
 
META TOPICPARENT name="MuCapRun8"

Revision 62006-09-29 - StevenClayton

Line: 1 to 1
 
META TOPICPARENT name="MuCapRun8"

Revision 52006-09-29 - PeterKammel

Line: 1 to 1
 
META TOPICPARENT name="MuCapRun8"

Revision 42006-09-28 - PeterKammel

Line: 1 to 1
 
META TOPICPARENT name="MuCapRun8"

Revision 32006-09-28 - StevenClayton

Line: 1 to 1
 
META TOPICPARENT name="MuCapRun8"

Revision 22006-09-28 - PeterKammel

Line: 1 to 1
 
META TOPICPARENT name="MuCapRun8"

Revision 12006-09-28 - PeterKammel

Line: 1 to 1
Added:
>
>
META TOPICPARENT name="MuCapRun8"
<--   
-->

Simple Questions

to Steve

  1. Define units used (ns, cm)?
  2. Also show zoom at cluster size distribution in Fig 3.4, should we check different cluster size cuts? The spark cuts have an amazing effect!
  3. Cathode or should be displayed in Fig 3.7, if that's your favorite.
  4. What don't we understand on the ePC autocorrelation plots?
  5. show 4 fold coinc autocorrelation.
  6. Show autocorrelation of mSC, mSCA and mPC as used in the PP
  7. Show mSC correlations of mSC, mSCA and mPC as used in the PP
  8. Determine drift time as function of TPC x z.

to Tom

  1. Show autocorrelation of mSC, mSCA and mPC as used in the PP
  2. Show mSC correlations of mSC, mSCA and mPC as used in the PP
  3. Determine drift time as function of TPC x z.

Advanced Questions

  1. Which studies of lambda versus ePC cluster parameter values were performed?
    Do we need more. Cluster size cut, incomplete gondola's?
  2. to Steve: Your ePC1xePC2xeSC pairing difference is a critical difference to the UCB analysis. What are the properties of the multiple ways pairing tracks. What is their fraction, what is their lifeitme etc. Could there properties be checked with the MC data?
  3. Justify that extra pixel>8 are ok, compared to > n, where n=1-10. Possibilities are
    lifetime vs pixel cut
    gondola effect vs pixel cut
    delta_y vs pixel
    pixel impact parameter
  4. Do we have a global consistent view of the PP inefficiency over the run?

Future Games

  1. Let's analyze higher energy muons to get the TPC y resolution, FADC, future.
  2. Let's determine the z resolution by comparing with FADC reconstruction.
  3. Determine TDC energy resolution.
 
This site is powered by the TWiki collaboration platform Powered by PerlCopyright © 2008-2019 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback