Alberto Lobo

1
DDS Data Analysis

• Alberto Lobo
• ICE-CSIC IEEC

2
DDS Data Management Diagnostics Subsystem

• Diagnostics items
• Purpose
• Noise split up
• Noise sources for LISA
• spot route to required
• sensitivity
• Sensors for
• Temperature
• Magnetic fields
• Charged particles
• Calibration
• Heaters
• Induction coils

• DMU
• Purpose
• LTP computer
• Hardware
• Power Distribution Unit (PDU)
• Data Acquisition Unit (DAU)
• Data Processing Unit (DPU)
• Software
• Boot SW
• Application SW
• Diagnostics
• Phase-meter
• Interfaces

3
Noise reduction philosophy
Problem to assess the contribution of a given
perturbation to the total noise force fint.
4
Various diagnostics items

• Temperature and temperature gradients
• Sensors thermometers at suitable locations
• Control heaters at suitable locations
• Magnetic fields and magnetic field gradients
• Sensors magnetometers at suitable locations
• Control induction coils at suitable locations
• Charged particle showers (mostly protons)
• Sensors Radiation Monitor
• Control non-existent

5
General scheme for DDS DA
(S2-IEC-TN-3031)

• For each diagnostic
• Measurement runs
• Controlled disturbance ON (if applicable)
• Controlled disturbance OFF
• Available data (in each case)
• Data Analysis Procedures

6
Thermal
22 NTC temperature sensors
16 heaters
7
Thermal
8
Thermal
Optical Window
9
Thermal
Optical Window Heaters
10
Thermal
Optical Bench Temperature Sensors
11
Thermal
Suspension Struts Heaters and Sensors
12
EH heaters activation scheme
Sensors response (CGS SW tool)
13
Heaters ON EH
Measurements Temperatures T1, T2, T3, T4 per
IS Accelerations a1,
a2 per IS Laser
Metrology x1, D
Main thermal signal DT (T1T3) - (T2T4)
per IS
Data Analysis fit data to
Transfer function temperature-acceleration ensues
14
Heaters ON OW
Measurements Temperatures T5, T6 in IS1, T11,
T12 in IS2 Laser
Metrology x1 for IS1, x2 x1 D for IS2
Thermal signals temperature closest to
activated heater
Data Analysis fit data to ARMA(2,1)

• Should be OK in MBW even beyond!, and for
  each OW
• Can easily be improved, if necessary, at lower
  frequencies

15
Heaters ON suspension struts
Measurements Temperatures Ti, i 1,...,6 all
struts Laser Metrology
x1 and x2 x1 D for each case
Thermal signals temperature closest to
activated heater
Data Analysis Transfer function is a 6x2
matrix Estimated by
standard methods
Cross correlations likely to show up (?)
Current shortage of
experimental data ?
? no sound a priori model available
16
All heaters OFF

• Temperature measurements to be translated into
  LTP
• signals (TM accelerations and/or laser
  metrology phase shifts)
• by transfer function scaling.
• Cross correlations between different channels
• Some can be (safely) discarded, e.g. OW-EH,
  etc.
• Others cannot, e.g., among different struts
• Global LTP system identification
• Some sensor readings used as housekeeping data,
  e.g., OB
• and redundant OW sensors
• Improved experimental characterisation needed
  and underway

17
Magnetic disturbances in the LTP

• Magnetic noise is due to various causes
• Random fluctuations of magnetic field and its
  gradient
• DC values of magnetic field and its gradient
• Remnant magnetic moment of TM and its
  fluctuations
• Residual high frequency magnetic fields

18
LCA
19
Magnetometer available areas
20
Magnetometers accommodation
21
(No Transcript)
22
Magnetic diagnostics coils ON

• B0 is calculated rather than measured with
  magnetometers
• Bbg is LTP background magnetic field

23
Magnetic diagnostics coils ON

• Data
• Laser Metrology x1 and x2 x1 D for each
  VE being affected
• a1 (a2) from IS1 (IS2) if possible
• Coil feed intensity and frequency

24
Magnetic diagnostics coils ON
From Fx,2w we can estimate c to 1 From Fy,2w
and Fz,2w we get error correction and cross check

• Fw can be useful to estimate remnant
  magnetisation M
• This is more complicated, though
• Fx,w has (max) SNR 100, but Fy,w and Fz,w
  quite less
• Yet all three components are needed, as M is a
  vector
• In addition, M needs to be disentangled from Bbg

25
Continuous magnetic field monitor
Data 4 3-axis magnetometers at fixed
positions in LCA 12 sampled
magnetic field channels

• Magnetic field and gradient must be known at
  TM locations
• Magnetometer data streams are fed to suitable
  extrapolation algorithms
• These algorithms are (so far) computationally
  demanding
• To be run offline
• They produce a magnetic field gradient map
  around TMs
• Magnetic map error estimates will be delivered
  by the algorithm, too

• Processed data directly yield magnetic transfer
  function.
• Extrapolation operation errors need tight
  control.

26
Magnetic Problem

• We need the magnetic field on the TMs region. For
  this, measuremrents provided by 4 3-axis
  magnetometers are available. There are (at
  present) 37 sources of magnetic disurbance
  identified (ASU). Magentometer information is
  thus insufficient to reconstruct the magnetic map.

• The nominal magnitudes (moduli) of the
  magnetic moments of the sources are reasonalbly
  well known moments but their directions are not.

27
(No Transcript)
28
Radiation Monitor
From S2-IEC-TN-3031 ...The radiation
monitor is primarily designed to help understand
and quantify these variable processes
modulations of CGR and fluxes of SEP by
monitoring the external particle fluxes and
allowing these to be correlated with the
test-mass charge measurements.
29
Radiation Monitor
30
Radiation Monitor
31
Radiation Monitor
32
Radiation Monitor

• Establish the charging-rate in the TMs due to
  cosmic-ray interactions. Compare with Monte Carlo
  simulations. Requires a long run with no UV lamps
  operating.
• Establish the cosmic-ray transfer function from
  the radiation monitor to the test-mass charge.
• Establish or limit the level of power spectral
  density of cosmic-ray modulations caused by solar
  activity. Provided by continuous operation of RM
  and other monitors available.
• Establish the solar-energetic particle (SEP) flux
  enhancement distributions (temporal and fluence)
  seen by the radiation monitor.
• Establish the solar-energetic particle transfer
  function from the radiation monitor to the
  test-mass charge. Done by cross-correlation of TM
  charge control data with RM (and other monitors)
  SEP data.
• Estimate the solar-energetic particle induced
  charging rate and compare with simulations.
• Demonstrate the closed loop charge control
  process and estimate its gain factor.

33
Radiation Monitor
Radiation Monitor data are formatted in a
histogram-like form. A histogram is generated and
sent (to OBC) every 614.4 sec.
34
Radiation Monitor

• Additional data required
• Test mass charges, Q1 and Q2 every 1000-10,000
  seconds to an accuracy to 104 elementary charges
  with sign.
• ULU time status including lamps on/off and
  commanded UV levels
• Inertial sensor noise power spectra
• RM calibration data channel to energy
  conversion
• RM calibration data efficiency factors for each
  spectral channel
• RM calibration data spectral resolution as
  function of energy
• Updated satellite geometry model
• Solar activity indicators

35
End of Presentation
36
Radiation Monitor
GCR SEP
37
Radiation Monitor
38
Radiation Monitor

• Data handling issues
• Front detector hits sent as flags
• Coincident events sent as energy deposed
• Electronics is able to cope with up to 5000
  c/s,
• so data compression will be eventually
  needed.
• Testing issues
• Artificially generated pulses
• Muon test
• Proton source exposition PSI, end of October