AURA - PowerPoint PPT Presentation

1 / 56
About This Presentation
Title:

AURA

Description:

4 Antennas. 1 Antenna Calibration Unit (ACU) ... Full Waveforms 4 channels x 512 bins 1 channel x 256 bins RAPCal records header ... – PowerPoint PPT presentation

Number of Views:27
Avg rating:3.0/5.0
Slides: 57
Provided by: icec
Category:
Tags: aura | channel

less

Transcript and Presenter's Notes

Title: AURA


1
AURA
In Ice Clusters 2 clusters were deployed last
year and are taking data. Is there a prospect
for Physics? What is our commitment to data
processing, analysis, calibration?
  • Future Detector
  • Design of a mini-GZK detector. Probably near or
    at surface.
  • Is it going to happen?
  • How active will UW be?
  • Simulation, Hardware, concept

Next AURA deployment 4 clusters were built for
deployment this year. They told us no. Maybe
next year. Is the hardware good enough? What
is our commitment to testing, Calibration,
deployment?
2
In Ice Cluster
  • AURA Inroduction
  • In Ice Clusters
  • Deployment geometry
  • Status
  • Data
  • External sources
  • Calibration and Hardware tests.
  • Prospects for analyses

3
AURA Radio Cluster
Counting house
surface junction box
surface junction box
  • Digital Radio Module (DRM) Electronics
  • 4 Antennas
  • 1 Antenna Calibration Unit (ACU)
  • Signal conditioning and amplification happen at
    the front end, signal is digitized and triggers
    formed in DRM
  • A cluster uses standard IceCube sphere, DOM main
    board and surface cable lines.
  • Use a DOM-MB as communication platform.
  • Advantage get a free design for power, comms
    and time stamping.

4
Waiting to be deployed
Antenna cables
Pressure vessels
Antennas
DRM
5
RF signal
  • Antennas
  • Broad band dipole antennas
  • Centered at 400 MHz
  • Front end electronics contains
  • 450 MHz Notch filter
  • 200 MHz High pass filter
  • 50dB amplifiers (20 dB in DRM)
  • LABRADOR digitizer
  • Each antenna is sampled using two 1GHz channels
    to a total of 512 samples per 256 ns (2 GSPS).

6
RF Signal
  • Nyquist Vrms (4KbTRDF)1/2
  • V3 mVolts RMS of 3-5 bins
  • Enviroment background
  • Average In Ice background up to 1 GHz
  • -86 dbm 2.5 E-9 mW
  • After 70 db amplification
  • 16 dbm ? 35mV RMS ? 30 DAC counts rms (for 2007)
  • 16 dbm ? 35mV RMS ? 60 DAC counts rms (for 2009)
  • Maximum signal
  • Dynamic Range1200 counts
  • ? 1320 mV RMS ? 15 dbm ? -55dbm 3E-6 mW
    before amps (07)
  • ? 720 mV RMS ? 10 dbm ? -60 dbm 1E-6 mW
    before amps (09)

2007 cluster mV/DAC is 1.1 2009 cluster mV/DAC
is 0.6
7
Deployment 2006-2007
Hole 57 scissors 2nd deployment, Shallow 4
Receivers, 1 Transmitters
Hole 78 rock 1st deployment, Deep 4
Receivers, 1Transmitters
Hole 47 paper 3rd Deployment, Deep 1
Transmitter
8
Triggering
Were Enough Antennas hit?
16 combinations of triggers
Band a 200-350 MHz Band b 350-500 MHz Band c
500-700 MHz Band d 600-1200 MHz
Antenna1
Antenna2
Antenna3
Antenna4
9
  • The threshold on each band on each channel is
    tuned independently.

Many triggers
Trigger rate
No triggers
Soft threshold
Strong threshold
DAC
10
Data, Offline analysis and logging
11
Data
  • Data include
  • Full Waveforms ? 4 channels x 512 bins 1
    channel x 256 bins RAPCal records header
  • Forced trigger runs
  • Real trigger runs
  • Constant DAC settings
  • DRM tunes it self based on a given rate parameter
    per freq band
  • DAC scans ? for every bands

12
Offline analysis
  • Offline analysis includes decoding the binary
    file, time ordering, Calibration, run summaries,
    FFTing, creating plots, root file and txt file
    for data analysis.
  • http//wiki.icecube.wisc.edu/index.php/Aura_Data_P
    rocess_status

13
Wave Forms
Time vs. voltage
Freq spectrum
Channel1 Channel2 Channel3 Channel4

14
DAC ScanShallow vs. Deep cluster DACs
A Shallow cluster channel
A Deep channel cluster
15
DAC scan Deep cluster
Ch 1
Ch 2
Ch 3
Ch 4
Less noise in higher freq steep slope
Band D High f
Band C
Band B
More noise in lower freq softer slope Lower DACs
Band A Low f
More noise in channels 3-4
Less noise in channels 1-2
16
The Bottom lineData taking and offline
processing
  • Data is being taken, and send to the north
  • Runs setting are being changed every few weeks.
  • Some work is needed to tune the thresholds above
    thermal/DRM noise and get stable running

17
A closer look at data
18
Run Summary
freq
time
mV
freq
mV
time
freq
mV
time
freq
time
mV
19
Solid- Real Trigger Dashed -Forced trigger
20
Where is it coming from?
t1
t2
t3
t4
21
Data Automatic monitoring
  • Rick Bruess

22
The Bottom lineA closer look at data
  • Data is being taken, and send to the north
  • We definitely Trigger on something
  • The channels below the DRMs are noisier than
    channels above it.
  • The shallow DRM is seeing more noise than the
    deep one
  • We have reasons to suspect that the In Ice DRMs
    are noise source.
  • Dave Besson working on WF filtering and timing
    analysis based on cross correlation.

23
Existing external sources
  • RICE
  • CW observed and measured by shallow cluster
  • Pulse not observed,
  • too weak.
  • Another RICE test is scheduled.
  • Other clusters ACU
  • ACU too weak Development of stronger ACU
  • Same ACU
  • Shows signal elongation (well get back to this
    point)

Xx add ray tracing xx
24
Antenna Calibration Unit
voltage
? 15 ns ?
? 40 ns ?
time
  • Single ping, or
  • Repeating pings (20Mhz or less)
  • Timing correspond to distances between antennas.
  • Transmitter probably too weak for inter-cluster
    triggering

voltage
time
25
RICE CW
  • Transmitter
  • at 500 Mhz
  • FFT of WF on
  • shallow

26
RICE CW
Linearity plot
27
Measurements we can do with a Pulser
  • Gain calibration
  • Linearity
  • Vertexing
  • Time resolution measurement
  • Ice properties

28
The bottom lineCalibration using external sources
  • RICE CW tests were great
  • RICE Pulser was not seen by AURA
  • A test is scheduled for next week.
  • Vertexing was not done on an external source
  • We need a stronger pulses source.
  • and then work on improving vertexing algorithms
    and WF cleaning.
  • Work in KU under progress to build a stronger ACU
    able to supply CW and pulses.
  • Another possibility Summer surface tests

29
Hardware studies and calibration Temperature
Clock Calibration
Nick Ballering
30
Clock correction
31
The bottom lineHardware studies and calibration
  • The digitizer chip was developed and programmed
    in Hawaii.
  • Some calibrations were implemented. Some were not
    implemented yet.
  • The hardware needs more understanding, studies
    and calibration
  • Timing
  • Temperature dependence
  • Gain calibration

32
The bottom line of all bottom linesIn ice
clusters
  • We have 2 working clusters in ice, and data is
    coming in.
  • There is a lot of work to do Data taking and
    handling, QA, calibration, WF studies,
    vertexing, timing, analyses.
  • Some studies were done, but this is not the real
    thing yet
  • Coincidence with IceTop
  • Constant Monitoring of the data, looking and
    studying interesting events
  • Strong pulser for vertexing, time resolustion and
    ice properties
  • Figuring out the time distributions of events
  • To get to the real thing we first need
  • Tune the run point
  • Understand the WF shapes
  • WF feature extraction, filtering, templating
  • It is a pity not to make use of the hardware

33
AURA
In Ice Clusters 2 clusters were deployed last
year and are taking data. Is there a prospect
for Physics? What is our commitment to data
processing, analysis, calibration?
  • Future Detector
  • Design of a mini-GZK detector. Probably near or
    at surface.
  • Is it going to happen?
  • How active will UW be?
  • Simulation, Hardware, concept

Next AURA deployment 4 clusters were built for
deployment this year. They told us no. Maybe
next year. Is the hardware good enough? What
is our commitment to testing, Calibration,
deployment?
34
  • This year deployment
  • Description
  • Main hardware issues
  • Things left to do

35
2008 clusters
  • General plan built 4 clusters
  • 3 sealed DRMs (2 with low freq channels)
  • 1 open DRM
  • Front ends, Antennas, cables

36
Changes from last year
  • Use only miteq amps
  • New Power Supply Unit for the DRM
  • New boards layouts
  • Smarter ACU
  • Low frequency channels
  • Decrease distances between clusters
  • Special EMI care when sealing the DRM

37
What was tested
  • Link to tests results wiki
  • http//wiki.icecube.wisc.edu/index.php/AURA07-08_
    Tests
  • Calibration
  • Testing
  • Vertexing
  • Concerns from last season
  • DRM noise
  • Elongation of signal
  • Better ACU

38
Gain Measurement
  • DRM 1
  • Full Cluster DRM only

Gain (db)
Freq (MHz)
Freq (MHz)
39
Vetexing
  • Cluster was spaced in the PSL production hall.
    Antennas 3 m high.

40
Is the DRM quiet
  • Screen chamber was built inside the old dfl
  • -174 dbm/Hz thermal floor translates into -108
    dbm/4Mhz.
  • DRM1 is watching DRM2

Gain corrected Average FFT Spectrum
Average FFT Spectrum
Ch 2
Ch 1
Ch 2
Ch 1
Ch 4
Ch 3
Ch 3
Ch 4
Low freq
Outside screen room
Outside screen room
Freq (4MHz)
41
DRM Response
Measured signal
Sharp Pulse in t
Frequency Response
Flat response Window 200MHz-900Mhz DRM
?

42
Tests left to do
  • Antenna response
  • EMI survey
  • More Vertexing
  • Complete Cold testing and calibration

43
The Bottom lineThis years clusters
  • The new clusters are working well
  • EMI noise needs to be carefuly measured. Tests
    using DRMs and using calibration antennas and
    scope indicates that the DRM are quite.
  • Antennas and cluster calibration in low noise
    enviroment.
  • Cold testing and calibration before deployment
  • ACU needs to be redesigned (too weak, too noisy)

44
AURA
In Ice Clusters 2 clusters were deployed last
year and are taking data. Is there a prospect
for Physics? What is our commitment to data
processing, analysis, calibration?
  • Future Detector
  • Design of a mini-GZK detector. Probably near or
    at surface.
  • Is it going to happen?
  • How active will UW be?
  • Simulation, Hardware, concept

Next AURA deployment 4 clusters were built for
deployment this year. They told us no. Maybe
next year. Is the hardware good enough? What
is our commitment to testing, Calibration,
deployment?
45
Future plans
  • Geometry
  • What depth? How many stations? Spacing?
  • Power? DAQ?
  • Full/Partial readout?

46
Simulation
  • Important for timing reconstruction.
  • There are many simulations out there we need
    one here.

47
Ray Propagation time
-10 m
-100 m
-600 m
-1400 m
48
(No Transcript)
49
Ray TracingThe bottom line
  • Simulation is crucial for vertexing and for
    geometry design of the next detector
  • Surface? Near surface? How deep? What spacing
    between stations? Between clusters?
  • We need access to good simulation tools
  • (I used my own hack for quick and dirty
    stuff-this is not a full ray tracing or god
    forbid neutrino Askaryan simulation, though it
    can be worked on and improved, I doubt that it
    worth the effort since there are more complete
    tools out there.)

50
Ray Tracing
51
(No Transcript)
52
Shallow Cluster
Channel 1 Channel 2 Channel 3 Channel 4
Channel 1 Channel 2 Channel 3 Channel 4
- Forced trigger runs RMS 20-30 counts 22-33
mV - Triggered runs RMS 90-110 100-120 mV -
More noise on Channel 4
53
Shallow cluster
Channel 1 Channel 2 Channel 3 Channel 4
Channel 1 Channel 2 Channel 3 Channel 4
  • More power in trigger runs than in forced runs
  • Channel 4 more power
  • Trigger runs show a peak near 200 MHz

54
Deep Cluster
Channel 1 Channel 2 Channel 3 Channel 4
Channel 1 Channel 2 Channel 3 Channel 4
- Forced trigger runs RMS 10-20 counts 11-22
mV - Triggered runs RMS 65-80 71-90 mV - More
noise on Channel 43
55
Deep Cluster
Channel 1 Channel 2 Channel 3 Channel 4
Channel 1 Channel 2 Channel 3 Channel 4
  • More power in trigger runs than in forced runs
  • Channel 34 more power - especially in lower
    freq
  • Trigger runs show a peak near 200 MHz

56
Suitability of IceCube environment
  • Channel and cluster trigger rates were compared
    when IceCube/AMANDA were idle and taking data.

Channel 1 Scaler rate vs. Discriminator value
IC AMANDA on AMANDA off IC AMANDA off
  • Noise from IC/AMANDA is enhanced in lower
    frequency on a given channel/band.
  • Combined trigger reject most of this noise.
  • Measurement only down to 200 MHz

band A (Lowest freq.)
band B (Low freq.)
band D (Highest freq.)
band C (High freq.)
Write a Comment
User Comments (0)
About PowerShow.com