MICROPHONES AT HRM
Post on 10-Feb-2016
37 Views
Preview:
DESCRIPTION
Transcript
Results of microphone tests in HiRadMat
Collimation WG Meeting15.10.2012
Daniel Deboyon behalf of the Collimation team,
thanks to A. Masi, J. Lendaro, C. Derrez and teamI. Efthymiopoulos and HRM team
1Collimation WG Meeting 15.10.2012
MICROPHONES AT HRM
Purpose of Sound Analysis• Can we (roughly) localize Impacts with
correlation measures between two or more microphone signals?
• Sound Pressure Level -> Amplitude of pressure wave
• Spectral Components -> Damage/ no damage?
• Investigate limitations of the system (EM noise, R2E)
Application: Impact detection and localization at LHC collimators!
2Collimation WG Meeting 15.10.2012
Microphone Positions
3
ca 6.5 m
ca 13.5 mca 13.5 m
ca 25 m
DownstreamUpstream
Collimation WG Meeting 15.10.2012
4
Localization with TDOA
∆T3
Mic 3 Mic 2 Mic 1
∆T2 ∆T1
Beam impact position
Beam
Speed of sound cs= 343 m/s (@20° C)Reference location sref = ∆T2 * cs = - ∆T1 * cs
Estimated location simp = sref + (∆T2-∆T1) * cs
Reference location
Collimation WG Meeting 15.10.2012
5
Microphone Signals
Typical radiation induced noise spikes during impact
Peak amplitude linear dependent on beam intensity-> saturation limit at signal amplitude equal to a sound pressure level of approximately 163 dB SPL (ca 2700Pa = 3V with 10mV/Pa output sensitivity)
Collimation WG Meeting 15.10.2012
6
Noise from Radiation vs. Intensity
Mic Downstream Measurements taken during HRMT 12 experimentExtrapolated and scaled assuming 1/r2 mitigation-> Estimate of distances needed to avoid signal overload
Upper limit of Mic sensor (eq. 163dB SPL)
Collimation WG Meeting 15.10.2012
Test 1 Mic Signals
7
ca 114 ms Sensor Saturation
Blue: 13.5m DownstreamGreen: 13.5m UpstreamRed: 24 m Upstream
Sou
nd P
ress
ure
(Pa)
High Intensity Shot – Spike from Radiation
Time (s)
Test 1:1 Nominal LHC Bunch @7TeV
Equ. 3.2E12 p @440GeV at HiRadMat
Shot on Collimator Jaw
Collimation WG Meeting 15.10.2012
Test 1 Sound
8
RMS Noise ca 70 dB
RMS Noise ca 60 dB
Delay ca 44 ms = 15 m
Delay ca 84 ms = 28 m
Peak Value ca 100 dB
Peak Value ca 90 dB
Delay ca 114 ms (sound lost due to sensor saturation)
Sou
nd P
ress
ure
(Pa)
Sound after High-Pass Filtering (Butter 3rd, 100Hz)
Downstream
Upstream
PatchRack
Time (s)
Amp Output Sensitivity: 3.16mV/Pa
Amp Output Sensitivity: 31.6mV/Pa
Collimation WG Meeting 15.10.2012
Test 1 Peak Estimate
9
Artifacts from Radiation Spike are cut for estimation of peak value.
Peak value of cut signal:Ca 93 dB SPL
Energy normalized to peak value!
RMS Noisefloor:Reached after 0.7s
Estimated Energy at expected onset (42ms):+4.6 dB
Estimated Peak SPL: 97.6 dB
Collimation WG Meeting 15.10.2012
10
Sound Pressure vs. Intensity
Collimation WG Meeting 15.10.2012
Test 2:No Damage
Test 1:Damage expected
Test 3:Severe Damageexpected
11
Algorithm: Localization and Damage
HighPassFilter
Peak Detection
CalculateDelay Time
CalculateMagnitude
HighPassFilter
Peak Detection
CalculateDelay Time
CalculateMagnitude
Estimate Location(TDOA)
EstimateImpact Strength
EstimateDamage Level
Mic 1
Mic 2
Reference Data
Collimation WG Meeting 15.10.2012
12
Remarks• Localization of impacts and damage estimation is
possible using only two microphones for a set of collimators!
• Radiation induced noise spike can be used as “trigger” signal, however…
• …Signal to Noise Ratio is bad due to radiation spike (long distances, low gain)– Improvement: Analogue HP Filter (allows higher gain factor)– Prototype Test of optical microphone solution (-> Balthasar Fischer
- TU Vienna)• Next steps:
– Analyze Sound Pressure vs. Distance– Find robust methods to determine Time difference of arrival
(TDOA)– Derive estimate for energy of pressure wave at outer shell of
collimator– Visual Inspection of HRM09 damage for reference classification
(2013)
Collimation WG Meeting 15.10.2012
top related