In Straw Tracker Prototype Tom-Erik Haugen, David Brown, Richard - - PowerPoint PPT Presentation

Fe55 Longitudinal Measurements In Straw Tracker Prototype

Tom-Erik Haugen, David Brown, Richard Bonventre, Andrew Edmonds.

Straw Tracker

• 5mm straw made of

15µm thick aluminum

• Filled with ArCO2 and

held at high voltage

2

Straw Tracker

• 5mm straw made of

15µm thick aluminum

• Filled with ArCO2 and

held at high voltage

• Signal amplified in gas

3 𝑓−

+ + + + + + + + +

Straw Tracker

• 5mm straw made of

15µm thick aluminum

• Filled with ArCO2 and

held at high voltage

• Signal amplified in gas
• Signal sent through

integrating amplifiers then to TDC and ADC

• End result of gain and

time difference

4 Earlier signal propagation Later signal propagation

Prototype Setup

• 8 straw prototype with a preamp
• n each side
• Straws 0 and 1 preamps

don’t work

• Particle crosses straw, signal

travels down to each end of straw

• Time difference from each

end gives longitudinal position of track through straw 5

Longitudinal Measurements

Measuring along z-axis (z = 0 at center of straw). Measure signal from Fe55 source with straws at 1250 Volts Signal size using Fe55 at 1250V is the same as MIPs at 1425V Measurements along z axis should show attenuation and resolution effects

6

Fe55 Signal Peak Minus Pedestal

Signal is recorded once it passes

• threshold. Both ends of straw are

summed It is recorded in 16 samples, the first 4 are presamples. Peak of signal minus pedestal (average

• f presamples) forms this characteristic

shape.

7

Means from Gaussian fit

Plotting the mean of the gaussian shows outliers on all straws except for straw 3 and 7

8

Asymmetric Signal

Measuring peak minus pedestal from

• nly +z side (HV side) and -z side (cal

side) of straw. This is physical and expected that signal is larger when measured closer to event Using the difference of the means and the sum of the means we can calculate the attenuation

9

Attenuation

Subtracting the mean value from the positive z side from the negative z side The sum of signals shows a clear dip at z = 0 is potentially from straw aging. Sum is expected to be independent of z to first

• rder.

Straw 3 and Straw 7 show very clear linear difference, straw 5 shows outlier at 87.5

10 Sum of signals Difference of signals

Attenuation Length

Modeling Attenuation as an exponential Dividing the sum of both ends divided by the slope of the difference gives the attenuation length. Straw 3: 4.84 ± 0.20 meters Straw 5: 4.97 ± 0.21 meters Straw 7: 4.33 ± 0.14 meters

11

𝐵𝑑𝑏𝑚 𝑨 = 𝐵0𝑕𝑑𝑏𝑚𝑓−

𝑀 2+𝑨 𝜇

𝐵𝐼𝑊 𝑨 = 𝐵0𝑕𝐼𝑊𝑓−

𝑀 2−𝑨 𝜇

𝜇 = Σ𝐵 𝜖 𝜖𝑨 Δ𝐵

One straw deltaT

Plot of the time difference between both ends of the straw. Taking the gaussian fit provides a mean and sigma

12

One straw deltaT

Plot of the time difference between both ends of the straw. Taking the gaussian fit provides a mean and sigma Straw 4 showed much worse resolution than all of the other straws

13 13

Gaussian Mean

Plotting the z position versus the gaussian mean acts linearly Slope of line is half of effective propagation velocity Data shown was taken at 1250V veff = 193.64 ± 39.24 mm/ns

• r 0.65c

with 6.5% RMS variation between straws

14

Effective velocity

Running the same analysis at other voltages shows very different propagation velocities (variations up to 50%).

15

veff= 0.43c veff= 0.67c veff= 0.76c

Signal Spread + Threshold Reduces Veff

16 HV Threshold Cal Threshold

Straw

Original current pulse Broadened pulse

55Fe source

Vprop

TThresh

Veff = DZ DZ Vprop +Tthresh <Vprop

Changes in Velocity

Effective velocity versus gain shows a clear drop off effect. Signal is expected between 400 and 500 peak - pedestal counts, in the region where variation is small.

17 17 17

Z resolution

Multiplying deltaT sigma by the velocity gives the longitudinal position resolution This does not accommodate asymmetric tails or any effect of slewing

18

Resolution vs Voltage

Plots of the core resolution: Black = 1425V Cyan = 1250V Red = 1200V Blue = 1150V All straws except straw 4 show better resolution at higher voltage We have not had time to study these variations in detail

Backup Slides

20

Tracker Design

21 Individual straw Prototype panel of 98 straws 6 panels to a plane 18 stations in tracker

Sum of both signals and difference of signals

22

5 Positions Along z

Measurement of peak minus pedestal at 5 positions along straw. Small variations in straw 3 are expected due to attenuation Straw 5 shows a clear outlier at Z = 87.5mm

23

Mapping out Straw 5 at z = 87.5mm

Measuring with 10 mm intervals around 85mm Anomalous region has length

• f about 40mm

24

Reducing Source Rate at z = 85mm

Moving the Fe55 source further away to reduce the rate from 187 kHz to 1.1kHz Peak minus pedestal returns to normal shape Rate dependent gain loss is consistent with charge accumulation inside the straw

25

187 51 21 11 3.5 1.1 rate (kHz)

G4 charge accumulation

Using Fe55 at 50kHz rate gives a current of 0.18 μA/cm Blue histogram shows full range of currents expected from backgrounds in Mu2e experiment 38% of straw-cms have large enough currents to produce significant gain losses

26

Andrew Edmonds:

Loss of Metalization

Second batch of straws produced showed loss of metalization. According to Bob Wagner the straws used in the LBL prototype were from batch 2 Batch 3 did not show this problem of metalization

27

inner liner from PPG batch 2 straw showing traces of metal removed from the straw inner wall (photo courtesy of Bob Wagner)

Calculating velocity

28 −𝑨𝑓𝑜𝑒 +𝑨𝑓𝑜𝑒 𝑨 = 0 𝑨

Δt = t1 − t2 Δt = 𝑒1 𝑤𝑓𝑔𝑔 − 𝑒2 𝑤𝑓𝑔𝑔 Δ𝑢 = 𝑨𝑓𝑜𝑒 − 𝑨 𝑤𝑓𝑔𝑔 − 𝑨 − 𝑨𝑓𝑜𝑒 𝑤𝑓𝑔𝑔 𝑤𝑓𝑔𝑔 = 2 ∗ 𝑨 Δ𝑢

𝑒1 𝑒2