Measurements on bare ASIC and full detector. M.Borri STFC Tests - - PowerPoint PPT Presentation

Measurements on bare ASIC and full detector.

M.Borri STFC

◮

Tests on bare asics v2.

◮

Tests on full detector.

Characterization overview.

Test on bare asic:

◮

Standard tests are completed on second board using asic v2: e-mode and h-mode.

◮

e-mode tests presented on 21 July 2014.

◮

h-mode tests present today, see my other set of slides.

◮

Board 2 did not show any particular signal of fault.

◮

It also never locked up (improvement w.r.t. board1). Tests on full detector:

◮

Detector received at DL on Thursday.

◮

Detector is connected and sits in air (lab.T9).

◮

Electrical tests are ongoing: IV and noise vs channel.

◮

Memory of DAQ allows for test of 1 side only: n side (or even-side).

◮

Detector is working with E-th at 0xe0. Other:

◮

New test has been implemented.

◮

Measures time-stamp (TS) jitter as a function of the threshold of TS discriminator.

◮

Need practice to possible optimization. Testing Sensor electrical test I-V measurement Front-end tuning Threshold Shaper Calibration Front-end electrical test: Noise and noisy channels Cross-talk Timewalk and jitter Detector performance Charge collection efficiency Signal-over-noise ASIC CMOS process AMS 0.35 µm Power consumption < 1.5 W/ASIC Channels per chip 128 Data rate < 5 kHz/channel Energy range 0-50 MeV Time stamp 100 MHz Sensor Type double sided Bulk doping n-type Bulk thickness 300 µm Strip pitch 50 µm Strip width 38 µm Strip stereo angle 16.2o Leakage current per strip 3.25 nA/cm (<100 nA) Strip capacitance 2.3 pF/cm (<80 pF)

Full-detector tests: IV

◮ IV different from that measured at Liverpool. ◮ It shows an increased ohmic behavior. ◮ Temperature effect: 30deg (from pt100), 43deg (from Ileak). ◮ If temperature effect then pt100 is not measuring the right

temperature.

◮ Other possibility would be mechanical damaging of sensor. ◮ No thermal run away.

[V] 10 20 30 40 50 60 70 80 90 A] µ [ 5 10 15 20 25 30 35 40 45 IV characteristic CHs OFF CHs ON IV characteristic [min] 10 20 30 40 50 60 70 80 90 A] µ [ 40.2 40.3 40.4 40.5 40.6 40.7 40.8 40.9 41 Ileak vs time Ileak vs time

Full-detector tests: noise vs strip length.

◮

Input capacitance ∝ strip length.

◮

I leak ∝ strip length.

◮

Noise increase for longer strips.

◮

Increasing shaping time makes the noise lower and homogenus across the channels. N.B. chain 1 (n side) is not working.

Channel[a.u.] 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 Std.Dev.[ADC] 1 2 3 4 5 6 7 8 9 10

Noise Vs StripLenght (Side0, PeakTime1.0us)

Channel[a.u.] 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 Std.Dev.[ADC] 1 2 3 4 5 6 7 8 9 10

Noise Vs StripLenght (Side0, PeakTime4.0us)

Channel[a.u.] 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 Std.Dev.[ADC] 1 2 3 4 5 6 7 8 9 10

Noise Vs StripLenght (Side0, PeakTime7.5us)

New test: time-stamp jitter.

TS-th [DN] 130 135 140 145 150 Mean [10ns] 800 850 900 950 1000 1050 1100 1150 1200

Mean Mean

TS-th [DN] 130 135 140 145 150 Std.Dev. [10ns] 100 200 300 400 500

Std.Dev. Std.Dev.

◮

GOAL: measure the time-stamp jitter.

◮

MEASUREMENTS: Loops over all the channels. Only one channel is powered up at each time. For this channel, loop over the time-stamp threshold. Inject N pulses for each time-stamp threshold. Each time calculates pulse period: time stamp new - time stamp old . N.B. At the moment, the mean is not the value of the real period. It requires to add an offset. TEST OUTCOME: Std.Dev. decreses by increasing the threshold: expected. As a default value we use 144dec (0x90).