The Leading Proton Spectrometer of ZEUS Design, Construction and - PowerPoint PPT Presentation

The Leading Proton Spectrometer of ZEUS Design, Construction and Performance Nicolo Cartiglia INFN, Italy and SCIPP (!) STD06, AbeFest, Carmel, 11 th September 2006 • Who, When and Why • Electronics • Beam • Performance 1 Nicolo Cartiglia, INFN, Turin, Italy September 11th 2006

The Leading Proton Spectrometer Group The LPS group was a collaboration of 3 Universities: Bologna, Turin and Santa Cruz for a total of more than 50 people. Santa Cruz designed and built the front-end electronics. Many people from SC took part: E. Barberis, N. Cartiglia, J. De Witt, D. Dorfan, T. Dubbs, A. Grillo, B Hubbard, W. Lockman, J. Ng, K.O’Shaughnessy, D. Pitzl, J. Rahn, B. Rowe, H.F.-W Sadrozinski, A. Seiden, E. Spencer, A. Webster, M. Wilder, R. Wichmann, D. Williams, D. ZerZion 2 Nicolo Cartiglia, INFN, Turin, Italy September 11th 2006

We built the LPS in 1989-91 Intel CPU 486: 33 MHz. 1.25 M transistor Windows 3.0 was released in 1989 Earthquake Abe was my professor of Q.M. There was a wall in Berlin 3 Nicolo Cartiglia, INFN, Turin, Italy September 11th 2006

HERA Physics The main goal of HERA is the study of electron – proton interaction both at high (DIS regime, Q 2 > 2-4 GeV 2 ) and low (Photoproduction) momentum transfer. 4 Nicolo Cartiglia, INFN, Turin, Italy September 11th 2006

The ZEUS detector The ZEUS detector is located on the HERA collider in Hamburg. It’s a general purpose detector: •Silicon vertex •Wire Chamber •Uranium calorimeter •Muon Chamber in the return yoke 5 Nicolo Cartiglia, INFN, Turin, Italy September 11th 2006

ZEUS forward detectors Forward Neutron S1-S6: 6 LPS station, 24- 90 m Calorimeter, 103 m S1-S3: horizontal bending S4-S6: vertical bending 6 Nicolo Cartiglia, INFN, Turin, Italy September 11th 2006

Forward Physics at ZEUS In some events the proton remnant is characterized by the production of a leading baryon (proton or neutron) These LBs carry a very high fraction of the incoming proton beam momentum: x L = p z /p’ z ~ 0.95-0.999 7 Nicolo Cartiglia, INFN, Turin, Italy September 11th 2006

Leading Proton Production The production of leading protons differs depending on the value of x L : fragmentation, reggeon, pomeron. Note: these names are overlapping processes without a clear distinction 8 Nicolo Cartiglia, INFN, Turin, Italy September 11th 2006

The ZEUS LPS •Six detector stations •Each station: 6 single sided Si detector planes •115 um pitch (0 degree) •115/sqrt(2) (+- 45 degree) •Inserted in Roman Pots Total: •54 detector planes •50,000 channels 9 Nicolo Cartiglia, INFN, Turin, Italy September 11th 2006

The silicon plane • 300 um tick Si wafer, p strips on n substrate • 3 manufacturing company: CANBERRA, MICRON, EURSYS • Oval cut, precision on cut- out better than 100 um • Yield ~ 85% • Capacitance ~ 1.2 pF/cm • Depletion Voltage: 30-50 V •Mounted on a 6 layers Cu-Invar support, water cooled 10 Nicolo Cartiglia, INFN, Turin, Italy September 11th 2006

The Front-end Electronics - I The LPS detector required a low-noise, low-power, radiation hard front- end electronics: it had to work near the beam with very little space available for the electronics and its cooling. It was decided to use a binary read-out to simplify the signal to noise problem and implement it using a combination of two chips: an analog amplifier comparator chip, the TEKZ, and a digital memory chip, the DTSC. Requirements: Low Noise: S/N ~ 22 Low Power: < 2 mW/channel Narrow width to match the microstrip Pipelined operation Radiation hardness up to 3 Mrad and 10 14 p/cm 2 11 Nicolo Cartiglia, INFN, Turin, Italy September 11th 2006

The Front-end Electronics-II The DTSC CMOS chip The TEKZ bipolar chip •DC Coupled to the detector •DC Coupled to the TEKZ •Shaping time = 32 ns •Clock freq 10 MHz •Gain = 150+-20 mV/fC •L1 and L2 buffers •S/N ~ 22 with 11pF load •FLT pipeline 5 us length •Power 2 mW/Channel •Power 2 mW/Channel •72 um •50,000 Transistor •1.2 um technology 12 Nicolo Cartiglia, INFN, Turin, Italy September 11th 2006

Roman Pot Detector Beam pipe Roman Pot: 3 mm with 3 movements: a 300 µ m thick window 1. Detector in/out 2. Pot in/out 3. Transverse Resolver Position accuracy: 5 µ m 13 Nicolo Cartiglia, INFN, Turin, Italy September 11th 2006

Roman Pot Motion At filling time the pots are retracted Approaching the beam Full insertion for data taking 14 Nicolo Cartiglia, INFN, Turin, Italy September 11th 2006

A LPS station Mechanical support Cooling pipe Filters TEKZ - DTSC Detector 15 Nicolo Cartiglia, INFN, Turin, Italy September 11th 2006

A real event 16 Nicolo Cartiglia, INFN, Turin, Italy September 11th 2006

Hit Maps Empty band: dead DTSC chips 17 Nicolo Cartiglia, INFN, Turin, Italy September 11th 2006

Detector Performance The LPS is a difficult detector to Year ZEUS LPS operate: it can only be used with 1994 3 0.9 perfect beam condition and no “creative” set-up 1995 6.6 3.4 1996 11 0 •Noisy or dead channels < 2% 1997 28 13 •Measured plane efficiency > 99.5 % •Noise < 0.3 ch/plane firing per bunch 1999 36 10 crossing 2000 47 35 Lum 132 62 (pb -1 ) 18 Nicolo Cartiglia, INFN, Turin, Italy September 11th 2006

LPS Physics Reach The LPS has been the used in many analysis: (unexpectedly we found ourselves in the middle of a strong competition with H1 on the topic of ‘diffraction’) •Clean tag of diffractive events, both at low and high mass •Measurement of the t- distribution in various processes (vector mesons, diffraction, high and low Q2 processes) •Clean tag of double diffraction •Leading proton production spectra 19 Nicolo Cartiglia, INFN, Turin, Italy September 11th 2006

Leading Proton Spectrum This plot shows the momentum spectrum of leading proton produced in e-p collision. It can be considered the summary of the LPS operation… Several models are also shown 20 Nicolo Cartiglia, INFN, Turin, Italy September 11th 2006

Conclusion The LPS was the first roman pot system operated at DESY. It took data for 7 years with a ~ 60% efficiency and integrated a luminosity of 70 pb-1 The LPS used 50000 channels of single sided Si detector read-out by two VLSI chips that provide a digital output. The 2-chip architecture that was developed for the LPS is now widely used in other systems 21 Nicolo Cartiglia, INFN, Turin, Italy September 11th 2006

Personal Conclusion It was and incredible occasion to witness the creation of a high tech lab: I arrived in Santa Cruz in 1989, before any probe station, chip analyzer or other sophisticated equipment. We had a rotary switch to change the gain on a amplifier…and we thought we should go and live in Texas. The lab was ‘the place to be’, long hours and a lot of new ideas. In one of my early day Abe asked me my favourite movies. Before I could come up with a really boring intellectual movie he said: I really like “The creature from the Black Lagoon”…and I understood I was in the right place! Thanks Abe, Thanks Hartmut! 22 Nicolo Cartiglia, INFN, Turin, Italy September 11th 2006