Dr Rob Bainbridge, Research Associate. Rob is holding a wafer of APV25 readout chips. The computer monitor shows a wafer map generated after the wafer has been automatically tested to identify which chips are good, and which chips have randomly-occurring manufacturing defects. Good chips are green, failures are red.
Dr Rob Bainbridge, Research Associate. Rob is sitting at the semi-automatic probe-station used to test the APV25 wafers. Each chip on the wafer (360 in total) is subjected to detailed tests to verify full functionality, so that when the wafer is diced, only good chips are mounted on the detector modules. The probe card (in the foreground) has approximately 40 tiny needles mounted on it, which are lowered onto each chip to provide power and control signals, and to read it out. After the test, which takes approximately 1 minute per chip, the card is raised and automatically moves to the next chip position. To test a whole wafer takes approximately 8 hours, so during mass production 2 wafers were tested per day. Approximately 650 wafers have now been tested to provide ~75,000 chips to CMS for ~15,000 detector modules.
As above.
As above.
Maria Khaleeq, Technician. Maria is a skilled ultrasonic bonding machine operator. The automatic bonding machine she is operating is used to make the extremely fine wire bonds needed to connect chips to the outside world.
As above.
Maria Khaleeq, Technician. Maria is inspecting one of the APV test boards, where you can see an individual APV25 bonded to its fineline test board.
Maria Khaleeq, Technician. Maria is a skilled ultrasonic bonding machine operator. The automatic bonding machine she is operating is used to make the extremely fine wire bonds needed to connect chips to the outside world. An APV test board can just be seen under the bonding head. The computer monitor shows a magnified view of the bonding area where some of the tracks on the test board can be seen.
As above.
An APV25 test board mounted in the X-ray irradiation setup. To monitor consistent quality during the production chips are sampled from the wafers and subjected to more detailed tests than can be performed at wafer probe time. These tests are performed before and after irradiation to 10 Mrads (100 kGray), the maximum level which they will reach after 10 years operation at LHC. The red laser light spot is used for alignment, showing the centre of the X-ray radiation field.
As above.
Dr Mark Raymond. Mark is setting up the X-ray machine where the APV25 chips are irradiated for the Quality Assurance tests. The X-ray tube can be seen in the middle of the picture. The APV25 chip sits, on its test board, on the platform underneath the tube (where Mark's hand is). The two photogrpahs immediately above are close-ups of what is on the platform.
As above.
As above.
Dr Yong-Jun Zhang. Yong is in one of our computer rooms beside a dedicated, high-specification farm of IBM storage, computing element and worker nodes used for CMS computing on the LCG Grid.
As above.
Stuart Wakefield and Dr David Colling. Stuart is submitting CMS analysis jobs to the Grid via a local User Interface machine. David is the Team Leader for CMS e-Science activities within the group.
The Imperial CMS group comprises ~40 people including academics, RA's, Engineers, PhD students and technical staff.
A significant number of staff are based at CERN. Approximately half the group are represented here, those who were at
IC on the day of the photography. Names:
Back row, left to right; Mark Pesaresi, Alex Tapper, David Wardrope, John Jones, Andrew Rose, Stuart Wakefield, Davy Machin, Dave Price, Mark Raymond.
Middle row, left to right; Rob Bainbridge, Jonathon Fulcher, Piera Brambilla, Kostas Petridis, Vera Kasey, Maria Khaleeq, Sarah Greenwood, Costas Foudas, Dave Colling.
Front left, sitting down, Yong-Yun Zhang. Front right, kneeling, Sebastien Greder, sitting down, Maiko Takahashi.
Front, Prof Geoff Hall.
As above.
Mark Pesaresi (right) and John Jones (left). In the CMS tracker the APVE (APV Emulator) is a VME board which sits in the trigger system, simulating the functionality of the APV25 chips down at the front end, and stopping excessively high trigger rates from being transmitted, which would cause overflow fault conditions. The board is FPGA based and has been designed with versatile features that allow it to be used in other applications. Here we see the APVE board laid flat on the bench, with a mezzanine card mounted on it, which forms part of a readout system for a medical imaging system for an EU-funded project.
As above.
Andrew Rose. The CMS group at Imperial has recently become involved in the re-design of the Global Calorimeter Trigger system for CMS, which will now be based on up-to-date FPGA technology. Here Andrew is inspecting a PCB layout of one of the new GCT boards. The computer monitor shows the software environment in which FPGA firmware is designed.
Dr Jonathon Fulcher and Dr Rob Bainbridge testing a rack of CMS readout electronics. The signals from the front end APV chips are transmitted optically to racks of electronics ~100m away in an adjacent underground cavern where they are fed into ~20 crates where 500 CMS FEDs are located. The FED inputs are 8 fibre ribbons, each ribbon consisting of 12 fibres, each fibre carrying the serially multiplexed data originating from 2 APVs. To test the FEDs special tester boards have been designed to produce simulated APV data in optical form. In the picture the yellow cables are the fibres which originate from the FED tester boards on the left hand side of the crate as 96 individual fibres, which are then combined into the 8 fibre ribbons feeding the FED board on the right hand side of the crate.
As above.
Dr Mark Raymond, Principal Research Fellow. Mark is inspecting a CMS silicon tracker detector module under test in an environmental chamber where it can be operated at temperatures as low as -20 degrees, similar to the operating temperature of the CMS silicon tracker. The CMS silicon microstrip tracker has an active detection area of ~200 square metres, made up of ~15,000 detector modules. Approximately 10 million microstrip channels are read out by ~75,000 APV25 chips. All the APV25 chips were wafer tested at Imperial College.
Dr Mark Raymond. The MGPA (Multi-Gain Pre-Amplifier) is the front end readout chip for the CMS Electromagnetic Calorimeter. It was designed at Imperial College and RAL, to meet stringent noise and linearity requirements, and there will be ~80,000 chips in CMS. Here we can see a 5 channel VFE (Very Front End) card under test.
As above.
Close-up of a VFE (Very Front End) card.
As above.
As above.
Close-up of an MPGA card.
As above.