From settles@mppmu.mpg.de Tue Jan 16 12:45:45 2007 Date: Tue, 16 Jan 2007 12:44:51 +0100 (CET) From: Ronald Dean Settles To: Dan Peterson , David Nygren , desch@physik.uni-bonn.de, Joachim.Mnich@cern.ch, jpmartin@lps.umontreal.ca, kaoukher@mail.desy.de, keisuke.fujii@kek.jp, Leif Jonsson , Luciano.Musa@cern.ch, msd@physics.carleton.ca, Paul.Colas@cea.fr, Stefan.Roth@physik.rwth-aachen.de, sugiyama@hep.phys.saga-u.ac.jp, takeshi.matsuda@kek.jp, ties.behnke@desy.de Subject: Agenda:16/01/2007 phonemeeting on LP electronics Dear friends, Three pieces of info for today's meeting are in the prel. agenda below. Anything else to be included? Cheers, Ron WP#2 Discussion 16/01/2007 -------------------------------------------- Tuesday 16 January 2007 West Coast East Coast W.Europe Japan 06.30 09.30 15.30 23.30 To join, dial +494089981390, after a few seconds into an answer in German, type the code number 52872# (LCTPC#). Agenda: -------------- 1. Discussion on the evolution in number electronic channels for LP1. For reference, here are three pieces of input: a. Leif's overview at the Eudet meeting in Munich in October can be downloaded from http://mppmu.mpg.de/~settles/tpc/lp/wpmtg/leifeudetmpi.pdf b. Here is what was written in the summary of WP#17 on 20 Dec. (with one small modification): "Leif would like a dedicated discussion on the number of electronic channels for the LP. As a reminder, Eudet foresees 2000 by the end of 2007, while many of us think around 10000 will be necessary to carry out the LP R&D program. This has several implications, e.g. where/how to instrument the endplate pads, what is necessary for the DAQ to inplement this number which goes beyond the Eudet plans, and so on. At first we thought to have this discussion during the next WP meeting on 10 January, but now it turns out that 16 January is better." c. Luciano's electronics report to the WP#16 on Dec.6 is here: "Development of the electronics for the LPTPC - PASA A first prototype of a charge amplifier/shaper circuit, has been developed in the course of the year 2006. This circuit has the following characteristics: - process: IBM CMOS 130nm - nr. channels: 12 - conversion gain: 9.5mV/fC - equivalent noise charge: 270e (Cin = 10pF) - dynamics: 4600 - low-pass filter: 4th order - power consumption: 10mW - package: CQFP 144 We developed this circuit with the following objectives: 1) verify the suitability of the IBM CMOS 130nm process for low-noise analogue applications; 2) study the performance of a circuit that represents a first step, in terms of building blocks and architecture, towards a programmable circuit for the LPTPC. The results above indicate that both goals have been met and are very encouraging for the development of the final version. The next submission is planned for the 2nd quarter of 2007. Hereunder are the specification (UNDER STUDY) of the new circuit: - process: IBM CMOS 130nm - nr. channels: 16 - conversion gain: programmable in the range 10-30 mV/fC - dynamics: ~2000 - architecture: CSA + PZ cancellation + 2nd order low-pass filter (the filter has the function of limiting the bandwidth before ADC) - peaking time: programmable in the range 30-100 ns - decay time: programmable in the range <1us - ALTRO chip The ALTRO chip exists in two versions. The two versions differ in the ADC bandwidth. A polarisation current, provided for each channel by an internal resistor, defines the ADC bandwidth and power consumption. The polarisation resistor is divided in multiple taps such that only one metal layer has to be changed, in order to optimise the power consumption to the required bandwidth. These two versions of the ALTRO chip have been produced with maximum rates of 25MSPS (Engineering Run for ALICE) and 40MSPS (Multi-Project Run), and power consumption of 12.5mW and 43mW respectively. The resolution of the ADC as function of the sampling rate (Fs) and frequency of the input signal (Fin), for the two ALTRO versions, is shown in the figure in attachment. About 125 ALTRO chips (2000 channels) of the 40MHz version are available for the LPTPC. A new production of 25-MHz ALTROs has the following costs: - wafers (minimum quantity 25 wafers): 66000 EUROs - packaging of 1000 chips (16000 channels): 4000 EUROs A new production of 40-MHz ALTROs has the following costs: - re-working of the mask set: 200000 EUROs - manufacturing and packaging of 1000 chips 20000 EUROs All prices quoted above are cost estimates and have to be confirmed by official commercial quotations."