W.J. Llope
Skip was charged by Lee on 8-9-95 with providing to the EMC Task Force (by roughly 8-18-95) a proposal for the institutional responsibilities for the EMC electronics. Gary has made the point that it is important to promptly define the institutional responsibilities for the entire EMC effort, including of course the EMC electronics. Steve made the suggestion that each of us - Gary (with Skip), Steve, and Bill - formulate such an institutional strategy as if each of us were making the final decision. This idea has been discussed casually with Dave and Tom, who both felt it was a worthwhile exercise that they wanted to participate in. For this exercise, Gary, Dave, and Tom participate as institutional representatives, not as task force members. The three definitions of institutional responsibilities made by Gary, Steve, and Bill would then be compared and discussed amongst us. I note that institutions other than those represented by Gary, Steve, and Bill (i.e. ANL, IHEP, PPL, UCLA, and WSU) have also made electronics related proposals, so we need to include these other institutions in this discussion.
I think that this scheme will be particularly efficient, which is important given the ~1 week deadline that was imposed on us to develop such a plan. The success of this exercise relies on the series of open discussions that will follow the circulation of the various definitions of the institutional responsibilities.
I need to make a few comments about what I am assuming and neglecting in the discussion below. It is clear that the assignment of an institutional responsibility does not preclude the participation of another institution in the same effort. In this case though, it is the assigned institution that is made publicly responsible for the overall effort and all of the final decisions that must be made in the course of this effort. On the other hand, each institution is clearly interested in being made responsible for at least one identifiable effort that will have a major impact on the EMC project in general. It should be our intention to assign each electronics effort to institutions in such a way that all of the necessary work is being done, that there are clear avenues for other institutions to help out if possible, and that each institution has clearly identifiable project(s) that are well matched to their manpower.
It is possible that some efforts will be requested by only one institution. If such a request is well-matched to the available funds and the capabilities of the institution, then this institution should tentatively be granted the request. Iteration to arrive at a consensus will be necessary. This is because some institutions have proposed several different efforts. An institution might rescind a part of their proposal if it is clear that this institution is the best suited to another part of their proposal which would demand the majority of that institution's resources. Furthermore, some institutions considered here have proposed efforts outside of EMC electronics, which could similarly affect an institution's interest and capability of completing a particular electronics effort. I will assume that such aspects will be worked out during the ensuing discussions.
I note that the categories as defined in the ANL breakdown, i.e. in the list we all put our institution name next our categories of choice, implicitly include the possibility for conflicts of exactly the kind that this exercise is intended to solve. For example, the fact that items 4.5.1.4 (Calibration systems) and 4.5.1.13 (EMC Controls) are listed separately occludes the fact that these two efforts are so closely tied that it might be best if one institution did both. There are other correlation's of this kind. I will assume that such aspects will also be worked out during the ensuing discussions.
With the institutional proposals as I understand them at the moment, this document is my first pass definition of an institutional strategy for performing EMC electronics R&D, made as if the decision were up to me. I note that I am presently just looking at accomplishing only the electronics R&D, not the building the entire electronics systems for STAR. Of course, the institutions that develop a particular prototype must certainly be considered first as the lead institution to build all of the necessary final versions. At this stage, I am only trying to define a strategy to advance the design of the EMC electronics components to a level of detail sufficient enough for us to know (and defend) if our plan is viable enough to be funded. I am sure others of us will also consider each institution's capabilities on the longer time scales up to Day 1 of RHIC. I believe that it is also important to consider this; I simply choose to abstain from this consideration at this time.
I consider the projects to be assigned by our consensus are the following. These are the particular items in the Jim's excerpt of the ANL WBS worksheet that I believe are related to EMC electronics. Some small revisions were made to this list by Lee at the end of Wednesday's session. The institutions that expressed an interest in each topic are also listed (alphabetically).
| 4.5.1.3 | PMT System | IHEP, RICE, UCLA |
|---|---|---|
| 4.5.1.4 | Calibration Systems | ANL, RICE, WSU |
| 4.5.1.9 | EMC FEE | MSU, RICE |
| 4.5.1.10 | EMC Contribution to DAQ | MSU |
| 4.5.1.11 | EMC Contribution to On-line System | IHEP, MSU |
| 4.5.1.12 | EMC Level 0 Trigger | ANL, IHEP, MSU, PSU, RICE |
| 4.5.1.13 | EMC Controls | IHEP, RICE, UCLA |
| 4.5.1.14 | EMC Level 1 Trigger | MSU, PSU |
| 4.5.2.4 | SMD FEE | ANL, MSU, PPL, RICE |
This is the list I will work with below, although I realize that institutions may decide to move about (and come and go) from within the list above as these discussions continue.
For completeness, a listing of the non-electronics efforts that were proposed by institutions also making an electronics-related proposal are listed below. If these institutions are eventually assigned one of these efforts, this could affect their capabilities for the electronics part of their proposal. If items below were proposed by only one institution, then I presently assume that they will be the lead institution for that effort. One question to be answered by our discussions is the extent to which such considerations affect the electronics part of that institution's proposal.
| 4.5.1.1 | Compressed Modules | ANL, IHEP, WSU |
|---|---|---|
| 4.5.1.2 | Optical Readout System | ANL, IHEP, RICE, UCLA |
| 4.5.1.5 | Module Final Assembly and Tests | ANL, IHEP, WSU |
| 4.5.1.6 | EMC Prototypes (other than simulations) | ANL, PPL, WSU |
| 4.5.1.7 | Transportation and Handling Systems | ANL, PPL, WSU |
| 4.5.1.8 | EMC Installation and Test | ANL, IHEP, WSU |
| 4.5.1.16 | Integration and Project Management | ANL, WSU |
| 4.5.2.1 | BSMD Chambers | IHEP, PPL |
| 4.5.2.2 | BSMD Prototype (full scale) | IHEP, PPL |
| 4.5.2.3 | Transportation and Handling Systems | IHEP, PPL |
| 4.5.2.5 | BSMD Conventional Systems | MSU |
| Simulations | Basically every institution |
Institutional Strategy for EMC Electronics R&D
The following presents a proposed institutional strategy for performing the EMC electronics R&D. I start below by tabulating the EMC electronics topics. Each topic is also labeled with a proposed lead institution for this work, followed by a proposed list of contributing institutions. Some comments on why the efforts were divided in this way are included below the table.
| WBS No. | WBS Item | Lead Institution | Contributing Inst. |
|---|---|---|---|
| 4.5.1.3 | PMT System | UCLA | RICE, IHEP |
| 4.5.1.4 | Calibration Systems | ANL | ANL (Radioactive) |
| UCLA (LEDs) | |||
| RICE (LaDs, LEDs) | |||
| 4.5.1.9 | EMC FEE | RICE | ANL, MSU |
| 4.5.1.10 | EMC Contribution to DAQ | MSU | PSU, RICE |
| 4.5.1.11 | EMC Contribution to On-line | MSU | |
| 4.5.1.12 | EMC Level 0 Trigger | MSU | ANL, PSU, RICE |
| 4.5.1.13 | EMC Controls | UCLA | RICE |
| 4.5.1.14 | EMC Level 1 Trigger | MSU | ANL, PSU, RICE |
| 4.5.2.4 | SMD FEE | RICE | ANL, PPL |
4.5.1.3 PMT systems
Here I consider only the CW bases and the associated controls, ADCs, etc. I see the possible IHEP contribution of overseeing the MELZ construction of the PMTs as a hardware effort, although the number of stages and other specs of these tubes might affect the CW design.
UCLA made the original prototype CW, which was tested at Rice. Some significant revisions to the UCLA design were made, resulting in the CWs for the CTB and TOF. Rice made changes to the feed-back loop and the low voltages, and also optimized the circuit to lower the oscillator frequency noise. UCLA and Rice should discuss these changes. IHEP, Rice, and UCLA should trade schematics and discuss the best design of the EMC CW prototypes. It has been assumed that we can sufficiently shield the bases from the fields. UCLA should present the costs to build ~2 prototype CWs, then ~10 more after one revision. These should be in the final mechanical assembly. The effort involves board layout, PCBs, and fabrication. Rice has already proposed to develop these prototypes. UCLA, Rice, and IHEP should therefore decide if the project is best performed via some division of effort.
The UCLA(IHEP) quote for 1200(1320) channels was 175$(~154$) each. Rice estimates 80-100$/channel for the EMC CWs. It will be easier to understand these price differences once the schematics and spread sheets are passed around. This should be done a.s.a.p.4.5.1.4 Calibration Systems
The development of the radioactive source system is closely tied with the development of the mechanical design of the stack. ANL should thus lead the development of the radioactive source system, working closely with WSU. This involves the sources, the controller, the encoder, as well as the sources, the wires, and the tubes. A prototype of this system should be included in the LPEMC, including the software needed to run the system and extract the calibration results.
ANL should also oversee the development of the LED, LaD, and (perhaps) UV-Laser calibration systems. Prototypes of these systems have already been developed at several institutions. UCLA has developed LED and other optical calibration systems, while Rice has developed LED, LaD, and UV-Laser-based systems. Schematics and other information should be exchanged amongst ANL, Rice, and UCLA and then discussed. ANL should organize these institutions to make available both an LED system and the radioactive source system for the LPEMC. Bench tests of these systems should be performed, possibly in parallel, to learn as much as possible about these systems before the prototypes are made. ANL should develop the costs and schedules for the calibration system prototypes, for whatever institutional division of this work is appropriate for its timely completion.
4.5.1.9 EMC FEE
First, the dual-range integrator+ADC and the associated controls. This system is very similar to the (single range) integrator+ADC developed at Rice for the CTB and TOF systems. Both ANL and Rice have already investigated this board. Rice should lead this effort, working closely with ANL, MSU, and other interested institutions. Second, Rice should also lead the development of the electronic multiplexing, pedestal subtraction, and demultiplexing, as it will be similar to the CTB and TOF needs. This effort must be carefully coordinated with the development of the controls systems (UCLA, Rice), the calibration systems (ANL, Rice, UCLA), and the trigger and DAQ-related systems (ANL, MSU, PSU, Rice). Prototypes should be ready for runs of the LPEMC, perhaps also for any SPEMC running.4.5.1.10 EMC Interface to DAQ
MSU proposed to concentrate on many aspects of all of the electronics "from the digitizer to STAR", which is well-matched to MSU's experience in the development of modern data acquisition systems. A transputer-based DAQ system was developed at the NSCL, which is presently being used for all runs of the MSU 4¹ Array. This system is highly portable (it was also used in both of the SPEMC test runs), so MSU should be able to provide DAQ prototypes for the LPEMC tests.
4.5.1.11 EMC Interface to On-line
I consider this effort so closely tied with the 4.5.1.10, (EMC interface to DAQ) that the same institution should do both. Therefore, MSU should be the lead institution, working closely with ANL, PSU, and Rice. I assume that it is not necessary to have a prototypical interface to On-line ready for the LPEMC or SPEMC tests. For example, we have survived quite nicely in the SPEMC test runs by simply using a global section approach based on the CERN software. However, the institutional responsibility for this effort should nonetheless be defined now to insure that it is cleanly integrated with the rest of the system.4.5.1.12 EMC Level 0 Trigger
The presently considered design of the system is generally based on the "analog+digital" scheme described in the STAR Note by D. Underwood, so ANL and MSU should work together closely on this effort. PSU has also expressed an interest. These three institutions should carefully coordinate some division of efforts, as both MSU and PSU are both ~1 FTE at the moment. ANL, MSU, and PSU should consider ANL to lead this effort until such a time that MSU and PSU obtain the additional manpower that they expect.
At the same time, Rice should continue to investigate the feasibility of the 3D-Flow approach. There were several important questions raised during Dario's presentation that need to be answered. A major one involved the time scales in getting the EMC information to the 3D Flow system soon enough so that the computations are finished in less than a few µsec. Rice will answer these questions, and continue simulating the performance of the 3D Flow system for realistic simulation events. This effort is already funded by Dario's SBIR grant.
This effort, as well as 4.5.1.14 (EMC Level 1 Trigger) should include MODSIM simulations, of the kind now being used to study the performance of the STAR trigger system. J. Whitfield is leading this effort for the STAR trigger system. This is an expensive package, which is already installed at BNL and is basically underutilized.
4.5.1.13 EMC Controls
UCLA is the proposed lead institution for the PMT system, and a contributing institution for the calibration systems. This institution should therefore lead this effort, as it is so closely tied with these other efforts. Rice should also contribute, having worked through similar applications to the CTB and TOF systems.4.5.1.14 EMC Level 1 Trigger
This closely tied with 4.5.1.12 (EMC Level 0 Trigger), so the same institutions involved there should also be involved here. ANL, MSU, PSU, and Rice should organize an efficient division of this work.
4.5.2.4 SMD FEE
ANL lead the effort to simulate the STAR magnet noise at the most recent SPEMC test run. Therefore ANL should contribute significantly to the development of the transimpedance amplifier cards and the related on and off-detector SMD FEE. IHEP has also expressed interest in this work. Close cooperation between the Russian institutions (that I assume will build at least the prototype BSMDs and ESMDs) and ANL/Rice is required. Rice has developed these circuits before, and is currently working on similar applications to the STAR-MWC, as well as the DDC for E896. The E896 and MWC applications will involve Dorfan chips, which may be directly applicable to the EMC SMD needs. Prototypes of these chips are expected in 1-2 months.
summary of projects proposed by institution
The following are notes I took during the Monday afternoon session, which included the institutional proposals for EMC efforts.
WSU (cormier)
contribute 6 FTE. no SVT overlap
description of E864 hadronic calorimeter, tour will be scheduled
proposal
get materials
mechanical fabrication
calibration
handling and shipping
numbers (based on 7/18/95 ANL rollup)
12.6 M$ reduced by 7.8 M$ at WSU by much lower labor rates
and free labor contributions from the university
total cost 4760.3 K$
machined scintillator, molded scintillator reduces cost by ~900K$.
28.2 man-years divided over a three year period.
UCLA (igo)
cockroft waltons: 210K$ for 1200 channels (175$ ea.)
scintillator tiles
frames for the tower modules
testing PMTs
LED calibration system
3 FTE physicists, 5 FTE tech/eng, students
MSU (westfall)
emc level 0 trigger
fast analog sums
patches
emc level 1 trigger
fast tower comparisons
fast digitial SMD analysis
characterization and zero compression
emc interface to star triggers
emc interface to star daq
data monitoring
RICE (llope)
I passed out a document during my presentation. This included (in no particular order):
SPEMC analyses
SPEMC simulations
STAR-EMC simulations
SPEMC, future test runs and maintenance
fiber splicing and optical read-out chain construction
calibration systems
FEE prototyping and fabrication
PSU (heppelman)
he is NSF, and is 0.5 FTE for STAR
negotiating with PSU and STAR for seed money to hire 2 FTE's
will give 2.5-3.5 FTEs total
have a highly subsidized shops
four FTE machinists for 29 $/hr, electronics techs are 20 $/hr
want to do electronics development
IHEP (vasiliev)
BSMDs: 240 chambers, 32 wires X 100 strips, 32 K channels total
432 K$
injection molded scintillator tiles
machining parts
PMTs, FEU 115M (being used in PHENIX) made by MELZ
1260 pieces, 218.9 K$
total cost to build EMC (stacked modules option) is 4691.6 K$
Cockroft Waltons, 1320 pieces, 202.9 K$
23 physicists and 10 technicians, all could be 100% STAR