HEP Project Status Report August 2018
High Luminosity LHC CMS Detector Upgrade Project
HOST LABORATORY: FNAL
FEDERAL PROJECT DIRECTOR: Robert Caradonna
PROJECT MANAGEMENT EXECUTIVE: J. Stephen Binkley
FEDERAL PROGRAM MANAGER: Simona Rolli
CONTRACTOR PROJECT MANAGER: Vivian ODell
NSF PRINCIPAL INVESTIGATOR: Anders Ryd
NSF PROGRAM MONITOR: Mark Coles
Explanation: The narrative parts of this report reflect the up
to date status. The financial data, EVM metrics, and milestones
reflect the status at the end of the previous month.
SCORECARD AS OF August 31, 2018
Current CD:
0
Date of Current CD approval:
April 13, 2016
Next CD:
1 (mini)
Forecast
April, 2019
Baseline:
NA
Percent Complete:
Pre-baseline
Planned:
NA
ETC:
NA
TPC or Cost Range (CD-0):
$125-155M
Contingency:
NA
Float to CD-4 in Days:
NA
Cumulative CPI:
NA
Cumulative SPI:
NA
NEAR-TERM MILESTONES
The table below shows all the Project Office milestones for
2018. Milestones in bold have been met.
WBS Area
Milestone
Forecast Date (month/year)
Date Achieved or Current Forecast
402.1 Project Management
Full project rollup and drafts of all CD-1 documentation ready
for review
6/17
2/18
Ready for Fermilab Directors Review
3/18
3/18
Ready for CD-1 Independent Project Review
9/17
5/18
US-DOE Risk Workshop
9/18
9/18
US-MTD scope/deliverables decision
10/18
10/18
QA and ESH ready for Project Review
10/18
10/18
Project documentation reviewed and synced
11/18
11/18
MTD Technical Review
11/18
11/18
MTD Cost/Schedule/Risk Review
12/18
12/18
NSF pre-FDR schedule frozen
12/18
12/18
DOE pre-CD-1 Directors Review schedule frozen
12/18
12/18
NSF pre-FDR Review
31 January 2019
31 January 2019
MTD TDR complete
2/19
2/19
Ready for pre-CD-1 Directors Review
2/19
2/19
MTD TDR submitted to LHCC
4/19
4/19
Ready for CD-1 mini-Review
4/19
4/19
Table 1: List of milestones for the Project Office during FY18.
Milestones in bold have been met.
Technical Project milestones for 2018 are embedded in the
relevant status highlights for easier cross-reference and are also
listed in the table below.
WBS Area
Milestone
Forecast Date (month/year)
Date Achieved or Current Forecast
402.2 Outer Tracker
Completion of prototype MAPSA test system design iteration 1
3/18
3/18
Completion of prototype burn in system
4/18
3/18
Delivery of Upgraded Gantry Robot
6/18
9/18
Completion of Sensor QC Site apparatus procurement
7/18
12/18
Completion of Prototype Plank Design
7/18
7/18
Completion of Single Module Test system
8/18
11/18
Completion of Prototype Ring Design
9/18
9/18
Completion of prototype MAPSA test system design
9/18
4/18
Completion of burn in system final design
9/18
2/19
Complete fabrication of initial functional 2S modules
(significantly delayed from last year due to delays from iCMS)
9/18
8/18
Completion of Conceptual Design of Flat Barrel Services
9/18
9/18
402.4 Endcap Calorimeter
Construct demonstration 8 sensor modules
7/17
12/17
Complete testing of silicon-only mockup cassette and prepare
report on results
1/18
2/18
Construct scintillator for mixed-mockup cassette
4/18
8/18
Assemble mixed-mockup cassette
8/18
10/18
Complete design of first prototype of concentrator ASIC
(HGCCC_p1) ready for international review
9/18
1/19
402.6 Trigger/DAQ
(L1TCal) APd1 conceptual design complete
12/17
3/18
(L1TCal) APd1 full design complete
7/18
8/18
(L1TCal) Emulated e/g performance
8/18
8/18
(L1TCal) APd1 initial software release
10/18
10/18
(L1TCal) first fully assembled APd1 board
10/18
10/18
(L1TCor) Emulated performance and efficiency for Jets/MET/HT
10/18
10/18
402.8 MIP Timing Detector
Baseline plan solidified with US and international CMS
11/18
11/18
CDR ready for external review
12/18
12/18
Cost and Schedule ready for external review
12/18
12/18
MTD ready for Directors Review
3/2019
3/2019
Table 2: List of milestones for the technical part of Project
(DOE part) during FY18. Milestones in bold have been met.
STATUS HIGHLIGHTS
WBS 402.1: Project Office:
We are preparing for CD-1 mini-Review in Spring 2019, planning
for CD-3a in Summer 2019 and planning for CD-2/ CD-3 in Summer/Fall
2020. The dates may evolve as we learn more.
This month we have been focused on several areas in order to
prepare for the CD-1 Mini-Review:
Out of the two recommendations with scheduled close in August
2018, we have closed the one requiring an update of the PHAR, and
partially addressed the one requiring the allocation of sufficient
expert ASIC engineering resources to meet our schedule. The CD-1
IPT Tracking Document can be found in:
https://cms-docdb.cern.ch/cgi-bin/DocDB/ShowDocument?docid=13604
.
In the context of re-accessing the technical risks of the
Project, we are organizing a Risk Workshop for the entire DOE scope
of the Project, to take place between Sept. 4 and Sept. 13 at
Fermilab, in 4 separate sessions for each L2 system. The Reviewers
are colleagues mostly external to the Project. The NSF part of the
Project will have a similar workshop, likely in October 2018.
We have brought on board a dedicated ESH&Q professional (TJ
Sarlina) to help review the ESH and QA/QC plans, to ensure they are
complete and take the lead on implementing them. We appointed TJ as
our QA/QC Manager in July 2018 and, in addition, as ESH Manager in
August 2018. We have rewritten our QA plan with the help of the QA
professional and internally reviewed it. We had a mini-workshop on
August 1 to document the QA/QC plans and ensure that all the
subprojects understand what is needed and expected from them. We
are in the process of collecting specific data from the subsystems
on validation/ verification and calibration planning activities, in
preparation for an ESH&Q Project Office initiated review
expected to take place in October 2018.
We will be triggering a review of the full documentation of the
Project in November 2018, in order to address the issues that our
CD-1 reviewers pointed out.
We continue to flesh out the MIP Timing Detector (MTD) schedule
for CD-1 mini-Review preparations. Within the US, we have filled in
most of the lowest level (L4) management and we are converging on
having the L2/L3 leadership in place very soon. We continue having
a series of workshops to get the technical, cost, schedule and risk
areas into shape. We have developed a bottom-up schedule for when
the MTD will be ready for a CD-1 Directors review, and which
technical decisions are necessary for it. We are aiming for a
Technical Review in November 2018 and for a cost/schedule/risk
review in December 2018.
Because we are now down to a single vendor for silicon for ATLAS
tracker, CMS tracker, CMS endcap, a team of CERN / ATLAS / CMS
experts are negotiating with the company and expect to have a site
visit to Hamamatsu in late September. We will then start learning
exactly when the silicon will be delivered for each subproject, and
when the US will need to pay for its share. The silicon sensors for
both Outer Tracker and Endcap Calorimeter are items of a CD-3a. We
are collecting additional CD3-a candidate items from all L2 systems
and we can firm up the CD-3a date when the international schedule
is clearer.
Meanwhile, we are moving the full project towards EVMS
reporting. We have defined Control Accounts and are making progress
towards defining Chargeable Task Codes and Project Measurement
Techniques (PMTs/EVTs); items needed in COBRA to establish the
reporting. Outer Tracker and Endcap Calorimeter continue statusing
their schedules, using this as an opportunity for their CAMs to
understand what level of project planning is necessary to be able
to collect reliable EVMS data. We have also organized EVMS training
for all CAMs which has started in August and will continue through
September 2018.
We had a fruitful workshop for the Endcap Timing Layer (ETL)
ASIC on August 7, 2018 and we expect a follow-up workshop on
September 6. In the meantime, we have reworked the ASIC schedule
and integrated additional ASIC engineering resources from a
University. We are also working with Lab Management on obtaining
additional ASIC engineering resources and the Lab is in the process
of interviewing qualified candidates.
The U.S. floats to international CMS need by dates have not
changed since the July report currently the tracker subproject has
the least amount of float at 7 months. The other subproject floats
to iCMS need by dates are between 9-10 months, with the exception
of the DAQ here there is only 5 months float, however this is by
construction. The DAQ contribution is a commodity item that is
bought as late as possible in order to capitalize on computing
technology price/performance. We continue to monitor the floats as
external deliverables / milestones are updated, or our work plan
evolves.
WBS 402.2: Outer Tracker (L2 Manager: Steve Nahn, Fermilab)
One of the major U.S. deliverables for the Outer Tracker is the
Macro-Pixel Sub Assemblies (MaPSAs), essentially the bump bonding
of ASICs to sensors. The project considered three potential vendors
for this process and has been producing dummy components for vendor
qualification. On the basis of the tests of the first round of the
MaPSA dummy modules received in July from two of the less expensive
vendors, only two out of the three companies are retained as
potential vendors (AMTec and Hamamatsu). Orders have been placed to
both companies for the next set of fully functional MAPSAs.
Aerotech announced an additional delay to the delivery of the
Gantry machine due a backlog of orders; the robot is now expected
on Sep 5th. The items needed to complete the gantry have just
arrived, so the Sept. 5 date looks firm. The delay so far does not
impact our international milestones.
Tests on the functional prototype 2S module assembled at FNAL
earlier this year were concluded successfully. Issues we had past
month with one of the hybrids not responding were addressed by
protecting better the module from light in a well-sealed box.
Details
Sensors 402.2.3
402.2.3.1 Sensor QC Centers
Brown measured diode characteristics to understand effect of
connecting them to 10x2 flutes of probe pads. There are some
disagreements with measurements at Vienna and we are
investigating.
Brown designed and ordered a new PCB for the probecard to
contact the process QC flutes. Connectors have a long lead time -
14 weeks. Designed holder to mount probe card on probe station.
Brown and Rochester compared test results for some old ST
sensors. There is broad agreement in the observed trends but there
are some differences in the details of the measurements which we
are investigating.
402.2.3.2 PS-P Sensors
Prototype
Brown received prototypes from Novati/NHanced
402.2.3.3 PS-S Sensors
Prototype
Brown received prototypes from Novati/NHanced
Electronics 402.2.4
402.2.4.1 MaPSA
Prototype
Dummy MAPSA parts have been received and tested at Fermilab.
Parts with underfill have unacceptable bow. The current underfill
uses a heat cure. We are now obtaining underfill material which has
a better thermal expansion match to silicon and has a room
temperature cure.
Parts from two assemblers, I3 and AEMtec, were tested. The I3
parts had low bump yield, especially the parts with underfill.
AEMTec parts look better and have good bump connectivity. As a
result of these tests, and increased quotes from I3, we are
dropping I3 as a vendor and will have the next set of parts
assembled by AEMtec and Hamamatsu. Hamamatsu will also be
fabricating the sensor components.
Orders are being placed for the next round of prototypes, which
will include the first fully functional MAPSA assemblies.
402.2.4.2 Test Systems
Hybrids
Nothing yet from iCMS side.
Module
2xSSA module components are in hand (PCB and interposers),
assembly will start next week. DAQ (firmware and Ph2DAQ) for the
2xSSA module is ready.
Software for the cold box slow control is almost done. Minor
tweaks in the design are ongoing based on the prototype
experience.
MaPSA
Final version of interface board in hand. Final version of probe
card still needs to be designed.
Modules 402.2.5
402.2.5.1 Module Assembly Consortia
East Coast
Brown: The cleanroom on the 7th floor of the Barus and Holley
building is being retrofitted for module assembly use.
Rutgers: hired a person to supervise the clean room, so theres
been a lot of set-up work
on room layouts, fixtures, equipment calibrations, etc.
Fermilab
Aerotech announced yet a third delay for the gantry robot, now
indicating that it will ship on September 5. The problem is a
backlog of critical components from a third party. Even if they do
ship on the 5th this will make it difficult to have gantry ready
for a demonstration during the Modules workshop at the end of
September.
Infrastructure
FNAL: Components for a gantry-based readout hybrid gluing
fixture are being machined at MAB. A design for 2S bias tab and
kapton insulation fixture is nearly complete. We will initially
fabricate two of these - one for the Brown group and one for us. An
order has been placed for 20 2S carrier plates. Our lead engineer,
Greg Derylo, has been contributing to the design of the plate as
they will also be used in the burn-in system, which he is also
designing in part. A coupling plate for the encapsulating robot in
Lab D was fabricated this period. Steel inserts for top and bottom
2S bonding jigs, initially fabricated at Wayne State, are being
machined in the Village machine shop, and the jigs will be sent to
Princeton once they are complete.
402.2.5.2 Module Components
Hybrids
We dont expect to receive CBC3 hybrids until next year and at
FNAL we are producing stiffeners for these hybrids (see below).
Mechanics
Brown: PS module assembly jigs have been modified to incorporate
more assembly steps into fewer jigs. This will also improve the
repeatability of assembly by removing some of the more manual
steps. New jigs with these modifications are in progress at Brown's
machine shop.
Brown: A second iteration of a PS module carrier design has been
fabricated. This includes all the components for shipping two PS
modules on one carrier. This prototype will be used with dummy PS
modules to further iterate on the PS module carrier design.
FNAL: Work has started on machining 180 sets of 2S stiffeners in
response to a request from the CERN group. The layups of the K13D2U
carbon fiber prepreg were done at both FNAL and Purdue, with the
Purdue layups still in progress. Purdue recently completed
measurements of the K13D2U thermal conductivity and these will be
included in a CMS Detector Note that Stephanie Timpone initiated at
the beginning of the summer.
402.2.5.3 Module Assembly
PS Modules
PS Prototype
1. Brown: The new wire bondable PS module, PW1803, has been
through additional rounds of inspection and will be shipped to
Princeton for bond testing, programming, and evaluation of a new
wedge bonding tool. This module also has more realistic readout
hybrids, allowing for a more accurate bonding experience.
1. Princeton: The wire Bonding stage for the PS module is
complete; alsosuccessful bonding of the first mechanical model
PW-1802 PS module
1. FNAL: Arnab Ghosh has completed layouts for dummy PS sensors
to be laser diced from 150mm blank silicon wafers.
. 2S Modules
2. 2S Prototype
1. Brown: SA1804, a dummy 2S module, has had more realistic flex
hybrids placed using a modified jig. After inspecting the module
for alignment of hybrids to silicon and checking for condition
issues, it was sent to Princeton for bonding practice, fixture
design, and programming.
1. Rutgers: encapsulation studies are converging, we should be
ready to pot a dummy module by the end of the month.
1. FNAL: We are joined this month by a new RA, Fabio Ravera, and
relocating the module test system from FCC to SiDet has allowed
Lorenzo and Fabio to read both the left and right hybrids on our
first functional 2S module for the first time.
Mechanics 402.2.6
402.2.6.1 Planks Mechanics
Prototype
1. [text]
. 402.2.6.2 Ring Mechanics
2. Prototype
1. [text]
Integration 402.2.7
402.2.7.1 Flat Barrel Design
[text]
WBS 402.4: Endcap Calorimeter (L2 Manager: Jeremy Mans, U. of
Minnesota)
The endcap calorimeter full-silicon mockup cassette has been
assembled and successfully tested. The measured temperature
distributions show good agreement with simulation, and the
performance of the PCB-based baseplates has been confirmed in the
mockup cassette. This technology will be considered the baseline,
and studies are underway to leverage the PCB design to reduce noise
sensitivity in the detector. The thermal performance of the
scintillator tileboard thermal mockup has also been demonstrated,
and the observed temperature distributions meet the requirements
for silicon photomultiplier cooling after addition of thermal
interface pads. A development board for the slow-control chip has
been produced, to provide a path to make progress on module
electronics in advance of the international first digital version
of the Endcap Calorimeter readout chip being available. The Endcap
Calorimeter readout chip and the Concentrator ASIC remain on
schedule. We have finalized the conceptual design of the
Concentrator ASICs trigger path first prototype, and presented it
to the ASIC PMG on August 20, 2018.
Details:
Silicon Sensors, and Silicon Modules:
TTU fabricated additional mockup silicon modules so we could
test a variety of baseplate materials: PCB (5), carbon fiber (5),
and copper/tungsten (2). These were all thermally cycled (-30 to 30
C) modules at TTU before shipping to Fermilab, and at Fermilab to
investigate the degree of warpage and mechanical stability.
Fermilab performed mechanical and thermal FEA calculations on
modules with these various baseplates, and these simulations showed
good agreement with measured mechanical (physical deflection) and
thermal measurements. TTU and CMU ordered the needed wire bonders
in preparation ahead of modules that will be needed for prototype
1. The group completed neutron irradiation of test samples at Brown
and UC-Davis, and the IV/CV and TCT measurements are now ongoing
with some results to be shown at UCSB module workshop that will
take place at the end of August.
Scintillator, and Scintillator Tile-modules:
We have been working on defining the decision criteria for
making the baseline scintillator choice including consideration of
blue vs. green scintillators. We continue to develop a slow
controls test system, and we are working with the international CMS
TCDS group on fast controls specification.
Cassette:
We performed cold testing of all-silicon mockup cassette with
new modules with various baseplates. We measured module warping
with a CMM, and compared the results to FEA studies that our group
has done. We also measured with a CMM the CTE of baseplates made of
various materials, and also compared with FEA calculations. We
designed a PCB baseplate with electrical shielding layers, and
worked on the design of the cassette mounting scheme into the
absorber.
ECON:
The organization of the lpGBT design transfer is in progress.
Fermilab engineers have scheduled a trip to CERN for lpGBT design
meeting on Sep 24-28. G Deptuch and J Hirschauer have finalized the
conceptual design of the ECON-TRG prototype 1 and presented it to
the ASIC PMG on Aug 20.
Project:
The project team is continuing to work on some issues raised at
the CD1 DOE IPR review in June. We are implementing our revised
plan in the Resource Loaded Schedule to take into account a 5-6
month delay in the next HGCROC submission, by planning for a faster
production schedule for the modules and cassettes. We had started
to plan for deliverables and goals for a new Major System Prototype
0, between the mockups and Prototype 1, so the needed designs,
prototyping, and tests can be done prior to a CD-2 review. However,
since the CD-2 review is now pushed back to late-2020, this
additional prototype 0 is now no longer necessary. Due to the delay
caused by international CMS revisiting the scintillator tile-module
geometry, the mixed cassette mockup deliverable is also delayed.
Work for the mixed mockup cassette is needed ahead of prototype 1.
We are reviewing a revised schedule for the mixed mockup cassette.
We are also reviewing our risk register to ensure it is updated
with the new plan of having a faster module and cassette production
schedule. We have been determining possible deliverables for CD-3a:
production sensors, and production wrapped scintillator tiles.
Other:
We have no changes at this time for the subproject float to
need-by dates. We have no updates for the risk registry or the IPT
tracking registry.
WBS 402.6: Trigger/DAQ (L2 Manager: Jeff Berryhill,
Fermilab)
The Trigger/DAQ activities during the past month (August 2018)
focused on finalizing the design of the initial prototype of an
Advanced Processor development trigger board (APd1) and on
developing algorithms related to calorimeter-only Jets/MET/HT
triggers as well as track-matched muons, forward electrons, and
particle-flow taus. A first implementation of the particle flow and
pile-up mitigation algorithms for the Correlator Trigger was
demonstrated this month on Amazon Web Services (AWS) FPGA
instances. We have also met the August milestone for emulating the
e/gamma performance. The project expects to complete the APd1
design milestone by the end of August, and is on track to complete
the APd1 produced milestone this Autumn. Finally, the project is on
target to complete the Jets/MET/HT algorithm milestone in
October.
Details:
L1 Trigger Hardware:
The layouts of the APd1 and its Rear Transition Module (RTM)
PCBs are nearing completion with the last few routes being
optimized and the overall routing being reviewed, and parts are
being ordered. We have 9 Rev B IPMCs in our testing lab now, with
the first one of them fully checked out prior to authorizing the
assembler to run the remaining 8 PCBs. Firmware development
continues on the UW-IPMC (ZYNQ-7000 based processor mezzanine to
provide Linux-based control), and the PO for Rev B UW-IPMC PCB has
been issued for 25 units of which we will initially assemble
20.
L1 Calorimeter Trigger algorithms:
The GCT algorithm, which completes clustering started in the
RCT-level and classifies them as electromagnetic or hadronic in
origin, has been reimplemented learning from the initial version to
optimize multiple levels of loops. The change was necessary to cope
with changes in the Xilinx Vivado HLS package. The new code meets
the timing requirement. A reduced version is ready for testing on
the CTP7 test stand at CERN. The classification process is followed
by identification of highest transverse energy objects,
electron/photons, taus, jets, sum of higher transverse energy
objects (HT) and energy imbalance in transverse plane (MET) based
on calorimeter information only. Note that the RCT code, which does
initial stages of clustering without sharing across the boards, has
been completed and tested. Further improvements to edges of the
geometric acceptance of the cards is underway.
L1 Correlator Trigger algorithms:
Studies updating the particle flow algorithm resource usage with
the latest version of Xilinx Vivado HLS found an optimization for
higher logic resource usage than previous versions. The follow-up
action item is to study how to improve resource control in HLS. A
first implementation of the particle flow and puppi algorithms on
Amazon Web Services (AWS) FPGA instances was demonstrated. The AWS
demonstration has lower prototyping costs and includes examples on
how to interface with the cloud FPGA instances. A study of the
vertex finding matching window effect on the particle flow + puppi
HT calculation found an optimal window around 0.5 cm. Progress
continues on the track-muon linking demonstrator including a
displaced muon algorithm with track-unmatched muons. Input
injectors for the demonstrator are complete and firmware (HLS)
algorithms of both track propagation and track matching have been
developed and are in testing. A first comparison of standalone tau
identification algorithms (hadron-plus-strips) against tau
identification with particle flow inputs was performed. Similar
performance is achieved except in the 3-prong tau case where
particle flow shows an inefficiency. The follow-up action item is
further investigation into collimated charged hadron efficiencies.
Particle flow taus are natively implemented in the endcap while
work is in progress for the standalone tau algorithms. First
studies of the HGCal trigger inputs to the correlator are studied
for electron identification in order to improve performance and a
first study of multivariate classifiers was presented. Performance
studies of the current particle flow electron identification
algorithm over all eta was performed showing 20-30% efficiency loss
due to lost tracks from the input tracking trigger. Currently this
is recovered with standalone calorimeter algorithms but loosening
track requirements and bremsstrahlung recovery has been identified
as next steps.
Milestones Progress for 2018:
Milestone: Emulated performance and efficiency for Jets/MET/HT,
October 2018
Progress since last report: The GCT algorithm, which completes
clustering started in the RCT-level and classifies them as
electromagnetic or hadronic in origin, meets the timing requirement
and a reduced version is ready for testing on the CTP7 test stand
at CERN. The classification process is followed by identification
of highest transverse energy objects, electron/photons, taus, jets,
sum of higher transverse energy objects (HT) and energy imbalance
in transverse plane (MET) based on calorimeter information only.
The project is on target to complete this milestone on time for
both the standalone calorimeter trigger and the correlator trigger
algorithms.
Milestone: Design complete for APD1 board, July 2018
Progress since last report: The layouts of the APd1 and its Rear
Transition Module (RTM) PCBs are nearing completion with the last
few routes being optimized and the overall routing being reviewed,
and parts are being ordered.
Milestone: APd1 Produced, 30 September 2018:
Progress: Wisconsin have 9 Rev B IPMCs in their testing lab now,
with the first one of them fully checked out prior to authorizing
the assembler to run the remaining 8 PCBs. Firmware development
continues on the UW-IPMC (ZYNQ-7000 based processor mezzanine to
provide Linux-based control), and the PO for Rev B UW-IPMC PCB has
been issued for 25 units of which we will initially assemble
20.
Change log for "Float to CMS Need-by-date", risk register, IPT
tracking:
None
WBS 402.8: MIP Timing Detector (Acting L2 Manager: Chris Neu, U.
of Virginia)
The MTD continues to progress in its technical areas as well in
the areas pointed out to be lacking during the June 2018 CD-1 IPR
Review.
Several technical milestones have been achieved on schedule:
e.g. the specification for the BTL Concentrator Card was finalized,
a decision was taken on the power provision to the BTL, the
procedures for BTL assembly of trays and ETL assembly of modules
have matured significantly and agreed upon by the international
collaboration, etc.
The WBS structure and organization chart have been updated and
refined to reflect the increased clarity in the areas of activity.
Preliminary versions of BoEs for every L4 area have been collected
and the process of building scenarios for the scope of US-MTD
deliverables has begun.
The outline for the CDR document has been updated to reflect the
project organization and areas of activity within the US. Overall
co-editors for the CDR have been identified and weekly meetings
with the content providers have begun.
A Cost/Schedule/Risk Workshop was held at CERN 23-24 August.
Risks for all L4 areas have been enumerated and the process of
entering these into the Risk Register has begun. Risk mitigation
strategies continue to be pursued, including the feasibility and
cost impact of an optimized redesign of the ETL.
We are preparing for a Technical Review in mid-November 2018 and
for a Cost/Schedule/Risk Review in mid-December 2018.
Details:
The MTD project 402.8 has made significant progress in the last
month. The most important task at this time is the definition of
the US scope. Input to this decision is cost, schedule and risk
evaluations from every L4 activity area. L4 managers have been
working on gathering this information for the last 8 weeks. This
has culminated in the collection of preliminary BoEs for every
activity area, a major milestone.
With this information in hand Frank Chlebana, Chris Neu and Jeff
Spalding are working on building different scenarios for the US MTD
scope. Given funding constraints, not every L4 activity area will
necessarily by part of the ultimate suite of deliverables provided
by the US. All US stakeholder institutes are represented in this
exercise, and the US MTD community understands the constraints. A
US MTD All-Hands meeting will be held Sept 5-6 at FNAL where a
proposal will be defined for the US scope. This proposal will
subsequently be presented to the USCMS HL-LHC Advisory Board (Max
Chertok, chair), interior to which a dedicated MTD advisory
sub-committee has been formed. The final definition of the US MTD
scope is expected by mid-October.
Other aspects of the requisite documentation are maturing as
well. Risks for each L4 activity area have been enumerated and the
process of formalizing these risks in the Risk Register has begun.
Developing the RLS in P6 has started with the major internal and
external milestones; now with the preliminary BoEs in hand, the
plan is for every L4 manager to work with Lucas Taylor, Bill
Freeman and Chris Neu to enter their activities into P6. A series
of dedicated P6 workdays are planned for the next two weeks. We
expect to have first full versions of BoEs, RLS in P6 and risks
formalized in the RR by the end of September. We expect subsequent
iterations as our scope comes into better focus, but we are on
track to have full versions frozen and ready in advance of the OPSS
Cost/Schedule/Risk Review, scheduled for mid-December.
The CDR document is maturing in parallel to all of the
activities described above. A re-organization of the document has
been completed and content contributors have been defined for every
subsection. Two overall co-editors (Lothar Bauerdick and Toyoko
Orimoto) have been identified. The co-editors have instituted
weekly meetings among the content contributors to discuss and
present progress; the associated milestone is to have v0 of the CDR
distributed to the US MTD community for comments by 1 October. A
subsequent v1 will be prepared by 1 November, two weeks before the
MTD Technical Review, scheduled for mid-November.
Several key technical milestones have been achieved on schedule.
The specification for the BTL Concentrator Card (402.8.3.3) was
finalized, allowing for an accurate costing exercise for this
component. An important point was the settling of the issue of the
power provision to the BTL SiPMs, which will be handled through the
BTL FE board, a non-US deliverable. Also, clarification was
received from the international MTD project on the requirements for
the materials for on-detector PCBs. The procedures for both the BTL
assembly of trays (402.8.3.4) and ETL assembly of modules
(402.8.4.4) have matured significantly in the last 4 weeks and
working versions of these models, developed by key US leaders in
this effort (Adi Bornheim and David Stuart) and agreed by the
international project, are now in hand.
The issue of risk mitigation in the ETL continues to progress.
In the reference design for ETL, the highest risk exposure areas
are (1) the potential for additional engineering design cycles and
engineering labor associated with a design that meets the stringent
time resolution requirement of the FE ASIC and (2) the possibility
of low wafer yield of functional large-size LGAD sensors. It was
shown that in an optimized redesign of the ETL with smaller
sensors, the per-wafer silicon use-efficiency increases by a large
amount hence reducing the exposure to cost risk due to wafer
losses. However, the smaller sensors introduce additional
inter-sensor gaps, requiring a second measurement layer to maintain
high per-track efficiency. Such a 2-measurement design would not
only preserve efficiency but also help mitigate the risk associated
with the FE ASIC, since in such a design the ASIC time resolution
specification is relaxed, which lessens the demand on the
engineering. Further, a small-sensor 2-measurement ETL would allow
for continuing and deepening our collaboration with the ATLAS HGTD
group on sensors and FE ASIC. An important final point: It was
shown in the MTD General Meeting on 24 August that the silicon
sensor cost for such a design would be within 5% of that of the
current reference design, mostly due to more effective wafer
utilization and sensor placement optimization. Further studies
continue on the cost implications on other system aspects, as well
as the feasibility in terms of provision of services and cooling
for such a design. A decision on the viability of the redesigned
small-sensor 2-measurement ETL is expected in early October,
involving the international MTD project and international CMS.
We have to wait for the cost and schedule information to be
assembled and evaluated before we can decide if the MTD can be
ready for baselining together with the rest of the Project in
2020.
ISSUES
Major issues are to bring the MTD up to a CD-1 level of
maturity, to fully define the US scope for the MTD, and to ensure
the overall project fits within the DOE funding profile.
We continue to address the comments and recommendations of the
June IPR and to raise issues to Lab management with respect to
procurement resources, ASIC engineering resources, the need of an
irradiation facility, etc. Weekly meetings between the project and
the procurement department are continuing. They have succeeded in
moving procurements along and have helped the procurement
department understand the needs of the project. Monthly ASIC PMGs
are continuing as well. The project manager and relevant managers
for the HL-LHC CMS ASICs attend these monthly PMGs. For the
irradiation facility, the project is getting regular updates from
lab management. After discussions between Lab Management and the
stakeholders an official proposal for an Irradiation Test Area ITA
at the end of the Linac has been submitted to the Lab Management on
August 20, 2018.
Within international CMS, the current need by dates leave
roughly on average about 9 months of float, which is small compared
to the size of the project. We monitor these dates monthly (see the
beginning of the report).
We have assigned people responsible for all items in the IPT
tracking document and will be reporting on them as we update the
schedule or close them out. Twenty-one of the recommendations are
expected to be closed before the for the CD-1 mini-review. We have
closed one of them this month. Table 3 below summarize the IPT
tracking areas, breaking them down by scheduled close dates on the
left, and breaking them down by responsible people on the right. We
will be updating these tables each month.
Table 3: Table of IPT recommendations by (left) scheduled close
date and (right) project responsible.
COST AND SCHEDULE SUMMARY
The financial status (DOE part only) is shown in Table 4.
The DOE OPC table contains the first $500k that was reprogrammed
from operations during CY16 and the second $4M which arrived in
FY17. Note that the small amount in the FPIX row is for work
performed at Fermilab for FPIX and should be reimbursed by Cornell
/ NSF. We are still missing $564k funding from DOE to receive our
full $12M funding for FY18 (see last line of table).
Table 4: DOE OPC Budget vs Actual Summary as of July 31,
2018.
The guidance / profile we have been working toward is shown
below. This is the guidance we were asked to use in June, shortly
after the CD-1 review.
Fiscal Year
FY 17
FY 18
FY 19
FY 20
FY 21
FY 22
FY 23
FY 24
FY 25
Total ($M)
Total Project Cost ($M)
$4
$12
$17.5
$24.5
$27.05
$30.7
$25.
$17.8
$3.5
$162.05
In the meantime, on June 14, 2018 we received a revised profile
for the out-years under the assumption that the House and Senate
marks will be confirmed in the final FY19 appropriation. By the
mini-review the goal would be to fit within this profile.
Fiscal Year
FY 17
FY 18
FY 19
FY 20
FY 21
FY 22
FY 23
FY 24
FY 25
Total ($M)
TEC
$27.50
$30.45
$30.45
$21.30
$18.00
$7.85
OPC
$6.70
$2.00
$0.80
$0.50
Total Project Cost ($M)
$4
$12
$27.5
$30.45
$30.45
$28
$20
$8.65
$0.5
$161.55
For the time being we are updating two different cost rollup
spreadsheets using the two different guidance/profiles.
We continue to refine the cost estimate of the MIP Timing
Detector. Our current plan is to keep our contribution to the MTD
to $12M (fully loaded) in order to stay within the overall cost
envelope. We have been holding a series of workshops and meetings
to get a crisper cost estimate and profile in order to define the
US MTD deliverables in this cost envelop scenario. We expect to
reach a decision on the US scope in October 2018. In the meantime
we will continue compiling profiles with and without the MTD
detector as it cost matures in order to assess the level of
contingency and risk to the Project completion.
2