ATLAS Project Document

Document Sample
ATLAS Project Document Powered By Docstoc
					                                               PRR of the MDT ASDAMT chip. Report

ATLAS Project Document. No.                 Institute Document No.             Created : 07-sep-0208 July Page        1 of 7
                                                                                                02
    ATC – RM – ER – 00221                                                      Modified: 03/10/1225/09/0 Rev.No.       11
                                                                                         219/09/0204/09/
                                                                                           0203/09/02



                         Report of the Production Readiness Review




                ATLAS MDT read-outASD chip (AMT)




                                                             Abstract
                        The PRR of the MDT read-outASD chip (AMT) board has been held at CERN
                        on AugustJune 230th 2002. The chip is ready for productionA few minor
                        modifications have to be introduced and extensive tests of the modified chip
                        must be done before starting the production.




                 Prepared by :                             Checked by :                        Approved by :

           Ph. Farthouat, Cern                       F. Anghinolfi, CERN                         M. Nessi
                                                    J. Christiansen, CERN
                                                   M. Dentan, CERN/Saclay
                                                    Ph. Farthouat, CERN
                                                       P. Jarron, CERN
                                                        R. Richter, MPI
   for information,
                             Ph. Farthouat
                                                        Tel.
                                                      W. Riegler, CERN Fax.                            E-Mail
   you can contact :                              +41.22.767 6221       +41.22.767 8350          Philippe.Farthouat
                                                                                                      @cern.ch
Distribution:          EB Members, MDT Members, all participants mentioned in the report.
                                                           ATLAS Project Document. No.     Page       2 of 7
                                                                                           Rev. No.     2


PURPOSE OF THIS REVIEW
          The purpose of this review was to check that the production of the AMTSD chip can be launched.


MEMBERS OF THE REVIEW COMMITTEE
    Review Committee                                       MDT AMT Team
    F. Anghinolfi, CERN                                    Y. Arai, KEK
    J. Christiansen, CERN                                  T. Dai, Michigan
    M. Dentan, CERN/Saclay                                 T. Kondo, KEK
    Ph. Farthouat, CERN
    R. Richter, MPIP. Jarron, CERN
    W. Riegler, CERN

    Ex officio                                             For Information
    M. Nessi, CERN
    G. Mikenberg, CERN/Weizmann
    G. Herten, Freibourg
    R. Richter, MPI


AGENDA
          The agenda and the documentation can be found at: http://documents.cern.ch/age?a021111
                 1.Specifications of AMT (Yasuo Arai)                                                          Formatted: Bullets and Numbering

                 2.Description of Circuit (Yasuo Arai)
                 3.Design Verification (Yasuo Arai)
                 4.Circuit Test results
                         a.Single chip test (Yasuo Arai)
                         b.Mezzanine card test (Tiesheng Dai)
                 5.Radiation tests results (Yasuo Arai)
                 6.Production schedule, and QC procedures (Yasuo Arai)
          The documentation is available at : http://atlas.kek.jp/tdc/PRR/


REVIEW OUTCOME
          Design
          The presented peaking time, linear range, bipolar shape and Wilkenson ADC parameters perfectly
          match the specified values.
          The programmable deadtime, which is used to get rid of multiple hits from the muon signals, was
          specified to be programmable up to the maximum drift time of the baseline MDT gas Ar/CO2 93/7
          which was measured to be 700ns. The maximum deadtime number of 700ns presented at the
          review therefore satisfies this requirement.
          Threshold variations of about 5mV r.m.s. were quoted from a measurement on the octal ASD chip.
          Cutting at +-10mV in the production testing procedure gives a yield of 90%. This means that
                                                 ATLAS Project Document. No.         Page       3 of 7
                                                                                     Rev. No.     2


instead of the 60mV (6fC,20 primary electons) nominal threshold one finds a threshold of 50-
70mV (5-7fC, 17-23 primary electrons). This variation is well within specifications.
The question of power dissipation in the cavern is still being studied. In case a cooling system is
required it will impact the design of the front-end board (mezzanines) but not the ASD itself.The
limited readout speed of 40 Mbits/s is the only limiting factor in the use of the AMT.
It seems that it is only considered to use the AMT in triggered mode; the AMT can also be used in
a non-triggered mode which can be very useful during testing and verifications. The muon group
should decide if they also want to support this mode during testing and calibration.
The data driven architecture of the AMT makes it very flexible ( programmable latency,
overlapping triggers, triggered or non triggered, etc.) but also makes the internal function of the
TDC quite complicated. The late discovery of the search pointer bug clearly illustrates this and a
serious testing of the chip for extended periods under different conditions must be made before
starting production. Extensive simulations must be made after correcting the known bugs to insure
that new bugs are not introduced.
The error reporting bug should be corrected as nothing is more confusing than getting wrong error
messages. The correction of this should be simple and not imply a significant risk of introducing
new bugs.
The production of the AMT must not start before the new version has been extensively tested.


Radiation hardness assurance
Pre-selection tests were thoroughly carried out in accordance with the ATLAS standard test
methods, and a precise PRR report was provided to the ATLAS Technical Coordination.
TID tests : the required RTCtid of 223Krads for preselection is satisfied by test carried up to
300krads total dose. No significant change in parameters as well a current consumption
(technology is HP-Agilent 0.5um).
SEE test : the required RTCsee of 1.3x1012 for preselection is satisfied by test carried up to
4.4x1013 accumulated fluence.
It is required that each of the production lots are evaluated against TID up to 112Krads.                Formatted

Pre-selection tests were thoroughly carried out in accordance with the ATLAS standard test
methods, and a very precise PRR report was provided to ATLAS Technical Coordination.
Unfortunately, radiation levels taken into account for radiation tests were excerpted from obsolete
tables given average values in ATLAS macro-domains. Some domains of MDT electronics range
from small radius to large radius. In small radius regions, local radiation constraints are much
higher than the average value of the full domain. For instance, SRLtid in the inner regions of MDT
EC2 (SRLtid = 5,1 Gy/year at R=200 cm) is about 7 times the average SRLtid computed for the
whole MDT EC2 domain (0,74 Gy/year on average between R=200 cm to R=1100 cm).
To properly achieve the pre-selectionqualification step, the ATLAS RHA Coordinator asks the
MDT collaboration to complies with the following requirements (see details in Appendix I):
1/ Before launching the production and asAs part of the tests of the modified chip, the MDT
collaboration must should perform additional SEE tests in order to get statistics large enough to
extrapolate results to the whole MDT System, and make sure that these extrapolated results are
acceptable at system level. The requirement is to attain a minimum accumulated fluence of
1E13p/cm2. This goal can be achieved with for example 10 chips receiving each a fluence up to
1E12p/cm2 (thus limiting the TID per chip to ~260Krads).
                                                       ATLAS Project Document. No.         Page       4 of 7
                                                                                           Rev. No.     2


        2/ Before launching the production, the The MDT collaboration must should perform additional
        TID tests up to the worst case SRLtid in order to confirm the value of the threshold TID from
        which Idd starts to increase rapidly (preliminary tests performed on a too small number of samples
        without annealing nor aging showed a threshold of about 500 Gy). The requirement is to test 10
        samples of the new version up to minimum 75Krads.
        For the mass production, the ATLAS RHA Coordinator asks the MDT collaboration to complies
        with the following requirements:
        3/ When launching the production, the The MDT collaboration must purchase a quantity of spares
        large enough to allow the replacement of the chips as soon as the expected TID at their effective
        location (all SF included) will reach the threshold TID from which Idd starts to increase.
        4/ All the AMT chips necessary to the MDT System (spares included) will be produced in one
        single batch. This batch must be qualified with respect to radiation as explained in Appendix I of
        this report.
        The MDT team has to makeis assuming sure that the replacement of the electronics parts in the
        most irradiated zone of the end-cap is will be feasible.


        Production and QC
        The production quantities (including spares) are well defined and the expected yield value is
        conservative enough to avoid bad surprises.
        The production tests and cuts are well defined. The infrastructure is in place (including the data
        base).
        In order to evaluate the possible problems of reliability, it is recommended to perform and
        accelerated lifetime test.


RECOMMENDATION
        The production of the ASD chip can proceed.
        The schedule presented for bugs fixing (August 2002), delivery of prototypes (mid-October 2002)
        and extensive test of prototypes (2 months) is valid.
        The amount of chips to be ordered (20000) compared to the needed quantity (16000) is large
        enough to cover all the needs during the lifetime of the experiment.
                                                            ATLAS Project Document. No.        Page       5 of 7
                                                                                               Rev. No.     2



APPENDIX I : RADIATION HARDNESS ASSURANCE ISSUES
                                                                                                                   Formatted: Bullets and Numbering
 1.1 PRE-SELECTION TESTS
         The official radiation constraints requested for pre-selection tests are:


         Worse location                     Constraint                    Value
         (Zmin, Zmax, Rmin, Rmax)
         EC2 (1340, 1350, 190, 200)         RTCtid                        2230 Gy (10 years, SFsim=3.5,
                                                                          SFldr=5, SFlot=2)
         EC1 (730, 740, 210, 220)           RTCniel                       1.6E13 (10 years,        SFsim=5
                                                                          SFlot=2)
         EC1 (730, 740, 210, 220)           Maximum hadron                2.7E12 (10 years, , SFsim=5
                                            fluence for SEE tests         SFlot=2)


         The effective radiation constraints applied during pre-selection tests are:


         Constraint                            Value
         RTCtid                                3000 Gy (10 chips)
         RTCniel (facultative)                 Not tested
         Hadron fluence for SEE tests          4.4E13h/cm2 (effective fluence during SEE measurements, 4
                                               chips)


                                                                                                                   Formatted: Bullets and Numbering
 1.2 QUALIFICATION TESTS
         The official radiation constraints requested for qualification tests are:


         Worse location                     Constraint                    Value
         (Zmin, Zmax, Rmin, Rmax)
         EC2 (1340, 1350, 190, 200)         RTCtid                        1120 Gy (10 years, SFsim=3.5,
                                                                          SFldr=5, SFlot=1)
         EC1 (730, 740, 210, 220)           RTCniel (falcutative)         0.8E13 (10      years,   SFsim=5,
                                                                          SFlot=1)
         EC1 (730, 740, 210, 220)           Maximum hadron                1.3E12 (10      years,   SFsim=5
                                            fluence for SEE tests         SFlot=1)
                                            (falcutative)
                                                  ATLAS Project Document. No.           Page       6 of 7
                                                                                        Rev. No.     2




1.1Pre-selection tests                                                                                      Formatted: Bullets and Numbering

The official radiation constraints requested for pre-selection tests are:


             Worse                               Constrai                     Value
             location                            nt
             (Zmin, Zmax,
             Rmin, Rmax)
             EC2         (1340,                  RTCtid                       715 Gy (10 years, all
             1350,         190,                                               SF included)
             200)
             EC1 (730, 740,                      RTCniel                      3.2E13 (10 years, all
             210, 220)                                                        SF included)
             EC1 (730, 740,                      Maximu                       5.3E12 (10 years, all
             210, 220)                           m hadron                     SF included)
                                                 fluence
                                                 for SEE
                                                 tests


The effective radiation constraints applied during pre-selection tests are:


             Constraint                            Value
             RTCtid                                200 Gy (11 chips) and 2 kGy (few chips)
             RTCniel                               1E13n/cm2 (8 chips) and 1.6E13n/cm2 (4 chips)
             (facultative)
             Hadron fluence                        6E11h/cm2 (effective       fluence    during    SEE
             for SEE tests                         measurements, 4 chips)


TID test results and conclusions
TID pre-selection tests were carried out in accordance with the ATLAS extended test method for
TID pre-selection tests, except the applied TID constraint. Eleven chips were irradiated up to 200
Gy and were found within the acceptance criteria at the end of the irradiation as well as after
annealing and after aging. Several chips tested up to more than 2 kGy show no significant failure
up to ~ 500 Gy, but no annealing nor aging was performed after this threshold TID was reached.
Their current consumption Idd start to increase rapidly from ~ 500 Gy onwards. More information
must be provided to the PRR reviewers on the spread of the threshold from which Idd start to
increase rapidly. How many devices were irradiated up to 2 kGy? Eleven chips must be tested up
to more than 500 Gy and followed by annealing and aging in order to confirm the threshold from
which Idd starts to increase. When a threshold TID will be strongly established, it will represent
the maximum TID acceptable for MDT chips (all SF included).


NIEL tests results and conclusions:
                                                 ATLAS Project Document. No.         Page       7 of 7
                                                                                     Rev. No.     2


Althought NIEL tests are not requested for pure MOS chips, neutron pre-selection tests were
carried out in accordance with the ATLAS test method for NIEL pre-selection tests, except the
applied NIEL constraint. Eight chips were irradiated up to 1E13 n/cm2 and five chips were
irradiated up to 1.6E13 n/cm2 (~ 0.5 RTCniel). After deactivation, not any apparent change was
observed. This test being facultative, additional NIEL tests are not requested.


SEE test results and conclusions:
SEE pre-selection tests were carried out in accordance with the ATLAS extended test method for
SEE pre-selection tests, except the applied proton constraint. Four chips were irradiated up to
6E11 p/cm2 (~ 0.05 times the maximum fluence to be applied during SEE tests) and were
measured on-line during irradiation. One SEU was observed on one chip only. According to the
ATLAS policy, if the test was performed up to two times the maximum fluence expected during 10
years of operation (i.e. 1E13 p/cm2), this result would be sufficient to conclude. However, due to
the mistake on the foreseen hadron constraint, this is not the case. Additional tests must be
performed in order to get relevant statistics. These tests must be pursued up to a maximum fluence
equal to two times that expected in 10 years of operation (all SF included) at the worse location.
During the irradiation, test chips must be replaced each time they die from TID effects. The test
can be stopped at a lower fluence if relevant statistics are obtained.
1.2Qualification tests                                                                                   Formatted: Bullets and Numbering

1/ Quantities: The ATLAS RHA Coordinator asks MDT collaboration to purchase a quantity of
AMT spares large enough to allow the replacement of chips as soon as the expected TID at their
effective location (all SF included) will reach the threshold TID determined during pre-selection
step (~ 500 Gy, to be confirmed). For example, in the most inner regions of EC2 (at R=200),
assuming a TID tolerance of ~ 500 Gy, AMT chips would have to be replaced after ~ 7 years.
Practically, only chips located between 200 and 240 cm will be subjected to more than 500 Gy;
their replacement could be decided during ATLAS operation on the basis of measurements of Idd
increase.
2/ TID qualification: Eleven chips excerpted from the production batch must be tested according
to ATLAS standard TID test method, up to the threshold TID determined during pre-selection tests
(~ 500 Gy, to be confirmed). These tests, if successful, will qualify the production batch of AMT
chips for use up to that radiation level.
3/ SEE qualification: Four samples (or more if necessary to get relevant statistics) chips excerpted
from the production batch must be tested up to a maximum proton fluence of two times that
expected in the worse location (all safety factors included). During the irradiation, test chips must
be replaced each time they die from TID effects. The test can be stopped at a lower fluence if
relevant statistics are obtained.

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:0
posted:10/3/2012
language:English
pages:7