Transcript ELQASpliceReviewTalkx

ELQA Campaigns for SMACC
TE/MPE/EE
LHC Machine
CONTRIBUTIONS BY ELQA
prior to
/ during
/ at the end of LS1
K. Dahlerup-Petersen TE/MPE/EE – 5 November 2012
Presentation to the 3rd LHC Splice Review
12-14 November 2012
by K. Dahlerup-Petersen -on behalf of the complete ELQA team: M.Bednarek, G. D’Angelo, R. Mompo, G. Seweryn,
S. Pemberton, M. Dominguez-Martinez, E. Nowak, P. Dubert, R. Kulaga and K. Pereira.
ELQA Campaigns for SMACC
TE/MPE/EE
LHC Machine
K. Dahlerup-Petersen TE/MPE/EE – 5 November 2012
Agenda:
•
General overview of Electrical Quality Assurance Activities related to LS1
•
PAQ - objectives for the SMACC-related ELQA campaign
•
Details of a Partial Assembly Qualification with respect to LS1
• For synchronized follow-up of the busbar consolidation
• For replacement of 15 + 4 main cryo magnets
• Safety aspects.
•
•
•
•
•
Short presentation of the new TP4 measurement system.
The AIV procedure.
Resources: Equipment and Manpower.
Human resources related to the LS1 ELQA campaigns.
Revision of HVQ test voltages:
• Motivation for applying changes
• The new values
Conclusion
•
•
The new, distributed voltage-to-ground monitor: Can it be used for early detection of
weak insulation points?
ELQA Campaigns for SMACC
TE/MPE/EE
LHC Machine
Electrical Quality Assurance is required in basically all phases of the LS1project:
 @ COLD @ THE END OF 2012-13 LHC OPERATION:
1) Verification & Consolidation of non-conformities of different origins in circuitry and
components - during the period of Powering Tests
 quenches at low current, open circuits, shorts, low voltage withstand, missing magnets
- 10 cases in 5/ 8 LHC sectors to be investigated/cured
K. Dahlerup-Petersen TE/MPE/EE – 5 November 2012
2) ELQA Standard Campaign TP4 E - with new HV parameters and cryo instr. disconnected
incl. the MIC-C & DOC-C tests
- TP4 E complete Q.A. verification @ cold (incl. transfer function and H.V.)
- MIC-C individual magnet checks, incl. resistance measurements and H.V. test
of coil, busbars and all quench heater circuits ( 6’000 units) @ cold
- DOC-C resistance measurements of coil, transfer function, H.V. and instrumentation
on all dipole orbit corrector circuits @ cold
 All this according to the updated test procedures – EDMS doc. 788197 –rev. 1.2 (April 2012)
 DURING WARM-UP:
- Continuous checking of short-circuits to ground (ELQA) and broken voltage taps (QPS)
 QPS/ELQA ACTIVITIES @ WARM PRIOR TO THE SPLICE CONSOLIDATION:
1)
ELQA Standard Campaign TP4 A+B
incl. MIC-W and DOC-W
2) Resistance measurements of all interconnecting busbar segments with 5-30 A –
200 V powering , - use of mDQQBS boards and nQPS data handling.
On the 3 Main circuits in all 8 sectors (2 sectors/day in average)
ELQA Campaigns for SMACC
TE/MPE/EE
LHC Machine
Electrical Quality Verification during LS1 (contd.)
 ELQA ACTIVITIES DURING LS1:
1) AIV (Arc Interconnection Verification ) applied for magnet replacement
-concerns all Spool circuits and all N-line circuits: integrity, polarity and HV tests
2) PAQ (Partial Assembly Qualification) for Main busbars and Spool circuits: electrical
continuity proof (Spools) and H.V. Qualification (Mains & Spools) – related to SMACC
- Non-conformity follow-up.
- Test Procedures: AIV unchanged, PAQ updated with specific requirements for LS1.
 See next slides
K. Dahlerup-Petersen TE/MPE/EE – 5 November 2012
 @ WARM AFTER COMPLETION OF THE SPLICE CONSOLIDATION AND THE
REPLACEMENT OF 19 CRYO MAGNETS:
1) Complete ELQA Standard Campaign TP4 A & B, MIC-W, DOC-W
2) Repetition of the warm resistance measurements of all interconnecting busbar segments
with 5-30 A powering - identical to the exercise made before the LS1 interventions.
 DURING COOL-DOWN:
TP4 C and QPS continuous monitoring of possible short-circuits and integrity of the voltage
tap connections
 @ COLD - PRIOR TO POWERING FOR HARDWARE COMMISSIONING:
Complete ELQA campaign TP4 E, MIC-C, DOC-C - based on the updated procedure
ELQA Campaigns for SMACC
TE/MPE/EE
LHC Machine
Particular emphasis on the ELQA Follow-up of the MSC Splice Consolidation ‘Train’:
Objective:
K. Dahlerup-Petersen TE/MPE/EE – 5 November 2012
To assure a prompt verification of the relevant electrical and di-electrical properties
of the dc powering circuits contained in the M1, M2 and M3 lines, i.e. the three
main powering circuits and the spools, during the consolidation period.
This added value is provided through a daily (nightly!) ELQA campaign (PAQ) of
circuit verifications, on-line transmission of the test results and collection of a
maximum of information for facilitating localization and correction of disclosed
non-conformities.
ELQA Campaigns for SMACC
TE/MPE/EE
LHC Machine
 The PAQ test
-Two-step procedure introduced during the initial installation and connection
of the LHC for application to M1, M2 & M3 Main busbars and M1, M2 spool
conductors. One TP4 mobile test bench is used per PAQ test.
-Set of electrical tests applied to a powering segment, originally corresponding
to one half-cell (3 or 2 (DS) dipoles + 1 SSS), however, other segments can
be defined and treated.
K. Dahlerup-Petersen TE/MPE/EE – 5 November 2012
-Procedure now adapted to the particular conditions of the less systematic
splice consolidation work of LS1 and documented in the document
‘LS1 PAQ Test Procedure’ (soon on EDMS). H.V. test voltage will be 500 Vdc.
-Required temporary conditions to allow correct application of the PAQ
procedure:
1) Disconnection of the warm DC cables & the instrumentation cables @
the level of the DFB/CL’s
2) Temporary interruption of each of the 2 x 10 Spool piece busbars @
an interconnection close to the odd-point end of the chain.
3) Temporary removal of the conductor insulation from the RB, RQF, RQD
busbars @ the same magnet interconnection.
ELQA Campaigns for SMACC
TE/MPE/EE
LHC Machine
 The PAQ test (contd)
-The places of temporary spool rupture and Main busbar insulation removal
have been coordinated with MSC-group so to minimize impact:
Point 1 & 5 left and right: The IC corresponds to the place of RQF/RQD busbar
shortening. NOTE: It is not necessary to check ‘upstream’ of this point as there
will be no risk of damage as no repair work there.
Point 3 right: The IC is at the Q7 quad which is listed for replacement.
In the remaining points (P7 left and right as well as P3 left) the open place
is also near Q7 or Q8.
K. Dahlerup-Petersen TE/MPE/EE – 5 November 2012
LHC sector
Temporarily open interconnect for ELQA
S12
QEQI.11R1
S23
QBQI.8L3
S34
QDQI.7R3
S45
QQEI.11L5
S56
QEQI.11R5
S67
QBQI.8L7
S78
QQBI.7R7
S81
QQEI.11L1
ELQA Campaigns for SMACC
K. Dahlerup-Petersen TE/MPE/EE – 5 November 2012
TE/MPE/EE
LHC Machine
Other requirements:
- At all points undergoing repair work all busbars must be individually insulated wrt
each other and to ground at the moment (17H00 daily) when the ELQA team takes
over the sector.
Only exceptions are the points in the above table, which shall be opened prior to
the beginning of the consolidation train and remain open until they, as the last IC’s,
will be closed.
Furthermore, also as from 17H00 the busbars must be left for continuous conduction
of the test currents.
The consolidation train will be in charge of installation of the agreed temporary insulation
and the circuit continuity measures at all opened interconnects at the end of the
working day.
ELQA Campaigns for SMACC
TE/MPE/EE
LHC Machine
 The PAQ - safety aspects
-Daily @ 17H00 the ELQA team will carry out a patrol of the complete sector(s)
or subsector(s) where the Q.A. campaign will take place on that evening.
People, not related to the ELQA activity, will be asked to leave the sector.
Also the AIV tests will require patrol and guard.
-Warning signs (flash lights) will be placed and activated at the two extremities
of the sector(s) where H.V. qualification will take place.
K. Dahlerup-Petersen TE/MPE/EE – 5 November 2012
-One ELQA team member will be positioned at each extremity of the test
sector(s) or subsector(s) for the complete duration of the H.V. testing.
They will assure that only members of the ELQA team can enter during the
H.V. tests.
-During the patrol, the team will verify that the busbars of all IC’s under tests
are (at least temporarily) insulated.
-At the end of the tests the WISH tool will be used by ELQA to transmit the
global signature. The pending ELQA steps will be cleared if tests were successful.
All disclosed non-conformities will be forwarded to the project coordination
office.
ELQA Campaigns for SMACC
TE/MPE/EE
LHC Machine
The new TP4 measurement system:
-A joint venture between MPE and the Institute HNINP, Cracow
K. Dahlerup-Petersen TE/MPE/EE – 5 November 2012
features:
-Source of stable current for ohmic resistance measurements obtained through voltage
source with thermally stabilized serial resistor.
-Temperature regulation based on -processor connected to an on-board computer.
Two-stage regulation: ‘internal’ with heaters in thermally-insulated housing, externally by
proportional regulator with heaters and fan. Stability typically 0.1 deg.
Feed-forward system for stabilization of resistor temperature (known resistor power).
-4-wire resistance measurements
-Voltage measurements with 84 inputs / multiplexing.
-Two PXI DMM cards have replaced the Keithley DVM’s. Advantage: absolutely simultaneous
pairs of measurements.
-Transfer function measurements with choice of reference impedance.
-Labview software with LHC circuit data stored. Test results stored locally – transferred regularly
to Oracle EDMS database.
-All instruments are mounted on trolleys so to constitute 8 systems. 2 spare systems are
available without trolleys. New 3 kV, 4 mA H.V. generators are ordered.
-The systems can be powered from 24 VDC battery power of the Pefra tractor or from the
Mains, now through a UPS which provides 30 minutes autonomy.
K. Dahlerup-Petersen TE/MPE/EE – 5 November 2012
ELQA Campaigns for SMACC
TE/MPE/EE
LHC Machine
ELQA Campaigns for SMACC
TE/MPE/EE
LHC Machine
 The AIV test
AIV is a Labview-based procedure for continuity & polarity checking, combined
with H.V. qualifications (@ 1 kV), for the LHC arcs and DS areas.
-Highly automated procedure with recognition of type of corrector magnet and
polarity through voltage tap measurements upon injection of dc current.
-Application: Spool circuits and the s.c. corrector circuits of the ‘N’-line.
-Was originally developed for the initial machine assembly , NOW used in case
of magnet replacements. Results are daily communicated to the Project Office.
K. Dahlerup-Petersen TE/MPE/EE – 5 November 2012
-Well-proven procedure, successfully used at several occasions (S34, S12)
-Existing test procedure can be directly applied to LS1 magnet replacements.
-AIV for ‘N’-line is applied in two steps:
1) With 46-wire cable inserted - before ultrasonic welding
2) After completion of the US welding
ELQA Campaigns for SMACC
TE/MPE/EE
LHC Machine
 The AIV test (contd.)
AIV test procedure generally requires electrical access to 4 Interconnects around the
removed and replaced ‘N’-line (in yellow) of the half-cell in which a cryo magnet is
being replaced.
K. Dahlerup-Petersen TE/MPE/EE – 5 November 2012
The cryo magnet replacement precedes the busbar consolidation by the ‘train’
ELQA Campaigns for SMACC
TE/MPE/EE
LHC Machine
K. Dahlerup-Petersen TE/MPE/EE – 5 November 2012
Resources: Equipment & Manpower - for all ELQA LS1-related activities.
PAQ:
One TP4 system is required per PAQ test. 8 TP4 systems (+ 2 Spares) are available.
We foresee up to 2 simultaneous PAQ campaigns, each requiring a team of 3 specialists.
Typical duration of one PAQ campaign: 90 minutes (once the equipment is in place)
One further team is kept as reserve.
AIV:
One AIV test system is required for a magnet replacement. 2 AIV testers are available.
For each campaign a specialized team of 3 people is required. A second team is kept as
back-up.
The test duration is 2 hours (once the tester is in place).
ELQA Campaigns for SMACC
TE/MPE/EE
LHC Machine
Human Resources related to the ELQA Test Campaigns:
4 collaboration agreements have been defined and signed, all with the institute HNINP, Cracow
1. KE1818/TE/LHC: Upgrade of two AVI systems and manufacture and assembly of
mechanical, electrical and electronic systems for constitution of 10 TP4 test systems, including
software upgrades and upgrade of existing TP1, TP2, TP3 benches to be TP4 compatible.
K. Dahlerup-Petersen TE/MPE/EE – 5 November 2012
2. KE2058/TE/LHC:
5 work-packages related to execution of standard ELQA
measurements during LS1 (TP4E, TP4 A & B, TP4 C, MIC-W, MIC-C, DOC-W, DOC-C).
3. KE2059/TE/LHC:
proximity equipment.
9 work-packages for AIV and PAQ as well as consolidation of
4.
KE2060/TE/LHC:
4 work-packages for treatment of non-conformities, data
handling during the ELQA campaigns, ELQA-related software maintenance and upgrades as well as
HNINP team coordination.
This in addition to the CERN ELQA ‘core’ team of nine people: four staff, one fellow, two PJAS, one
TS and one FSU.
ELQA Campaigns for SMACC
TE/MPE/EE
LHC Machine
K. Dahlerup-Petersen TE/MPE/EE – 5 November 2012
Resource planning for ELQA-LS1 activities
Total envelope: 516 man-months (43 man-years). Jan 2013 – Dec. 2014.
Team leaders, Engineers & Technicians.
New HV Test Specification
TE/MPE/EE
LHC Machine
Revision of HVQ Test Voltages:
Reasons for revision of voltage withstand figures used during
HV Qualification in the LHC:
- Reference data for cold tests are still originating from the
original document of Felix Rodriguez Mateos from Oct. 1998,
revised once (Nov. 2004).
- In this reference document each circuit is considered
alone. For the busbar insulation of adjacent and nested
circuits there’s a wish to consider the highest differential
voltages.
K. Dahlerup-Petersen TE/MPE/EE – 5 November 2012
- The after-LS1 discharge time-constant of the Main Quad.
Circuits has been fixed to 15 s (17.9 mΩ/19.0 mΩ) (if no
quench-back) whereas τ of the Main Dipole Circuits will go
back to the original figure 105 s (2 x 70 mΩ) (from today’s
52 s - 2 x 140 mΩ).
- Because of the particular topology of the Main Dipole
Circuit the present test level does not cover the case of an
earth fault. For all other circuits operational figures will be
used with 20 % safety margin.
- For the 3 Main Circuits voltage figures shall be related to
nominal current, all other circuits to ultimate current.
Introduction of earth fuse supervision in the 6’000
DQHDS quench heater power supplies.
ELQA Campaigns for SMACC
TE/MPE/EE
LHC Machine
Some additional comments:
• RB/RQF/RQD: Rated currents (7 TeV): 11850A /11162A/10677A rounded to 12000A/11200A/11200A
• Vquench is obtained from quench calculations. For the main quadrupoles we assume that quenches in
magnet coils do not propagate to the busbars.
• The present test voltage for RB (1900 V) was calculated from operational scenario with 20%
overshoot of extraction voltage. With capacitive snubbers the overshoot is eliminated. From worstcase considerations the RB test voltage shall be 2100 V.
K. Dahlerup-Petersen TE/MPE/EE – 5 November 2012
• New Quadrupole test voltage is 500 V (with 20% safety margin). BUT nearness of spools will require
busbars tested to much higher value (= 1.2 x (VEEQuad + VEESpool) = 800 V).
•For the spools we consider worst-case neighborship, i.e. spool-to-spool, giving 1000 V test voltage.
•For nested magnets without EE: Worst-case consideration will give 2 x VQuench plus 20 % safety
margin.
•For the MIC-test of Quench Heater –to- coil (grounded) the test voltage can be reduced to
1.2 (Vquench + 450 V) -see next table.
New H.V. Test Specification.
TE/MPE/EE
LHC Machine
Proposed new test voltages:
MS
IT
K. Dahlerup-Petersen TE/MPE/EE – 5 November 2012
N/N'
M1/M2
M3
Magnet coils
Ultimate/Nominal
Current [A]
EE res.
[mΩ]
MB
12000
MQ
11200
MQ (Alone)
Bus-bars
Up to now Proposed
1.2 x
tested at
new
VGPA [V] Vmax [V] Vmax [V]
[V]
voltage [V]
VEE [V]
Vquench
[V]
VEEq [V]
V2q [V]
150
900
1200
2100
--
900
2100
N.A.
1900
2100
19
213
200
413
--
213
413
495
240
500
11200
19
213
200
413
--
633
633
759
240
800
MCS
600
700
420
400
820
--
840
840
1008
480
1000
MCD
600
700
420
400
820
--
840
840
1008
480
1000
MCO
110
--
--
--
low
400
420
420
504
480
500
MS
600
700
420
500
920
--
840
920
1104
600
1100
MSS
600
700
420
500
920
--
840
920
1104
600
1100
MO
600
700
420
500
920
--
840
920
1104
600
1100
MQS
600
700
420
300
720
--
840
840
1008
360
1000
MQTL9
600
700
420
300
720
--
840
840
1008
360
1000
MQT/L
600
--
--
300
300
--
840
840
1008
360
1000
MQTD/F
600
700
420
300
720
--
840
840
1008
360
1000
IPQ (1.9K)
5820
--
--
400
400
--
--
400
480
480
480
MQM (4.5K)
MQY
4650
--
--
400
400
--
--
400
480
480
480
MQY
3900
--
--
100
100
--
--
100
120
480
480
IPD
6000
--
--
500
500
--
--
500
600
600
600
MQ6(IP3, 7)
600
700
420
300
720
420
720
864
360
900
Undulator
600
700
420
--
--
--
420
420
504
500
500
IT quads
6000
--
--
350
350
--
--
350
420
420
420
IT 600A corr.
600
--
--
500
500
1000
--
1000
1200
600
1200
IT 120A corr.
120
--
--
500
500
1000
--
1000
1200
600
1200
ELQA Campaigns for SMACC
TE/MPE/EE
LHC Machine
Suggested new test voltages for HVQ in all cryo magnets with heaters.
All lower than equal the present test levels.
MIC-W
Vquench
Proposed Test
voltage
Test
duration
Limit
[mA]
[V]
[V]
[V]
[s]
[mA]
60
1
1200
2000
2000
60
1
500
60
1
200
900
800
60
1
Individually Powered Quadrupoles (MQM, MQMC, MQML)
500
60
1
400
1300
1000
60
1
Individually Powered Quadrupoles (MQY)
500
60
1
100
700
700
60
1
500
60
1
500
1500
1100
60
1
500
60
1
350
1100
1000
60
1
Test
duration
Limit
[V]
[s]
Main Dipoles (MB)
600
Main quadrupoles (MQ)
CIRCUIT type
Separation Dipoles
(D1, D2, D3, D4)
Low Beta Quads
(RQX)
K. Dahlerup-Petersen TE/MPE/EE – 5 November 2012
MIC-C
Previous Test
voltage
Test
voltage
Consequences of the higher test voltages for coils and busbars:
- It may be necessary to disconnect temporarily certain monitors such as the cryo
sensors (insufficient rating of connectors).
- The existing HV test generators of ELQA are limited to 2 kV. Replacement needed.
- It is vital to perform tests with these higher levels before LS1 so to identify weak
points (the individual magnets have been tested to higher voltages in SM18).
- Breakdowns may cause voltage waves which could damage other equipment.
ELQA Campaigns for SMACC
TE/MPE/EE
LHC Machine
K. Dahlerup-Petersen TE/MPE/EE – 5 November 2012
Conclusion:
• All required test procedures are now available. The LS1 adaptation of the PAQ procedure is in
circulation for approval. All test procedures will be registered as EDMS documents.
• 10 TP4 systems are at the point of being ready, with new Labview software loaded, one station has
been successfully tested in SM18 on s.c. load (Q5).
• The TP4 systems are being equipped with new H.V. test generators and UPS’s.
• Three contracts have been signed with the Polish Institute HNINP for supply of engineers, team
leaders and technicians for the various LS1-realted ELQA jobs.
• The MPE/EE/ELQA team will now begin an ‘internal’ training campaign for becoming familiar with
the new TP4 system.
• In parallel, the teams at NHINP, which are selected for the LS1 tasks at CERN, will undertake similar
training using the two TP4 stations which are still in Cracow. The AIV teams will also repeat training.
• The majority of selected people for the three contracts are ‘veterans’, many of them having
experienced several ELQA campaigns at the LHC.
• The new, higher test voltages will prevent operational breakdowns under worst-case conditions, in
particular for adjacent busbars and nested magnet coils as well as for the main quadrupole circuits
where extraction voltages will exceed 200 Vpeak (new arc chambers to be installed on the extraction
switches during LS1).
• It is important to test the LHC circuits at these higher test voltages at cold before LS1 so to identify
at an early stage any weak elements.
DQQDE Voltage-to-Ground Monitor.
TE/MPE/EE
LHC Machine
DQQDE board for voltage-to-ground measurement
Communication part
Memory
Analogue part with
voltage divider
K. Dahlerup-Petersen TE/MPE/EE – 5 November 2012
ADUC
controller
DC-DC supply for
+/- 15V and 5V
-The system was designed for low-precision voltage measurements with the purpose of
-Verifying the correct behavior of the Energy Extraction systems in the 3 Main Circuits
after each discharge – through a series of voltage plots as above.
-Facilitating the identification of the location of an earth fault with voltage
breakdown.
-The system is a part of the nQPS systems installed under the MDB dipoles, i.e. in 54(55) locations
per sector, 1308 monitoring boards in total.
-All nQPS crates were prepared for installation of the 3 DQQDE boards. A pre-series of 250 boards
are now available but not yet installed.
-ACQ software was developed.
-It is the intention to install completely the system during LS1.
DQQDE Voltage-to-Ground Monitor.
K. Dahlerup-Petersen TE/MPE/EE – 5 November 2012
TE/MPE/EE
LHC Machine
The question was:
Could this network of ground monitors be used for early detection of precursors for voltage
breakdowns?
The facts:
•As the system does not measure leakage currents, detection must be based on voltage
variations.
•With its present ADC the resolution is 61 mV. However, if replaced by a 24-bit version the
resolution can be reduced to 0.2 mV.
•Because of its high leakage current the ground connection of each board is opened during ELQA
HVQ (test mode). The leakage currents during energy extraction (in normal operational mode)
attains typically 0.1 – 0.5 mA .
The difficulties:
-During its normal operations mode the ground voltages varies from 0 to full extraction voltage
depending on the place of the magnet in the chain. Only a few magnets experience the full
extraction voltage (Rdump x Ipeak).
-If a dielectric breakdown is preceded by a rise in leakage current this can today only be
detected at the grounding system of the power converter.
-Dielectric breakdowns are strongly non-linear phenomena (avalanche). The DQQDE system is
slow: 5 Hz sampling frequency.
- The capacitive earth currents will complicate the precursor detection during voltage ramps.
BUT:
- PSpice simulations are ongoing in the MPE –Performance section to determine the voltage
fluctuations in case of local degradation of the insulation resistance.
DQQDE Voltage-to-Ground Monitor.
TE/MPE/EE
LHC Machine
K. Dahlerup-Petersen TE/MPE/EE – 5 November 2012
Proposal:
– if simulations suggest a measurable voltage modification:
To equip the boards with 24-bit ADC’s and operate them in ‘normal mode’ in combination
with ELQA HVQ . Test both ramped mode and steady-state voltage application.