Transcript Test Procedure 4 - Indico

TE-MPE-TM 25/10/2012, ELQA team
TP4 System –
the main tool
for ELQA
Mateusz Bednarek, CERN, Geneva
Jaromir Ludwin, IFJ-PAN, Krakow
TE-MPE-TM 25/10/2012, ELQA team
ELQA prior to the Powering Tests
Main part of ELQA activities was marked in black.
All possible ELQA scenarios are described in EDMS document 788197
ELQA campaign
Tests are divided into 3 groups:
• TP4 – Test Procedure 4
o
o
HV and LV tests of circuits powered via DFBs
743 circuits
• DOC – Dipole Orbit Corrector check
o
o
HV and LV tests of circuits powered locally
914 circuits
TE-MPE-TM 25/10/2012, ELQA team
• MIC – Magnet Instrumentation Check
o
o
HV and LV tests done locally on each magnet – it includes quench
heater qualifications
1763 magnets.
TP1, TP2, TP3 were the test procedures suitable for testing DFBs at different stages of
assembly. TP4 is the last test procedure that shall be applied once the DFB are
connected to the magnet chain.
Typical circuit tested with TP4
• 2 current leads
• Magnet chain
TE-MPE-TM 25/10/2012, ELQA team
We need to check:
•
•
•
•
Electrical insulation of the circuit,
Continuity and resistance of the circuit,
Continuity and resistance of V-taps,
Check if V-taps are properly distributed along the CLs.
• Make sure that the Pt100 is properly attached to the CL potential (to be
implemented)
Complexity issue
• For the MIC test there are 23 different types of
magnets.
• The TP4 and DOC tests are even more varied
TE-MPE-TM 25/10/2012, ELQA team
o
o
o
2-4 current leads in one circuit,
7 types of current leads,
6 types of connectors on DFBs
TE-MPE-TM 25/10/2012, ELQA team
Complete system overview
• A universal measurement system is needed to cover
all configurations
• Measurements have to be automated
• Results must be stored in the central database
• Follow-up web page should be available as well
TE-MPE-TM 25/10/2012, ELQA team
Complete system overview
The TP4 System
• The TP4 system was recently
upgraded by the team from IFJ-PAN
(aka HNINP), Krakow
o
o
o
TE-MPE-TM 25/10/2012, ELQA team
o
o
The hardware of the old system was
unmaintainable
There were not enough spares
The old system was entirely dependent
on the Keithley multimeter
The stability parameters of the old
system were acceptable, but there was
a lot of room for improvement
Many additional features were added
in the upgraded version
• In this presentation we will focus on
the current version of the system
TE-MPE-TM 25/10/2012, ELQA team
Hardware – general architecture
Hardware
TE-MPE-TM 25/10/2012, ELQA team
• Measurement of the ohmic resistance of the highly inductive
superconducting magnet, requires extremely stable current.
• Using voltage source with stable serial resistance gives better
results than using current source.
• In order to stabilise the resistance of the current path within
the measurement system, we have decided to thermally
stabilise the crucial elements.
TE-MPE-TM 25/10/2012, ELQA team
Hardware
• The resistors are used to set the working point of the
power supply and for current measurement.
• 18 Internal voltage taps are routed to the inputs of the
multiplexers.
The temperature regulation
• The temperature regulator is based on the microcontroller,
connected to the computer via USB port.
• Regulation is done in two stages:
TE-MPE-TM 25/10/2012, ELQA team
o Temperature of internal part is controlled by proportionalintegrating regulator using 4 heaters. The current relays PCB
and resistors are enclosed in thermally insulated box.
o Temperature of external part is controlled by proportional
regulator using set of heaters and fan
• Parameters of regulators are stored in the EEPROM, and can be
modified from the level of LabVIEW software.
• Since we know the power that will appear on the resistors
during the measurement, we can use feed-forward method to
stabilise the temperature of the resistors. The feed-forward
control is done by altering the integral term of PI regulator.
Hardware
The system is equipped
with 84 inputs:
•
•
18 internal,
66 available on the front
panel.
Having 4 independent
1 of 84 multiplexers gives
following possibilities:
TE-MPE-TM 25/10/2012, ELQA team
•
•
•
Measurements of any
two voltages inside the
system and on the
device under test
4-wire resistance
measurements
Transfer function
measurements using
freely chosen reference
impedance
PCBs arrangement
TE-MPE-TM 25/10/2012, ELQA team
• Multiplexers are fitted
into a 3U crate,
• Power supply is located
in a separate 1U case
TE-MPE-TM 25/10/2012, ELQA team
Shielding
Current status of the system
TE-MPE-TM 25/10/2012, ELQA team
• All multiplexer crates are assembled and
tested,
• Measurement instruments are mounted
on the available trolleys,
• Most of the measurement cables are
ready,
• During the development of the system
emerged an idea of replacing Keithley
multimeter with two PXI DMM cards.
o
o
The migration is ongoing
SM18 tests have shown that the migration is
not going to be too complicated
TE-MPE-TM 25/10/2012, ELQA team
Debug console
Apart of the standard measurement applications (TP4, DOC, MIC)
there is a “debug console”
• Access to all the measurement instruments in the system,
• Low level control of the multiplexers and the thermostats,
• Usage of functionalities inaccessible for standard applications,
• System self testing capability,
• Advanced circuit
diagnostics,
• Designed to be
used by experts
only.
Software
Database is the central part of the system. It stores:
TE-MPE-TM 25/10/2012, ELQA team
•
•
•
•
•
•
Reference data describing circuits
Test content
Validation conditions
Configuration parameters
Calibration parameters
Results
Signal translation in low voltage test applications.
Measurement results validation process.
TE-MPE-TM 25/10/2012, ELQA team
Open question
During the MIC tests, we are powering the
measurement system form 24 VDC batteries of
Pefra tractor.
There are rumours, that the tractor type at CERN
may change.
In such case we would have to
adapt our hardware and
the work flow.
Source: Model 507 datasheet
TE-MPE-TM 25/10/2012, ELQA team
Conclusions
• TP4 is a very versatile system perfectly suited
for the complex LHC circuits,
• Upgrade works are almost finished,
• If needed the TP4 platform may be used for
new types of measurements,
• The database system makes it a reliable and
scalable system,
• A test on a highly inductive superconducting
magnet is still missing
TE-MPE-TM 25/10/2012, ELQA team
Acknowledgements
The new TP4 is a very complex project.
Many details had to be carefully designed in order to
achieve expected performance, both from electrical and
mechanical points of view.
Credit for solution of those problems goes mainly to the
IFJ-PAN team:
• Piotr Jurkiewicz,
• Marek Ruszel,
• Ryszard Tarczoń
• and others, who are not mentioned here, but realisation
of this project would be impossible without them.
TE-MPE-TM 25/10/2012, ELQA team
Thank you for your attention!
Back-up slides
TE-MPE-TM 25/10/2012, ELQA team
For the discussions
Keithley -> PXI controversy
TE-MPE-TM 25/10/2012, ELQA team
• Pros of having a PXI on board
o Reduction of time needed for the
measurements,
o Increased precision as the current can be
measured at the same time as the voltage across
a component of the circuit,
o Decreased wear of the multiplexer relays,
o Increased versatility of the system
• Cons
o Increased complexity of the user interface,
o Communication with the central DB will need to
be modified
TE-MPE-TM 25/10/2012, ELQA team
Example screenshots