Choose timezone
Your profile timezone:
MP3 meeting
Participants
Adriaan Rijllart
Amalia Ballarino
Andrea Musso
Antonio Perin
Antonio Vergara Fernandez
Bernhard Auchmann
Christian Giloux
Edward Nowak
Francesco Bertinelli
Gerard Willering
Gert-Jan Coelingh
Giorgio D'Angelo
Glyn Kirby
Howard Pfeffer
Hugues Thiesen
Jean-Philippe Tock
Jim Strait
Karl Hubert Mess
Knud Dahlerup-Petersen
Luca Bottura
Marco Zanetti
Markus Zerlauth
Matteo Solfaroli Camillocci
Michele Modena
Nuria Catalan Lasheras
Paolo Fessia
Reiner Denz
Richard Mompo
Robert Henry Flora
Rüdiger Schmidt
Sandrine Le Naour
Scheuerlein Christian
Vittorio Remondino
Walter Venturini Delsolaro
Zinur Charifoulline
MP3 meeting 19 August 2009
Nuria informed on the insulation fault that has been found at the beginning of the week in sector 67. The cool down had started on Saturday. Now the ELQA team is trying to localize the fault.
The naming of the signals is such that a stretch including 3 segments is named after the last one
Signals are doubled: with a reduced sampling frequency, the copy goes directly to the analysis software
Signals are recorded on change with no dead band.
The LSB for U_RES is 1.5 nv, for U_MAG is 305 nV
The threshold is 300 microV for 10 s, i.e. it triggers when the voltage integral over the last 10 s reaches 0.3 mWb.
the system will create PM data, although this is not yet implemented and for the time being the source of data is the logging DB.
To compensate for the busbar self inductance, the system uses the upstream magnet for each segment
Walter asked whether this means that all the detectors must be tuned separately.
Reiner answered that this will probably not be necessary, there are two segment types with different lengths. It is likely that after tuning of the first sector the values can be propagated to the others.
Karl Hubert Mess remarked that previously we had only two reference magnets which needed to be recycled in case of quench to re establish the superconductor magnetization (affecting the inductance)
Now it seems that this will be needed after any magnet quench
Reier replied that in the new system the effect of the magnetization error should not unbalance the detectors, as it is not multiplied by 77 as in the previous system.
Nuria asked how the inductance of the last segment is calculated (in absence of an upstream magnet )
Reiner said that one of the reference magnets is used twice to cover the last segment
Concerning the question of maintaining or not the old global busbar detector, it was noted that this may be used for another application, although Reiner has shown that it will be fully redundant.
The drawback of keeping it alive is the abovementioned need of recycling when a reference magnet quenches.
The most prominent issue were in A22A1, where almost all high field heaters have a high leakage current; and B19R1, which has a weak high field quench heater.
For A22A1 it was decided to reconfigure the protection and use the low field heaters to minimize the risk of a short circuit
K. H. Mess observed that the leakage current still corresponded to 2 GOhm, which is quite a high resistance, but R. Mompo gave a further detail which lead to take the safest course: the fault was absent before the first powering in SM18, and has degraded ever since.
One sample was prepared reproducing the "silent killer" defect in a quadrupole busbar.
The lack of solder material between cable and the copper stabilizer was evidenced by a radiography which looks exactly as those taken in the tunnel on the defective junctions.
The R16 value of the sample was 60 microOhm,which is at the high end of what is belived to be the distribution in the tunnel.
The RRR of the copper was 180-200.
As values as low as 100 were found in the tunnel, the effective RRR of the sample was reduced using the a magnetic field to increase the resistivity.
Taking into account the combined effect on the elecrtical and on the thermal conductivities a magnetic field of 2 T is equivalent to RRR=100.
The sample was equipped with temperature sensors, heaters to induce quenches, and voltage taps to observe the quench propagation.
In the experiment the current was ramped up and the heaters were used to induce resistive transitions.
As expected, cryostability and thermal runaway regimes were found as a function of temperature and of current.
The results were presented in form of plots where the time for a thermal runaway was plotted as a function of the current. The curves are fit with hyperbolae and the vertical asymptotes are the safe currents i.e. the currents at which a thermal runaway will need an infinite time to develop. However a cutoff of 30 s was applied to define the safe current.
The most striking result was that, for 2 T (equivalent to RRR=100) and at 1.9 K, the safe current defined as above was as low as 5.5 kA. Moreover, the corresponding value at 4.4 K was about 4900 A.
In the conclusions Luca pointed out that the results are preliminary, in particular additional analysis is required to assess the representativeness of the sample as regard to the use of the magnetic field to simulate a low RRR, to a discrepancy which was found between temperature measurements and the temperature deduced from the voltage taps records, and finally to the cooling conditions, which seem less favorable than in the tunnel, due to the presence of the instrumentation around the junction.
H. Pfeffer observed that 30 s is rather generous as a cutoff
J. Strait said that it looks reasonable, as it is comparable with the time constant of the current decay
A. Ballarino remarked that joints that work at 4.4 K are present in the DFBs
A. Perin said that the discrepancy in the temperature reading is understandable in terms of the perturbation introduced by the termocouple wires
K.H.Mess agreed observing that the temperature gradient observed is in the right direction
K.H.Mess added that, when a magnet is quenching, the busbar will be heated and this heat must flow toward the junction.
This additional heat was not present in the experiment, which in this respect underestimates the heat load on the joint
Luca said that this could be easily added in the next experiment.
He concluded asking for feedback to prepare the test program for a second campaign
General information
Powering in sector 12 is imminent.Nuria informed on the insulation fault that has been found at the beginning of the week in sector 67. The cool down had started on Saturday. Now the ELQA team is trying to localize the fault.
QPS Layout for busbar protection (R. Denz)
Reiner detailed the layout of the new QPs for the busbar. The new system covers all the busbar segments including the return on the quadrupoles.The naming of the signals is such that a stretch including 3 segments is named after the last one
Signals are doubled: with a reduced sampling frequency, the copy goes directly to the analysis software
Signals are recorded on change with no dead band.
The LSB for U_RES is 1.5 nv, for U_MAG is 305 nV
The threshold is 300 microV for 10 s, i.e. it triggers when the voltage integral over the last 10 s reaches 0.3 mWb.
the system will create PM data, although this is not yet implemented and for the time being the source of data is the logging DB.
To compensate for the busbar self inductance, the system uses the upstream magnet for each segment
Walter asked whether this means that all the detectors must be tuned separately.
Reiner answered that this will probably not be necessary, there are two segment types with different lengths. It is likely that after tuning of the first sector the values can be propagated to the others.
Karl Hubert Mess remarked that previously we had only two reference magnets which needed to be recycled in case of quench to re establish the superconductor magnetization (affecting the inductance)
Now it seems that this will be needed after any magnet quench
Reier replied that in the new system the effect of the magnetization error should not unbalance the detectors, as it is not multiplied by 77 as in the previous system.
Nuria asked how the inductance of the last segment is calculated (in absence of an upstream magnet )
Reiner said that one of the reference magnets is used twice to cover the last segment
Concerning the question of maintaining or not the old global busbar detector, it was noted that this may be used for another application, although Reiner has shown that it will be fully redundant.
The drawback of keeping it alive is the abovementioned need of recycling when a reference magnet quenches.
Quench heater NC (G. D'angelo)
Giorgio illustrated some anomalies and non conformities that ELQA has found in the QH circuits of the Main Dipoles of sector 1-2The most prominent issue were in A22A1, where almost all high field heaters have a high leakage current; and B19R1, which has a weak high field quench heater.
For A22A1 it was decided to reconfigure the protection and use the low field heaters to minimize the risk of a short circuit
K. H. Mess observed that the leakage current still corresponded to 2 GOhm, which is quite a high resistance, but R. Mompo gave a further detail which lead to take the safest course: the fault was absent before the first powering in SM18, and has degraded ever since.
Tests of a bad joint in FRESCA (L. Bottura)
Luca presented preliminary test results of a badly stabilized connection in the FRESCA test facility.One sample was prepared reproducing the "silent killer" defect in a quadrupole busbar.
The lack of solder material between cable and the copper stabilizer was evidenced by a radiography which looks exactly as those taken in the tunnel on the defective junctions.
The R16 value of the sample was 60 microOhm,which is at the high end of what is belived to be the distribution in the tunnel.
The RRR of the copper was 180-200.
As values as low as 100 were found in the tunnel, the effective RRR of the sample was reduced using the a magnetic field to increase the resistivity.
Taking into account the combined effect on the elecrtical and on the thermal conductivities a magnetic field of 2 T is equivalent to RRR=100.
The sample was equipped with temperature sensors, heaters to induce quenches, and voltage taps to observe the quench propagation.
In the experiment the current was ramped up and the heaters were used to induce resistive transitions.
As expected, cryostability and thermal runaway regimes were found as a function of temperature and of current.
The results were presented in form of plots where the time for a thermal runaway was plotted as a function of the current. The curves are fit with hyperbolae and the vertical asymptotes are the safe currents i.e. the currents at which a thermal runaway will need an infinite time to develop. However a cutoff of 30 s was applied to define the safe current.
The most striking result was that, for 2 T (equivalent to RRR=100) and at 1.9 K, the safe current defined as above was as low as 5.5 kA. Moreover, the corresponding value at 4.4 K was about 4900 A.
In the conclusions Luca pointed out that the results are preliminary, in particular additional analysis is required to assess the representativeness of the sample as regard to the use of the magnetic field to simulate a low RRR, to a discrepancy which was found between temperature measurements and the temperature deduced from the voltage taps records, and finally to the cooling conditions, which seem less favorable than in the tunnel, due to the presence of the instrumentation around the junction.
H. Pfeffer observed that 30 s is rather generous as a cutoff
J. Strait said that it looks reasonable, as it is comparable with the time constant of the current decay
A. Ballarino remarked that joints that work at 4.4 K are present in the DFBs
A. Perin said that the discrepancy in the temperature reading is understandable in terms of the perturbation introduced by the termocouple wires
K.H.Mess agreed observing that the temperature gradient observed is in the right direction
K.H.Mess added that, when a magnet is quenching, the busbar will be heated and this heat must flow toward the junction.
This additional heat was not present in the experiment, which in this respect underestimates the heat load on the joint
Luca said that this could be easily added in the next experiment.
He concluded asking for feedback to prepare the test program for a second campaign
There are minutes attached to this event.
Show them.
The agenda of this meeting is empty