BlogHccMay07 < HCC

TWiki>

HCC Web>BlogHccMay07 (2007-05-02, JacekSzkutnik)

EditAttachPDF

TWiki home

Point 1

Point 2

Point 3

Point 4

Point 5

Point 6

Point 7

Point 8

Minutes

Create new blog entry for Point 8

Created by Main.BorisBellesia on 2007-05-20 - 22:20

Powering Tests - 1000A in Q5 -

Done on Friday 18.05.2007

9:30	Powering Procedure for RQ5B1(B2).L8 started
11:30	PLI1.1-4-7 I_inj: 350A PLI2.1-3-4 I_int_1: 1000A procedures performed manually and NOT using the sequencer
13:00	Boil-off test on DFBMA-C

Remarks on the work of the last week (14-18 May)

The current was ramped to 1000 A, and the magnet was quenched by triggering the firing of the quench heaters.
The different teams performed a preliminary analysis of the data that had been taken, and some observations were reported:
The DECOUPLING BETWEEN THE TWO CONVERTERS WAS EXCELLENT. By applying a modulation on one power converter of 10A at a DC current of 1000A it was measured that the coupling is negligible (ppm level). The modulation was seen by the quench detectors, but was below the threshold. For future tests, it was suggested to include the current modulation into the procedures, if required (Action: D.Nisbet).
The balancing of the QUENCH DETECTORS was very good. The THRESHOLD was set to 20mV, and the detectors did not trigger. However, it was noted that the PEAK AMPLITUDE OF THE BALANCED SIGNAL WAS SLIGHTLY ABOVE 20mV, BUT FOR A TIME LESS THAN THE VALIDATION TIME OF THE QUENCH DETECTORS OF 10ms. This effect depends on the speed of the ramp, and decreased between the first ramp at 13A/s AND THE SECOND RAMP AT 10A/s.
FOR THE FUTURE COMMISSIONING OF Q5 A RAMP OF 5A/S WILL BE APPLIED, in order to reduce the noise level for the quench detectors. The modified value should be integrated into the sequence. In the second phase the sensitivity of the quench detectors as a function of ramp speed will be further investigated. It might be required to increase the quench detector threshold from 20mV to a higher value. The objective is to ramp the magnet with the nominal rate of about 10A/s.
At the first glance it appears that all required data after the quench were recorded and written into the PM system.
The heater discharge worked as expected. It was correctly recorded, but due to one point with the wrong time the start of graph was wrong. This will be corrected (Action: R.Denz and A.Rijllart).
THE DATA FROM THE CURRENT LEADS (VOLTAGE ETC.) LOOKED OK. It was noticed that the voltage was high during the trigger of the heaters. This is due to the fact that the quench heaters are triggered using a test signal, and that the effect does not indicate an increased voltage across the leads.
AFTER TRIGGERING THE QUENCH, THE HELIUM LEVEL IN THE Q5 MAGNET CRYOSTAT DECREASED IN A SHORT TIME (~MIN) TO ZERO, and increased in about the same time to the nominal value. No change of temperature and pressure were observed. The valve opened after the quench. Since similar changes of the helium level were recorded at other occasions without quenching magnets, it is questions if this effect is real. This needs follow-up, since such drop of helium level would trigger a power abort (Action: ACR). OTHERWISE NO CONSEQUENCES OF THE QUENCH WERE OBSEREVED IN THE CRYOGENIC SYSTEM.
Since the procedures were executed without the sequencer, that data that was recorded needs to be put into MTF manually (Action: HCC)

The current was ramped to 1000 A, and the magnet was quenched by triggering the firing of the quench heaters.

The different teams performed a preliminary analysis of the data that had been taken, and some observations were reported:

The DECOUPLING BETWEEN THE TWO CONVERTERS WAS EXCELLENT. By applying a modulation on one power converter of 10A at a DC current of 1000A it was measured that the coupling is negligible (ppm level). The modulation was seen by the quench detectors, but was below the threshold. For future tests, it was suggested to include the current modulation into the procedures, if required (Action: D.Nisbet).

The balancing of the QUENCH DETECTORS was very good. The THRESHOLD was set to 20mV, and the detectors did not trigger. However, it was noted that the PEAK AMPLITUDE OF THE BALANCED SIGNAL WAS SLIGHTLY ABOVE 20mV, BUT FOR A TIME LESS THAN THE VALIDATION TIME OF THE QUENCH DETECTORS OF 10ms. This effect depends on the speed of the ramp, and decreased between the first ramp at 13A/s AND THE SECOND RAMP AT 10A/s.

FOR THE FUTURE COMMISSIONING OF Q5 A RAMP OF 5A/S WILL BE APPLIED, in order to reduce the noise level for the quench detectors. The modified value should be integrated into the sequence. In the second phase the sensitivity of the quench detectors as a function of ramp speed will be further investigated. It might be required to increase the quench detector threshold from 20mV to a higher value. The objective is to ramp the magnet with the nominal rate of about 10A/s.

At the first glance it appears that all required data after the quench were recorded and written into the PM system.

The heater discharge worked as expected. It was correctly recorded, but due to one point with the wrong time the start of graph was wrong. This will be corrected (Action: R.Denz and A.Rijllart).

THE DATA FROM THE CURRENT LEADS (VOLTAGE ETC.) LOOKED OK. It was noticed that the voltage was high during the trigger of the heaters. This is due to the fact that the quench heaters are triggered using a test signal, and that the effect does not indicate an increased voltage across the leads.

AFTER TRIGGERING THE QUENCH, THE HELIUM LEVEL IN THE Q5 MAGNET CRYOSTAT DECREASED IN A SHORT TIME (~MIN) TO ZERO, and increased in about the same time to the nominal value. No change of temperature and pressure were observed. The valve opened after the quench. Since similar changes of the helium level were recorded at other occasions without quenching magnets, it is questions if this effect is real. This needs follow-up, since such drop of helium level would trigger a power abort (Action: ACR). OTHERWISE NO CONSEQUENCES OF THE QUENCH WERE OBSEREVED IN THE CRYOGENIC SYSTEM.

Since the procedures were executed without the sequencer, that data that was recorded needs to be put into MTF manually (Action: HCC)

Program for Monday 21.05.2007

Continue commissioning of Q5, manually without the sequencer.
Start commissioning of the 120A circuits using the sequencer. This can be done in parallel to Q5 commissioning, although there might be some interference.
When Q5 is commissioned, Q4 and D2 will follow. Before starting to commission D2,y the QPS team needs to intervene in order to correct a problem that was reported after PIC2.

Continue commissioning of Q5, manually without the sequencer.

Start commissioning of the 120A circuits using the sequencer. This can be done in parallel to Q5 commissioning, although there might be some interference.

When Q5 is commissioned, Q4 and D2 will follow. Before starting to commission D2,y the QPS team needs to intervene in order to correct a problem that was reported after PIC2.

Comments:

Edit, attach file or add comment