FREIA Magnet Test Facility Wiki

2. Post-quench recovery of the test bench

Document information

Tags/Project: Operational-Notes-wiki

Metadata FieldDocument Information
StatusDraft
Created2026-09-24
Last Updated2026-09-29
AuthorDavide Cuneo
Rev. #0.4
ReviewersTBD

Introduction and Context

This note describes what happens on the test bench after a quench is detected during the training of a superconducting magnet, and which actions are needed to recover the authorization to power the magnet again. It is not a procedure in the strict sense, but a description of the operational practice, meant to be used as a reference and as a basis for future procedures.

The note is based on the explanation given by Jean Luc Guyon at SM18, CERN, during the training campaign of the MCBRD corrector magnet. The recording was made right after a quench event, and Jean Luc used the occasion to comment on the signals and on the steps needed to operate the test bench again.

Few words about the test bench

The MCBRD is a two-aperture magnet, and each aperture is powered by its own Power Converter (PC). Each PC is associated with an Energy Extraction (EE) switch, so the bench has two EEs, one per aperture.

Both apertures are protected in parallel by two quench detection systems:

  • the PotAim cards, housed in their crate;
  • the uQDS (universal Quench Detection System).

All the triggering signals are collected by the Safety Matrix, which is the central element of the protection chain and in charge of interacting with the various interlocks in the facility. The whole bench is supervised by a PLC, which checks every interlock and gives the authorization to power the magnet only when each component is in a safe state.

What Happens During a Quench

When one of the detection systems (PotAim card or uQDS) detects a quench, it sends a trigger (trig) to the Safety Matrix. The Safety Matrix then acts in two directions at the same time:

  1. It triggers the DAQ, which saves the signals acquired over a preset time window starting a few seconds before the event. This is what makes the post-mortem analysis of the quench possible.
  2. It triggers the interlocks, which fast-abort the PCs and open the switches of the EEs. The current is thus diverted into the extraction resistors, and the energy stored in the magnet is extracted.

The sequence is summarised in the diagram below.

flowchart LR
    Q[Quench] --> D[PotAim / uQDS<br/>detection]
    D -->|trig| SM[Safety Matrix]
    SM --> DAQ[DAQ records signals<br/>incl. pre-trigger window]
    SM --> IL[Interlocks]
    IL --> PC[Fast abort of the PCs]
    IL --> EE[EE switches open]

The consequence is that, after the quench, the authorization to power the magnet is completely lost. It must be recovered before the test can continue.

Recovering the Authorization to Power

To operate again, the PLC must check each interlock and give a green light on each component. The components must be reset in a precise order, from the protection side to the powering side: first the Safety Matrix, then the EEs, then the PCs. the logic in FREIA is the same but the PLC system is build in a different framework and we use custom Labview-based GUI to coordinate the operations of the facility.

flowchart TD
    A[Safety Matrix:<br/>no trig on any card] --> B[EE switches:<br/>back to CLOSE]
    B --> C[PCs: no alarms,<br/>reset in NewPower application]
    C --> E{OKAY CRYO?}
    E -->|yes| F[PLC: authorization<br/>to power the magnet]
    E -->|no| G[Wait for / recover<br/>cryogenic conditions]
    G --> E

Step A: Clear the triggers on the Safety Matrix

The system cannot be reset as long as any card on the Safety Matrix still shows a trig. All the triggered cards must therefore be reset first, before any attempt to power the magnet again.

In the case observed, the trig came from the uQDS, which was wired to slots 15 and 16 of the Safety Matrix. Jean Luc cleared it by pressing the reset button on the Safety Matrix (specifically on the Intermem card).

Step B: Close the EE switches from the PLC software

Once the Safety Matrix is clear, the PLC software can be used again. Since the EE interlock triggered during the quench, both EE switches are in the open state and must be set back to close.

This is normally done from the PLC software, which runs an automatic process that tries to reset both switches. If this does not work, the operator must go to the EE units and act from the hardware panel:

  1. switch to LOCAL mode;
  2. press RESET;
  3. CLOSE the switch.

Step C: Reset the PCs

After the EEs, the PCs must be reset. They are controlled with the NewPower software application at CERN SM18, which is used both to reset each PC and to set the powering cycle for the following operations. Before resetting the PCs, the operator must make sure that:

  1. no alarm is active; if an alarm is present, it can be cleared manually with a hardware button on the PC;
  2. everything on the PC side is in order.

Once this is done, the whole system can be operated from the PLC and the magnet can be powered again.

The “OKAY CRYO” Condition

The electrical chain is not the only condition required to power the magnet. The PLC also needs the green light from the cryogenic system, the so-called OKAY CRYO. If it is lost, this is clearly shown on the interface of the PLC software, and the authorization is not given until it is recovered.

Losing the OKAY CRYO is a common consequence of a quench. The energy stored in the magnet is dissipated as heat, partly in the coil and partly during the extraction, and this heat can boil off part of the liquid helium, changing the conditions inside the cryostat. Typical causes are:

  • a pressure in the cryostat higher than allowed;
  • a helium level too low to properly cover the Current Leads (CLs) above the lambda plate.

In these cases the operator has to wait for, or act to recover, the nominal cryogenic conditions before the test can continue.

Data Analysis After a Quench

As described above, the signals of the quench are recorded because the Safety Matrix triggers the DAQ at the same time as the interlocks. The recorded window starts a preset number of seconds before the event, so the development of the quench can be analysed from its very beginning.

A few practical points about the data:

  • Analysis software. The signals are analysed with DIAdem (National Instruments). Before the analysis, the file must be converted to the .tdms format. The conversion is done directly in DIAdem, by loading the file and converting it within the software.
  • File naming. The recorded files are named with the format yyyymmdd-hhmm (e.g. 20260924-1430).

Quick Reference: Post-Quench Recovery

The list below summarises the recovery sequence in the order in which it must be carried out.

  • Checked the Safety Matrix and identified which cards triggered
  • Reset all the triggered cards on the Safety Matrix (no trig left)
  • Closed both EE switches from the PLC software (or manually: LOCAL → RESET → CLOSE)
  • Checked that no alarm is active on the PCs (cleared with the hardware button if needed)
  • Reset the PCs in newpower and set the next powering cycle
  • Checked the OKAY CRYO status on the PLC (pressure, helium level above the lambda plate)
  • Authorization to power recovered on the PLC
  • Converted the recorded file to .tdms and analysed it in DIAdem

Feedback

Found an error or something outdated in this note? Please contact Davide Cuneo (davide.cuneo@physics.uu.se).

Recently updated