MAGNETOM Avanto

• Faulty RFSU power supply (confirmed in one case)
• Tune-up out of date, or adjustment not being run for each patient (image quality case)

1. Read the error log and the test reports before swapping parts
2. Swap in a known-good TAS to rule it out
3. Go through the modulator – RFPA – TAS – BCCS – TALES – body coil chain using the results of the first three test steps
4. Look at the TALES voltages reported by the tests
5. Check when tune-up was last run and whether adjustment runs for each patient; run TuneUp and Quality, then the RF characterisation tests if the fault remains

In the PALI / RF test case, replacing the RFSU power supply fixed the fault. For the image quality complaint, the advice was to run TuneUp and Quality first, then the RF characterisation tests; that case had no confirmed outcome.

• A cable on the BDAS that was not fully tightened again after the connector check

1. Re-check every RF connector that was disturbed, especially on the BDAS, and make sure each one is fully seated and tightened
2. Repeat the Max kV test and the Rx test
3. Scan a phantom to confirm

Probably a BDAS cable that was not fully tightened after the connector inspection (colleagues' assessment, not confirmed). Re-tightening the connections and repeating the tests is the likely fix.

• Static build-up from ordinary linoleum and a synthetic office chair in the control room and the cage, made worse by dry winter air

1. Test all coils to see which are actually damaged
2. Look at the working conditions: floor covering, furniture, clothing, air humidity

Replacing the linoleum and the chair with antistatic versions solved the problem.

• Bent metal shield inside the coil

1. Inspect the coil for mechanical damage after any drop
2. Compare where the artefact sits in the image with where the coil is damaged

Straightening the shield solved it in this case. Colleagues doubted that a bent shield alone could cause a local artefact, so treat the cause with caution.

• Worn or broken rollers or bearings on the vertical drive guide rails
• S30 trips when the table meets an obstacle while moving down (anti-crush protection)

1. Check whether something blocked the table on its way down before treating an S30 trip as a fault
2. Inspect the whole vertical drive while you are in there: bearings, brake, ball screw spindle and nut
3. Look up the original part numbers in the manufacturer's spare parts catalogue

Probable fix: replace the rollers or bearings as a complete assembly with their rails. The official procedure needs the magnetic field to be ramped down for access; that must be planned with the manufacturer. Replacing the rail assembly under field was described as possible but extremely difficult.

• One of the chiller's two compressors failed, leaving too little cooling capacity
• Return line pressure too high (above 1.3 bar) on the secondary circuit pump (probable cause of error 42)

1. Check the chiller's compressors and the primary circuit temperature trend
2. Check the heat load in the technical room and the control room, including ventilation
3. Measure the return line pressure, not just the general circuit pressure
4. Check the pump's indicator and LED status

Temporary measure for the chiller case: turning off ventilation in the technical room and control room brought the primary circuit back to a stable 16°C. An extra split air conditioner in the technical room was recommended, and the failed compressor still needs repair. For error 42, high return pressure is the likely cause; no fix was confirmed.

• A built-in start delay, by design: it prevents frequent on/off cycling during power cuts and lets the SEP cabinet get into its operating mode. The secondary circuit pump starts after 3 minutes, which can also make the helium compressor cut out

1. Confirm which scanner the compressor is connected to
2. Check the MSUP signal (DB25), which is involved in controlling the start delay

This is probably normal behaviour, not a fault. Changing the factory delay is not recommended. Tell operators to wait about 20 minutes after power returns. Colleagues said cycling the MSUP supply breaker starts the compressor at once, but because it involves the cryo system, only do this if the manufacturer's procedure allows it.

• Database clean-up is slow when many studies are still present
• The study is still open in another task card, or it is protected
• A lower-case 'S' means only part of the study was transferred
• Metric units not enabled in configuration

1. Do not reboot before the database clean-up has finished; an early reboot left the database uncleared
2. Close the study in the exam, viewing, filming and 3D task cards before editing
3. Check whether the study transfer is complete
4. Look for the 'metric' option in options/configuration; where it is depends on the software version

Database: delete the studies first, then run AllDBRemove (on Avanto this cut the time from about 35 to 15 minutes), or let it run for about 30 minutes and then reboot manually. That fix was confirmed. Editing data: in the patient browser select the folder, then Edit > Correct; if it is blocked, right-click and use Remove Protect (likely fix). Units: enable 'metric' (likely fix).

• Not determined in these cases
• Dirty contacts on the teslameter probe and the MPS connection (measurement problem during ramp-up)

1. Record the helium level before and after the event and the current at which the quench happened
2. Record the events and hand them to the manufacturer

No confirmed fix. Ramping, quench investigation and any work on the magnet or helium must be done only by the manufacturer or under its direct guidance. In one ramp-up case, cleaning the probe and MPS contacts with alcohol resolved the measurement problem.