Sequence abort on MAGNETOM Prisma (XA60) with gradient amplifier errors

[ste-phi]

Hello,

We are encountering a problem when running the cmrr_mbep2d_bold sequence on a MAGNETOM Prisma system running syngo XA60. The sequence started normally but aborted after some time (~10 minutes) due to gradient amplifier errors.

Error messages (I had to translate them from German, sorry if they are not 100% accurate):

• “Gradient amplifier error Z‑axis: PWM – end stage frequency too high”
• “Gradient amplifier error X‑axis: Controller loop error or controller error signal too high”
• “GPAK2309 error”

This sequence, with exactly the same parameters, has been running reliably for approximately one year, with hundreds of successful measurements.
However, a few months ago, we observed two possibly related failures during measurement where the reported fault was that the gradient voltage was too low (contrary to the current 'too high' error).

System details:

• Scanner: MAGNETOM Prisma (Siemens)
• Software: syngo XA60
• Sequence: cmrr_mbep2d_bold, MB = 8, TR = 700 ms, TE = 37 ms

Given the long history of stable functioning under identical parameters, we are puzzled as to what could be the issue.

Has anyone else encountered similar behaviour on a Prisma system or with the XA60 software version when using this sequence variant? Are there known compatibility issues or recommended parameter adjustments for this hardware‑/software‑combination?

Any insight, suggestions for diagnostic checks, or workarounds would be highly appreciated. Please let me know in case any more information could be helpful.

Best regards!