Gradient amp error w/ HCP rsfMRI protocol

[markymarkymark]

Recently our VD13D Prisma has sporadically crashed when running the Lifespan HCP rsfMRI protocol with R015 of the MB EPI sequence. This occurs in about 20% of the scans, in the middle of the run. No software reboot is required to continue scanning. The error dialog is pasted below.

This does not occur for other MB protocols (e.g. the HCP dwi). Reducing the number of slices by one MB factor (i.e. by 8) appears to resolve the issue.

Siemens is aware of the problem, but has not been able to fix it, despite swapping out the gradient amp.

Has anyone else encountered this?

[markymarkymark]

Update on this issue: reducing the number of slices from 72 to 64 does NOT resolve the issue.