RF Reference Amplitude on Siemens 3T GUI

[RikRolling]

Hi,

I've been loading my sequence files onto the scanner and wanted to know whether the reference amplitude option within the Tx/Rx menu on the scanner GUI, changes the RF properties defined in the sequence file.

We noticed that a 45 degree and 90 degree flip angle for otherwise identical sequences both ran successfully, however when we halved the reference amplitude using the GUI for the 90 degree flip angle, we encountered a SAR exceeded error we did not expect. (We also had a SAR exceeded error when we doubled the reference amplitude, i.e. a 180 degree flip angle - which we would have expected).

Has anyone else experienced this? And is this reference amplitude function actually 'connected' to the pulseq sequence file itself, and if not why is it triggering this SAR exceeded error?

[btasdelen]

Can't explain why decreasing reference voltage triggers a SAR error, but this is how reference voltage is linked to your RF design: When you design your RF pulse, you design it for a specific flip angle (calculated by pulse's integral). When you put a subject in the scanner and hit scan, scanner does "adjustments" (calibration scans). One of the adjustments is to figure out how much voltage to apply to achieve 90 flip angle, for the specific subject inside. That voltage is basically your reference amplitude now, i.e., if you want a 30 flip angle, scanner just divides the reference voltage by 3 to achieve it. It also calculates SAR from the reference amplitude. When you halve the reference voltage, you are essentially telling scanner that you achieve 90 flip angle with this halved voltage, so every RF pulse in the sequence are halved in flip angle in reality (unless you are changing individual pulse voltages). Scanner still "thinks" it is playing a 90 pulse, but in reality it is close to 45 pulse.

[schuenke]

Not exactly the same, but also a strange behaviour that could be related:

pulseq/pulseq#156