In MRI systems, the gradient amplifier increases the energy of the signal before it reaches the gradient coils, in a way, that the field strength is intense enough to produce the variations in the main magnetic field for localization of the later received signal. Three gradient coils normally require three gradient amplifiers. As all amplifiers, gradient amplifiers produce heat, which requires cooling.

See also the related poll result: 'Most outages of your scanning system are caused by failure of'

(GRASE) A hybrid sequence with a combination of gradient and spin echo sequences. If multiple image lines are obtained during a single echo, the imaging pulse sequence type is a GRASE sequence.

See Gradient Echo Sequence and Spin Echo Sequence.

Current carrying coils designed to produce a desired magnetic field gradient (so that the magnetic field will be stronger in some locations than others).
Proper design of the size and configuration of the coils is necessary to produce a controlled and uniform gradient. Three paired orthogonal current-carrying coils located within the magnet that are designed to produce desired gradient magnetic fields, which collectively and sequentially are superimposed on the main magnetic field (B0) so that selective spatial excitation of the imaging volume can occur.
Gradients are also used to apply reversal pulses in some fast imaging techniques. Gradient coils in general vary the main magnetic field, so that each signal can be related to an exact location. The gradient coil configuration for the z-axis consists of e.g., Helmholtz pair coils, and of paired saddle coils for the x- and y-axis.

See also the related poll result: 'Most outages of your scanning system are caused by failure of'

(GE) An echo signal generated from a free induction decay by means of a bipolar switched magnetic gradient. The echo is produced by reversing the direction of a magnetic field gradient or by applying balanced pulses of magnetic field gradient before and after a refocusing RF pulse so as to cancel out the position dependent phase shifts that have accumulated due to the gradient.
In the latter case, the gradient echo is generally adjusted to be coincident with the RF spin echo. When the RF and gradient echoes are not coincident, the time of the gradient echo is denoted echo time (TE) and the difference in time between the echoes is denoted time difference (TD).
Gradient echo does not refocus the effects of main field inhomogeneity and therefore is generally used with a short echo time. Disadvantages of gradient echo imaging are compromised anatomic details and artifacts in regions with varying susceptibility e.g. between the air-containing sinuses and brain and especially between haemorrhages and normal tissue.

See also Susceptibility Artifact.

(GEMS) This pulse sequence uses a changeable flip angle instead of a 90° pulse and a gradient instead of a RF pulse to rephase the FID.
T2*, T1 weighted and proton density images can be acquired. The flip angle in combination with the TR determines the T1 weighting and the TE controls the amount of dephasing. To minimize T2* the echo time should be short.

See also Gradient Echo Sequence.