[en] Bolometric, thermocouple, Langmuir probe and infrared camera data are used to deduce the power flow in neutral beam-heated (P/sub input/less than or equal to 5 MW), dee-shaped, limiter discharges. Power loss measurements typically account for less than or equal to 60% of the total input power, with the majority either deposited on the primary limiter (20 to 40%) or radiated away (10 to 25%). Infrared camera measurements indicate that the heat flux to the ion drift side of the primary limiter is significantly greater than that to the electron drift side. Charge exchange neutral analysis, diamagnetic loop data and photodiode observations are consistent with a non-thermal flux of fast, beam produced particles to the limiter. If such an effect is indeed present, it could be partially responsible for the degradation in energy confinement observed during beam heating. Analysis of data comparing neutral beam injection from the two beamlines indicates that there is a variance in the energy confinement time which is beamline specific; this variance may be correlated to the limiter location relative to the beamlines