In conveying pulverized coal pneumatically to a blast furnace using compressed nitrogen, it was observed that the conveying system would plug with passage of time due to coal deposition. A pilot test rig was designed and operated in order to probe whether specific types of bends are responsible for this process and to identify problematic bend types. The test rig was operated to closely mimic actual plant conditions. Three different bends: a long radius bend (LR bend with R/D = 7.2) actually used in the plant, a short radius bend (SR bend with R/D = 2.4) and a blinded T bend both fabricated from glass were tested in a horizontal to vertical orientation. Among the bend types tested, the LR bend was found to produce maximum buildup. In a test conducted over 90 min, buildup was observed after around 20 min at the first point where the direction of flow changes, ultimately leading to blockage of approximately 50% pipe cross-sectional area at the end of the run. The blinded T bend was found to be self-cleaning, with negligible deposits even after 30 min of continuous operation. With the SR bend, some fine deposits were observed on the outer edge of the bend where considerable sliding of particles occurs. A solids flow-metering device based on pressure drop measurements in dilute phase conveying systems was developed earlier. This meter is based on the linear relationship between specific pressure drop ratio between suspension and single-phase conveying gas flow and the suspension loading. This meter was tested for metering the pulverized coal feed and found to give satisfactory metering performance over the range of conditions tested.