While　a　redundant　robot　is　still　able　to　continue　the　desired　task　in　the　presence　of　locked　joint　failures,　its　end-effector　error　both　at　the　instant　of　locking　joint　and　in　post-failure　operations　will　increase　obviously.This　work　discusses　the　path　error　of　a　redundant　robot　in　fault　tolerant　operations　for　locked-joint　failures.First,　the　analytical　formula　of　the　optimal　joint　velocity　with　minimum　jump　is　derived,　and　a　new　fault　tolerant　algorithm　with　the　minimum　jump　is　proposed.Then,　the　jump　difference　between　the　minimum　jump　solution　and　the　least-norm　velocity　solution　is　mathematically　analyzed.Finally　numerical　simulations　and　experimental　studies　are　implemented　with　a　planar　3R　robot　for　the　two　algorithms.Study　results　indicate　that　the　increase　of　the　path　error　at　the　instant　of　locking　joint　mainly　depends　on　the　joint　velocity　jump;　the　increase　of　the　path　error　in　post-failure　operations　mainly　depends　on　the　joint　velocity　of　the　reduced　robot.The　new　algorithm　proposed　in　this　paper　can　be　used　to　reduce　the　path　error　at　the　instant　of　locking　joint.