In the process of GFE(L)1 Secondary Wind finding Radar sounding, the difference between geopotential and geometric heights shown on the monitoring interface can be large, sometimes up to 4000 m or 5000 m, far beyond the normal range, thus causing problems for observers in record analysis. Various sources of the bias and possible reasons are discussed, and the possible error ranges are estimated and categorized. Some practical suggestions are given about bias correction approaches.