I noticed at least a 20% difference in flows computed using HEC HMS to TR 55 and TR 20 using the SCS method.
Has anyone else had this problem and if so were you able to get the models calibrated more closely?
Sorry to suggest the obvious, but make sure you check ALL the input data including:
Area
CN
Tc
calculation time step (dt)
Unit Hydrograph
Rainfall distribution
Differences in rainfall tables are often the cause. In particular, make sure both tables use the same increment. (that is, have the same number of points in the table)