HydroCAD Modeling assistance - subsurface infiltration bed

I am trying to model a subsurface facility that works in conjunction with a detention pond. The goal was to utilize as small of a bed as possible to receive runoff from about 2 acres, infiltrate the 2-yr, and then let the rest of the water flow out to Basin 1, which provides rate control. The bed is provided with an outlet manhole, with a weir plate set at an elevation such that the 2-yr storm is kept within the bed. Everything else will be above the elevation of the plate and ideally SHOULD flow freely out of the bed. In order to model this, the bed has a starting elevation (under the Advanced tab)set at the elevation of the weir plate. This initially didn't work and I received errors about the max WSEL exceeding the top of the bed. To fix this a prismatoid was added as storage to model the conditions in the outlet manhole (storage above the weir plate). 

This seemed to fix it though now I am having two issues: 1.) peak outflow slightly exceeds peak inflow for the bed for 25, 50, and 100-yr storms and 2.) if I re-run the critical duration analysis at the POI for the storms smaller than 100-yr, the bed will again not work due to the varying duration for these other storms (i.e. 100-yr duration is 60 min, 25-yr is 42. model will show bed top being exceeded if I use a duration of 42 for 25-yr, but works fine with 60 min). The model I have set up is attached (as is the rainfall data file for it), the bed in question is BMP 4. Ideally I'd like to figure out how to make the bed work at the size it is at since the size is optimized for the 2-yr storm, while providing enough of an orifice in the OS to discharge the 100-yr storm inflow rate. If it is not possible to make it work as it is currently sized, I would like to have a good explanation to satisfy my boss. Hopefully someone here can help because I am stumped. Thanks!

View attachment PA -Springettsbury Township.hci

View attachment Post Rate FINAL rev bed design.hcp