Measure a background ADU of a few frames, average it, then input the information into here -
http://www.ccdware.com/resources/subexposure.cfm
This is give you a starting point but you'll probably fine you'll need longer.
Question however...are you sure the stars are really blown out? Is the ADU count of the bright stars already at 65k in only a 150s exposure. Unbinned will be less sensitive so I wouldn't judge 2x2 as the baseline. If you are using tools like MaximDL etc, the screen stretch function can give the appears that the stars bloated and plain ugly. The truth however is in the star centroid values. MaximDL's line profile tool can be very handy in diagnosing star related problems.