I'm still very dubious this will actually make any profit.  At the current rate of difficulty increase (may was 74k low and june was 124k low) 32 chips, 9 g/hs would make generally nothing later in the year.... anyone else done the math?