Homebrew PCBs

I am using a 275W Sun Lamp from GE, one of the kind that was used 
years ago to get a sun tan. It's 15 inches above the sandwich of 
positive PCB and 'flimsy' image. I tried times from seconds to 6 
minutes and each time, after having it developed in NaOH (5%) for 30s 
to 1 minute - there is no image left, all emulsion is gone. I've done 
so many tests that a test wedge method would become redundant. I 
think that since all image is gone that either the whole PCB is 
overexposed or the developer is too strong or a cobination of the 
two. Maybe you have some suggetsions. I also tried a laboratory clean 
NaOH and a new batch of PCBs- no change. The transfer method with its 
quirks of laminating, appears sweeter and sweeter.