Update:6-18-2014 This was not the real issue, the real issue was overheating the transistor, i added a second zener to the circuit to add some feed back so it is not blindly dumping power.

I specked in the MPSA42 npn transistor so if i need to to later it would also work as a high voltage transistor in case i need more power for driving SCRs/Triacs  anyways i need to read data sheets better because its not a very good chose for the power supply circuit.  It responsible for driving the fet low however in practice it create an oscillating gate pin that is just amplified by the fet. SO now i need a transistor that can drive the fet low faster without oscillating.

http://www.fairchildsemi.com/ds/MP/MPSA42.pdf

The big issue is the DC current gain aka: how well the transistor amplifies signals in this case its 25-40 so i happen to have a few other tranisters unfortunately in other pin orders and foot prints, anyways so i had a pack of tip120 transistors sitting around. TO220 package i think they fit just fine:)

http://www.onsemi.com/pub_link/Collateral/TIP120-D.PDF

Anyways on the TIP120 it has a gain of 1000 much better at driving the signal to the fet low. 

Here is what the corrections  look like:)