Update,
Thought this may help anyone else with the same problem.
Added a Schottky diode between the coil to help dissipate the voltage spikes. The FET is now so cool I can hold it. Checked the job on the scope. See attached results before and after fitting the diode.
Thanks all
Thanks for the update. We all appreciate knowing what happens after our advice.
I assume you mean "across the coil" rather than "between the coil".
Some users just take the advice and are never heard of again (unless they have another problem >:( ).
Weedpharma
chucktodd:
Put a fast Schottky diode from Source to Drain. Cathode (banded side) to Drain.
Every time the PWM cycle turns off, all of the energy stored in the coil is being dissipated in the body diode of your MOSFET.
If the body diode is a Zener as implied in your diagram.
chucktodd:
That diode is not very efficient. so it heats up. Use something like a 1N5822 (3A 40V Schottky Fast Recovery Diode).
What does "efficiency" have to do with it?
The diode across Q3 is unnecessary as it is already contained in the FET.
chucktodd:
Try this circuit:
I really would not advise that! 
Paul__B:
If the body diode is a Zener as implied in your diagram.
What does "efficiency" have to do with it?The diode across Q3 is unnecessary as it is already contained in the FET.
I really would not advise that!
The body diode in the N-MOSFET exists because of the structure of the MOSFET, It was not built to optimized with the lowest Vfw. This increased forward voltage creates the 'heat'.
Take for instance a BUZ11 N-MOSFET, (Fairchild Semiconductor):
as compared to the 1N5822
You can see that the Vf of the the 1N5822 is only 0.525V with 3A of current.
They don't rate the Body diode of the MOSFET in the same range, but what they do rate:
typical 1.7V with a max of 2.6V at 60A.
The 1N5822 has a graphic that show up to 15A:
That 0.8V to 1.7V difference produces a lot of heat.
The flyback diode across the inductor may have been enough for @Jim_Cliff, but relying on the Body Diode will create situations were repetitive stresses will destroy the MOSFET. A few pennies for a Schottky diode v.s. failure. I know which one I will choose.
Chuck.
The question is - what conditions in that circuit would ever cause the body diode to conduct?
And there is a major blunder elsewhere.
