Disclaimer! This is just my opinion and I have no experience, education or knowledge to back it up. 
I peruse this forum often and came across this FAQ: Laser not turning off between cuts? (burning during travel moves), and buried in there was this line:
The drive circuit that turns the laser power off and on tends to go first, before the part that controls the overall power level.
Now I’ve often seen on DIY lasers the LPSU on/off pin controlled by the PWM signal, and the control pin by a constant voltage set by a potentiometer. I could never understand why and I chose to wire mine such that the on/off pin is switched by M3/M5 and the power is varied by the control pin by the S value in the Gcode line, (with both a pot and trimpot in place). Simply felt better to me. I never wanted to get in to a discussion on the merits of this until I read the above line. It doesn’t surprise me that this circuit fails considering that the specs on my power supply, (a Cloudray DY13), suggest that it’s stress tested at the factory to 500 times over 7 seconds, (approx 70Hz), where a modest 200 microsecond PWM signal would be switching it 5000 times per second, (5kHz). Doesn’t sound like it’s designed for it to me. Now this is not to say that all LPSU’s are designed the same way, and I sure don’t know anything about what makes them tick. Just offering up an opinion.