Disadvantage is speed. Try adding a dual transistor driver, one PNP one NPN, both emitter followers. That will boost speed somewhat which will lower switching losses for one.
No isolation between the Vss on the uP and on the Power supply that powers the high-current load.
A common ground between uP Vss and the power circuit is a sure invitation to noise/spikes coupling into the uP circuits, especially if it has any low-level ADC inputs.
Look at the datasheet (https://www.electro-tech-online.com/custompdfs/2012/12/6N135.pdf), the following points may be notable:
1) there should be a load resistor connected (to the base) to increase the speed and to make the rise and fall times symmetrical (Figure 6),
2) CTR is <30%, so the output current will be only 5mA even if you put 15mA in.
5mA drive would take 1us to put 5V in a 1nF gate... if using a resistor pull-down/up there will be less current available and therefore rise/fall time will be worse. Depending on the switching speed, you may need to think about some form of buffer to drive the FET.
sorry I wasn't available to reply. My requirements are to control speed of a DC motor 24V, 2.25A, 33W, 2500 rpm. If anyone have any suggestions please include with a small diagram so I can understand. Thanks
I finally got it to work. The 6N136 was damaged and I had to replace it. I use Vcc connect to output and G of mosfet. When LED of opto is on, Vg is down to zero so it turns off as opto is on. So am I doing the right way? I tried Mike's suggestion but the current from GND was too low and it was before I replaced the opto though.