Hello.

I was thinking of trying out something that relates to robotic. I am still a beginner.

I've been reading up a lot through the web recently...
Okay, so here's the idea.

I understand the theory in having a h-bridge for dc motor, so I tried simulating using 4 NPN transistors and it works fine. I checked around and found that 2 NPN and 2 PNP transistors can also be use. The question is what's the difference?

And for configuration wise, if I use MOSFET to replace the normal BJT transistors, is it the same? Gate as the base? Drain as the collector? source as emitter? No extra thing to take note about? I can add a diode for extra protection even if there's one inside a MOSFET.
Another concern is how does a optocoupler works actually? I read it up but still have some doubts. An LED lighted up and forces an electrical contact with to the transistor base so that current can flow through Bce??

Next the dc motor speed will be control by 555 timer... varying the duty cycle..

And lastly how do I use a remote control (those for toy car) to control my dc motor? There must be some IR receiver? TSOP1738? Or any other suitable one?

I do not plan to use MCT to draft out this.. maybe just a analogue based design is enough. Are there any device to control the direction of motor? Let's say to control it to move left or right. Servo can but need programming...

Everything's feasible??

Yup I'm still learning. Thanks all.

 

Quoted text from the O/P is in blue text.

I understand the theory in having a h-bridge for dc motor, so I tried simulating using 4 NPN transistors and it works fine. I checked around and found that 2 NPN and 2 PNP transistors can also be use. The question is what's the difference?

Generally, changing the two top NPN transistors (those going to the positive power supply) to PNP transistors will allow the high side switching devices to operate in the common emitter configuration rather than the common collector configuration. The benefit is that in the common emitter configuration, the voltage drop across the emitter and collector can be much lower, resulting in less power being dissipated by the transistor.

And for configuration wise, if I use MOSFET to replace the normal BJT transistors, is it the same? Gate as the base? Drain as the collector? source as emitter?

In the configuration of two N-channels and two P-channel transistors, it can be an easy substitution. You have correctly identified the correspondence between electrodes on the MOSFET and bipolar transistors.The main difference that you will see is that the MOSFET requires higher gate-to-source voltage than a bipolar transistor's base-to-emitter voltage, and that the gate on the MOSFET compared to the source, is a capacitive load with some leakage current, while the bipolar transistor's base will look like a low resistance (actually a diode) between the base and emitter. When properly designed, MOSFETs are more robust because they do not succumb to secondary breakdown. Just blindly substituting MOSFETS for bipolars probably won't work.

No extra thing to take note about? I can add a diode for extra protection even if there's one inside a MOSFET.

You won't need an external diode because "flyback" current from the motor will not exceed the current you put into the motor, and the integral drain-to-source diode is rated for at least as high a current as the transistor mode current (while the channel conducts).

Another concern is how does a optocoupler works actually? I read it up but still have some doubts. An LED lighted up and forces an electrical contact with to the transistor base so that current can flow through Bce??

http://en.wikipedia.org/wiki/Opto-isolator

Next the dc motor speed will be control by 555 timer... varying the duty cycle..

Have a look at this tread for starters:
http://forum.allaboutcircuits.com/showthread.php?t=43340


I do not plan to use MCT to draft out this.. maybe just a analogue based design is enough. Are there any device to control the direction of motor? Let's say to control it to move left or right. Servo can but need programming...

That is left as an exercise for the student.

Everything's feasible??

Not really, but with some study and experimentation, you should be able to sort it out.

Good luck with your experiments.