For the reasons that Erik and Andy state - using time and current for sensing motor position is fraught with problems - and we all repeat the same lesson! I've seen these techniques fail on two occasions: one was a garage door opener - any variation in the friction of the door would cause the system to fail and second was a release mechanism. Again, a bit of corrosion, more friction and the system becomes unreliable - just to save a dollar or two on some microswitches. The initial saving was undone manyfold.

It can be made to work beautifully in the workshop - 100% reliability, but put it out into the real world and watch it fail. Even the auto makers have learnt this lesson. Keep it simple - use a switch! Have the micro sense the current and have a timeout to protect the motor and to provide diagnostics by all means.

Mehdi - as for the current sense resistor - use ohms law to calculate the resistance needed to produce about 0.7V with the maximum current you expect. The 2051 has a comparator, you can use a silicon diode as the 0.7V reference - be aware that it changes at about 10mV/C. Even with the micro sensing the current, I would suggest you also put a fuse (slow blow) or a polyswitch - you can never trust the micro in an emergency.

You haven't mentioned how critical the operation of this valve is - if it fails what happens? Risk of explosion, property damage, death? This needs to be considered.