I will never do it without debouncing, what happened was I had no data on the differential dc motor and gear part, I had to open it and see how it works. It had a 3 slip rings contacts giving some sort of code when it is completely in the right hand position and a different code in the middle, another code in the full right hand position, done C code with state machines to control the motor taking into account all possibilities even power interruption. I noticed it did work well without debouncing, guess slip rings are good as steve indicated in his post.


Sure thing, I put current sensing resistor and a comparator for sensing the motor current and included software in the microcontroller.


To be honest in development phase I tried to stop the motor with no success because it had a gearbox built in and gives huge torque which i couldn't beat. Also the japanese are good, they made the motor move completely in the engage or disengage position with no problem, the diff lock engages only when the gears get alligned, must be spring loaded somehow so there is realy no preassure on the geared DC motor. So I guess I have to implement time out which sounds good idea and it costs nothing on hardware although it is unlikely to happen.


I implemented a timeout at first, but removed it after I figured out that there is no pressure on the geared DC motor, what happens is I have another sensor for diff lock, when the differential is locked I get a signal, so what happens is when you engage the diff lock switch, I start blinking dif lock led on the dashboard untill the diff lock sensor indicates diff lock so I make the LED go on perminantly. Maybe I should re implement the time-out thing in case of problems, better to be safe than sorry although I made my controller fused also.


The DC motor gets stuck if it reaches beyound the RH or LH positions, so it can't keep rotating, what happens if my realys fail and stay on all the time is the motor will heat up causing current sensing circuit to trigger indicating a fault at the diff lock indicator with some blink code. Hopefully the fuse will blow and no further damage will occur, if the fuse refuse to blow I guess the coil of the motor would overheat and blow up eventually.

see above.


I implemented state machines so as long as the microcontroller is alive most conditions have been taken care of.

I'm using 78S05 voltage regulator to power microcontroller board, it should have some sort of overvoltage protection, the car battery is going directly to the DC motor through a fuse and relays, the micro is controlling the relays, Geez Russel it is very good job you are not my boss!


As I said above if a sensor fails the dc motor will be pushed to the limit causing an over current condition, the state machines stops the operation giving blink code alarm and keeps in this state till the diff lock switch is toggled again in case it was a glitch fault.


Well I never thought of that, do I need to implement temp sensing you think?
Do I sense the temp of dc motor? quite difficult because it has a thick shell around it. Do I need to sense temp of my control circuit?
Although i like to make my design solid but remember this is not a life saving equipment, if the worst happened the diff lock gears would break down and the vehicle works with normal differential. It must be fixed at garage by professionals.


True i guess all my testing is beeing done in the field using brand new DC motor and diff lock mechanisms, Thats why everything is working perfectly, but this is straight forward and easy stuff, all I have to do is control the motor one way or another till I get sensors readings then I implement dynamic breaking connecting an 18 ohm 30 what power resistor across the dc motor to make it stop fast.


You are absolutely right, thank you very much for your great comments.
Mahmood