[visky]

Hello!

I would like to start discouse how to design circuit for large brush dc servo motors with PWM & direction control signals and IR compensation mode.

So, PWM & direction signals are converted to analog .

[samco]

I have goofed around with a simple h-bridge design that uses pwm+pwm for input. The latest version has been lightly tested with good luck. Would not take much to make it pwm+dir. (a bit of logic). It uses lowside current limit and blanking. (designed for about 20khz pwm)

http://www.cnczone.com/forums/showthread.php?t=61437
http://www.cnczone.com/forums/showth...=25929&page=24

http://www.electronicsam.com/images/...tcurrentlimit/

(use at your own risk )

sam

[visky]

Would this shematics work? It is taken from UHU servo controller.
I would like to avoid ic4071 .

I would go up to 180 -200 Volt and around 12Amp

Of course i would add opto for bouth pwm channels and enable to flip-flop data pin

[samco]

They are using diodes instead for the or gates. (I just didn't like it - just me ).


I would really look at blanking. Otherwise for higher current you have to filter the crap out of the current limit sensing to get rid of the mosfet/igbt turn on spike.

sam

[visky]

i made some modification - check it

You will se ? becouse i do not understand what it does.

Could i use 0,1R 5Watt resistors insted of 0,015 ( R11 ) ?

[visky]

samco,

could you export pcb circuit into gerber files ?

[samco]

what you have questioned is the blanking. The flipflop fires an aprox 5us pulse during each time the pwm signal is rising which causes the transistor to turn on for 5us. That inhibits the current sense for that amount of time (as the mosfets/igbts first turn on).

I am using eagle - I will have to look to see what kind of export I can do.

sam

[visky]

Shematic to check? Would current limit work?