Ask a way to pass high frequency PWM trough PC817（PC847）？How to realise this goal trough push-pull？
I am using PC847（datasheet ）to isolate a PWM control signal from CPLD. I will describe my design .When duty cycle changes from low to high, the forward current changes from low to high, so I have to keep Vce a stable vaule. I refer to Fig12 on page 7,
I will keep Ic=0.5mA and even if forward current is very low ,the Vce is not very high. As a result the duty cycle of output signal can have relations to input PWM. Then, I refer to Fig15 and 16
In order to improve frequency response, I have to keep Rload small. because my Ic can not be high, I have to use smaller VCC and I choose 1.8V. At last, below is my final design: However, the frequency response is pool when I use 10Khz PWM. the output is not a symmetrical wave. and the Tr is about 15.6us，Tf is about 1.4us.Can anybody give some suggestions? I hear that a push-pull way can improve frequency reponse of PC847 becuase Tpdh (propagation delay high) and Tpdl(propagation delay low) are always different. Does anybody know how to realise it?
I am using PC847（datasheet ）to isolate a PWM control signal from CPLD. I will describe my design .When duty cycle changes from low to high, the forward current changes from low to high, so I have to keep Vce a stable vaule. I refer to Fig12 on page 7,
I will keep Ic=0.5mA and even if forward current is very low ,the Vce is not very high. As a result the duty cycle of output signal can have relations to input PWM. Then, I refer to Fig15 and 16
In order to improve frequency response, I have to keep Rload small. because my Ic can not be high, I have to use smaller VCC and I choose 1.8V. At last, below is my final design: However, the frequency response is pool when I use 10Khz PWM. the output is not a symmetrical wave. and the Tr is about 15.6us，Tf is about 1.4us.Can anybody give some suggestions? I hear that a push-pull way can improve frequency reponse of PC847 becuase Tpdh (propagation delay high) and Tpdl(propagation delay low) are always different. Does anybody know how to realise it?
No comments:
Post a Comment