Half bridge dc-dc circuit with PIC16F1827

[jean12]

Hello,so how may I have to do all signals might have the same polarity,I think they are not because what I inputted from the PIC is the same with what I have at the outpu(with the same polarity).

Have you any idea on another alternative,conside the ICL7667 pinout herehttps://www.google.rw/search?q=ICL7667&biw=1024&bih=633&tbm=isch&source=iu&imgil=ywQh6EMjVeD9LM%253A%253Bhttps%253A%252F%252Fencrypted-tbn1.gstatic.com%252Fimages%253Fq%253Dtbn%253AANd9GcSkzC-2D-Dvn5v1kwbKF6I5CafQBc7SH8tG6WrcAH0rPgdhDBUltQ%253B1000%253B554%253Bg--QZJA-GGya2M%253Bhttp%25253A%25252F%25252Fwww.intersil.com%25252Fen%25252Fproducts%25252Fpower-management%25252Fmosfet-drivers%25252Flow-side-fet-drivers%25252FICL7667.html&sa=X&ei=fnuxUurcMseG0AXCnIDgDw&ved=0CEQQ9QEwBQ#facrc=_&imgdii=_&imgrc=ywQh6EMjVeD9LM%3A%3Bg--QZJA-GGya2M%3Bhttp%253A%252F%252Fwww.intersil.com%252Fcontent%252Fdam%252FIntersil%252Fdocuments%252Ffn28%252Ffn2853.jpg%3Bhttp%253A%252F%252Fwww.intersil.com%252Fen%252Fproducts%252Fpower-management%252Fmosfet-drivers%252Flow-side-fet-drivers%252FICL7667.html%3B1000%3B554

Please help.Thanks

 

[FvM]

I think the problems of not showing an actual schematic are quite obvious.

Instead of further guessing, these are the facts so far:
- ICL7667 is an inverting gate driver
- the waveform in post #19 looks like the output of ICL7667 controlled by the code in post #16
- the waveform has overlapping high phases, creating a temporary short of the output stage. The problem can be easily solved by changing the code.
- the gate driver supply is apparently 5V, as in post #9. That's too low for low FET losses, 10-15V would be good
- the output frequency is below 10 kHz, according to post #16 it should be 1o kHz. There might be a difference in microcontroller oscillator frequency.

 

[jean12]

I think the problems of not showing an actual schematic are quite obvious.

Instead of further guessing, these are the facts so far:
- ICL7667 is an inverting gate driver
- the waveform in post #19 looks like the output of ICL7667 controlled by the code in post #16
- the waveform has overlapping high phases, creating a temporary short of the output stage. The problem can be easily solved by changing the code.
- the gate driver supply is apparently 5V, as in post #9. That's too low for low FET losses, 10-15V would be good
- the output frequency is below 10 kHz, according to post #16 it should be 1o kHz. There might be a difference in microcontroller oscillator frequency.

Hello Sir,here is the schematic of the PIC and the MOSFET I would like to control.

Thanks.

 

[FvM]

Yes, that's what I assumed in my previous post (ignoring the 555 symbol you used in the schematic).

To repeat my suggestion:
- invert the PIC output signal by changing the code, essentially writing HIGH instead of LOW and vice versa.
- use a higher gate driver supply of 10 - 15V

 

[jean12]

But if I modify the codes there will be nothing new because where was HIGH will be low and where was LOW will be HIGH,unless I don't get well the point,I think that is what you are saying by changing the codes,and also the problem is that when using 10-15V the drive become very hot also.So I choosed to use 5V,but here I have another oncern can 10V be applied on the gate of the MOSFET? This is because when applying more than 5V on the drive the output becomes in 8.5V-10V.

Is it possible that you can propose by schematic the form of signals which might be used to control the MOSFET?

Thanks,please help.

 

[FvM]

and also the problem is that when using 10-15V the drive become very hot also.So I choosed to use 5V,but here I have another oncern can 10V be applied on the gate of the MOSFET? This is because when applying more than 5V on the drive the output becomes in 8.5V-10V.

A transistor datasheet tells you about suitable gate drive voltage, also the driver IC documents give some hints. I already told my suggestion, no need to repeat it.

There's no problem with your circuit (as far as we can guess from the incomplete schematic).

But if I modify the codes there will be nothing new because where was HIGH will be low and where was LOW will be HIGH,unless I don't get well the point,I think that is what you are saying by changing the codes.

My last to attempt to visualize the problem. With the gate waveforms according to your screenshot, the output is shorted for 10 µs during each half periode. The interval is marked red below. By inverting the polariry, you get 10 µs deadtime instead of 10 µs short. (1 µs deadtime should be more than sufficient, by the way)



There's however a relation between wrong gate waveform and gate voltage level: Short circuit with higher gate voltage results in higher short circuit current.

 

[jean12]

Hello Sir,I now see where the error is,thank you for the help;now I would like to ask you if you can help me to make calculations if for any H-bridge which must output 220VAC.

Thanks.Please help.