AleXYZ
Junior Member level 3
My end goal is to produce 170-200 volts DC at up to 5 watts. The circuit on my bench at the moment is a simple push-pull using a CCFL transfomer I bought off Digikey. However, I cannot seem to tune the circuit. It is drawing too much current and the MOSFETs get warm after a while.
Here's the circuit (stripped down):
The power supply is an Agilent unit with 6.00v dialled in and a maximum current of 0.25 A. When I apply power, the current limit pegs out and voltage drops to about 3.34V. The output is about 100v, but it is so weak that it barely powers a neon bulb connected in parallel to the capacitor.
The MOSFETs are being timed by a PIC chip in advanced PWM mode with a dead-band delay of about 1 usec. Between the PIC and the MOSFET gates is a TC4424 MOSFET driver. I can vary the frequency between 5kHz and 60 kHz, however changing the frequency does not appear to change the results.
The end goal is to make 170-200 volts efficiently. Weight and size are crucial factors which is why I decided to try a CCFL transformer. Yes, I am aware that they are intended for 1000+ volt outputs, but I only need it to make 200 volts max. The final circuit will have feedback and regulation (not shown or indicated in above diagram).
Details on the transformer can be found here:
**broken link removed**
If you have an idea how to tune the above circuit so it works efficiently, or if you have a suggestion for a different kind of commercially available transformer for this application, please let me know.
Thanks for your help!
---------- Post added at 07:57 ---------- Previous post was at 07:16 ----------
Hmm. Found a PowerPoint presentation from National Semiconductor which very nicely explains push-pull theory. Correct me if I'm wrong, but it seems that my problem has to do with having almost a 100% duty cycle?
It's late (2am) and I've spent all day on this project. Tomorrow I'll try increasing the dead-band delay and see what happens.
Here's the circuit (stripped down):
The power supply is an Agilent unit with 6.00v dialled in and a maximum current of 0.25 A. When I apply power, the current limit pegs out and voltage drops to about 3.34V. The output is about 100v, but it is so weak that it barely powers a neon bulb connected in parallel to the capacitor.
The MOSFETs are being timed by a PIC chip in advanced PWM mode with a dead-band delay of about 1 usec. Between the PIC and the MOSFET gates is a TC4424 MOSFET driver. I can vary the frequency between 5kHz and 60 kHz, however changing the frequency does not appear to change the results.
The end goal is to make 170-200 volts efficiently. Weight and size are crucial factors which is why I decided to try a CCFL transformer. Yes, I am aware that they are intended for 1000+ volt outputs, but I only need it to make 200 volts max. The final circuit will have feedback and regulation (not shown or indicated in above diagram).
Details on the transformer can be found here:
**broken link removed**
If you have an idea how to tune the above circuit so it works efficiently, or if you have a suggestion for a different kind of commercially available transformer for this application, please let me know.
Thanks for your help!
---------- Post added at 07:57 ---------- Previous post was at 07:16 ----------
Hmm. Found a PowerPoint presentation from National Semiconductor which very nicely explains push-pull theory. Correct me if I'm wrong, but it seems that my problem has to do with having almost a 100% duty cycle?
It's late (2am) and I've spent all day on this project. Tomorrow I'll try increasing the dead-band delay and see what happens.