-2
\$\begingroup\$

I need idea to design 800-1000V to 5V 1W buck regulator.

I have a very narrow space so I don't want to use flyback controller. Maybe monolithic ones.

Do you know any ic or any idea?

\$\endgroup\$

closed as off-topic by pipe, Chupacabras, PeterJ, Harry Svensson, laptop2d Dec 20 '17 at 5:48

This question appears to be off-topic. The users who voted to close gave this specific reason:

If this question can be reworded to fit the rules in the help center, please edit the question.

  • 6
    \$\begingroup\$ 1 kV and very narrow space sounds like a good mixture for fireworks. \$\endgroup\$ – Arsenal Dec 18 '17 at 14:20
  • 7
    \$\begingroup\$ By the way. Maybe you could explain a little bit about the application, because people raise eyebrows and downvote, while you may have very good reasons for such specs. \$\endgroup\$ – Gregory Kornblum Dec 18 '17 at 14:25
  • 4
    \$\begingroup\$ "I have a very narrow space" - exactly how much space do you have, and why? \$\endgroup\$ – Bruce Abbott Dec 18 '17 at 14:39
  • 1
    \$\begingroup\$ Another by the way: circuit may be submerged to oil to be more compact under higher voltage. If you do this, don't forget to tell how it goes. \$\endgroup\$ – Gregory Kornblum Dec 18 '17 at 14:57
  • 3
    \$\begingroup\$ @GregoryKornblum Assuming he's still alive. :) \$\endgroup\$ – pipe Dec 18 '17 at 15:51
1
\$\begingroup\$

Here's what I thought:

1- TRANSFORMERLESS SOLUTION

If you go for a transformerless power supply approach (i.e. reducing the voltage with series capacitor) then the series capacitor will be quite big and power dissipation for other series elements (if any, e.g. regulators, resistors etc) will be high for required output current.

2- FLYBACK CONVERTER

Even if you design a flyback converter, the spikes at primary side will be extremely high and snubbing them will be hard. Choosing the primary MOSFET and primary-side snubbers will be problematic.

3- CONVENTIONAL BUCK CONVERTER

Duty-cycle will be around %0.625 at 800VDC input. From \$V_L = L \ di/dt\$, required L will be high. You can increase the frequency to decrease L but what about the pulse width when D<<1?


So,

  • You can make a converter consisting of cascade-connected sub-converters (can be any combination of the ones above), or
  • You can go for the conventional "step-down transformer + rectifier + capacitor + linear regulator" approach.
\$\endgroup\$
  • \$\begingroup\$ Low duty cycle just means constant pulse width and frequency modulation. \$\endgroup\$ – Gregory Kornblum Dec 18 '17 at 16:03
  • \$\begingroup\$ Thanks for answer it's look like these are only options for me . Now i need to make some calculations for which one is best suited. Thanks😉 \$\endgroup\$ – Emin Asım Yılmaz Dec 18 '17 at 16:12
  • \$\begingroup\$ @Gregory You're right but I meant that the pulse width will be extremely low. For example, if f=20kHz and D=0.625% then the pulse width will be 313ns. I think that it will not be a pulse, it will be a triangular "thing" instead. \$\endgroup\$ – Rohat Kılıç Dec 18 '17 at 16:36
  • \$\begingroup\$ Make it 1kHz. Should be enough. Besides, thank you, i needed someone to say i am right after that day... \$\endgroup\$ – Gregory Kornblum Dec 18 '17 at 16:39
1
\$\begingroup\$

Here the solution i found. First divide voltage by 2 with caps(around 10uF) after that using st's integrated 800V mos buck controller Viper06 to make 12V and than standart regulator to make 5V. St not essential many companies have 600-700V controller.

\$\endgroup\$
  • \$\begingroup\$ So capacitor divider, bridge (or diode?) then viper? \$\endgroup\$ – Gregory Kornblum Dec 19 '17 at 9:12

Not the answer you're looking for? Browse other questions tagged or ask your own question.