Tapped buck inductor design for LED Driver

[Pato2010]

Dear All,
I'm going to develop LED Driver using tapped buck topology i want to know how to select a proper inductor for it?
I'm now in need to develop this specs:
Input 220
Output 4.95 W, 9 V, 550 mA
Ic Control LinkSwitchTM-PL LNK457DG

Also If I need to change the output power from the same IC how to change the inductor design?
Thank you, :smile:

 

[BradtheRad]

It needs to be wound with wire thick enough to carry an average 550 mA without overheating.

Its current rating for magnetic saturation must be greater than 550 mA (because the waveform peaks will be greater than 550 mA).

It will be a short duty cycle, because you are bucking 220 V down to 9 V.

I guess you will arrange the tapped inductor to create automatic oscillations. It will choose its own frequency based on its Henry value. A typical frequency to design for is 20 kHz. This is above the range of human hearing, and so it will not be distracting.

You must make sure oscillations never stop with the switching device in a conducting state, because something will fry if that happens.

 

[Pato2010]

Dear I just need a formula to get it's specs and then i'll send it to a custom designer to develop it

 

[mtwieg]

Are you sure you intend to use the LNK457DG? It's meant for flyback, I don't see how it could work as a buck converter. And the parts that are meant to function as buck converters won't work with a tapped inductor since they need to be referenced to the switching node.

 

[Pato2010]

Are you sure you intend to use the LNK457DG? It's meant for flyback, I don't see how it could work as a buck converter. And the parts that are meant to function as buck converters won't work with a tapped inductor since they need to be referenced to the switching node.

This ai an application note from Power Integration : DER-398 Please downlowd it and see

 

[mtwieg]

Okay, so they use a completely split inductor rather than an actual tapped inductor. But I think there's something wrong with their schematic, the polarity of the windings are opposed... I'm pretty sure it can't possibly work like that.

So you just want to replicate that design but make it work at 220VAC, correct? In that case you basically just want to double number of primary turns to ~80 and use thinner wire for it. So long as the operating frequency and power is not changed, the core specifications should not need to change either.

 

[Pato2010]

Okay, so they use a completely split inductor rather than an actual tapped inductor. But I think there's something wrong with their schematic, the polarity of the windings are opposed... I'm pretty sure it can't possibly work like that.

So you just want to replicate that design but make it work at 220VAC, correct? In that case you basically just want to double number of primary turns to ~80 and use thinner wire for it. So long as the operating frequency and power is not changed, the core specifications should not need to change either.

Ok mtwieg but exactly i need to know the way that he calculated this specs of the core, winding , inductance and turns ratio in details Can you tell me how? As I need to use Specific PI IC to get a range of powers with different currents and voltages.

 

[mtwieg]

Well to get transformer specs you have to start with specs for the overall power supply. At a minimum you need output voltage, input voltage (max and min), switching frequency, output current, and output ripple current. Those are needed to define the required core reluctance and turns. Once you have that you can pick a specific core and gap size, but that is usually a more arbitrary choice (maybe just make the effective gapped length ten times the ungapped length).

 

[Pato2010]

Ok I have the total Specs Already
Can you provide me with materials to do the design with formulas step by step