Need help selecting inductor for boost regulator.

Other Hardware Motors, Mechanics, Power and CNC
1

Hi guys,

I'm working on a little regulator board which can boost 5V to 12V and be switched on and off. It will need to output 3W (250mA) peak power.

For this board I have selected this regulator, which, according to the convenient schematic for a 5-12V boost, appears to be able to handle the current I need:

I also found this datasheet detailing the layout of the regulator's evalulation board:

(The evaluation board datasheet says it's for the LM2731, but this regulator is part of a family of regulators which all have different power output capabilities, and they all use the same components in their schematics. So they're basically interchangeable.)

Anyway, I've gone through all the components and selected appropriate substitutes for those suggested, all except one... the inductor. I don't know a lot about inductors, having never used them before, so I want to make sure I select an appropriate one.

The datasheet for the evaluation board lists this one:
http://parts.digikey.com/1/parts/641224-power-inductor-10uh-4-0a-smd-cdrh125-100mc.html

But Digikey doesn't carry it, and I don't know how much it costs. I can see though that it is rated 10uH 20%, 4A, 25mOhm DCR, and 100C max temp. It is also shielded, and I believe it has a ferrite core, but I'm not certain about that.

I would like to use an inductor that will work with the LM2731 shown on the evaluation board just in case I decide later that I need to use the beefier regulator.

Here is the datasheet for the LM2731 btw:

If I used that regulator, its 6W max output should correspond to 500mA @ 12V which seems a lot lower than the 4A rating on the inductor they suggested. And if you look at page 13 it shows that the X model's load current when boosting 5V to 12V maxes out around 600mA, which still is far less than the 4A rating on that inductor they suggested.

But, if I do the calculations on page 12 to see what the peak current would be if the output is 500mA and the duty cycle is 62.5%, I get 1.42A. And though that is much higher, it is still well below the current rating of the inductor they suggested. (It does indicate to me though that I should choose the current rating of the inductor based on the input current rather than the output... I would calculate 1.5A with a 5V input to get 500mA @ 12V output. That's really close to the 1.42A that equation gave me.)

So, what I want to know is do I need an inductor that can handle 4A, or would one which can handle 2A do? And what about this one here, which can handle 3A? Would it be fine?

Also, how's the DC resistance and self resonant frequency on that look? The datasheet doesn't seem to specify what range of values these should be within, so I'm not sure if those are appropriate values or not. I notice the DC resistance is a bit higher than the part they recommended but I don't know how much that matters. I also don't know if the resonant frequency is important. I'm guessing there's some important relation between that and the frequency at which the boost regulator does its PWM.

Oh, and the reason I chose that particular inductor is it's surface mount, ferrite core, shielded, 10uH, the DCR didn't seem too high, and the current rating was pretty close to the one they suggested. DigiKey also has a ton of them in stock and they were among the cheapest available. I'd prefer not to choose a part where they have less than 2K in stock.

2

Hm, well I see why that one particular inductor is cheap. It's pretty large. 8mmx8mm. I'd prefer to use something a bit smaller.

3

Thinking about using this one:

Digikey's got 100K of them, they're cheap, they're shielded, and they have a ferrite core. They're also 6mmx6mm which is reasonably small.

I calculate for that 6W boost regulator that the peak inductor current would less than 1.7A. (And that's if pushed well past 6W.) That inductor is rated for 1.9A and has a saturation of 1.9A. I think that means it's within spec.

And if that's within spec, then I should have nothing at all to worry about if I go with the 3W regulator.

I would like to know if the DC resistance and self resonant frequency matter at all, but I see inductors with much much higher DCR ratings and the resonant frequency is far higher than the switching frequency of the regulator, so even if matching the resonant frequency is bad, that can't happen.

So I guess unless anyone has any objections, that's the one I'll go with.

4

Inductors are problematical - they are nowhere near perfect ideal components and they saturate. Saturation is a killer for switch-mode operation - the current will skyrocket if saturation is reached.

You also have to consider core losses and how hot it wlll get - too hot and its effciency will decrease, leading to more heating and then saturation and failure.

Typically you can get much more linear behaviour from an inductor with a gapped core or incomplete magnetic circuit (the air gap is a very linear magnetic material and it doesn't saturate :wink:

To play safe meet or exceed all the specs of the coil they recommend: >=4A current rating, <=0.025ohm resistance. The high current rating is probably to allow saturation-free startup and may be crucial.

5

what did you end up using for the inductor scswift?