Selecting inductor for buck converter.

I'm trying to find a suitable inductance for the TI TPS54283
"2A Dual non-synchronous converter with integrated high-side mosfet".
Calculating the inductance value in [H] and the current [A] is no problem.
The DCR [ohm] should aswell be kept low. So far so good, but then there seems
to be few other parameters to need to be determined:

* Allowable DCR range
* Q @ Frequency
* Self resonant frequency
* Shielding
* Tolerance
* Type:
Drum core
Ferrite core
High power, High current
Metal composite core
Powdered iron core

How does one find suitable values for these characteristics ..?

Links:
http://focus.ti.com/lit/ds/symlink/tps54283.pdf
http://search.digikey.com/scripts/DkSearch/dksus.dll?Cat=196627

 

Apply ohms law and hot much temperature rise you can stand.

You don't usually think in turns of Q for such inductors. Many makers
will spec a suggested AC current level. That is a more direct
indication than Q. You need the Q to be good all the way from the
switching frequency up to about its tenth harmonic. Think of the Q as
a resistor in parallel with the inductance and consider the power into
that resistance.

Way above the operating frequency of the switcher.

This is needed mostly of EMI reasons.

Assume about 25%

 

You must calculate the current waveform that will be going through the
inductor. You want both the average DC level and the p-p current ripple.
Find an inductor that can handle the DC+peak current expected on the output
+ ~20% for margin.

Look at the frequency range the current is rated for and compare to you
switching frequency. You want these to overlap so the switching losses will
have been measured and guaranteed by the manufacturer. Self resonant
frequency should be at least 5 times the switching frequency preferably
higher.

Shielding is very nice so the PSU noise is less likely to radiate into
adjacent circuits.

For tolerance, make sure the supply can handle the max and min inductance.

Type is typically linked to the degree of shielding availale. Drun core
with shields, torroids and ferrite and composite cores are common.

 

Also check the saturation effect at the peak. You want to not be too
far into saturation at the peak of the waveform. It is not uncommon
for the peak current rating of an inductor to be where significant
saturation is happening. This will increase the rate of rise in the
current. Stay short of this point if you can.

I say 10 times. A lot depends on the exected duty cycle. Do an FFT
on the expected waveform.