While reviewing recently a Buck converter design, I came across the issue of the inductor losses in a PWM converters. I have found, perhaps not surprisingly, that magnetic core and device manufacturers are still testing our patience when it comes to the simple task of providing us with clear data and design information. While Googling to see what are the loss calculation methods suggested by the manufacturers, I have found some bizarre suggestions such as to first calculate the magnetic field strength (H) at the two ripple current extremes, or to look up plots showing the inductors’ ESR (?) as a function of frequency. Core losses are related to magnetic flux density variations (deltaB) and there is no need to involve the ripple current, let alone H. And for the “ESR”, since the losses are a function of detaB (which is of course linked to ripple current magnitude) you cannot define an inductor ESR, unless you will assign an ESR to each ripple amplitude and frequency.
Other vendors offer confusing loss calculation instructions which you need to decipher.
Here is one which I pose as a riddle:
Given a Buck converter with input voltage Vin and output voltage Vo having an inductor L (at no bias) and running at a switching frequency Fs under CCM. The inductor DC resistance is RDC and the inductor DC current is Idc. The datasheet shows that at the given Idc the inductance drops by K%. The vendor’s guidelines for calculating the core loss are to use the equation:
deltaB=K*Ipp*L (K is given in the datasheet),
and then to use the given loss plots (power loss as a function of frequency with Fs as a parameter).
And the questions are:
1. Which L and ripple current Ipp to use in equation? nominal or under Idc? And why?
2. What would be an alternative way to calculate the core loss?
And most importantly:
3. Which information is missing making the loss calculation only an approximation at best?