View Full Document

17 views

Unformatted text preview:

DESIGN TIPS High Frequency Power Inductor Design In this article we continue the theme of custom versus off the shelf magnetics design The relatively simple output inductor of a forward converter is used as an example to show the issues and pitfalls involved in trying to use off the shelf inductor designs for high frequency power supply applications By Dr Ray Ridley Ridley Engineering Power Inductor Design for overcurrent situations When I first started my career in power electronics over 25 years ago all magnetic components were custom designed and manufactured In our current power supply design workshops we dedicate an entire day to the theory and practice of making custom inductors for switching power supplies Attendees at the course design analyze and build their own inductors and transformers in the lab 1 In the last few years numerous companies have developed kits of standard parts aimed at the semiconductor companies to integrate into their designs This proliferation of parts has recently led me to wonder if custom inductor design was becoming a skill belonging more to the past than the future The off the shelf inductor also 47 H had a single number assigned to it on the parts kit a current rating of 5 A Since the parts kit is aimed at the nonexperienced power supply designer it seems reasonable to assume that the inductor is suitable for a 5 A converter application Figure 2 shows two inductor designs one developed in our workshop for a rugged 60 W converter and the smaller low cost off the shelf part You can see a significant size advantage of the standard component The 47 H custom inductor was designed to operate continuously at 5 A with plenty of margin Figure 1 Building custom magnetics in our design workshop 14 The parts are quite different in their design as shown in Fig 3 The custom made inductor has a singlelayer winding on an RM8 core with a gapped center leg The windings are a significant distance away from the center leg gap essential to avoid proximity losses in the windings at high frequencies The off the shelf part is very different a small drum shaped core with many turns wound on the core in multiple layers The air gap is large and the windings are all situated in the middle of the gap This violates all the rules Figure 2 a 47 H custom inductor and b off the shelf commodity component Power Systems Design Europe March 2007 DESIGN TIPS Figure 3 Cross section of a RM8 core and b drum core showing winding layouts Figure 5 Primary switch current waveforms for a forward converter with a 47 H inductor a proper operation with RM8 custom design b drum core after 30 seconds at room temperature ambient c drum core after 60 seconds d drum core after 3 minutes operation rose further shown in 5 c Testing was stopped after the waveform of Figure 5 d showed complete saturation and smoke started coming out of the windings Figure 4 Forward converter schematic Output is 12 V 5 A switching at 100 kHz The reset circuit of the transformer is omitted for simplicity of high frequency inductor design as I know them but it s hard to argue with the size cost and current rating of the part compared to a custom design Forward Converter Testing So how do these parts compare Each was inserted into the forward converter shown in Fig 4 and run at 4 5 A 16 Initially the waveforms for the two parts look the same as shown in Fig 5 a However after 30 seconds of operation the drum core design became very hot The primary current started showing curvature shown in 5 b indicating the onset of saturation of the core material The inductor temperature quickly Clearly the rating of 5 A listed on the parts package was not accurate As a power engineer knowing that a 5 A rating could have many meanings I referred to the data sheet on the manufacturer s website and found that this current referred to saturation current not continuous current The continuous dc current rating was only 2 8 A Power Systems Design Europe March 2007 DESIGN TIPS The next experiment used a 15 H part with a stated current rating of 9 A This was also a saturation current and the continuous dc current was listed as 5 A The waveforms of Fig 6 show the results Once again the inductor started out fine with a 4 5 A load but began to saturate after one minute of operation In this case the dc current rating should have been sufficient but another level of analysis is needed The inductor design is inadequate for the high frequency ac current imposed upon it and for this there was no proper specification provided Inductor AC Resistance Measurements The problem of the smaller inductor is clearly illustrated by measuring the frequency dependence of the resistance of the inductor windings Fig 7 shows how this resistance changes up to 1 MHz for both the drum core and the RM8 design Figure 6 Primary switch current waveforms for a forward converter with a 15 H off the shelf drum core inductor a initial operation b after 60 seconds at room temperature ambient c after 3 minutes operation The dc resistance of the windings for both designs was approximately 55 mOhms However at 100 kHz this rises to 1 6 Ohms for the drum core and to 100 mOhms for the RM8 core design In other words while the dc resistance and inductance of the parts are the same the resistance of the small off the shelf part is ten times higher at 100 kHz The rapid rise of resistance of the drum core design is due to the proximity loss with the multiple layers of windings located in the gap area of the core 2 The inductor is basically unusable with any appreciable amount of ripple current at the switching frequency Summary By all means take a look at standard off the shelf parts when designing power inductors Be aware however that the apparent ratings may greatly overstate the capabilities of the part in your application You need to dig deep into the data sheets to find out whether that part is truly acceptable in your circuit If the inductor looks far smaller that you anticipated it s probably inadequate for the job In a rugged power supply you should never under any conditions of line load and temperature and transients see curved waveforms like those shown in this article Those waveforms indicate failure of the magnetic core material and will lead to breakdown of the winding insulation or semiconductor failure And finally if you need the best inductor for your application go ahead and design it properly yourself Off the shelf parts are optimized for cost in high volume


Access the best Study Guides, Lecture Notes and Practice Exams

Loading Unlocking...
Login

Join to view Inductor Design and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view Inductor Design and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?