heating in inductor ( LC Filter )

[yanuarbob5150]

hi all, i am in final step to design 100 W rms class D power amplifier. Recently, i am in experiment to design low cost power mosfet, now i have problem in inductor ( part of LC filter ). The inductor being warm and at several minutes the inductor hot!!!!!!!!!!!

in LC filter, the inductor design i use toroidal core ( dimension not measured ) with AWG cooper - wire with diametre 1.2mm. I use LCR meter and get the inductor values 21.7uH.

can anybody help me to design the inductor with proper concepts?????
escpecially in relation with heating of an inductor.

thanxxx

[dweekie]

You may want to browse this for some ideas.

http://www.41hz.com/Forums/topic.asp?TOPIC_ID=1769

[Ouroboros]

I have found that inductor core losses (not copper losses) usually dominate the total loss (at idle), especially if you use the wrong type of core. Many commercially available off-the-shelf inductors are not designed for operation at the 300kHz or so that a Class-D amplifier will be working at. You usually have to use an inductor with a much larger core than you might expect to keep the losses low.

I have had very poor results using dust-iron toroids. The best toroidal cores I've used for the output filter have been the fairly recently introduced gapped ferrite toroids from Ferroxcube that are designed for this sort of operation. These are made of a high-grade ferrite with a precision gap sawn in them, which is then resin filled to restore the mechanical strength. Alternatively, the traditional RM8 and RM10 cores work well, and are available in a range of different ferrite types. The documentation for RM cores from different manufacturers is easy to find on the web.

[Pafi]

Generally: the higher the number of turns, the lower the core loss. A bad iron powder core also can be made better by cutting a narrow gap (or two, 4, etc...) into it. (Gap->low Al->higher number of turn->smaller dB/dt->less core loss.)