From: Winfield Hill
Subject: Re: TI's low-cost 20W "filterless" class-D amplifer IC
Date: 23 Dec 2002 04:24:35 -0800
Organization: Rowland Institute
X-Newsreader: Direct Read News 4.20
John Woodgate wrote...
> Kevin Aylward wrote:
>> So, do you notice the "creative" specmanship. They show graphs
>> of THD, with 20Khz around 0.01% and 1Khz around 0.06%. Wonder
>> how they do that :-)
> It's called 'insufficient bandwidth'.
Right, a sharp filter in the path of the audio analyzer, justified
to keep the switching waveforms out of the instrument's sensitive
circuits. :>) They discuss these filters in Claus Neesgaard's
slaa114 app note, "Digital Audio Measurements."
They use AES17 filters, named for "AES Standard Method for Digital
Audio Engineering—Measurement of Digital Audio Equipment," AES17.
"Audio Precision (AP) has implemented a new low-pass filter type
in accordance with AES17 §220.127.116.11 which, in AP’s terminology, is
referred to as the AES17 filter." If a new-style AP system isn't
available, an "existing 20-kHz bandwidth-limiting filter, together
with a passive low-pass filter between the device under test (DUT)
and the audio analyzer is recommended."
So it would seem useless to show any data for distortion at 20kHz;
TI should have shown 5kHz instead, to reveal if the distortion is
increasing with frequency, e.g. 0.06% at 1kHz and 0.3% at 5kHz.