LM3886 Done Right

The performance measurements of the LM3886 Done Right are shown below. The THD+N vs output power matches the performance specified in the LM3886 data sheet.

The THD+N vs frequency is shown below. The performance of the LM3886DR is actually better than the data sheet performance of the LM3886! This speaks volumes to the quality of the layout of the LM3886DR.

Repeating these measurements with a 4 Ω load provides the following results. As expected, the THD+N is slightly higher for a 4 Ω load.

The THD+N versus frequency plots above were both measured with a 60 kHz measurement bandwidth to include at least three harmonics of 20 kHz. However, in particular for Class D amplifiers, it has become increasingly common with measurements of THD+N versus output power at three frequencies: 100 Hz, 1 kHz, and 6.67 kHz. This allows for capturing at least three harmonics of all test frequencies within a 20 kHz measurement bandwidth. I measured the LM3886DR accordingly and the result is shown below.

The THD at 1 W output into an 8 Ω load is quite low as seen in the harmonic spectrum below. It is interesting to note the mains-related IMD spectral components. At nearly 110 dB below the fundamental signal, they’re hardly an issue. Just an interesting testament to the power supply rejection of the LM3886.

The mains hum itself is actually very low as shown in the graph below. 

The multi-tone IMD is shown below. The tallest offender checks in 112 dB below the peak of the signal, which was set to provide 45 W into 8 Ω.

The 18 kHz + 19 kHz (1:1) two-tone IMD test is often used as an indicator of the loop gain available at 20 kHz. As shown below, the IMD of the LM3886DR is quite good.

The 60 Hz + 7 kHz SMPTE IMD, on the other hand, is often used to tease out thermal issues in the amplifier. As shown below, the LM3886DR passes with flying colours.

Siegfried Linkwitz argues that the 1 kHz + 5.5 kHz intermodulation distortion (IMD) measurement is one of the measurements which is more indicative of the perceived sound quality. He bases this argument on the fact that IMD products in this measurement fall in the frequency range where the ear is the most sensitive (see the Fletcher-Munson curves for more detail). I think this argument carries a good amount of weight, so I measured the LM3886DR accordingly. The measurement is shown below. Note that due to a limitation in the DFD IMD source of the APx525, the frequencies used must be an integer multiple of each other. Thus, I measured at 917 Hz (5500/6) + 5.5 kHz. I performed this measurement at 1.0 W. The result is shown below. 

Finally I measured the gain and gain flatness within the audio range. As seen below, the gain is very flat, with a slight rolloff towards the upper end of the audio band.