Poor THD at low frequencies - more a hint than a question

Hello RLutz215,

You never mentioned which eval boards you are using but we use these on a lot of our boards. The reason has to do with cost and size. If we place a large electrolytic or a more expensive type like a tantalum, some customers will think they HAVE to use something like that for their design and rule out the DSP for that reason. So for many audio applications this distortion at 20Hz will not be important at all. So it is the usual story of tradeoffs. It would be great to use a great big film cap!

You are certainly correct that these caps are not the best for very discriminating users. For those users we would catch flak for whatever we used. So it is good that you know this and are working on a great sounding design.

Dave T

Hi Dave,

I used the VAL-ADAU1452REVBZ-UG-1662 board. In the meantime we have developed our own board which reaches -100 dBTHDN over the full audio frequency range.

The problem stems from the series connection of 10µF and 5 kOhms. At low frequencies the impedance of the capacitor is high and so the voltage drop, which in turn modulates the capacity. It would make sense to go for a non-inverting stage, in case the eval board is re-done one day. This allows to use small capacitors with higher quality.

Thanks for your reply.

All,

Yes, high-gain ceramic caps are quite nonlinear -- the worst are so bad that it's difficult to measure their capacitance with certainty. We had to switch from ceramic to film in one application for a different reason -- the ceramic was so microphonic that we could hear vibrations loud and clear.

Best regards,

Bob