DC Accuracy of the SGTL5000

[recursinging]

I'm prototyping a breakout board conceptually similar to the euroshield using a Teensy, and the CS42448 CODEC from Cirrus Logic. I see you've DC Coupled the SGTL5000 and I am curious what kind of DC accuracy you have seen with that codec. I've seen the calibration sketch, how ist the tracking for V/Oct after being calibrated? Is there a lot of DC offset, and does it drift over time/temp much? Does it vary much between channels? (I've considered donating an input and output channel of the CS42448 for constant closed loop calibration - which only helps if the offset error is essentially the same between channels). How is the linearity? The data sheets for these audio CODECs are pretty sparse with DC related details, and I'm just looking for some first hand experience.

Cheers, and thanks for bringing this cool product to market. (I'll be purchasing one myself soon).

[Aaron]

These are all good questions and I don't have many answers for you due to the nature of the product. Since the euroshield is meant for prototyping and is open source, we didn't go through the same level of calibration and evaluation we normally do. Euroshield customers have the tools to make their specific board work as well as they want. Designing for mass production involves comparing several units, characterizing the performance across the entire range, building in safety margins, and developing robust testing and calibration routines.

I hope this is of some help. I wish you the very best with your project.