Reply To: BEOTOOTH 5500 : A Datalink Bluetooth Receiver

[B3OHACK3R]

pilatomic wrote:

That’s interesting ! Any insight on the mechanism that lowers the THD here ? Too much drop through the resistors ?

Only noticed when measuring a PCM5122 on the audio analyzer (pretty much the same chip on the analog output). Never made much assumptions why this is actually the case. In-field you will either see the 470R from the datasheet / EVK being used or a value around 100 … 75R. The ones with a lower value were likely be tuned with a proper analyzer, the others are probably just copy-paste. Just make sure you have proper ESD protection. Also sometimes they don’t like a ferrite bead (FB3?) in front of the analog power input. Better to have the whole system powered by a buck converter and only have a small LDO for the analog part of the DAC.
With exception of the PCM5242 (convenient if you need symmetrical output) I stopped using those TI DACs a while ago because they also seem to have a higher variability in quality than usual. E.g. ESS has a part that is pretty similar to the PCM5102 and has no such issues.
Not that all that would make any difference when connected to vintage Beo gear… 😉

pilatomic wrote:

and the recommended output cap is a 10µF tantalum

You mean 100 uF, right? Which is likely also the reason why they are recommending tantalum caps. Ceramics in that range are quite a bit more expensive (and probably weren’t even available back when that LDO was designed – same goes for low-ESR electrolytics).
Personally I don’t like tantalum much. Only use it if there is no way around. Although they still have the highest density with lowest ESR they also really dislike repetitive and excessive in-rush current. They tend to go up in flames if they are unhappy with their environment. Then there is also the fact that it’s a conflict resource…
Alu-Poly is a great alternative if normal electrolytics or ceramics won’t cut it.