taipan254
100+ Head-Fier
That is extremely ironic given the stance of Rob Watts on linear power supplies:
Yup - I was one of the guys that asked questions at that lecture.
First off - Rob is an extremely nice guy and was very kind to offer up his time and perspectives to me before and after both lectures he gave. I was so appreciative of his time.
Rob has very strong opinions on linear power supplies. And Rob believes that a car battery is among, if not the very best, power source for most devices. I personally find his battery implementation in Mojo 2 sufficient.
My understanding of the counter-argument to removing the battery and using an external power supply (say a supercapacitor-based design or a high-quality LPS) is that it disables the battery management features in the FPGA, thereby reducing noise. God knows if that works and frankly I'm not at the point where I want to experiment with that. Maybe some day. That being said, Rob even admitted in this lecture that were he designing Dave today, he would not use the same power supply that he included in the Dave. My guess is that his next DAC will have a very hardcore, highly designed and spec-ed SMPS that is highly resilient in the face of, if not immune to, EMI and RFI.
My basic approach to maximize the sound quality I'm getting out of Mojo 2: run as much as I can off of batteries. Mojo 2's battery works for my usecase. My headamp runs off a battery + inverter. When I'm mobile, I'll run off USB to my phone (sufficient for RFI / EMI mitigataion as both are "floating" / powered off batteries). If I'm not upsampling and on a PC, I'll use the Mojo 2's optical input - nearly immune to source jitter and full galvanic isolation so no nasties coming through. When I upsample, I use a SRC-DX, but that doesn't have full galvanic isolation from the source.
My next project: building a battery-based dedicated music server after I get my home network set up (wish me luck: I'll be running ethernet cables in my walls and terminating them!).