Converting the stock switcher to a linear will take care of the RF emissions.
SMPS not encased in metal, and whose input/output lines not sent thru feedthru caps will generate RF at the fundamenatal and third harmonics, as it is a square wave signal rapidly turining on/off a power device. In some systems this RF may be a problem, although I would thing it would be obvious like increased noise floor noticed when system is on and idle but volume turned up.
I would think the most vulnerable systems would be ones with point to point wiring.
Now general convention when designing power supplies for mixed signal (analog/digital) components suggests that the regulators ought to be placed as close to the components being powered...as possible. Having an external power supply with even a superb linear regulator, and then connecting that to a device over 2,3,4,5 feet of umbillical wire does not improve matters (inductance, slows things).
Considering the design of the SB3:
The Squeezebox 3 has a cascade of internal linear regulators...
It takes the raw 5 volt line and does DC-DC conversion in a few places:
1) display - to generate higher dc voltage for tube
2) DAC chips and Analog outs - both derive the source DC voltage from the output OF THE DISPLAY REGULATOR.
The regulators feeding the analog section and the digital section are separate low dropout (LDO) types, and the input ripple is not affected by the external supply (assuming the external supply has decent ripple and decent regulation).
I think the main benefit from a linear power supply is for the RF issue. You can get a very capable one for under 50 bucks if you research carefully.
However, since it is a free market, you can also buy 800 dollar squeezebox supplies :) as long as your "golden ears" and high res system are up to par you might actually hear a difference -lol
I would personally not mod the squeezebox. Fundamentally the electrical design is what it is (it's a mass market consumer product, not a design with high end aspirations from the start)...
I would save for the transporter, a much improved design, better noise rejection, one that was meant for much higher performance...
BTW, I bought a welborne labs diy kit, (before I learned all this stuff from the slimdevices forums)... anyway I guess it explains why i could not detect any difference.
The notion that there are 800 dollar power supplies for a 299 dollar mass market device is why I love being an audiophile (or is that un-reasonable-phile) :)
just my 0.02
SMPS not encased in metal, and whose input/output lines not sent thru feedthru caps will generate RF at the fundamenatal and third harmonics, as it is a square wave signal rapidly turining on/off a power device. In some systems this RF may be a problem, although I would thing it would be obvious like increased noise floor noticed when system is on and idle but volume turned up.
I would think the most vulnerable systems would be ones with point to point wiring.
Now general convention when designing power supplies for mixed signal (analog/digital) components suggests that the regulators ought to be placed as close to the components being powered...as possible. Having an external power supply with even a superb linear regulator, and then connecting that to a device over 2,3,4,5 feet of umbillical wire does not improve matters (inductance, slows things).
Considering the design of the SB3:
The Squeezebox 3 has a cascade of internal linear regulators...
It takes the raw 5 volt line and does DC-DC conversion in a few places:
1) display - to generate higher dc voltage for tube
2) DAC chips and Analog outs - both derive the source DC voltage from the output OF THE DISPLAY REGULATOR.
The regulators feeding the analog section and the digital section are separate low dropout (LDO) types, and the input ripple is not affected by the external supply (assuming the external supply has decent ripple and decent regulation).
I think the main benefit from a linear power supply is for the RF issue. You can get a very capable one for under 50 bucks if you research carefully.
However, since it is a free market, you can also buy 800 dollar squeezebox supplies :) as long as your "golden ears" and high res system are up to par you might actually hear a difference -lol
I would personally not mod the squeezebox. Fundamentally the electrical design is what it is (it's a mass market consumer product, not a design with high end aspirations from the start)...
I would save for the transporter, a much improved design, better noise rejection, one that was meant for much higher performance...
BTW, I bought a welborne labs diy kit, (before I learned all this stuff from the slimdevices forums)... anyway I guess it explains why i could not detect any difference.
The notion that there are 800 dollar power supplies for a 299 dollar mass market device is why I love being an audiophile (or is that un-reasonable-phile) :)
just my 0.02