SB3 digital mod – significant improvement?

Can anyone comment on the difference between stock SB3 using linear regulated power supply and SB3 with digital modifications using linear PS?
Are there significant benefits to digital mod, as with going from switching (stock) PS to linear PS?
Or digital mod does not represent such an improvement as linear PS?
Let me quote from the Bolder Cables website for you. It is only one of two or three places you can get this mod done, and the only one I have first hand experience with.

Short answer is that probably the most important thing to do is clean-up the power supply. Besides Wayne's very reasonably priced mods, you can also get one from Wellborne Labs.

After that if you want to get the most out of the unit, because you have golden ears or a highly resolving system, the digital and or analog mods both add a lot and IMHO are well worth it.

The SB3 digital only mod has many of the main input power supply capacitors and the digital buffer's power supply capacitor replaced with higher value, low ESR versions. BlackGate NX caps are used throughout the circuits.

Our latest version of the SB mod now includes changes to the internal switching supply that provides voltage to the display. In the past, this supply has added quite a bit of noise to the positive voltage and ground paths. We have changed the circuit to reduce this noise. This results in a quieter "Blacker" background

The power supply cap to the digital buffer chip is replaced by a BlackGate NX. The signal path to the digital output jack is rerouted. The digital output jack is replaced with a 75 ohm, WBT NextGen, gold plated, copper RCA.

There are options to install a Bybee Slipstream Purifier in the digital signal path. You have your choice of Copper or Silver Slipstream Purifiers.
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 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

Here's my response to a similar post in AC......
"IME, one thing that stood out between Bolder's SB3 digital only mod versus stock SB3 digital out is details are presented without glare & because of this, more of it can be heard as well. As I start to break in the Bolder's SB3 unit, the perceived soundstage began to widen, has greater depth & the LF tightens considerably.
FYI, I was using a modded Rotel 1070 CDP (Superclock II, DBE digital output, Rubycon's caps, deadening damping sheets & diodes upgrades) before it got replaced by the SB3s feeding digits to a DPA DAC. Based on a scaling of 1 to 10 (10 for best), I would rate the stock SB3 to be 4, my modded Rotel CDP/transport 5 & Bolder's SB3 digit only mod to be 7. The SB3s are being powered by Bolder's Basic Rev.1 PS for this comparison.
The Ulitmate MKII PS are more of "macro" improvements ie: providing more life to the finer/"micro" improvements of the SB3 unit digtial mods.
Hope this helps & make sense......"