[Puma Cat]

Originally Posted by W9TR

Here is an article that succinctly defines the problem with line operated switchers.

https://www.eetimes.com/power-tip-47...chers-part-1/#

It is from 2012, btw, so this phenomenon has been known for quite awhile. I believe it is the same mechanism John Swenson describes on his website.

In my case there is no path to create any such currents. My battery powered tablet floats wrt to everything and any common mode leakage currents it might impart to my Yggy would be based on an external signal coupling through the air to the tablet’s 14 pF free air capacitance. I have removed all nearby radiators so that’s not an issue.

There are other noise sources. For example the USB power coming from the tablet might be dirty. That noise is not common mode, and not referenced to any grounds in the Yggy, so does not have a path into downstream equipment.

There could be noise from the onboard WiFi transmitters. I don’t have a way to measure this directly, but I have run some comparisons with WiFi on and off while streaming content from the tablet’s SSD. I can hear no differences.

The notion that tablets or laptops are intrinsically inferior to dedicated end points is simply not true. The answer is ‘it depends’ and there are a multitude of interference sources in play when you connect any digital device to your lovely analog stereo system.

When you hardwire your mains powered endpoint to any DAC, you have a multitude of interference sources to mitigate. I think Uptone is doing a fantastic job building effective products that mitigate this type of noise.

This is the frontier of excellent digital playback and deserves more airtime.

Tom

Hey Tom,
Thanks for the link, I will read it with interest.

Sounds like you've got good setup by using battery power.

What the article referenced doesn't speak to high-source leakage current and its impact on clock phase noise. That's because this class of leakage current was only discovered in Oct, 2017 by John Swenson. In fact, he had to build specific equipment just to test for it; nothing on the market had high enough impedance.

As for tablets or laptops being intrinsically inferior, here's what I've learned from my researches and experiences: any high-bandwidth CPU or GPU will put a lot of high (gigahertz band) noise; this noise can be picked up by unshielded speaker cables acting as antennas, and be fed backwards into the power amp, only to be amplified again as noise. John Atkinson wrote an article about this some time back on Stereophile.

Also, the real culprit in any type of computer is the darned switch-mode power supply. They are a big source of the dreaded high-source leakage current. What I can also say is that one of the biggest improvements I've made for my digital front-end was simply getting the music server out of the main audio rack and putting it 40 feet away.

Last, there will be a document that should be published by tomorrow or thereabouts that will also provide some really useful information about "noise", jitter, and leakage current. Can't say more at this time, but as soon as it goes live, I'll post a link here.

Good discussion, Tom, thanks for sharing. Cheers, mate.