Yeah I don’t get it at all.
Benchmark pulls it off super well with their DACs also.
But with amplifiers , headphone amps etc, it needs to go away.
You almost need for a linear power supply to be feet away from and amp to not influence it somehow. Not to mention it’s not efficient at all
I guess it’s the Audiophile , bigger is better thing.
Another question regarding measurements
You measured the Zen Can and came up with these results. Of course using your AP
I measured my Zen Can but of course using a Focusrite Scarlet 2i4 1st Gen. I used a 250ohm load made by myself using a headphone chord. It actually doesn’t change if I don’t use a load for some reason. Am I doing something wrong or is the Focusrite noise floor just not that great?
Why do they differ so much , especially the noise floor?
Ideas?
Thanks!
Noise from a linear power supply is two things the first is a function of the transformer and in some cases the wiring acting as an antenna and broadcasting it into the box. The second is just a lack of filtering with noise making it to the supply line.
Generally you can tell if a noise is RF or actually making it onto the PSU rail by the frequency, RF noise will be 50 or 60Hz, noise from a linear PSU making it onto the power rail will be 120Hz because it’s after the rectifier.
For RF noise the only fix is transformer design (Toroid’s and R-Core transformers are better), shielding (you’ll often see separate compartments inside the case to isolate the transformer), or physical distance if the PSU is separate as stated.
RF noise in headphone amps with sensitive headphones was one of those things I grossly underestimated the impact of when I designed and built my first amp.
Most good Linear PSU’s can trivially hit -100+dB’s on the power supply rail and not all of that makes it to the signal path depending on the designs PSRR that could be attenuated another 60+dB’s (or a lot less depending).
As an aside this is why good tube amp PSU’s get expensive, the combination of High Voltage and the fact the standard SE tube stage has basically 0 PSRR, mean the PSU has to be VERY good at eliminating noise.
The nice thing about linear power supplies and why many designers prefer them is that 60Hz noise is either audible or it isn’t, and there is nothing messing with high frequencies.
Switching mode PSU’s have smaller transformers so broadcast less noise and the noise they put on the power rail (into the actual signal) are at much higher frequencies than we measure amplifiers. The downside of that is that amplifiers are not always stable in the presence of high frequency noise, so while the noise itself is outside the audible range it can impact the the way the amplifier works.
Designers can add filters to remove the switching mode noise as well. And some designers swear by Switching mode power supplies.
Switching mode power supplies tend to just get thrown into the “not audiophile” bucket, but it’s mostly based on poor implementations.
Like a lot of things in audio, it’s all tradeoff’s, there are good switching mode designs, and good linear designs.
Here is another measurements from a headphone amp. Maker shall not be named.
It sounds good. But has a linear transformer
Funny think about this one is that if I put some mu-metal around the transformer some of those PSU spikes lower a bit.
Go figure
How do you connect the Schiit Valhalla 2. Mojo and the rest.
Thanks and enjoy 
Any response to the post?
Pardon I know you are busy
Im just here about the observation I made about everyone making amps to power the susvara which I find interesting. It almost as that headphone artificially inflates the price. As far as I can tell most headphones don’t need as much as we think it does?
I’m leaning towards an MJ3 and Kenzie ovation, velo or woo audio wa22(2nd gen).
If you get the WA22 on a susvara let me know. 
I don’t like the susvara!! Or most hifiman headphones. The Arya and the nano would be my two I’ve demoed.
I just don’t know what the amps are doing that much differently to justify the price.
