[time-nuts] "top cover effect" &c

[Eric Scace]

Warning: websites cited here can be seriously time-consuming rabbit holes!

   An anonymous engineer using the pseudonym NwAvGuy <http://nwavguy.blogspot.com/> got fed up with the subjective nature of discussions about DACs and amplifiers in the world of high fidelity audio. To clear away the smoke, in 2011 he started publishing independent measurements, along with some blog posts about what measurements and goals were detectable. As one might imagine, this was very controversial.

  Later that year he (or she) concluded it was possible to build a replicable DAC and a headphone amplifier <http://nwavguy.blogspot.com/2011/08/o2-summary.html> that could exceed the performance of so-called high-end units being sold for 10e3 - 10e5 dollars, yet cost the consumer an order of magnitude less. His blog posts describe the design process from requirements to finished result.

   Relevant to this thread was NwAvGuy’s long list of PC board routing and enclosure topics that significantly affect performance of otherwise identical parts and circuit:
ground routing, ground planes, floods & fills
EMI loops
inductive coupling
parasitics
failure to follow the PC layout developed by the component manufacturer
component placement, aka “rows and columns may not be your friend”
track placement: moving a single track can change increase distortion [by 26 dB] was one example given
all this at audio frequencies.

   So, yes, I cringe at the thought of developing such systems at frequencies and performance levels we talk about here on Time-Nuts. There are not enough lifetimes to learn how to do this well, I feel. :-(

   Hats off to all those people who’ve mastered and advanced this craft.

— Eric K3NA

p.s.: NwAvGuy’s resulting DAC/amp has been available in kit or assembled form since 2012 from companies such as JDS Labs <https://www.jdslabs.com/products/48/objective2-odac-combo-revb/> ($280), built exactly according to the design spelled out by NwAvGuy.
   NwAvGuy’s identity has been closely guarded by the few individuals who know it. (I am not one of those.) She stopped posting in 2012 and disappeared from the scene.
   More recently, Audio Science Review <https://www.audiosciencereview.com/forum/index.php> has started publishing objective measurements of commercial audio equipment, including DACs <https://www.audiosciencereview.com/forum/index.php?threads/review-and-measurements-of-wesiontek-khadas-tone-board-dac.4823/> and headphone amps <https://www.audiosciencereview.com/forum/index.php?threads/review-and-measurements-of-massdrop-thx-aaa-789-amp.5001/> that somewhat exceed the performance of the NwAvGuy’s solution but remain in similar price ranges. The site’s admin and principle author employs a pseudonym amirm <https://www.audiosciencereview.com/forum/index.php?members/amirm.2/>, claims to be in the Seattle area — and some wonder if this is the same person as NwAvGuy.

> On 2018 Dec 20, at 01:41 , ed breya <eb at telight.com> wrote:
> 
> Yup, I've seen plenty of "top cover effects," and other strange things. That sort of thing is evident even when you don't have to reach E-12 resolution levels - like a million times less. The instrument cabinet and hardware are indeed a part of the electronic system and circuits. So are what it's hooked up to, and its various characteristics, and so on.