Digital that sounds like analog
Posted 10th December 2012 at 04:18 AM by abraxalito
Updated 15th February 2013 at 06:11 AM by abraxalito
Updated 15th February 2013 at 06:11 AM by abraxalito
For those who missed Frank (fas42)'s link on a thread I started then here's where I'm continuing my minimal oversampling DAC developments for the time being : Digital that sounds like analog
Total Comments 108
Comments
-
I wonder whether the inductor is giving a roll off of treble distortion, "noise", being fed to the speaker: 2.5uH is about 1R at 20kHz. If the treble driver has audible IMD when fed ultrasonics every bit of attenuation will help ...
FrankPosted 2nd April 2013 at 09:37 PM by fas42
-
As a corollary, I imagine you've seen this - but if you haven't: [url]https://www.diyaudio.com/forums/solid-state/230492-audio-power-amplifier-design-book-douglas-self-wants-your-opinions-41.html#post3437919[/url]
FrankPosted 2nd April 2013 at 10:25 PM by fas42
-
I had been following that thread but wasn't completely up to date so thanks for the heads up. Yes, most prescient remarks from the beach bum. I note that Bob Cordell's estimate of inductor reactance differs from your own - 1uH is 0.13ohms so then 2.5uH should be about 0.33ohms no? Certainly my LCR meter only last night told me 17uH measured about 2ohms at 20kHz.Posted 3rd April 2013 at 01:17 AM by abraxalito
-
Okay, sloppy looking up of a reactance chart! I only worry about orders of magnitude in such matters, once something approaches a level where it is of the same order as another parameter which is crucial, like the impedance of a speaker then I take it very seriously.
FrankPosted 3rd April 2013 at 04:01 AM by fas42
-
Yes I agee - within an order of magnitude of the speaker would be 0.8ohms, I don't take it seriously because its under half this.Posted 3rd April 2013 at 04:38 AM by abraxalito
-
One of "worst" aspects of tweaking I've found over the years is nailing the optimum values, configurations for everything. You try a value of 1 unit for something ... a-ha!, it's improved the SQ, which means that units of 0.1, or 10 may make things worse, or better, or just barely make a difference. Since Murphy's Law dictates that every tweak will interact with every other tweak, the situation rapidly spins out of control, in a multi-dimensional nightmare -- which is the right combination of values, parts and setups to give the best result?
To retain my sanity, I just work towards a known quality of sound, that I'm certain is possible from experience; once achieved, I heave a sigh of relief, and leave everything for a goodly run of time before thinking of fiddling further ... ;)
FrankPosted 3rd April 2013 at 05:47 AM by fas42
Updated 3rd April 2013 at 05:51 AM by fas42 -
That's exactly what makes tweaking so much fun - its an N-dimensional optimization problem. Optimization is where its at - I have a good book on the topic :
Amazon.com: How to Solve It: Modern Heuristics (9783642061349): Zbigniew Michalewicz, David B. Fogel: BooksPosted 3rd April 2013 at 08:49 AM by abraxalito
-
You young whippersnappers might reckon it's fun, ;) -- I'm a bit weary of such activity, did too many years of fine needlework optimising and mental juggling in the computer game ... these days I would be happy to just push the big green button and have it all just happen in front of me!
That said, I couldn't tolerate the sound not being as good as I know it can be, so would always be motivated to give it another go. Especially irritating is relatively expensive and competent gear sounding pretty dreadful for want of some TLC ...
FrankPosted 3rd April 2013 at 09:13 AM by fas42
