georgehifi,
Your most recent post, posted last night at 10 p.m., consists entirely of a quote from the author of an article on
www.audiophilereview.com, Skip Taylor, that is dated 9/21/2016 and is titled Why We'll Soon Be Living In A Class D World. You and I have both previously linked to this article on this thread; you initially then me subsequently, mistakenly and redundantly.
I'm not sure why you felt the need to also post that specific portion of the article again, this time in quote form rather than a link, since your post lacks any pre or post quote comments.
atmasphere and I previously expressed concern with the author's inaccurate terminology and his knowledge of how class D actually functions.
Now, I want to freely admit upfront again that I do not claim to be technically knowledgeable of electrical engineering generally or how class D functions specifically.
Even a class D fan but technical layman such as myself, however, is able to find fault with some of Skip Taylor's article and question whether he has a firm grasp on how class D actually functions. I'll just critique one sentence you quoted from his article to explain my concerns from my layman's perspective:
"The greater switching speed of
Efficient Power Conversion's (EPC's) eGaN®
FETs allow amplifier designers to increase PWM switching frequencies, reduce dead-time, and drastically reduce feedback; in turn, producing a sound quality previously limited to large, complex, heavy
Class A amplifier systems."
My concerns with this quote are:
1. I think most are aware that MOSFETS are used as electronic switches in class D amps, required to turn on/off very quickly as the PWM (Pulse-Width-Modulation) circuit's introduced ultrasonic switching frequencies dictate, rather than as linear gain devices as in other traditional linear amps. Given this, I think everyone understands that the new
eGaN®
FETs, having the ability to turn on/off much faster than MOSFETS, will allow amplifier designers to increase PWM speed thus reducing dead-time, something called 'shoot through' and distortion.
But how do these new
eGaN®
FETs allow amplifier designers to increase PWM switching frequencies? He mentions this as if it's a direct result but without any further explanation.
He fails to mention whether current PWM switching frequencies are too low and whether they cause any proven audible sonic anomalies in current class D amps' outputs. He also gives no explanations or proven scientific facts as to why higher PWM switching frequencies are desirable or would be beneficial.
2. Skip Taylor states that the new
eGaN®
FETs will allow amp designers to drastically reduce feedback in class D amplification, something that may be beneficial in traditional linear amp design but I'm not certain he realizes that reducing feedback in class D amps has no relevance since high levels of feedback is a major factor/tool utilized in good class D amp design. As the inventor of Philips Ice and Hypex NCore class D power modules, Bruno Putzeys, often states: "there is no such thing as too much feedback in class D amp design".
While I agree with Skip Taylor's article title that we'll all soon be living in a class D world since many of us already are, I also got the impression from his article that he doesn't have a complete grasp of class D amplification.
Please take the above only as the reduced value opinions of a declared layman who isn't as concerned with how class D functions as with how good it already performs.
I've been listening to some computer audio music files
as I write this post; Two very good albums from Carmen Gomes that were recorded direct to 96kHz/24bit
high-resolution digital: the highly recommended Thousand Shades of Blue and Torn(SE). These downloaded files are one-to-one exact copies of the masters recorded live, unedited, unmixed and without the use of compression in front of a small audience at the legendary Dutch Studio Eleven with just left and right main mics and a few additional mics for the minimal instrumentation. Most songs are intimate- jazz-club like tracks with the beautifully soulful, smooth and sexy lead vocals of Carmen Gomes backed up by a small but very skilled group of clearly jazz/blues influenced musicians on guitar, drums/percussion and double bass.
I mention this because I believe the combination of very low distortion, low noise, extremely detailed, completely uncompressed and accurate source material that direct to hi-res digital recording provides paired with the very low distortion, very low noise floor, very neutral tonal quality, powerful dynamic ability, highly detailed and accuracy that good class D amplification allows produces the most life-like and natural reproduction of live music that I've ever been able to attain in my system by a wide margin.
Now georgehifi and randy-11, are you both claiming I'm currently attaining this extreme level of system fidelity despite the presence of sonic anomalies in the audible frequency range caused by my class D switching amps' switching frequencies being too low?
Have you guys even considered the possibility that your unproven switching frequency theory is unproven for a reason?
Do I, and all of the thousands (millions?) of other class D users and fans just have tin ears?
Why do only you two contributing thus far on this thread apparently buy into this unproven class D switching theory?
Do you both have canine-like hearing and inexplicably obey dog whistle commands?
I actually hope you two pooches are correct and class D switching frequencies in the 3-5 mHz range really do eliminate the sonic anomalies that it seems most users cannot currently hear.
I'm just wondering how we'll be able to perceive the absence of something as a sonic improvement when we didn't perceive the presence of that something in the first place.
But heck, let's look on the bright side, at least you two good doggies will appreciate and possibly even salivate at the suspect improvement if your theory is correct.
Sorry guys, but I think you two 'woofers' may be barking up the wrong tree.
Later,
Tim
