View Single Post
  #14   Report Post  
Posted to rec.audio.high-end
Trevor Wilson[_3_] Trevor Wilson[_3_] is offline
external usenet poster
 
Posts: 139
Default Introducing a New Horse to the Stable

On 12/09/2019 11:32 pm, Peter Wieck wrote:
https://www.eeweb.com/tools/parallel-wire-inductance

This website will allow you to calculate inductance by giving it the gauge, type and nature of the wire you are using. You will find, pretty quickly, that the distances involved in the typical component of say 40 cm (16") square are such that the actual inductance realized will be infinitesimal in "real life".


**Some of the products I work on are significantly larger than 40cm. One
(badly designed) amplifier, from a well-known 'high end' manufacturer,
used power supply wires which were approximately 60cm long! Whilst it
made servicing easier (allowing the output device heat sinks to be
removed from the amplifier and still be fully operational, the
additional inductance damaged the ability of the output stage to deliver
fast transients. A bunch of capacitance close to the output devices made
things much better.


Again, getting to practical matters: there are common-sense applications and techniques for wiring electronics, much dependent on the nature and use intended. Part of my hobby is the restoration of vintage Zenith TransOceanic tube radios - and wire location/component location can be critical for high-band Short Wave sensitivity. It is common sense to shield power-supplies in pre-amplifiers, especially those that contain phono or NAB pre-amp sections. And, wire-dressing is always good practice. But worrying about straight-wire inductance at audio frequencies is much akin to worrying about skin-effect...'


**No. Skin effect is only a worry at RF and for power line companies,
where VERY long cables can lead to significant losses (hence the rising
popularity of DC transmission systems).



Now, Trevor clearly has a 'thing' about negative feedback, which is entirely his choice, and doubtless for sufficient and good reasons.


**Let me be very clear about several things:

* NFB is fine. In fact, NO audio amplifier can work without it.
* GLOBAL NFB is also fine. When properly applied.
* I have a personal preference for the amplifiers I use, which employ
lots of local NFB and no global NFB. Others may have a different opinion.
* As part of my education into the world of zero global NFB amplifiers,
I subjected myself to a couple of single (unfortunately) blind tests,
between two, otherwise identical, amplifiers. One employed zero GNFB and
one employed a modest amount of GNFB. I preferred the zero GNFB one.
Since that time, I subjected several (10) of my clients to the same test
(DBT). The zero GNFB models was preferred every time. Except one.
* Once mo I would posit that part of the reason why some listeners
prefer valve amplifiers, is due to the fact that global NFB levels are
very low, or non-existent.


But, again, in the real world, negative feedback, done properly, has
many more advantages than disadvantages.

**Again: No issue with NFB. In fact, no issue with GNFB, when done well.


Done badly - Ouch! Keep in mind that in its most practical
application, it dates back to 1927, and was patented by Bell Labs in
1937. So, it is a pretty well established technique, such that any
thoughtful designer not totally strangled by the bean-counters will get
it right very nearly all of the time.

**Sure. I learned about GNFB back when I was a teenager, having just
built my second amplifier. A mighty 10 Watt/ch, push pull amp using 6V6
output valves. It had no GNFB. It also had a gain control before the
phase splitter. After reading an old article in a local electronics
magazine about NFB, I decided to try it. With the in-loop gain control,
I found I could vary the amount of NFB right up to the point of
oscillation. Backed off a fraction, I found that GNFB improved the sound
quality significantly.


https://en.wikipedia.org/wiki/Negati...is_of_feedback

Go down to the "distortion" section. As brief as it is, it conveys some very good information.


**Indeed. All good and well, but that does prompt the question as to why
your preference is for an old valve amp, which employs far less GNFB
than a typical SS amp?


In point-of-fact, part of the TIP-Mod for your 120 involved increasing capacitance within the feedback loop to reduce bass roll-off.


**Huh?


--
Trevor Wilson
www.rageaudio.com.au

---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus