Sound Quality Vs. Measurements

[cliffforrest]

Originally Posted by dvv

From my school yard, I could just barely see Watchet. And, across the Bristol Channel, Swansea.

But if you want something really impressive, do go see Ilfracombe. It sits on top of a gigantic boulder, very surreal. When I was there, there were dark storm clouds gathering, which made it even more surreal.

For 10 years in the 80's our family went every year to the summer school at Taunton School. I led a walking group across every which-way of the Quantocks, from the Black Hills to the Blue Ball.

I have just googled the summer school - all gone

[Wavebourn]

Originally Posted by dvv

There is no other singular semiconductor manufacturer in the world who has anything like Motorola's 40+ years of experience in power transistors.

I know one. It is called Motoshiba.

[ThorstenL]

Hi,

Originally Posted by Wavebourn

I know one. It is called Motoshiba.

Are you sure they are not called Toshirola?

Ciao T

[dvv]

Originally Posted by cliffforrest

For 10 years in the 80's our family went every year to the summer school at Taunton School. I led a walking group across every which-way of the Quantocks, from the Black Hills to the Blue Ball.

I have just googled the summer school - all gone

So is my old school - from 1692, but it's gone now.

All that's left is a forum of ex-students.

[Wavebourn]

Originally Posted by ThorstenL

Hi,



Are you sure they are not called Toshirola?

No, I am not sure.

[bcarso]

Ah yes I remember that cross-licensing with Moto and Tosh. I actually did advise on a discrete power amp and initially wanted the slightly larger devices (3281 etc.) but the Chinese manufacturer pleaded for the next size down. As I was attempting to make incremental changes to a not-entirely-excremental design, so as to not wholly alienate said manufacturer, I didn't do a complete tearup. The design already had complementary bipolar pairs at the input. I think I enhanced the second stages, used "better" output devices, but otherwise did not implement a whole host of ideas I might have considered.

The overall product, a home theater system, was a critical success and a complete flop commercially, owing much to the stupidly poor AV preamp/DVD player sourced from a famously terrible vendor (can you say 41% defects in initial shipments?), but also due to the economic collapse. Either people had money for a 99 dollar Whizzio, or >50k for a full-fledged room. But at the ~6k price "point" as they say, there were few customers. The system included, besides the bad apple piece, a 5.1 channel amp system in the subwoofer, a motorized screen with LCR speakers, a top-line LCD projector for ceiling mounting flanked with rear channel speakers, a universal remote that communicated through a receiver/controller in the screen. The sub electronics cried out for a good DSP-based programmable EQ, but at that point it represented too much of a sea change for the managers, and instead there were five switch-selectable complex EQs for different positions of the screen relative to the wall and ceiling, all done in the analog domain with a sea of modified state-variable biquads. The boards are something to behold.

Oh well, it paid the bills for a while.

As far as high end semantics, I guess my definition is do the best you know how to do. To say simply to "do the best" might presume one knows all and sees all, and I make no such claims. Indeed I am undergoing a huge shift and cascading series of revelations that are quite bracing and challenging. I read part two of the Rees pieces last night and thanked myself for my background in philosophy I can understand SY's impatience with it. But there is something going on, and I don't want to risk violating the Geneva conventions by airing some of the woolier things that could be considered.

Oh these smilies. Who thinks of this stuff?

The first product I'm considering is "yet another" phono preamp, which will employ among other things synthetic "cooled" terminations as at least an option for the cartridge damping resistor, done in a somewhat new way. There's little advantage with MC cartridges with very low inductance, but substantial noise reduction for highish inductance ones. I'd also like to provide variable input capacitance loading, including negative capacitance, very high overload margins with however the option of a very clean symmetrical clipping function, to prevent downstream overload. Fully balanced might be an option, despite the ~fourfold increase in silicon area (or other device proliferation!) for the same noise performance, and despite the very few tonearms etc. that are wired for balanced operation.

Just some thoughts. Actually I've been working on aspects of this for some years now. How the time flies.

[a.wayne]

Sounds good Brad, just one question, what's a phono pre-amp ..... ? synthetic "cooled" terminations...

[jcx]

Self describes it in his Small Signal Audio Design book - his example phono preamp app only gave 1.5 dB noise improvement

basically you use a quiet input op amp and a higher value resistor, with a fixed negative feedback loop gain reducing the effective value of the big R seen by the cart

[bcarso]

Originally Posted by a.wayne

Sounds good Brad, just one question, what's a phono pre-amp ..... ? synthetic "cooled" terminations...

Synthetic terminations (or more generally synthetic impedances), which "look" like an impedance with a resistive component but have less than a physical resistor's thermal noise over a limited (but usefully large) bandwidth, go way back.

Someone pointed me to a possibly-earliest reference in van der Ziel's classic book from 1954, simply titled Noise. The reference was to a considerably earlier article by Percival in Wireless Engineering, Vol. 16, 237, 1939, which was followed up with a scholarly paper with a whole bunch of other stuff in Physica by Strutt and van der Ziel (Vol. 9, 513, 1942). Oliver mentions impedances with less-than-thermal noise over a restricted bandwidth in his IRE tutorial paper Thermal and Quantum Noise. I saw it first in the Arbel book I like a lot, Analog Signal Processing, and his discussion is of its application to nuclear science instrumentation by the great instrumentation guy at Brookhaven, Veljko Radeka. Pease mentioned it in a column some years ago, and recently Self included the notion along with his particular realization for cartridge loading in his Small Signal Design (along with a remark that he doesn't like the "cooled" part --- but this is the "standard" terminology among the Nuc Sci brethren).

Brad

[bcarso]

Originally Posted by jcx

Self describes it in his Small Signal Audio Design book - his example phono preamp app only gave 1.5 dB noise improvement

basically you use a quiet input op amp and a higher value resistor, with a fixed negative feedback loop gain reducing the effective value of the big R seen by the cart

Thanks jcx. When applied to some high L low R cartridges that like to see 47k, done right it can yield a substantial improvement. Of course it will be argued that surface noise spoils all of this as soon as the needle is in the groove. But I still think, as does Self, that's it's worth doing. If one is going "all out". Self's particular example in the book is also woefully suboptimal. Good book though, although like much of his writing he's light on historical references. But filling those in is a whole lot of work.

Brad