Sound Quality Vs. Measurements - Page 695 - diyAudio
Go Back   Home > Forums > Member Areas > The Lounge

The Lounge A place to talk about almost anything but politics and religion.

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 3rd August 2012, 06:23 PM   #6941
dvv is offline dvv  Serbia
diyAudio Member
 
Join Date: Sep 2004
Location: Belgrade, Serbia
Oh, woe is me ... !

THE criterion for life on a shelf is its size, it has to be small enough ...

Perhaps a DiscMan?
__________________
Per Aspera Ad Astra.
  Reply With Quote
Old 3rd August 2012, 08:04 PM   #6942
tvrgeek is offline tvrgeek  United States
diyAudio Member
 
Join Date: Dec 2009
Location: Md
Round object is a ships clock. Sharp thing my grandfathers army saber used to cut my mothers wedding cake. Keyboard is an Underwood 1894. My father found it in his uncles insurance company attic. Got him and myself through college. Types well but that "z" takes some real pounding! Box is a Meng 6P1 half way to becoming a RLD. Got it as a cheap way to learn about tubes. A cheap amp and copy of Jones goes a long way.

Man cave. My stuff.

Yes, environmental conditions are valid constraints. Explain how everyone seems to need a 14 inch box to hold a 5 inch disk. Ego, mine is bigger, bigger box must sound better etc. Open them up and they are empty. When I built the shelves I was running a Creek intigrated, Kenwood, tuner and Denon CD that were all less than 10 deep. Advantage to the DH-120 as it sits on a lower shelf under the amps with the remote sequencer, crossover, and lines filters. Sub is in the next bay. So I could buy a big empty box, cut it in half and weld it back together ( thinking about a matching Nak actually) or tweak the NAD, go external DAC, or modify the speakers a bit more. I need more projects like a hole in the head.
  Reply With Quote
Old 4th August 2012, 06:51 AM   #6943
dvv is offline dvv  Serbia
diyAudio Member
 
Join Date: Sep 2004
Location: Belgrade, Serbia
@tvrgeek

I was kidding, of course, but seriously, size was never even a consideration with me. True, I never considered any case that was outlandishly high, wide, deep or all of it to cause any concern. The number of stored devices is a concenr, by now they are overflowing, but that's entirely my fault since most of them are of very reasonable, even modest propotions.

Depending on where you live, space may well be a valid, even critical importance, agreed. Nevertheless, it still seems a bit odd to me that your choice of CD players seems to be dictated by size, probably because I never had that problem myself.

On the subject of typewriters, I also have an old one, I think my dad bought it in 1956 or so, which is the root of much of my well being. I used it throughout my studies for seminars, papers and eventually graduation paper, then to earn my first honest dollar, and eventually it generated enough income for me to buy my very first computer, an IBM PC, in 1984. I don't use it any more, but I am extremely attached to it, and it's not going anywhere from my room until I am dead and gone, period. Mine is a locally manufactured unit, based on a German Olympia licence, it's called Pearl, and to me, it really was a pearl.
__________________
Per Aspera Ad Astra.

Last edited by dvv; 4th August 2012 at 06:56 AM.
  Reply With Quote
Old 4th August 2012, 08:00 AM   #6944
diyAudio Member
 
Join Date: Mar 2007
Location: Oxfordshire
Quote:
Originally Posted by tvrgeek View Post
I know there are far better CD players. But NONE of them fit on a 12 inch shelf.
Quad CD67 was good if price is right .
  Reply With Quote
Old 4th August 2012, 11:06 AM   #6945
tvrgeek is offline tvrgeek  United States
diyAudio Member
 
Join Date: Dec 2009
Location: Md
I think I remember the Olympia. Before the Smithsonian took half the "stuff" off display in the History and Technology building to make it the American History and Technology building, they had a great display with dozens of typewriters ( including one just like mine). I think I remember one. I like machines.

My first business computer was a Kaypro 4 because it came with Word Star and The Word spelling checker. Life changed. I held out in the CPM world until I bought a 386 running Windows. As I sit in front of what would have been a multi-million dollar supercomputer, I am not sure I am any more productive. Well OK, Spice in text was not fun. My first computer was a wire-wrap 6502, Hex pad, and 7 segment LED's. I had 4K!

See, my trolling works, Nigel recommends a CD I would never have looked for. I bet they are rare on this side of the pond. I'll look around.

On the subject of this thread (originally), I was playing with constant current sources again. Limiting myself to two devices. I also have been thinking about what I am modeling and why. This goes with a comment I heard about "measures well, but sounds bad". Anyway, I have this big sheet with a dozen or so common designs. Basically a conventional LTP input stage, current mirror etc. So the only thing changing is the current source in the tail. It is easy to see the differences in the current, ripple, and how constant it is over frequency. But does this matter? If I look at the voltage across the resistor that simulates the load, and look at the distortion there, is that not really what I am concerned with? If a simple resistor tail does no different that a FET-cascode on the output, is all this a moot point? I should be looking at PSRR not ripple or impedance over frequency of the tail? Is the consistency with frequency not more important that the ripple? Aren't I trying to get a difference in the input pair, not some absolute total value?
  Reply With Quote
Old 4th August 2012, 12:34 PM   #6946
diyAudio Member
 
Join Date: Mar 2007
Location: Oxfordshire
I have played with CCS and no CCS with LTP . DVV tells me I go too far when being a minimalist sometimes . I like to use as few pieces as possible especially if an active device . However a CCS or current mirror must be the most benign active device use ? I have used a voltage doubler and filter to provide a cheap low current supply to LTP . That is the capacitor input Cockroft-Walton type as used in colour TV . I would guess a very high voltage supply to be very interesting and safe enough . The LTP will never swing to the rail in typical application ( op amp power amp ) .

The CD 67 was the first CD I heard with a vertical quality to the sound stage . In Oxford the local orchestra practiced in the Old Fire station next door . I became very aware of the vertical quality baroque music has . The CD67 had it . It used Crystal 20 bit Delta Sigma chip . CD 67 has I think 5 clean PSU lines for various sections . he also designed the Naim CDI etc . The Quad had more detail , the Naim more blood and thunder . I don't know the exact size of the Rega ? It looks small ( I didn't read all the text , it might say ) . Read it now , too wide .

http://www.rega.co.uk/html/Isis.htm

Last edited by nigel pearson; 4th August 2012 at 12:45 PM.
  Reply With Quote
Old 4th August 2012, 10:04 PM   #6947
tvrgeek is offline tvrgeek  United States
diyAudio Member
 
Join Date: Dec 2009
Location: Md
DANGER! Naive modeling results:
Modeled 9 simple ( no more than two pieces of silicone) constant current sources.

In a traditional LTP with current mirror. 2 mA, only the CCS changing. No optimizing for a topology. No thermal tracking, no testing PSRR, just within my limited comprehension.

Values are reference to diviation from the DC operating point, standard LTSpice AC analysis of current thriugh the CCS.

Resistor -97 dB
A single self biased JFET as -127 dB.
BJT and Red LED -147 dB,
BJT current feedback pair -159 dB,
FET/FET cascode at -184 dB. Humph, getting pretty silly.
BJT/FET cascode, -214 dB. Sorry, looking sideways at a part is 1000 times worse than this.

Conclusion, no wonder I was not seeing effects on the overall amplifier simulation. For an IPS, just get a nice fast JFET. I guess the CFP could be helpful in the VAS, which is where I see it sometimes used.

Last edited by tvrgeek; 4th August 2012 at 10:08 PM. Reason: fat fingers
  Reply With Quote
Old 4th August 2012, 10:26 PM   #6948
diyAudio Member
 
Wavebourn's Avatar
 
Join Date: Sep 2006
Location: Pleasant Hill, CA
Send a message via Skype™ to Wavebourn
Another conclusion: models differ.
__________________
The Devil is not so terrible as his math model is!
  Reply With Quote
Old 4th August 2012, 11:54 PM   #6949
tvrgeek is offline tvrgeek  United States
diyAudio Member
 
Join Date: Dec 2009
Location: Md
I need to add that to the one correct answer to anything: "It depends"
and the answer to everything: "42", and of course the ultimate question: "Why"

I have a pair of clone MX50 boards that I can bugger with different versions and do A/B tests. It will have to wait until I finish my SEAS project. I started putting them together just now. They should be breaking in tomorrow.

As I march through my CD collection, on the better CD's I am hearing things I never heard before. Actual intentional music things. Bad recordings are still really bad. I am not brave enough to try my old Telark set of the 9 symphonies. Those were really early.
  Reply With Quote
Old 5th August 2012, 07:24 AM   #6950
dvv is offline dvv  Serbia
diyAudio Member
 
Join Date: Sep 2004
Location: Belgrade, Serbia
Nige is right, I do criticise his far too simplistic approach on occasion, when I feel he has overdone it. Like using a simple resistor for a CCS in the input stage.

Some years ago, I played quite a bit with input stage configurations. It has become a standard of sorts to use a cascode; I know most of its benefits in theory, but life is practice, not theory. My experience is that assuming a well matched pair of tranistors is used, and that the gain stage is 20 dB or less, I cannot hear the difference between a cascode and a straight differential pair. I should note here that I take great pains to match the diff pair and in final form, always use 0.5% resistors all around it, so in general, it's as well matched as it's reasonable to expect. Typically, my input stage gain, a diff amp, has a gain of 15 to 18 dB, no more. Since each tranny is biased with 1 - 1.4 mA, the slew rate is usually rather good,

For current sources, typically a transistor, MPSA 56/06 being my all time favorites for the job, has two simple 1N4148 diodes on ts base, but I do throw in a parallel capacitor, just in case. Or two joint back to back transistors, with thermal compund joining them. Since full regaulation of the voltage gain stage is a given with me, this is not a problem for me.

On the other hand, I've "discovered" that it is MOST opportune to use a buffer or decoupling stage between the say input differential pair and the say cascode second and final voltage gain stage; these transistors I typically bias at 2 mA or so, because by that value, most of them have already reached at least 80% of their absolute peak performance. Favorite devices are 2N5551/2N5401 or BF 422/423, which have an uncommonly low output capacitance factor (max. 3 pF, typically < 2 pF) and are 250V devices to boot.

I find this makes the sound rather easy, unforced, free, so to speak. The downside is that in a fully complementary ciruit, this is another pair which needs to be closely matched, meaning more work for me. But that's the standard bugbear of such topologies anyway, you really need to have it closely matched from input to output if you want the best from it.

Lastly, I save 0% on the output stage and its power supply; in my mind, these two are joined into one, and only together can they do their very best. Skimp on either and your chances of missing the mark increase exponentially. My simple rule of the thumb is that the theoreticakl (nominal) power capability of the output stage must be such that its 8 Ohm rating falls below the 10% mark of its theoretical capabilities. So, if I want say a 100W/8 Ohm power amp, the nominal power handling of my output stage must be at least 1 kW. More would be better. Currently, the said 100W/8 Ohm stage has four pairs of 200W devices, theoretically 1,600 Watt power capability.

I realize that this is a very pessimistic outlook which does cost money, time and effort, but I am spared the economic outlook as I do this for myself, not commercially. On the other hand, if that stage should run into a really evil load, it will cope better than another stage which uses half that many transistors in its output, and rated at 150 Watts rather than 200 W, as used in the industry.

I realize that the industry has used 4 150W devices to squeeze 150W/8 Ohms from their amps, but that was at a time when power devices were very expensive and of course, price was a high importance factor with them. Today, power devices have never been better and were never cheaper, even in retail, let alone bulk purchases.
__________________
Per Aspera Ad Astra.
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Quality Control differences = variations in sound quality? KT Class D 3 4th June 2014 12:02 AM
Sound Card for Measurements Marik Solid State 2 2nd January 2012 08:59 PM
Sound Card Recommendations (For Audio Measurements) dchisholm Equipment & Tools 5 16th July 2011 09:40 AM
How to protect sound card during amp measurements? okapi Everything Else 13 2nd September 2008 03:06 PM
Sound cards - test and measurements jackinnj Everything Else 2 5th July 2003 03:02 PM


New To Site? Need Help?

All times are GMT. The time now is 02:15 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2