Sound Quality Vs. Measurements - Page 874 - 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 26th March 2013, 05:37 PM   #8731
dvv is offline dvv  Serbia
diyAudio Member
 
Join Date: Sep 2004
Location: Belgrade, Serbia
Am I to understand from the above that it's really just art for art's sake? Just an engineering exercise?

Fine, I can understand that we all sometimes just go silly for the fun of it, or because we're simply curious. I cannot deny that I also do this when I have some free time on my hands, which is very rare unfortunately.

But, if so, why build a whole therory around it?

Exactly what do we gain by making an amp all NPN? And why should we do that, in this day and age, when we have rather good matching PNPs?
__________________
Per Aspera Ad Astra.
  Reply With Quote
Old 26th March 2013, 08:08 PM   #8732
diyAudio Member
 
PlasticIsGood's Avatar
 
Join Date: Mar 2012
Location: Bradford
Quote:
Originally Posted by nigel pearson View Post
...
I noticed recently that the KEF LS 50 speaker has below 1 % distortion . As long as an amp has minimal crossover distortion and reasonable damping factor the 1947 standard of hi fi is still more than good enough ( 0.1 % ) . There " should " be no human on the planet able to say differently ?
The change from valves to transistors was accompanied by much discussion about the quality of distortion. Conventional wisdom appears to be that the distortion of early transistor amps was more intrusive, or at least less acceptable. Once quality enters the argument, it can go on forever.

I can see a good reason to build old designs if they are representative of the best practice of a particular period. Exploring a path that history escaped from at the earliest opportunity seems a bit far out even for a hobby. The possibilities are endless...

It's an interesting question though. Where would we be without PNP?
  Reply With Quote
Old 26th March 2013, 08:23 PM   #8733
diyAudio Member
 
Join Date: Mar 2007
Location: Oxfordshire
The point is we lost a lot over the years . Bootstrapping almost abandoned . Here is a conventional feedback amplifier . Except it is far from conventional . The transformer I find has undreamt of performance . The idea of using the transformer as if it was a gain stage is very clever . The output stage as in bootstrapping provides the power to drive the circuit . The big downfall of this circuit is the availability of the output devices I beleive . I have some very disgusting fet's and a mains transformer . See how far I get .

All NPN , it is always worth considering . The op amp I gave ( MC 1530 ) is I feel worth trying as a power amp . It looks to me to be a recipe for good performance , especially if the sliding bias can be made to work at amps rather than mA . Somewhere he claims GBP of 10 MHz , 33 V / uS when carefully compensated ( better than LM 301 if so ) . A dinosaur with attitude ? Even the output voltage is OK . Most modern op amps don't go high enough to claim so much more . With some 2N4403 as inputs it probably would be OK as a phono stage ? As far as I know NPN in chip form is far easier . Thus even today it is the device of choice . PNP usually gives low noise .

I have some work to finish so the above is one day perhaps . I hope to knock a TDA 2040 into class A for fun soon . If in the unlikely event it survives I will post the distortion if worth knowing . I hope to run 1.25 A .
  Reply With Quote
Old 26th March 2013, 08:49 PM   #8734
diyAudio Member
 
Join Date: Mar 2007
Location: Oxfordshire
BTW . The folk law about older amps is mostly untrue . 99% untrue if designs which passed the scrutiny of the technical magazines . The Leak stereo 30 for example . Harold Leak made fabulous valve amps . In fact his was the first mass production hi fi amp in 1947 ( I gave one away , yikes ) . Crossover distortion was troublesome as was too little current in the input to the VAS in early transistor amps . However Leak was dogged in making sure his class AB amps were reasonably identical to the previous designs . This was done with live verse hi fi demonstrations and included the previous amps . I became addicted to hi fi through Leak transistor amps . I don't think hearing the very best amps in the world was as big a leap as hearing the Leak Delta 30 . Even the Germanium versions pass muster . Most of the amps ever sold were rubbish , the valve ones especailly ( hum , hiss , distortion , explosions ) .

Google Image Result for http://obrazki.elektroda.net/65_1283442944.gif

These are the sort of amplifiers people think of when talking of the bad old days . My friend John singles them out as his pet hate .

Click the image to open in full size.
  Reply With Quote
Old 26th March 2013, 09:35 PM   #8735
bcarso is offline bcarso  United States
diyAudio Member
 
Join Date: May 2005
Location: Canoga Park, California
Quote:
Originally Posted by PlasticIsGood View Post

It's an interesting question though. Where would we be without PNP?
We'd pursue designs that favor balanced topologies, probably. Even without using transformers an all-NPN restriction can still permit a lot of variety.
  Reply With Quote
Old 27th March 2013, 10:33 AM   #8736
diyAudio Member
 
Join Date: Mar 2007
Location: Oxfordshire
The one strange thing I see is this .What would be standard practice for valve designers is the plague to transistor designers and vice versa . Valve amps do not respond well to vast amounts of feedback , they get ugly for want of a better way to say it . It also makes them more expensive for no good reason . Transistor amps seem to tolerate almost any amount of feedback as long as the rules are followed . The temptation is to go where it goes and add parts . I would say it is much harder to go the other way . Have a goal in mind and see how economically it can be done . I remember seeing old Norton single cylinder 500 racing against the Yamaha TZ 350 two strokes . The Norton was not left for dead . I had a friend who worked on them . He said it is a seriously fast motorcycle . They made replica motors so as to allow the preserved Norton's to race . Simplicity with high performance is a joy to behold .

Although I never had this conversation I suspect Bob Stuart was saying this . When a feedback loop is closed we have no real notion of what harm is being done to the signal inside the amp , in 1980 especailly . Bob's suggestion was although undetectable by usual measurements the copying of the signal must be not the best thing that ever happened to it .

I said something previously that was not uniquely true . I said the Pass amplifier used the output stage to supply current for the feedback loop . Well that is always true . This time I feel the significance was rammed home to me .

Here is a design at the point when it worked . A voltage doubler PSU was added later to get better output ( driver ) . It was disliked and used at below rail voltage as a clean supply ( rail - 2 V ) . Note how ridiculously small the bootstrap is ( 2 u 2 ) . The point is , it was enough . 10 uF was then used as a small increase in distortion could be see at 1 kHz ( not below or above ) . The point is the loop gain of the amp was enough up to 1 kHz . None of what you see was for other than for me to read , it's just my notes . It is a mixture of designs I like , Hitachi , Leak ,JLH . The output into 3 R was great ( goes into AB ) . The output capacitor ( 4 x 1000 uF non polar ) was so as to be certain it was OK for DC offset , I never had a problem . The MOSFET's were marginally inferior to a complimentary feedback pair . The difference was by deduction cured . The 50 kHz graph was the design goal . All graphs were identical down to far below one watt . Note the biasing of capacitors . < 400 mV would be ideal . The transformer in the Pass design usefully avoids that . I had no J Fet to try .

Note there is no significant distrotion of any kind . OPA 604 with complimentary feedback pairs was marginally inferior in measurements and sound . What I suspect is buffering an op amp is not 100 % what one would hope for . It was close run . One criticism , 0.7 A is above the optimum for this FET ( 0.1 A ) . It is OK . An A/AB bias resistor ( < 470 R ) would have been interesting for listening tests . I suspect AB would be more punchy ?

Click the image to open in full size.
  Reply With Quote
Old 27th March 2013, 08:16 PM   #8737
dvv is offline dvv  Serbia
diyAudio Member
 
Join Date: Sep 2004
Location: Belgrade, Serbia
In my view, each and every capacitor is a can of worms. The best one is the one which isn't there, especially at the output of an amp.

For the wold of me, I cannot see why are you even thinking about output caoacitors, Nige, and I don't know anyone else who's doing the same. I admit I love some vintage products, but even I look only at direct coupled DC designs. Which, BTW, do not necessarily demand DC Servo ciruits for pure DC operation - see Marantz 170 DC, 1152 DC, 1180 DC, etc schematics, to mention just one manufacturer.

Unless one is absolutely crazy for capacitor induced phase shifts.
__________________
Per Aspera Ad Astra.
  Reply With Quote
Old 28th March 2013, 11:27 AM   #8738
diyAudio Member
 
Join Date: Mar 2007
Location: Oxfordshire
I think I said before , the capacitor is simply so as not to loose the experiment during tests . Interestingly the least obvious sound impediment was the output capacitor . I couldn't measure any real distortion from it . Most of my listening was done without it . If you look it is a non polar cap . If wanting a cap I would have used the old style PSU .

The amp started as a speculation about repairing a stack of Leak Germanium amps . I decided to build something new so as to know as much as I could . The FET's almost mimic the Germanium's in many ways apart from speed . If I remember correctly the Leak Stereo 20 was TO3's on a bracket . So just a simple rewire and resistor bias ( get rid of Darlington half ) .

I learned a lot from this .

1 / I don't like single VAS for it's lack of symmetry . No amount of current fed into the VAS cures it . 10 kHz is OK . Maybe this is why we see the need for ridiculous slewing rates ? A one legged cyclist always has difficulty .
2 / Single input stage is not obviously more stable ( folk law ) , those of the past that were had high capacitance transistors . If a BD139 is substituted then perhaps no VAS cap needed ( it worked with considerably more distortion above 10 kHz ) .
3 / Playing with the VAS emitter resistor can reduce distortion !!! My best guess is the impedance matching to the input is better . If so it will show up as IM distortion . A high gain transistor was not the complete cure ( about trans-conductance ) . 16 R seems a good choice here . This is not simple local feedback as many think . It is usually a trans-impedance stage ( I in V out ) . If you every have tried to use a pre amp with a high output impedance you will know it is not the best idea you ever had . Especially if it almost matches the next stage ( higher can work ) . So why do people assume it is OK here ? I know the arguments and have to agree that many amps designed this way are fine . I just suspect what I have done here is worth a thought . Note 0R was the design goal . 47 R was to hand . It was the first time the 50 kHz distortion was better than perfect ( < 0.1 % ) .
4/ If class AB is mostly A then the application of feedback is easier . This amp will pump current as it then in class AB when loud . 8 mA will drive at least 3 sets of outputs on real music .
5 / I like long tail pair input as it makes life easy . Apart from that a single looks better . Such distortion as it as is harmless enough and invisible in this test .
6 / A complimentary feedback pair input for the very pedantic would be the optimum input . Very low distortion and noise . Slightly better stability . As I like double VAS I think LTP is a no brainer .

I am told that in 1938 Edward Turner by using a bit of thought managed to make a twin cylinder motorcycle for the price of a single , it saved the company . Although plenty of things were wrong with it ( balance factors , no centre support ) , it came to be the one to clone . I feel my amp I show although fundamentally excellent ( that is praise of it and not of myself ) is stupidly simple . A caprice . What I show previously in the MC1530 is better than 90% of what I see . How it accepts feedback especially . Well done J T .
  Reply With Quote
Old 28th March 2013, 01:11 PM   #8739
godfrey is offline godfrey  South Africa
diyAudio Member
 
Join Date: Nov 2009
Location: Cape Town
Quote:
Originally Posted by PlasticIsGood View Post
Where would we be without PNP?
Hopefully designing nice simple class A amps even I can understand.
Attached Images
File Type: gif simple a.GIF (4.9 KB, 64 views)
  Reply With Quote
Old 28th March 2013, 02:44 PM   #8740
diyAudio Member
 
Join Date: Mar 2007
Location: Oxfordshire
That looks ideal to make a big MC1530 . Any specs ?
  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 11:22 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