New Posts New Posts RSS Feed - 1970s Design Indulgence
  FAQ FAQ  Forum Search   Events   Register Register  Login Login

Welcome to the Graham Slee Audio Products Owners Club

 

Open to all owners plus those contemplating the purchase of a Graham Slee audio product wishing to use our loaner program: join here
Subscribe to our newsletter here (Rules on posting can be found here)


1970s Design Indulgence

 Post Reply Post Reply Page  <1 828384
Author
Graham Slee View Drop Down
Admin Group
Admin Group
Avatar
Telling it as it is

Joined: 11 Jan 2008
Location: South Yorkshire
Status: Offline
Points: 8652
Post Options Post Options   Thanks (0) Thanks(0)   Quote Graham Slee Quote  Post ReplyReply Direct Link To This Post Posted: 04 Sep 2019 at 9:47am
"Celeb" amp tester, James Kelly, called in yesterday, and whilst he was here he put the amp through its paces using version 4.4 of Audio Precision's APx software.

Prior to that he had a listen, and "On Every Street" just happened to be the file open (CD quality FLAC). Just like the Proprius which he'd heard a few years back, James liked the sound.

Why I call him celeb is because he was tester for some of the best known manufacturers including: Arcam; Tag McClarren; Quad, Wharfedale, Mission and Audiolab - and he knows how they sound too.

I didn't expect 40 watts since including the output resistor, and we didn't get 40 watts - we got 38 watts per channel (both channels driven) at 1% THD - which is the measure James uses to establish maximum power output.

At 0.1% THD it was 36 watts 20Hz - 20 kHz and reducing input signal to -2dB gave the best distortion: 0.046% THD except for the quite normal, and expected rise to 0.1% at 20kHz - this was at 24 WPC.

James commented that the last amplifier he'd tested designed by Doug Self had 10 times better THD (an extra nought after the decimal point) but it was a whole lot more complicated design than this.

He also commented that valve amps could be 10 times worse (100 times worse than Doug Self's) but still sound good.

Frequency response was 5Hz - 43 kHz (-3dB) and signal to noise 94dB.

He said I'd probably get better results replacing my clip-on scope leads with bespoke test cables, and we were in agreement that the board with all its track cuts and bodges to try all the modifications would not be helping.

Noticing the filename, he had to laugh (DAKW55). James asked if it began as a 55 watt amp to which I replied yes, and that it had reached it, measured well, but sounded crap.

However, for a tuned up vintage transistor design (only 6 transistors per channel), 38 WPC compared with the Leak Stereo 30 (15 WPC), the Mitsubishi Teleton SAQ-206B (16 WPC), and even the early (but more complicated design) NAITs (13 WPC), must be considered an achievement.
Not simple enough for Google-Bot to understand...
Back to Top
Dave Friday View Drop Down
Senior Member
Senior Member


Joined: 07 Apr 2011
Location: Spain
Status: Offline
Points: 166
Post Options Post Options   Thanks (0) Thanks(0)   Quote Dave Friday Quote  Post ReplyReply Direct Link To This Post Posted: 04 Sep 2019 at 12:16pm
Well done,when will a kit be available?
lp12,oc9mk3,ca610p,krimson40watt pa,kef105.4
Back to Top
BackinBlack View Drop Down
Senior Member
Senior Member
Avatar

Joined: 05 Feb 2012
Location: Hinton, N'hants
Status: Offline
Points: 1359
Post Options Post Options   Thanks (0) Thanks(0)   Quote BackinBlack Quote  Post ReplyReply Direct Link To This Post Posted: 04 Sep 2019 at 12:23pm
A hearty slap on the back for your achievement Graham, well done.
I imagine that the same problems with conformance labelling will apply as with the 5v PSU, even if it's a kit.

Ian
Just listen, if it sounds good to you, enjoy it.
Back to Top
Graham Slee View Drop Down
Admin Group
Admin Group
Avatar
Telling it as it is

Joined: 11 Jan 2008
Location: South Yorkshire
Status: Offline
Points: 8652
Post Options Post Options   Thanks (0) Thanks(0)   Quote Graham Slee Quote  Post ReplyReply Direct Link To This Post Posted: 04 Sep 2019 at 1:11pm
I will be giving it some more "Sir Alec Issigonis" treatment before I commit.

The LS3/5A's are back on, and I will keep it playing whenever I can, and it will remain on for another week.

After that, and providing it is still sounding as good, it will be stripped down and fully documented, after which I will set about tidying it all up.

The case maker will get involved and there will have to be another bare-metal prototype, which would have to be built up using a new and updated set of boards. The transformer manufacturer will then be asked to make a sample batch, but this time with 230V and 240V primary taps. If it performs as good running on the 230V tap on 240V mains, the 240V tap will be ditched.

I can then commit to having case parts painted and heatsinks anodised, and then there's the printing. I don't think the UK will be leaving the EU anytime soon with Parliamentary remainers being the victors and denying any public say, and so it will be printed up as usual with the EU-only marks.
Not simple enough for Google-Bot to understand...
Back to Top
peterb View Drop Down
Senior Member
Senior Member
Avatar

Joined: 02 Feb 2017
Location: Cornwall
Status: Offline
Points: 137
Post Options Post Options   Thanks (0) Thanks(0)   Quote peterb Quote  Post ReplyReply Direct Link To This Post Posted: 04 Sep 2019 at 1:44pm
I hesitate to ask you Graham but with the many iterative changes you have made searching for the right sound, do you think that the mains transformer change you made did all that you think it did? 
One of my personal hates (one of many according to SWMBO), is background noise from the equipment that is providing the sound I really want to hear. In my very limited experience, a well made torroid has always been quieter, self noise and hum, than other designs, my knowledge limit is how they effect the music.
Would you consider a 'backwards' test, if only to prove a point one way or the other, or are you convinced?
Peter
---------------------------------------
Dual 505-1, Cyrus CD T, DIY 80W MosFet amp and PreAmp, 2xKEF 103.2
Back to Top
Graham Slee View Drop Down
Admin Group
Admin Group
Avatar
Telling it as it is

Joined: 11 Jan 2008
Location: South Yorkshire
Status: Offline
Points: 8652
Post Options Post Options   Thanks (0) Thanks(0)   Quote Graham Slee Quote  Post ReplyReply Direct Link To This Post Posted: 04 Sep 2019 at 4:06pm
Well Peter, I would welcome it if a toroidal transformer worked. I have Bruce's recommendation, the AnTek AS 3450, and although challenging to make it fit, it has the low voltage windings needed to make the amp into the integrated version I would really like to make. As more VA is available it could push the amp's power above 40 watts.

It is also low saturation and very quiet, but inter-winding capacitance ensures HF mains noise enters the amplifier's HT rails. No matter how much PSU filtering I've tried (and ran out of options) that problem will not go away. The same happened by putting a flux band around a split bobbin laminated transformer: with it inter-winding capacitance increased and led to similar undesirable results.

Now, to answer one of my recent posts (as promised), I strongly believe that input stage open-loop linearity has gone a long way to improving the sound, and is testament to the technique used in the Stereo 30 and SAQ-206B, as well as a number of others.

What stopped me trying this sooner was the DC biasing technique, which was also used in the Dynaco ST120 and Proprius, which seems to prevent the switch-on thump. Until I imagined and dared to use the DC level shifter (C5 || R14 page 81) the only way would have been to completely redesign the DC bias circuitry.

The Dynaco/Proprius/Dinsdale DC-coupling technique limits T1's collector-emitter voltage. The complimentary "singleton" as used in latter Leak amps; Armstrong 600 series; and many others, limits T1's collector resistor voltage. The DC level shifter allows more voltage on both, making them more like a symmetrically biased transistor stage.

This then, might be sufficient to compensate for failings elsewhere, and we won't know until we try. I shall do as you suggest after a few more days of "Issignosisicis", but have been told by my SWMBO not to "bugger about with it anymore", so we'll have to be secretive! Wink

Not simple enough for Google-Bot to understand...
Back to Top
Graham Slee View Drop Down
Admin Group
Admin Group
Avatar
Telling it as it is

Joined: 11 Jan 2008
Location: South Yorkshire
Status: Offline
Points: 8652
Post Options Post Options   Thanks (0) Thanks(0)   Quote Graham Slee Quote  Post ReplyReply Direct Link To This Post Posted: 13 Sep 2019 at 10:38am
Apologies for not trying a toroidal transformer yet

I am not sure the amplifier is "stabilising" in exactly the "right direction".

It appears that the class-A voltage amplifier has more influence on the sound than the output stage. Since making T1 more symmetrical the sound is much clearer and therefore more revealing.

If the stability capacitor (C7: 220 pF) is removed from the simulation, a high frequency spike is seen on the plot, and so this capacitor sort of "catches" it, and I'm not sure it is capable of doing so in the real world of man-made "stuff".

The signal does not contain any such frequency, but stimulus always exists (unless it's switched off). I once studied for the RSGB exam, and although I never sat it, I know that much in radio frequencies is temperamental. And therefore, an amplifier relying on a single electro-deposited film, which will otherwise flip in the mid-MHz, isn't ideal.

Most amps are like this anyway.

If C7 could be assisted with the spike it might help, so I simply gave T2 the miller compensation too. Starting at 100 pF and after the usual elapsed time, I think it was probably overkill. I think a value just able to kill the spike would be sufficient, something like 33 pF.

This however, has the effect of a reduction in phase margin, taking it down to 80 degrees, but what the heck; I'll give it a try.

Although miller compensation is local high frequency NFB which is good, it also reduces slew-rate, and with 100 pF it calculates as 14V/uS, so with 33 pF it should be around 40V/uS. C7 is still responsible for the dominant pole.
Not simple enough for Google-Bot to understand...
Back to Top
 Post Reply Post Reply Page  <1 828384
  Share Topic   

Forum Jump Forum Permissions View Drop Down

Forum Software by Web Wiz Forums® version 12.01
Copyright ©2001-2018 Web Wiz Ltd.

This page was generated in 0.090 seconds.