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 7576777879>
Author
John1479 View Drop Down
New Member
New Member


Joined: 05 Jun 2019
Status: Offline
Points: 33
Post Options Post Options   Thanks (0) Thanks(0)   Quote John1479 Quote  Post ReplyReply Direct Link To This Post Posted: 11 Aug 2019 at 9:11pm
It's been a unique experience following your R&D over the past year. Although much of the electronic technicalities are well beyond my understanding, I have enjoyed reading and catching up with each update. Certainly, ‎it's been a Graham Slee Master class, I am doubtful whether any other engineers would be as open and honest (capable!) in their endeavours for fear of revealing potential commercial advantage, or (and this is very much down to perspective) a failure. I hold no doubt that your discoveries and understanding will yield some benefits for future projects even if this amplifier design goes no further. 

Thanks ‎Graham ‎for taking the time (how many hours?.....! ) to document this  project. 
Looking forward to discovering what this open R&D leads to..... ‎

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: 8610
Post Options Post Options   Thanks (0) Thanks(0)   Quote Graham Slee Quote  Post ReplyReply Direct Link To This Post Posted: 11 Aug 2019 at 10:26pm
Yes, it's a bit disappointing I guess. It raises an awful lot of questions I suppose - how these things worked in the past and all that - but don't work very well now. But we can't change what has occurred, especially other technology which might be causing interference, and I'm not of a mood to research if it's so.

As those better versed will know, today's "performance" amps are a lot different. It's easy enough to copy a modern published design (after all, I paid for the books) but this topic is about me taking a trip down memory lane.

I think for the next part I'll take a look at the Leak Delta 70 (http://www.angelfire.com/sd/paulkemble/sound7h.html - third circuit down) seeing I have a board that should accommodate it. The devices won't be the same and neither will be the frequency compensation, but I can make it as near as damn it.
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: 8610
Post Options Post Options   Thanks (0) Thanks(0)   Quote Graham Slee Quote  Post ReplyReply Direct Link To This Post Posted: 12 Aug 2019 at 1:36am
I've not liked the squeaky treble which has resulted every single time each modification has settled-in, and we're talking about upwards of 50, if not nearly 100 ideas tried. I'd say that's no stone unturned with respect to the NPN DC coupled pair voltage amp.

It is obviously a distortion product which might have been cured along the way with the introduction of differential inputs, current sources, current mirrors, cascode stages, emitter followers, baker clamps, and output triples, but what a faff for something which should be quite simple.

Mr Kemble, who deserves tons of credit and has my admiration for compiling his website full of interesting old (and newer) amplifier designs, offers us what might be a clue. It's on the page linked to above, and describes the treble filter used in the preamp stage of the Leak amplifiers.

It's the operating frequencies I'm interested in, with 6kHz and 9kHz being mentioned. These are the sibilance frequencies of the human voice, male and female respectively. Whether Leak saw it that way is another thing.

Today we tend to use wide response speakers often stood on stands away from wall and corner positions, and so there is little bass reinforcement. The hi-fi industry also seems to have an addiction to brightness, or what I would simply call treble boost, and it seems to have been built into modern loudspeakers, as if to make up for the lack of tone controls on amplifiers.

Comparing speakers, I have the Harbeth M20's which I use to monitor the squeaky treble because it's very obvious with these; the LS3/5A which are simply dead-flat in the mids and treble; and the Castle Trent’s which although being bass reflex, are back to the wall for sensible bass.

I can live with all the amplifier incarnations using the Castle Trent’s, and the wall positioning is reminiscent of how the average user would set up his/her speakers back in the day: either hung/bracketed to the wall, or stood on bookshelf’s.

The LS3/5A on stands offer an incredible sound stage, and hint at the squeaky treble, but the Harbeth's lay it bare.

The early bass problems were solved by the power supply, by reverting to laminated transformers with windings on split bobbins which reduced interwinding capacitance. This tends to prove the existence of mains interference (at least at this location) which is readily coupled to the secondary side by toroidal windings being so mutual. I wrapped a flux band around a laminated split-bobbin transformer and got bass loss: what I'd done is to increase capacitance between primary and secondary. There you go then.

My opinion is that some form of tone correction is essential with perhaps the exception of "miraculous" power amplifiers which seem to be pitch perfect with everything. Not even my Proprius are that miraculous (IMO). Without once ever touching established tonal adjustment, this journey has so far managed to manipulate the tonal results by means most readers (and some designers) would never dream of. But real tone controls, if only they were accepted, would make things a lot easier.
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: 165
Post Options Post Options   Thanks (0) Thanks(0)   Quote Dave Friday Quote  Post ReplyReply Direct Link To This Post Posted: 12 Aug 2019 at 12:38pm
I've seen amp designs where the pre-driver stages are fed from a separate higher voltage tap ( on the mains transformer ) to get more voltage swing to drive the output stage...I can't find out if this was a 70s thing though!!
lp12,oc9mk3,ca610p,krimson40watt pa,kef105.4
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: 8610
Post Options Post Options   Thanks (0) Thanks(0)   Quote Graham Slee Quote  Post ReplyReply Direct Link To This Post Posted: 13 Aug 2019 at 7:18am
Operating the VAS at a higher voltage than the power stage only works on +/- supplies as opposed to +/GND, but the Bailey amp was +/- and that was published 1968. It is most likely that it was a 70s innovation having seen the NAD3020 schematic (1978), although it was only 0.2V higher, and only on the positive rail. However, being regulated it would remain at 30.7V whilst the unregulated output stage supply would have fallen below 30.5V when power demanded.

The 3020 circuit can be seen here: http://p10hifi.net/planet10/manuals/3020b_schematic.gif (not for the faint hearted!).

The existing design capable of 30 watts at less than 0.1% THD (0.2% at 20kHz) sounds pretty reasonable for its circuit's "age", and if around in the 70s would probably have given other designs a run for its money.

However, it has become quite complicated with so many "fixes" and I can't help thinking there has to be a simpler solution.

I think the next step will be to look at the PNP/NPN voltage amplifier pair, which was not only used in the Leak Delta, but also in many others, including the Bailey amp; the Armstrong 621; the NAD3020; and my own initial idea back on page 3. I wasn't able to get good results out of it then, but I don't think I gave it my best effort, and I was put off by Mr Self's "singleton" comments.

The PNP/NPN open-loop gain isn't easy to prove because it would involve shunting the first transistor's emitter to ground via a large capacitor to demonstrate. In that condition and without any AC degeneration the open loop gain simulates around 100dB.

Because of that condition IPS transistor degeneration effects are difficult to model, so we're left with VAS degeneration to play with in trying to up the open-loop frequency response as far beyond 20kHz as possible.

The reason for this was given in Otala's transient intermodulation distortion AES paper in 1972 - to improve the sound - and is used for that effect in the Armstrong 621 and NAD3020, but not in the Leak Delta or the Bailey/Radford. I can remember the HiFi press heaping praise on the Armstrong and NAD so perhaps Otala got it right?
Not simple enough for Google-Bot to understand...
Back to Top
Richardl60 View Drop Down
Senior Member
Senior Member


Joined: 04 Nov 2014
Location: Yorkshire
Status: Offline
Points: 1449
Post Options Post Options   Thanks (0) Thanks(0)   Quote Richardl60 Quote  Post ReplyReply Direct Link To This Post Posted: 13 Aug 2019 at 3:51pm

Graham, an absorbing thread temporarily paused.

 

I know you made reference to the squeaky treble which has resulted every single time each modification has settled-in and … it's the operating frequencies I'm interested in, with 6kHz and 9kHz being mentioned. These are the sibilance frequencies of the human voice, male and female respectively.

 

Observationally, when switching from a standard mains cable to very high quality and multiply shielded mains cables results have consistently shown a general reduction in HF hash; when switching back to supplied cables (to the PSUs or my amps, CDs etc) the standard cables amongst other things have sound thin, bass lacks weight and definition and there tends to be a degree of confusion and sibilance to vocals. 

 

Whilst I fully appreciate that the majority of issues you have faced appear to rear after the lengthy burn-in period, there does seem some commonality to my findings with mains cables; both mains and air-borne interference also appears a common and increasing thread since the 1970s.  Since all of my cables are extremely short, have high current carrying capacity and high quality gold plugs in theory have little length/capacity to absorb airborne interference; my results have proven consistent from say 2” to 2’ of external cable length.

The Source turntable, Audiomods V micrometer, Dynavector XX2-2, Accession M, Elevator, Leema Antila 2Seco CD player, TDL studio 1.
Back to Top
Dave Friday View Drop Down
Senior Member
Senior Member


Joined: 07 Apr 2011
Location: Spain
Status: Offline
Points: 165
Post Options Post Options   Thanks (0) Thanks(0)   Quote Dave Friday Quote  Post ReplyReply Direct Link To This Post Posted: 13 Aug 2019 at 7:13pm
In the early 80s I had a NAD3020 at home for a few days...
It didn't do anything nasty but seemed a bit flat and lacking power compared to my Goodmans 110 tuneramp.

lp12,oc9mk3,ca610p,krimson40watt pa,kef105.4
Back to Top
 Post Reply Post Reply Page  <1 7576777879>
  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 0.125 seconds.