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 113114115116>
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: 9567
Post Options Post Options   Thanks (0) Thanks(0)   Quote Graham Slee Quote  Post ReplyReply Direct Link To This Post Posted: 11 Oct 2020 at 8:29am
Well, Mick, I've had dreams of it being an equal but different product, and back at the start, I thought it might beat the Proprius. Still, this indulgence has highlighted a sort of wakeup call.

Having been a "see and do" designer ("Graham, see this?" "Yes" "Well, do it!"), I'd started questioning things, which probably made me unemployable and brought me to the place I am.

Although I have great respect for Self and Cordell, they specialize in what I call "pattern part amplifiers," which compete as equals, commoditizing transistor amplifiers, which sends music lovers rushing into the arms of the tube copycats. It's a great way of terminating thought, which is the Orwellian direction of society, and this indulgence is my two fingers salute to that!

I suppose the main problem I have is a lack of reference except for my memories. I could swear that this configuration used to sound great, so why should it be so hard to get it to do the same today? Although we are in a much-changed world to that 50 or 60 years ago, few admit it. Most don't want to see the obvious, and the rest don't even care. Today, everyone can transmit anything they wish by radio waves. Still, it took 53 years for Radio Caroline to get a license!

Analogue has been shoved into a little corner by legislation allowing an explosion of digital and control technology signals to land on the mains, and thanks to Cambridge for boasting its "invention" of the toroidal transformer, the laminated transformer has become a historical laughing stock amongst transformer manufacturers. Toroidals saturate easily, which couples mains borne comms straight into the power supply from which we are supposed to produce a higher-powered facsimile of the high fidelity signal.

This was largely overcome by the Proprius's use of switched-mode regulated power. Since then, power supply distributors have been gobbled-up, leading to supply difficulties for small manufacturers such as us.

The small manufacturer is now faced with only one design route, in that he must make the best of what he is allowed to use (read discrete). China's monopoly decides which parts are available to Western distributors, and the UK has mostly given up the fight, forcing us to source from the US. This situation will only last while buoyant US-China rivalry exists.

Effectively, an indulgence has become a race against time.
Not simple enough for Google-Bot to understand...
Back to Top
Aussie Mick View Drop Down
Senior Member
Senior Member


Joined: 21 Feb 2014
Location: Bendigo Aust.
Status: Offline
Points: 872
Post Options Post Options   Thanks (0) Thanks(0)   Quote Aussie Mick Quote  Post ReplyReply Direct Link To This Post Posted: 11 Oct 2020 at 10:26am
An excellent, and deeply appreciated answer. 
All the best,
Mick.
PS - can’t wait to hear it!
Rega RP8 - Apheta 2 - Accession MC Enigma PS -Solo ULDE (Focal Utopia) - Proprius - PMC Twenty5.22
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: 9567
Post Options Post Options   Thanks (0) Thanks(0)   Quote Graham Slee Quote  Post ReplyReply Direct Link To This Post Posted: 13 Oct 2020 at 10:58pm
So, how does it sound?

After 9 days on all the time since the addition of the isolating diode, my tinnitus has not returned.

It still has what some call a brightly lit treble. The bass and midrange are very close to the "Slee house sound". Around half my test material allows one to chill-out and be drawn into the music. It is mostly the lyrics where I get that "never heard that before" reaction.

The precision of percussion reproduction isn't exactly where I'd like it. The jingle stick on "One For The Vine" (Genesis - Wind and Wuthering) is too "hashy" and doesn't sound close enough to being a jingle stick, yet.

jingle stick

The only vid I could find featuring the jingle stick

Not simple enough for Google-Bot to understand...
Back to Top
BAK View Drop Down
Senior Member
Senior Member
Avatar

Joined: 14 Mar 2010
Location: Kentucky, USA
Status: Offline
Points: 1132
Post Options Post Options   Thanks (0) Thanks(0)   Quote BAK Quote  Post ReplyReply Direct Link To This Post Posted: 14 Oct 2020 at 12:08am
Bruce
AT-14SA, Pickering XV-15/625, Technics SL-1600MK2, Reflex M, Lautus, Technics SH-8066, Dynaco ST120a, Eminence Beta 8A in custom cabs;; Using Majestic DAC
Enjoy Life Your Way!
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: 9567
Post Options Post Options   Thanks (0) Thanks(0)   Quote Graham Slee Quote  Post ReplyReply Direct Link To This Post Posted: 14 Oct 2020 at 8:34am
Rob Townsend, left of stage, at 68 (now 71).

Ah, breathe in the burnt triethylene glycol particulates (disco fog) - gives a more patchy result than cigarette smoke - but more healthy (in contact with the eyes it can cause immense pain).
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: 9567
Post Options Post Options   Thanks (0) Thanks(0)   Quote Graham Slee Quote  Post ReplyReply Direct Link To This Post Posted: 17 Oct 2020 at 11:11am
Do you remember when I tried the laminated transformers? Didn't it give a more full-bodied feel to the music?

The trouble is, I can't remember if it took away the shrill ess sounds you get from sibilant singers, such as Timothy B. Schmit of the Eagles, most of Eric Clapton's output, some of Seal's, and bright recordings of Paul Simon. Everything else just sounds fine - I think.

(by the way, the only way my posts can look like I've still got an agile mind is thanks to Google)

Polishing out (as in the use of T-Cut), the last sharp scratches of the esses is getting a bit of a chore, but the most significant changes now come from messing about with the VAS between HT and ground. As this is the highest voltage gain part of the circuit and depends on the power supply to modulate the power supply (circles within circles), it is hardly surprising.

The VAS emitter circuit bypass capacitor is essential for there to be voltage gain. The considerable value of 1,000uF ensures cut off is well below the low bass. Such large electrolytics are bad for HF, so you bypass them with a small film capacitor, except that wasn't helping. The large capacitor's inductance alone acts as a series NFB with the VAS collector load's stray inductances. A high emitter inductance must therefore reduce the parasitic gain at very HF. So, out came the bypasses bypass, and the esses were noticeably improved.

The input impedance of the VAS acts directly on the gain of the input stage collector. The input stage collector has to have a capacitor to ground, or it starts to oscillate near 40MHz, this being an artifact of this configuration, which I remember from "Teach-in 76". Things like this have to be optimised because of the parasitics of the transistor. There is, therefore, a Goldilocks value that must remain unchanged. Global negative feedback has zero influence over that.

Coincidentally, it is the VAS collector capacitor that sets the slew rate in this design. I hope you remember this, as nobody else in the history of solid-state amplifiers will tell you.

Back to the VAS input impedance, and this acts in parallel with the IPS collector resistor to produce the real IPS load. The VAS input impedance is effected by the small VAS emitter "bleed" resistor, which also has a Goldilocks value, and I just went and spoiled it by changing it all so slightly.

Beware what the simulator reckons! An "ear tweak" at this stage will probably be more correct, and that's because I cannot see the stray inductance, which must be the cause. If I can't see it, I can't model it, and thus, the simulator can't show it.

Even at this Goldilocks stage, we still have the budgie whistle of Timothy B. Schmit on TGFY (Eagle fans will understand). It could all be actually true to life, and I might be gilding the lily, but I don't think so.

Thus, my attention is turning back to the power supply, knowing that at 240 volts and 50Hz, this 230V 50/60Hz transformer must be well into saturation. Saturation is akin to clipping, and therefore, harmonics are produced, and is there one near to male vocal sibilance?

It would be nice to prove such things without too much experimentation, so I put the amp on the Variac and wound it down to 205 volts. The transformer will now be nearer its designed saturation, but the circuit won't have as much open-loop gain, so the situation is far from ideal.

Tim's budgie wasn't as prominent, and Paul Simons Boy in a Bubble more listenable. Being so conscious of the slightest sibilance is really taking my mind off the music.

It might be possible to further filter the power supply - to filter out the saturation harmonics sufficiently - and if so, the smoothers will not be facing what seems to be the multiple tasking of what I can only entitle as ghosting or phantom ripple.

As usual, I cannot fathom a way to measure any of this. As Bruce knows, I've had the supply on the FFT, but then again, the budgie whistle only intensifies after a long on-time. It is challenging to measure something so elusive.
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: 9567
Post Options Post Options   Thanks (0) Thanks(0)   Quote Graham Slee Quote  Post ReplyReply Direct Link To This Post Posted: 18 Oct 2020 at 9:22am
The power supply pi circuit

Where multiple smoothers are used, a break in the positive connection between two of the smoothers, which results in equal capacitance on either side of the break, can be reconnected using an inductor.

pi filter

Cordell's book shows a small value resistor (0R22) instead of the inductor. I modeled an inductor value for the amp power supply's approximate capacitance and a Zo of 0R22 using an online calculator. I have gone to zero pains to work this out myself, mainly because most things are never that accurate in real life.

Zo should be the output impedance of the transformer/rectifier, and the load supposedly matches it. In this class-B amplifier, neither of these will be constant at all, so why bother trying to square an impossible circle?

After some manipulation (not quite cheating), the value given by the calculator was 300uH to attenuate frequencies above, say 200Hz. The inductor resistance had to be close to 0R22. So using my Peak LCR meter, I set about unwinding a 1mH 1-ohm loudspeaker crossover inductor. At 350uH on the meter, the DC resistance read 0 ohms - it's not that accurate - and I thought it good enough.

This was inserted after the 4th 1000uF smoother, leaving 2 x 1000uF plus the two decouplers at 680uF a piece, totaling 3360uF - not balanced by any chalk of the imagination, but close.

Some very slight hum was picked up by the inductor's proximity with the on-switch neon, but that can be tidied up later.

To my ears, some of the sibilances had gone, and the presentation became a tad more "holographic."
Not simple enough for Google-Bot to understand...
Back to Top
 Post Reply Post Reply Page  <1 113114115116>
  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 0.250 seconds.