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 Forum

 

Open to all owners plus those contemplating the purchase of a Graham Slee HiFi System Components audio product and wishing to use this forum's loaner program: join here (Rules on posting can be found here)

This website along with trade marks Graham Slee and HiFi System Components are owned by Cadman Enterprises Ltd


1970s Design Indulgence

 Post Reply Post Reply Page  <1 144145146147148 345>
Author
Graham Slee View Drop Down
Admin Group
Admin Group
Avatar
Retired

Joined: 11 Jan 2008
Location: South Yorkshire
Status: Offline
Points: 16298
Post Options Post Options   Thanks (0) Thanks(0)   Quote Graham Slee Quote  Post ReplyReply Direct Link To This Post Posted: 22 Jul 2021 at 10:27pm
There's always another tweak, or so it would seem.

It's just my opinion, but -3dB should be accompanied by a 45 degree phase shift, because many agree it sounds natural. It results in "flat phase" at 10 times the cut frequency (at the bass end - the top end is 1/10).

Stack a number of cuts at the same frequency and it sounds as natural as an original CD brick wall filter. The world moved on from that, and so at the bass end I thought I needed to revisit the frequencies and rates.

Assuming an 8 ohm load, the output caps do their cut at approx. 4Hz. It would be nice if the VAS would cut at 2Hz, the IPS at 1Hz and the input at 0.5Hz. The bootstrap is adjusted higher up to take it out of the way at about 16Hz, so it doesn't cause bass peaking.

A double the frequency spread as in a Sallen Key filter offers best two pole flatness, so we can do the same here, but with three orders.

One problem sticks out, and that's the VAS grounding capacitor, which is already large. I had tried to manipulate the lows using a 470u instead of a 1000u, but realise it was like putting the cart before the horse, and sounded so.

I think the required 4700u to obtain 2Hz is OTT as most speakers won't go anywhere near, and neither will our ears. I therefore chose to try 2200u.

The cut coinciding with the output capacitor, will result in 60 degrees, not 45 degrees, but it's closer than what I had.

By spacing the other frequencies at 1:2 or wider, so the IPS decoupling is 2Hz and input high pass is 1Hz, then interaction is minimised.

Only two components (per channel) needed changing. The 470u in the VAS emitter circuit becomes 2200u, and the input cap increases from 220n to 1u.

The resulting bass (so far) seems more natural. The -3dB frequency is around 6.5Hz with +55 degrees phase according to SPICE. It has also shifted the +10 degrees bass phase nearer to 35Hz from 50Hz, so it should be better on big floor standers.

That none should be able to buy or sell without a smartphone and the knowledge in how to use apps
Back to Top
Sylvain View Drop Down
Senior Member
Senior Member


Joined: 18 Jan 2010
Status: Offline
Points: 481
Post Options Post Options   Thanks (0) Thanks(0)   Quote Sylvain Quote  Post ReplyReply Direct Link To This Post Posted: 23 Jul 2021 at 2:49pm
''...The resulting bass (so far) seems more natural. The -3dB frequency is around 6.5Hz with +55 degrees phase according to SPICE. It has also shifted the +10 degrees bass phase nearer to 35Hz from 50Hz, so it should be better on big floor standers.''

And thank YOU 
......acoustic double bass '' Contre basse '' is still being heard in remastered CD's and Veuve Vinyl. 
Sub woofer in a box produce a good bass effect But do not do justice to a French Contre-basse operating between 36-42 hz. 

Back to Top
Graham Slee View Drop Down
Admin Group
Admin Group
Avatar
Retired

Joined: 11 Jan 2008
Location: South Yorkshire
Status: Offline
Points: 16298
Post Options Post Options   Thanks (0) Thanks(0)   Quote Graham Slee Quote  Post ReplyReply Direct Link To This Post Posted: 05 Aug 2021 at 8:35am
So much for the even harmonics of valve amps. I'd not realised how damaging second harmonic distortion could be to my ears. Luckily, after a few days, they recover.

At this point, I should apologise for the dashed hopes of those who thought it was done.

The function now seems apparent, and I don't know if I should celebrate or be sad.

I seem to remember that Quad (Acoustical mfg co) were relied upon in always-on studio installs. I can't speak for the 405, but the 303 has several second harmonic fixes (deny them if you wish).

It now looks evident that the smoothing capacitors are given a hard time being a practical AC short - given that they have class-B asymmetrical power supply distortion and second harmonic distortion to deal with at the same time.

Add to the smoothers woes that their ripple current pulses are asymmetrical 620 - 1000 Hz, and the immeasurable subjective mid-frequency distortion throws an onslaught of knockout punches at them.

To make things easier to understand, we'll take the frequency of 620Hz because, due to silicon rectifier characteristics, the current pulses will appear as a periodic 620Hz half cycle - the period being 0.01 of a second - very asymmetrical.

Should a frequency in music be at 310Hz, its second harmonic is 620Hz, but also it demands two "shots" of power supply current - due to push-pull - which is also 620Hz, but asymmetrical.

Now, choose a different music frequency - any, I don't mind - and see how they mix. Rod Elliott has two excellent pages on transient intermodulation distortion, and at one point, discusses measurements for RF. If you understand how AM radio works, you'll see why it's essential at RF.

Mixed frequencies always combine to produce new frequencies, as anybody sat in rush hour traffic with the window down will testify.

So, if the second harmonic is high enough to be audible (obviously unconsciously at the rate of musical change), it will mix in the smoothers and produce another alien frequency for that particular split second.

In the Quad 303, it had a hard job mixing within the smoothers because there was a regulated power supply in the way. Also, stereo-loop distortion was dealt with in the same way as the Tobey and Dinsdale amp. Additionally, the voltage regulator sensed the ground side.

The 303 was a combined effort between Peter Walker and Peter Baxendall - the latter having contributed to the quasi-output stage linearity.

The "Bax-diode", the output triple, and additional voltage amplifier gain stage all improved linearity, meaning a good reduction in second harmonic distortion.

Amplifier design then progressed to today's high negative feedback designs; otherwise, you have second harmonic distortion by multiple paths.

The difference between valves and transistors is the output transformer, which will always round off a sharp waveform.

That none should be able to buy or sell without a smartphone and the knowledge in how to use apps
Back to Top
Graham Slee View Drop Down
Admin Group
Admin Group
Avatar
Retired

Joined: 11 Jan 2008
Location: South Yorkshire
Status: Offline
Points: 16298
Post Options Post Options   Thanks (0) Thanks(0)   Quote Graham Slee Quote  Post ReplyReply Direct Link To This Post Posted: 05 Aug 2021 at 10:22am
What is 1/2VCC in an unregulated supply?

1/2VCC will vary under load, because of standing currents in different parts of the circuit.

However, we can set it at an arbitrary value, which turns out to be 37V under quiescent conditions, on a 245V ac supply!!!

The fictitious 230V supply is a very big problem in making an amplifier suit 220V to 250V outlets, but you know my feelings for the EU's level playing fields.

I think we must have a transformer with primary taps! That would enable us all to be on the same page.

Thanks to the incredibly stable voltage amplifier, we waste 8 volts of the negative swing, and so, in making 1/2VCC arbitrarily right, we also lose something like 8 volts on the positive swing. So, we need to add 8 volts to our HT supply.

If we use the existing 50V transformer we can only achieve 25 watts with best case symmetry - although it will do 47 watts asymmetrically.

At 25 watts we generally reduce the second harmonic by 10dB. Will -65dB be sufficient to result in pleasurable continuous uptime listening?

I'm not changing a thing till I find out.

That none should be able to buy or sell without a smartphone and the knowledge in how to use apps
Back to Top
Sylvain View Drop Down
Senior Member
Senior Member


Joined: 18 Jan 2010
Status: Offline
Points: 481
Post Options Post Options   Thanks (0) Thanks(0)   Quote Sylvain Quote  Post ReplyReply Direct Link To This Post Posted: 05 Aug 2021 at 6:51pm
What devotion to perfection ......unhindered for marketing specifications and indulgences 

A new generation listening to synthesise music reproduced through a 'D' class amp with it's punching bass and clear treble will be for a treat for true music when the 70'S amp would experienced. They would still access to the dozens of  MTV Unplugged Cd's or new Vinyl plastic produced on a digital lathe ........they would experience real and true not '' virtual'' when class AB exhibit the transient, the nuance the emotion of acoustic instruments harmonising with the soul..... 

Keep at it 
Back to Top
Graham Slee View Drop Down
Admin Group
Admin Group
Avatar
Retired

Joined: 11 Jan 2008
Location: South Yorkshire
Status: Offline
Points: 16298
Post Options Post Options   Thanks (0) Thanks(0)   Quote Graham Slee Quote  Post ReplyReply Direct Link To This Post Posted: 05 Aug 2021 at 10:49pm
I am tempted not only to specify an 80V HT, but also to replace the bootstrap, and lose 8V in the positive side to counter the 8V in the VAS emitter.

The VAS transistor load would then comprise an MPSA92 PNP with a 1k emitter resistor to rail, a 1k resistor in series with a 1N4148 diode from rail to base, and 10k from base to ground.

The approx. 10dB reduction in the second harmonic is providing good sound so far. The 80V rail would give us our target power output back and the DC prior to the blocking caps to go to 1/2VCC.

By then, the linearity would be like it is now - doing 25W at <0.04% THD1 - but at 40 watts (I hope).

However, the VAS loaded as above should reduce its second harmonic further. The transistor count then being seven.

A side benefit indicated in simulation would see bass cut at 6dB/octave rather than the bootstrap 9dB/octave, and improved bass phase.

We'll see.

That none should be able to buy or sell without a smartphone and the knowledge in how to use apps
Back to Top
Graham Slee View Drop Down
Admin Group
Admin Group
Avatar
Retired

Joined: 11 Jan 2008
Location: South Yorkshire
Status: Offline
Points: 16298
Post Options Post Options   Thanks (1) Thanks(1)   Quote Graham Slee Quote  Post ReplyReply Direct Link To This Post Posted: 10 Aug 2021 at 6:05am
The current source is in, and empirically adjusted to provide the 7mA VAS current. It has a slight problem on switch-on which causes a pop, and I'm working on mitigating that right now.
That none should be able to buy or sell without a smartphone and the knowledge in how to use apps
Back to Top
 Post Reply Post Reply Page  <1 144145146147148 345>
  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.