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 7778798081 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: 16 Aug 2019 at 8:53am
I once had a Ford Cortina. It was one of the best driving cars I ever owned.

Years later I bought another Ford Cortina and at 60 MPH the steering wheel and whole vehicle vibrated so much it was unbearable.

I replaced just about everything on it including engine/gearbox mounts, prop-shaft, wheel bearings, ball joints, void bushes, springs, shock absorbers, etc. I had the body checked for twisting and had it professionally tracked three times. I even made templates to align caster and camber. Over the period of 1 year it cost me a fortune. I'd have been better getting shut and buying a different car.

No matter what I did, it made no difference.

There's an analogy here somewhere...
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: 16 Aug 2019 at 11:01am
So today I'm trying to find out if the bias spreader is the culprit, and I'm doing that by replacing it with 3 series wired signal diodes in series with a 100 ohm trimmer between driver bases, and a 47uF bypass capacitor.

Two of the diodes receive thermal feedback from the output transistors to prevent thermal runaway - they're simply stuck on them using epoxy two pack adhesive - so as the base emitter junction of each transistor conducts harder and their temperature increases, the attached diode gets hotter and conducts harder, reducing base emitter voltage.

Unfortunately there will be thermal lag because of the plastic transistor case. The thermal track version of the MJL21193/4 transistors, prefixed NJL, has a diode inside the package for precise thermal tracking...

the data sheet exclaims

"Superior Sound Quality Through Improved Dynamic Temperature Response
Significantly Improved Bias Stability
Simplified Assembly
Reduced Labor Costs
Reduced Component Count
High Reliability"

...except you can't get them because they're not made anymore, because the amplifier industry didn't want...

Superior Sound Quality Through Improved Dynamic Temperature Response
Significantly Improved Bias Stability
Simplified Assembly
Reduced Labor Costs
Reduced Component Count
High Reliability

Wink
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: 16 Aug 2019 at 6:51pm
Circuit update 16/08/2019

Audio Power Amplifier 160819

(click view image to be able to read circuit clearly)

R5: the new collector load for T1 for it to assume 16V such that the transistor collector-emitter voltage is increased to 13.1V from 3.5V.

C5 and R14 form the level translator.

C1 increases to 10uF to preserve LF phase.

The bootstrap: R11, 12 C8, returns with R12 3k instead of 3k9. It wasn't guilty of distortion as thought.

D5 to D7 plus trimmer R13 replace the transistor bias spreader as it was thought, being an amplifier in its own right, could have been unstable.

D5 and D6 are thermally coupled (stuck) to T5 and T6 to provide thermal tracking.

The diode bias spreader impedance is shunted from mid-bass frequencies by C9.

R19 is increased from 180 ohms to 360 ohms to lower the DC load on the VAS.

R19 is shunted by C11 (1uF) to sweep charge from the power transistors at high frequencies (from about 16kHz when output is 1W). This value might be increased but ought not to have a longer time constant to that of C9 to avoid switch-on stress to T3 and T4.

So far the "icy-S/squeaky treble" problem has not manifested itself... (yet)


Edited by Graham Slee - 16 Aug 2019 at 6:53pm
That none should be able to buy or sell without a smartphone and the knowledge in how to use apps
Back to Top
peterb View Drop Down
Senior Member
Senior Member
Avatar

Joined: 02 Feb 2017
Location: Cornwall
Status: Offline
Points: 332
Post Options Post Options   Thanks (0) Thanks(0)   Quote peterb Quote  Post ReplyReply Direct Link To This Post Posted: 19 Aug 2019 at 1:55pm
I seem to remember that the Bailey amp also used thermal tracking by sticking one of its transistors on the output stages heat sinks. I tried unsuccessfully to find a conductive adhesive to achieve this but ended up making clamps out of a section of copper pipe and used heat sink compound to give them the best 'bond' I could !
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
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: 19 Aug 2019 at 4:22pm
How about class-A?

Before I begin, the output transistor base stoppers were changed from 6R8 to 2R2 to allow improved high frequency charge carrier sweeping.

I'm not talking about real class-A but the "pure class-A" a number of manufacturers claim.

Turn the wick up to 291mA standing current and the case delivers 40 watts of heat. How hot is that? Just bearable in a 25C ambient room (note to self: needs a thermal trip).

The amplifier measures like a valve amp from the point of view that distortion reduces along with signal:

38WPC 0.5% THD 1kHz
30WPC 0.5% THD 20Hz - 20kHz
10WPC 0.2% THD 20Hz - 20kHz
1WPC 0.05% THD 1kHz
1WPC 0.1% THD 20Hz - 20kHz

Output noise: -98dB (which is the same as saying 73dB S/N)

Now, that was with all the NFB from the output. It didn't sound so good.

Replacing local NFB making global 14dB resulted in

240mA standing current; 35 watts case dissipation.

40WPC just below clipping 1% THD 1kHz (990mV rms input)
32WPC 0.3% THD 1kHz
32WPC 0.3% THD 20Hz - 20kHz
10WPC 0.3% THD 20Hz - 20kHz
1WPC 0.07% THD 1kHz
1WPC 0.09% THD 20Hz - 20kHz

Output noise: -94dB (69dB S/N)

Standing current was adjusted for minimum harmonic peaks without bringing the noise up, and that's why it resulted in 240mA.

Advancing standing current further the 3rd harmonic disappears but the noise comes up to meet it, so more current doesn't improve matters, and just makes the case a lot hotter.
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: 19 Aug 2019 at 4:38pm
Originally posted by peterb peterb wrote:

I seem to remember that the Bailey amp also used thermal tracking by sticking one of its transistors on the output stages heat sinks. I tried unsuccessfully to find a conductive adhesive to achieve this but ended up making clamps out of a section of copper pipe and used heat sink compound to give them the best 'bond' I could !


The problem I have about transistor bias spreaders is their stability, or lack of it.

I have two designs which use them: one didn't matter too much because the boss man said not to worry; the other is the Proprius which has a regulated supply and so is stable.

A music signal using an unregulated supply amplifier causes the HT to vary considerably, and the gain of a transistor causes its collector emitter voltage to vary even with quite a stiff potential divider (and the usual 18R resistor doesn't help much). Capacitors help but cannot hold the correct voltage long enough with a big band playing.

Diodes don't have gain. The trimmer can vary with HT but not by much.

The thermal tracker transistors were a great idea except their sales were limited and so the manufacturers dropped making them, or only supply to order, mine not being big enough.

So it's back to stirring Araldite and sticking a 1N4148 to each power transistor, just like the bad good old days.
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: 19 Aug 2019 at 4:45pm
Another thing that helps is to reduce the power dissipated by R3,4 and 12 by replacing them with series/parallel combinations resulting in the same values. This is to reduce the dissipation to around 1/10th their rating which seemed to improve the sound.
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 7778798081 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 0.111 seconds.