Single Ended amplifiers

[acg]

18 minutes ago, Grant Slack said:

The main point that I wanted to mention, however, is the way that the PP distortion advantage is bigger at low output power. Proponents of SE amplifiers often claim that they are an ideal choice at modest volumes with reasonably efficient speakers. Not so, by that graph.



 

 

Of course that is likely to be true if one amplifier has audible distortion and the other non-audible, but if they are both inaudible then the THD difference could be thousands of times greater and it would make no difference to SQ.  Oh, forget the "modest volumes" bit of your statement...if it can't go loud in your system then the amp is crap regardless of topology (haha). 

 

My horns need 0.015w to make 92dB at my listening chair from 80Hz - 40kHz.  Likewise, 92dB at 20-80Hz requires 0.125watt.  Forget Class AB or Class D at those output levels, Class A it must be.  Also, there is no need for negative feedback or pushpull in that situation.  That is where SE really works well...when it does not have to work much at all (unfortunately).

[davewantsmoore]

45 minutes ago, Grant Slack said:

Proponents of SE amplifiers often claim that they are an ideal choice at modest volumes with reasonably efficient speakers.

The reason that is said, is not because they are "superior" to PP in that regard .... but because at anything other than modest volume and/or efficient speaker, then the typical SE amp might be over-driven.

[Grant Slack]

@davewantsmoore I honestly believe that the argument has been that they are superior in that regard. (I just discovered the 'Insert Nick' function.)

[Grant Slack]

On 07/08/2018 at 3:38 PM, Zaphod Beeblebrox said:

The blue line is SE and the red line is PP.

 

1 hour ago, Grant Slack said:

The main point that I wanted to mention, however, is the way that the PP distortion advantage is bigger at low output power. Proponents of SE amplifiers often claim that they are an ideal choice at modest volumes with reasonably efficient speakers. Not so, by that graph.

 

47 minutes ago, acg said:

Of course that is likely to be true if one amplifier has audible distortion and the other non-audible, but if they are both inaudible then the THD difference could be thousands of times greater and it would make no difference to SQ.  Oh, forget the "modest volumes" bit of your statement...if it can't go loud in your system then the amp is crap regardless of topology (haha). 

 

My horns need 0.015w to make 92dB at my listening chair from 80Hz - 40kHz.  Likewise, 92dB at 20-80Hz requires 0.125watt.  Forget Class AB or Class D at those output levels, Class A it must be.  Also, there is no need for negative feedback or pushpull in that situation.  That is where SE really works well...when it does not have to work much at all (unfortunately).

Hello, I am afraid I don't understand. Are you saying that SE is the one to choose at low power requirements, where the gap between the SE and PP distortion is at its greatest? Where the red (PP) line is at its lowest point and showing the greatest advantage? If only considering Class A, zero-feedback amplifiers, I would have chosen the PP, based on that graph, for low-power situations.

[acg]

31 minutes ago, Grant Slack said:

 

 

Hello, I am afraid I don't understand. Are you saying that SE is the one to choose at low power requirements, where the gap between the SE and PP distortion is at its greatest? Where the red (PP) line is at its lowest point and showing the greatest advantage? If only considering Class A, zero-feedback amplifiers, I would have chosen the PP, based on that graph, for low-power situations.

 

No, I'm saying that if they are both low enough to be inaudible that you would need measures other than THD to distinguish the sound.  Pass mentions IMD in that article and there would be others.  THD is so overrated as an indicator of audio quality.  Once it is "good enough" there is no need to try for "vanishingly low" because no-one can hear vanishingly low THD, but they can hear other things that may be stuffed up by chasing a silly number that does not mean much...Pass mentioned negative feedback and too many gain stages in that article as such things that can cause problems when chasing low THD.

[Grant Slack]

OK, thank you. Can you suggest any "measures other than THD to distinguish the sound" where SE outperforms PP, in principle? And where your very important criterion of being audible is met?

 

Regards

Grant

Edited August 9, 2018 by Grant Slack

[vivianbl]

On 31/07/2018 at 8:27 PM, davewantsmoore said:

It is an understatement to say that "this speaker will not be driven well by a SET amplifier"

davewantsmoore from an earlier post- just a clarification/expansion 

Did you mean that the Quad ESL 57s will not be driven well by a SET amplifier?

[Grant Slack]

I was very strongly counseled against using valve amps with ER Audio ESL speakers by their designer, but I can't remember exactly why. It may have been his conviction that they are best suited to high current amps.

[georgehifi]

56 minutes ago, Grant Slack said:

I was very strongly counseled against using valve amps with ER Audio ESL speakers by their designer

Probably because of this, the way it dives down at the HF, most ESL's are worse, ML's quad57's etc:  down to 1ohm this graph "may" be a little doctored.

 

http://www.eraudio.com.au/Mini_Panels/Slide1.JPG

 

http://www.audiomisc.co.uk/57and303/57imp.gif

 

https://www.stereophile.com/images/912Montisfig1.jpg

 

Cheers George

 

 

Edited August 9, 2018 by georgehifi

[davewantsmoore]

2 hours ago, vivianbl said:

Did you mean that the Quad ESL 57s will not be driven well by a SET amplifier?

Yes.   It has a very reactive impedance.    It wants an amplifier that has a low output impedance.

 

A typical SET amplifier will produce a (likely very) distorted frequency response... and is liable to be over-driven.

 

Due to this, a SET will sound very "different" to a more appropriate amplifier ... and so there may be substantial "wow" (that's different!) factor when a SET is swapped in.

[davewantsmoore]

1 hour ago, Grant Slack said:

I was very strongly counseled against using valve amps with ER Audio ESL speakers by their designer, but I can't remember exactly why. It may have been his conviction that they are best suited to high current amps.

Valve amps typically have:

• Low power
• High output impedance
• Inability to drive low impedance loads at full power without clipping and/or instability

 

ESLs want none of those things.

[davewantsmoore]

3 hours ago, Grant Slack said:

@davewantsmoore I honestly believe that the argument has been that they are superior in that regard. (I just discovered the 'Insert Nick' function.)

Yes, I realise that.  

 

What I am saying as that all the people who you have heard run that generalisation are wrong.    Like most things, it's based on a somewhat-truth, which has been twisted.

[acg]

1 hour ago, Grant Slack said:

OK, thank you. Can you suggest any "measures other than THD to distinguish the sound" where SE outperforms PP, in principle? And where your very important criterion of being audible is met?

 

Regards

Grant

 

Let me phrase my answer this way...I'm not aware of any situation where a PP amp will outperform a SE amp where a SE amp is highly suited to the task, which as has been discussed in this thread is only in limited situations.

 

I am someone that has gone to great lengths to get myself an acoustic system that is suitable for SET amplifiers because it produces a dynamic quality to the sound that I've not found elsewhere.

 

 

[FR DRew]

On 08/08/2018 at 4:20 PM, Zaphod Beeblebrox said:

There are no single driver speakers that can:

 

* Deliver a full frequency response (20Hz ~ 20kHz).

* Not "smear time". 

 

Like it or not, even single driver speakers damage the phase response of a signal.

 

There are only two proven methods (to the best of my knowledge) to preserve accurate phase response (ie: Non 'time smearing'). They are:

 

* Accurate time alignment of drivers, with good damping to reduce reflections. Duntech, VAF, et al.

* Full range electrostatic speakers (which are not without their own set of problems).

 

Triodes are an excellent amplification device. The best of any vacuum tube, in fact. They are almost as good as a modern bipolar transistor WRT linearity. You are right to claim that triodes are a good choice. 

 

 

 

 

I'll deal with the last, first.  I'm not aware of any solid-state bjt that operates linearly without feedback applied. (Something triodes have been easily able to do for 70+ years).

 

 Never claimed a 20-20k response.  Given that precious little beyond Duntech Sovereigns even try to claim a response like this, to criticise single driver speakers for not doing what 99% of other commercial speakers can't do is a long bow to draw...

 

I'll take my fast responding light cones in concrete enclosures and a 50 - 18k response at 94dB/W happily.  Less "time smear" than all the timber cabinet multi way speakers I've ever heard.  Sorry to disagree (or not go for the 20-20k straw man).

[Mickstuh]

My interest in single-ended amps was piqued recently by reading the review on 10audio of a 12 watt German solid-state SE power amp (Valvet E2). Having read many of the reviewer's reviews, I was fascinated by his conclusion. So I've enjoyed reading this thread too. Thanks to all. 

So my question: any SE SS amps people have heard that they have liked?

 

[Colin Rutter]

"Time smear"
You guys have got to get help.
Do you mean phase coherence or phase alignment?
"Time smear" isn't a scientific or audio engineering term.
Someone in marketing has made it up.

[Happy]

@Mickstuh i tend to favour class a single ended be it valves or state solid. I have a soft spot for my little Sugden A21a.

[Colin Rutter]

I forgot you probably are getting "time smear" from your "flux capacitors".

[FR DRew]

On 08/08/2018 at 4:20 PM, Zaphod Beeblebrox said:

There are no single driver speakers that can:

 

* Deliver a full frequency response (20Hz ~ 20kHz).

* Not "smear time". 

 

Like it or not, even single driver speakers damage the phase response of a signal.

 

There are only two proven methods (to the best of my knowledge) to preserve accurate phase response (ie: Non 'time smearing'). They are:

 

* Accurate time alignment of drivers, with good damping to reduce reflections. Duntech, VAF, et al.

* Full range electrostatic speakers (which are not without their own set of problems).

 

Triodes are an excellent amplification device. The best of any vacuum tube, in fact. They are almost as good as a modern bipolar transistor WRT linearity. You are right to claim that triodes are a good choice. 

 

 

 

 

I'll deal with the last, first.  I'm not aware of any solid-state bjt that operates linearly without feedback applied. (Something triodes have been easily able to do for 70+ years).

 

 Never claimed a 20-20k response.  Given that precious little beyond Duntech Sovereigns even try to claim a response like this, to criticise single driver speakers for not doing what 99% of other commercial speakers can't do is a long bow to draw...

 

I'll take my fast responding light cones in concrete enclosures and a 50 - 18k response at 94dB/W happily.  Less "time smear" than all the timber cabinet multi way speakers I've ever heard.  Sorry to disagree (or not go for the 20-20k straw man).

[Zaphod Beeblebrox]

31 minutes ago, FR DRew said:

I'll deal with the last, first.  I'm not aware of any solid-state bjt that operates linearly without feedback applied. (Something triodes have been easily able to do for 70+ years).

Triodes already employ internal, degenerative feedback. Tetrodes and Pentodes remove that feedback system, to acquire more gain. BJT amps can be built to operate with no global NFB, relying, instead, on a similar form of degenerative feedback that triodes rely on.  AKA: Emitter resistors.

 

Quote

 

 Never claimed a 20-20k response.  Given that precious little beyond Duntech Sovereigns even try to claim a response like this, to criticise single driver speakers for not doing what 99% of other commercial speakers can't do is a long bow to draw...

I should have stated that single driver systems cannot remotely approach 20Hz ~ 20kHz. As for a commercial speaker that attains 20Hz ~ 20kHz, there's this, very reasonably priced, one:

 

https://vaf.com.au/products/signature-i-93mkll

 

Quote

 

I'll take my fast responding light cones in concrete enclosures and a 50 - 18k response at 94dB/W happily.  Less "time smear" than all the timber cabinet multi way speakers I've ever heard.  Sorry to disagree (or not go for the 20-20k straw man).

"Lighting fast"? 

Only if it has a diaphragm of around 20mm. 

 

Edited August 9, 2018 by Zaphod Beeblebrox

[Zaphod Beeblebrox]

27 minutes ago, Mickstuh said:

My interest in single-ended amps was piqued recently by reading the review on 10audio of a 12 watt German solid-state SE power amp (Valvet E2). Having read many of the reviewer's reviews, I was fascinated by his conclusion. So I've enjoyed reading this thread too. Thanks to all. 

So my question: any SE SS amps people have heard that they have liked?

 

Sure. My very first SS amp was a 5 Watt, SE beastie. Before that time I was building exclusively valve amps. Push pull and SE. I preferred the push pull valve amps. So, why did I build a SE SS amp? Because, way back then, AD149 output transistors were very expensive for a school-kid. I could not afford two of the suckers and I wanted to muck about with transistors. I went back to valves, because I could get more power for less money in push pull. BTW: My old 5 Watt SE SS amp sounded very nice. It just didn't go very loud. In fact, it sounded very much like a SET amp. A modern SE SS amp would be much better, because there are far superior devices to the old AD149 available today. In fact, I would posit that one could build an SE SS amp, running on a (say) 300 Volt rail, with an output transformer coupled to the speaker. The result would be very, very close to a SET amp, as BJTs possess similar characteristics to Triodes. OTOH, MOSFETs possess similar characteristics to Pentodes. 

Edited August 9, 2018 by Zaphod Beeblebrox

[Zaphod Beeblebrox]

13 minutes ago, Happy said:

@Mickstuh i tend to favour class a single ended be it valves or state solid. I have a soft spot for my little Sugden A21a.

Your Sugden A21A was a push pull Class A/B amplifier, running a pretty hefty level of Class A bias. It was not SE.

[Guest Muon N']

17 minutes ago, Colin Rutter said:

"Time smear"
You guys have got to get help.
Do you mean phase coherence or phase alignment?
"Time smear" isn't a scientific or audio engineering term.
Someone in marketing has made it up.
 

 

14 minutes ago, Colin Rutter said:

I forgot you probably are getting "time smear" from your "flux capacitors".

You profess to know a lot, but do you know the meaning of the word condescending?

[FR DRew]

On 08/08/2018 at 4:20 PM, Zaphod Beeblebrox said:

 

Edited August 9, 2018 by FR DRew
Double post

[Mickstuh]

Thanks for that Trevor. Thanks too @Happy. I'll have to check out the Sugden. I know that I have to hear some SE SS class A to satisfy my curiosity. But in relation to the Sugden, the manufacturer claims they are SE. Are they being mischievious?