Jump to content

Single Ended amplifiers


Recommended Posts

On 31/07/2018 at 4:46 PM, Zaphod Beeblebrox said:

* Compared to an approximately equivalent push pull amp, a single ended amp delivers LESS power, as the load impedance falls. This is the opposite of the 'perfect amplifier'. A 'perfect amplifier' will always deliver more power as the load impedance falls. This can be a significant problem with ESLs, generally and Quads in particular, as their impedance varies wildly.

Thanks for your succinct summary of the key issues with SE, that was very helpful. Totally understand that there are limitations and that speakers need to matched to the SE parameters to benefit, as many others have said. Just want to seek further explanation (excuse my ignorance !) on one of your points quoted above. How will this effect be heard/show up in the audio/music output from the speakers?. 

  • Like 1
Link to comment
Share on other sites



15 minutes ago, vivianbl said:

How will this effect be heard/show up in the audio/music output from the speakers?. 

It is a very deep and pertinent question.

 

On the surface of it.... the answer is that the speakers frequency response will usually be changed.

 

This is the biggest difference that people normally notice (broad changes in frequency response tilt are very audible)... and the change in the frequency response if often not what the speaker designer intended.  Hence the "very far from the perfect amplifier" comment.

 

Once you look past this, the question goes to the very heart of 'how a speaker works'....  and take quite a few words to answer.

  • Like 2
Link to comment
Share on other sites

22 hours ago, A9X said:

I hope you are enjoying your new acquisition. If so, why did you start the thread? It seems to me that you are looking to justify your purchase.

 

As for the parallel SE, yes, in my experience, they are worse for a simple technical reason. The increased gap in the magnetic core required for the extra bias current of a pair of tubes, will further reduce the primary inductance de-loading the tubes and increasing the distortion in the lower end of the spectrum; where and how much is OPT dependent, but it will be there. You cannot argue with physics. Tube aficionados like to claim that the distortion is benign, but to generate said THD, the amplifier must have a highly curved transfer function which also generates high IMD in the presence of multiple tones as generated by multiple instruments playing together.

You must be a cynic from way back A9X!!?. But I am interested in getting a better understanding of the innards of what I am now using. This is the first SET amp I have had and not an aficionado of that area. My other valve amps that I have used with the Quad ESLs have been push/pull from 10wpc to 80wpc (and solid state amps).

 

Seeking further clarification from your explanation about the problem with PSE as compared to SE.  

1. How will that effect/flaw that you describe show up in the audio/music output. "blurring" of sound?

2. Can better design to compensate provide a workaround to minimise this.

3. Wouldn't there be a similar issue with push/pull since you are not only dealing with "mismatched" devices, but you have to add the "mismatched" driving circuits in parallel, and the "mismatched" iron that "adds" the parallel signals together?3.

 

thanks

 

Link to comment
Share on other sites

I have between 8W to 500 for my headphones, starting with SE at 8W.
In my case I never get into something that makes the sound audible differently.
Has two amplifiers with a built-in SE with KT88.

But I think it's fun that you hear a difference in your systems when it comes to amplifiers. Get a little applause for your different hearing !!!

But you can not complain that there is no difference, the only thing you can complain about is the cohabitation ?? And no one goes on the Crown best sound amplifier. Ok, it's not necessary because "everyone" can be heard well, just do not need the effect.

At the moment, here is too hot for the tube so it's getting QSC, nothing's crazy, it's only 325W for small headphones.DSCF865822.thumb.jpg.aa1722990d4c3a12a333770a92f64c68.jpg

Link to comment
Share on other sites

13 hours ago, 13000km said:

But I think it's fun that you hear a difference in your systems when it comes to amplifiers.

I think one important fact to keep out in the open is that this isn't a debate on whether there ARE differences or not.    The differences (when present) can be relatively easily quantified.

 

Whether differences are audible will depend on the details of the situation.   Usually yes, but sometimes no.

 

Whether someone likes the differences.... are another much higher level of complexity and subjectivity  (people like all sorts of things, for all sorts of reasons).

Edited by davewantsmoore
  • Like 1
Link to comment
Share on other sites



Going out on a limb here, but output impedance, damping factor, distortion and power output are nothing to do with being SE versus being PP...

 

All these characteristics are down to design and implementation, operating points, transformer quality etc.

 

I have a PP amp.  It is low power (approx 5 watts),  has a relatively high output impedance (due to not running negative feedback), but is low distortion and very clean sounding.

 

Badly designed amplifiers where corners are cut to save costs are a totally different beast, irrespective of topology.

  • Like 1
Link to comment
Share on other sites

4 minutes ago, FR DRew said:

Going out on a limb here, but output impedance, damping factor, distortion and power output are nothing to do with being SE versus being PP...

Yes, you are going out on a limb. A PP amplifier automatically reduces even order distortion, compared to a similar SE amplifier. Output impedance and damping factor are the same thing, expressed differently.

 

4 minutes ago, FR DRew said:

 

All these characteristics are down to design and implementation, operating points, transformer quality etc.

Certainly, design and component quality are important factors. However, if the same attention to quality is paid to a PP amp, as to an SE amp, then the PP amp will offer several tangible, measurable improvements. 

 

4 minutes ago, FR DRew said:

 

I have a PP amp.  It is low power (approx 5 watts),  has a relatively high output impedance (due to not running negative feedback), but is low distortion and very clean sounding.

 

Badly designed amplifiers where corners are cut to save costs are a totally different beast, irrespective of topology.

Of course. 

Link to comment
Share on other sites

3 minutes ago, Zaphod Beeblebrox said:

Output impedance and damping factor are the same thing, expressed differently.

Pardon my ignorance, but is there a relatively simple mathematical equation between them?

Link to comment
Share on other sites

Just now, Ittaku said:

Pardon my ignorance, but is there a relatively simple mathematical equation between them?

Sure is:

 

DF = Zload/Zsource

 

Where Zload is usually 8 Ohms.

 

However, damping factor has been the butt of considerable argument on SNA. And with some justification. For a couple of reasons:

 

1) No speaker presents a constant 8 Ohm load. A typical speaker might exhibit an impedance variation from 3 Ohms to 30 Ohms, depending on frequency and level applied.

2) Damping factor is a confusing figure, which seeks to explain an almost non-existent problem for a PROPERLY DESIGNED loudspeaker system. 

 

OTOH, output impedance is a VERY important figure, which can explain why some amplifiers sound the way they do.

  • Like 1
Link to comment
Share on other sites

7 minutes ago, Zaphod Beeblebrox said:

Sure is:

 

DF = Zload/Zsource

 

Where Zload is usually 8 Ohms.

Great thanks. I was wondering why a lot of amps don't specify their output impedance, preferring to write the damping factor. On a PP valve amp with multiple taps, if it's specified into a particular load, would it be fair to assume the absolute output impedance is similar on the other taps, but obviously the damping factor changes accordingly? I don't imagine output transformers dramatically change the output impedance between taps. So if a DF of 10 into 16ohms is specified, it's only 5 into 8 ohms since the output impedance remains about 1.6ohms?

Link to comment
Share on other sites



1 hour ago, Ittaku said:

I was wondering why a lot of amps don't specify their output impedance, preferring to write the damping factor.

Because it gives a bigger number to impress the uninformed.

 

1 hour ago, Ittaku said:

On a PP valve amp with multiple taps, if it's specified into a particular load, would it be fair to assume the absolute output impedance is similar on the other taps, but obviously the damping factor changes accordingly? I don't imagine output transformers dramatically change the output impedance between taps. So if a DF of 10 into 16ohms is specified, it's only 5 into 8 ohms since the output impedance remains about 1.6ohms?

Not really. The reflected plate to plate load will remain the same, but the impedance of the TRF secondary will be about 1.414x lower on a 4R tap vs an 8R tap improving the DF number slightly. NFB will reduce the difference between them even more as my example assumes no NFB around the secondary of the TRF. I've don't recall ever seeing a DF referenced to anything but 8R.

  • Like 1
Link to comment
Share on other sites

Zaphod,

 

Yes, a PP amplifier cancels out even order distortion (if of course, its phase splitter and output transformer are perfectly balanced, and there are at least 5 topologies for phase splitting all of which have their own pros and cons ). 

 

Of course, a well designed SE amp using linear devices operated properly shouldn't be generating much second harmonic anyway, so the argument is pretty moot, except for on short term peaks where, often, distortion from the speaker is significantly higher than anything being produced by the amplifier anyway.

 

Point is, people should be railing against sloppy design and implementation, not against SE.

Edited by FR DRew
Typo
  • Like 1
Link to comment
Share on other sites

15 minutes ago, FR DRew said:

Zaphod,

 

Yes, a PP amplifier cancels out even order distortion (if of course, its phase splitter and output transformer are perfectly balanced,

The vast majority of PP amps do not use output transformers. Only a minuscule number of amplifiers employ output transformers today. 

 

15 minutes ago, FR DRew said:

 

 

and there are at least 5 topologies for phase splitting all of which have their own pros and cons ). 

No, there are literally thousands of different ways of achieving push pull (SS and valve - I make no distinctions). 

 

15 minutes ago, FR DRew said:

 

Of course, a well designed SE amp using linear devices operated properly shouldn't be generating much second harmonic anyway, so the argument is pretty moot, except for on short term peaks where, often, distortion from the speaker is significantly higher than anything being produced by the amplifier anyway.

 

Point is, people should be railing against sloppy design and implementation, not against SE.

I am not railing against anything. I am just stating fact. Push pull amplifiers exhibit vastly lower levels of even order harmonic distortion. AND, the distortion produced by many loudspeakers is of a completely different type to that produced by amplifiers. I do not accept that just because many loudspeakers (certainly not all, BTW) produce relatively high levels of particular forms of distortion, then that does not mean we should ignore the distortion produced by other parts of the reproduction chain. It is, for instance, generally accepted that, under controlled conditions, a sensitive human ear can detect THD levels in the order of 1% easily and anecdotal evidence exists that some listeners may be able to perceive around 0.1%. Here is a paper published way back in 1957:

 

http://www.r-type.org/articles/art-143.htm

  • Like 2
Link to comment
Share on other sites

Pretty sure that the OP when posting about SE amps vs PP amps was talking valve amps with output trannies.

 

PP amps for a given presence of feedback also exhibit far greater higher order odd order distortion products than most SE amps do (where it is usually confined to a dominant second order distortion product which is often deemed more benign audibly).

Link to comment
Share on other sites

2 minutes ago, FR DRew said:

PP amps for a given presence of feedback also exhibit far greater higher order odd order distortion products than most SE amps do (where it is usually confined to a dominant second order distortion product which is often deemed more benign audibly).

Is this based upon your actual measurements of amps, or upon assumption based upon theory? See my earlier post re this (as I don't agree based upon a lot of actual measurements) and transfer function. Also the lack of primary inductance in an SE OPT re an equivalent PP.

Link to comment
Share on other sites



7 hours ago, Ittaku said:

I was wondering why a lot of amps don't specify their output impedance, preferring to write the damping factor.

Because "it's commonly understood".... which is hilarious as manufacturers do all sorts of "uncommon" things like use the DCR or nominal impedance, or don't quote frequency .....  it's the most commonly misunderstood thing going in audio.

 

7 hours ago, Ittaku said:

So if a DF of 10 into 16ohms is specified, it's only 5 into 8 ohms since the output impedance remains about 1.6ohms?

Correct.

DF is load impedance / source impedance (as a specific frequency).

 

EDIT:  As MLXXX/A9X have both said - your example of constant output Z, might not always be realistic.

Edited by davewantsmoore
  • Like 1
Link to comment
Share on other sites

1 hour ago, A9X said:

I've don't recall ever seeing a DF referenced to anything but 8R.

Mine references 16R which is why I used the example... It also specifies 8.8dB of NFB, for what it's worth.

Edited by Ittaku
Link to comment
Share on other sites

56 minutes ago, FR DRew said:

Pretty sure that the OP when posting about SE amps vs PP amps was talking valve amps with output trannies.

Again: Irrelevant. It doesn't matter what the active devices used are. SE solid state amps exhibit remarkably similar levels of distortion to that of SE valve amps (given similar topologies and feedback systems). The only real difference between solid state and valves, is that solid state devices don't wear out (IE: Their characteristics do not change over time). 

 

Quote

 

PP amps for a given presence of feedback also exhibit far greater higher order odd order distortion products than most SE amps do (where it is usually confined to a dominant second order distortion product which is often deemed more benign audibly).

Not so. For approximately similar topologies, PP exhibits similar levels of odd order distortion and much lower levels of even order distortion. This is worth a read:

 

https://www.passlabs.com/press/audio-distortion-and-feedback

 

Note the graph in fig 4. The blue line is SE and the red line is PP. 

 

And again: It makes no difference if it is valve or SS. The problems and solutions are largely the same. 

fig_4_distortion_in_single-ended_vs_push-pull.png

Edited by Zaphod Beeblebrox
  • Like 1
Link to comment
Share on other sites

1 hour ago, Zaphod Beeblebrox said:

THD levels in the order of 1% easily and anecdotal evidence exists that some listeners may be able to perceive around 0.1%

 

Most SE amps can't make a truckload of power by 1% .... so they come with heaps of caveats.

  • Like 2
Link to comment
Share on other sites



4 hours ago, Ittaku said:

Great thanks. I was wondering why a lot of amps don't specify their output impedance, preferring to write the damping factor. On a PP valve amp with multiple taps, if it's specified into a particular load, would it be fair to assume the absolute output impedance is similar on the other taps, but obviously the damping factor changes accordingly? I don't imagine output transformers dramatically change the output impedance between taps. So if a DF of 10 into 16ohms is specified, it's only 5 into 8 ohms since the output impedance remains about 1.6ohms?

Not quite, as @A9X has outlined in his response above.

 

The outcome you have described would be broadly true of a conventional transformerless solid state design.

 

With a conventional tapped output transformer valve design, the ability to connect to different taps to match the nominal speaker impedance tends to even out the damping factor when changing to different nominal impedance speakers.

 

So, to use your figures, it is actually not true that if a DF of 10 applies for a 16 ohm load a DF of only 5 will apply for an 8 ohm load. This is because when you change to an 8 ohm load you will connect to a tap involving fewer turns of the secondary winding. That connection will provide a much lower effective source impedance than the transformer tap intended for a nominal 16 ohm load.  (The source impedance provided by a loudspeaker transformer secondary winding is very much determined by what is connected to the primary winding, and modified by virtue of the turns ratio.)

  • Like 2
Link to comment
Share on other sites

1 hour ago, davewantsmoore said:

 

Most SE amps can't make a truckload of power by 1% .... so they come with heaps of caveats.

Yeah, the triode versions are generally gutless, heavy, expensive, inefficient, inconvenient, hot, high maintenance and completely unnecessary.  But they can sound pretty good in the right circumstances.

  • Haha 2
Link to comment
Share on other sites

1 hour ago, acg said:

can sound pretty good in the right circumstances

Most modern speaker designs (either wittingly or unwittingly) want an amplifier which "damps" the electromechanical resonances in the speaker.

 

If you don't provide that, then the speaker will sound all "over the place" compared to what the designer intended .... either through a wonky frequency response, or through clipping on hard amplifier loads.   Sometimes this distortion can be serendipity - and either way, "different" can polarise opinions.

Edited by davewantsmoore
Link to comment
Share on other sites

38 minutes ago, FR DRew said:

I guess this means we aren't all chasing convenient, cheap, cool running, efficient, light weight or high powered...

Most are though....  as they have speakers which need "damping" and power

Edited by davewantsmoore
Link to comment
Share on other sites



  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...
To Top