The single-ended Nature of things

[MBK]

I'd like to extend Nelson Pass's thoughts about the single ended nature of things such as air, loudspeaker cones and SE amplifiers. I have no answers, just questions.

Question one:

A speaker cone has 2 sides, front and back. The manufacturer likely optimized the front for optimal transduction, and especially for the conventional "positive" movement, i.e. pressure-producing, outwards. Likely the backstroke works less optimally. In a box then we have asymmetric operation with 2 transduction modes (in-out). The rear wave also works differently for in and for out movements, and has likely not been optimized, i.e. it works differently from the front. In a box speaker this likely matters less, but in a dipole speaker then we have asymmetric operation with 4 transduction modes: Front and back in and out.

Now: with this in mind, when we use a single-ended amplifier, should we use it in "correct" or in inverted polarity vs. the speaker's polarity? Meaning, "optimized" side amp to "optimized" side speaker, or inverted? In the first case, "conventional" connection, the speaker will be most "truthful" as a transducer on the out stroke, but combine both speaker and amp nonlinearities at the in stroke, mulitplying the nonlinearities. In the second case we combine the "good" side of the amp with the "bad" side of the speaker. So we never get the optimal transduction condition with the given materials, however, we get a more even condition by having a lesser difference in nonlinearities between in and out strokes.

I just tried this at home:

Using my Mini-Aleph prototype I compared correct polarity at pre and at driver, with inverted polarity at pre and at driver (so that absolute polarity stays the same). It seems to me that the "conventionnally correct" way (amp pos. out to driver pos. in) sounds a bit more dynamic, but a bit less pleasant than the reversed connection.


Thoughts?

[jh6you]

In addition, what about the single ended ears...?
Could we manage to make non-inverted or inverted ears...?

Anyway, I have tried the same with my Zen. I have hardly noticed any different sound. In my opinion, the sound wave is the same between the non-inverted and the inverted, with a merely half frequency misalignment.

[MBK]

Originally posted by jh6you

In my opinion, the sound wave is the same between the non-inverted and the inverted, with a merely half frequency misalignment.

Yes... but the distortion of both speaker cone and SE amps is different between the plus sides and the minus sides of center / ground, although surely in a different way. So, should we connect the more ideal sides of both speaker and amp, or should we connect ideal to non-ideal...?

[AudioFreak]

Ultimately, you are going to want to preserve absolute phase. It's been a well known scientifically proven fact for quite some time now that absolute phase is audible.

jh6you,
Air is single ended by nature and so are our ears. The very nature of that frequency dependant time delay results in the air thru which sound travels being more greatly compressed when it should be less greatly compressed and this also occurs in our ears. Due to the single ended nature of both air and our ears, the resultant nerve impulses delivered to our brain will be quite different (not just inverted) between correct absolute phase and when absolute phase is inverted.

Try listening to a signal with a fast attack / slow decay with absolute phase correct and inverted. It shouldn't take long to hear the difference.

[MBK]

Yes, I am talking about preserving absolute phase, by switching phase on both input and output of the amp; easy to do with balanced inputs, requires an inverting buffer for unbalanced inputs. But, the linearity of a SE amp's plus side is different from the linearity on the minus side, just as a speaker cone's linearity is different between its in and out strokes.

So, should 'plus amp out' go to 'plus speaker in', or 'plus amp out' go to 'minus speaker in' (to preserve absolute polarity one must reverse polarity of inputs in the second case).

[Bernhard]

What about classA single ended speakers ?

I think for some time about speakers that have a suspension which will pull the cone to the negative elongation limit and a voicecoil current that pushes the cone to the zero point @ idle.

So the classA idle current has to flow through the vc.

There will be no hysteresis caused nonlinearities of the suspension around zero no more.

By the way, plasma tweeters are single ended, the flame breathes around a medium size.

[MBK]

Bernhard,

Das noch dazu!

I thought of the same thing, though that means the coil has to have the properties and size of the typical class A heatsink. Urrrgh.

[jh6you]

Originally posted by AudioFreak

Due to the single ended nature of both air and our ears, the resultant nerve impulses delivered to our brain will be quite different (not just inverted) between correct absolute phase and when absolute phase is inverted.

Very intersting subject.

When I have different phases between left and right spekers, I immediately hear it as everyone does. Otherwise, if they are in the same phase, it seems for me no big matter of inverting or non-inverting between the SE amp and spekers even if I well understand the point of the greatly summed-up compressed air you mentioned.

I will re-try test of my high-end ears but stupido brain.



PS. Geeeee... I have no SE amp here in Shanghai...

[Bernhard]

Originally posted by MBK
I thought of the same thing, though that means the coil has to have the properties and size of the typical class A heatsink. Urrrgh.

Oh, I forgot something...

It should be a system with moving magnet, not moving coil...

[Jay]

Using my Mini-Aleph prototype I compared correct polarity at pre and at driver, with inverted polarity at pre and at driver (so that absolute polarity stays the same). It seems to me that the "conventionnally correct" way (amp pos. out to driver pos. in) sounds a bit more dynamic, but a bit less pleasant than the reversed connection.

Thoughts?

I once questioned this thing but got no answer. My asumption was that I wouldn't be able to hear any difference so I didn't bother to try. I don't think that this is so dificult to answer by Physics but may be no-one has enough curiousity to find out the answer. Well, at least some of you had shed me some lights.

MBK, your clear ability to spot such a difference performance made me think. Not hard ().
1. Could it be the crossover that exagerated the distortion?
2. Alephs do not have output capacitors. I don't believe that several mV DC cannot appear in the output. Could it have an impact on this?

So, should we connect the more ideal sides of both speaker and amp, or should we connect ideal to non-ideal...?

I think no speaker has been designed taking into account this phenomena. So, IMO we must try both and choose the best one for any given set of amp-speaker. So will I do that? Not yet... (**waiting for the experts to clarify**)

PS. Geeeee... I have no SE amp here in Shanghai...

Whhaatt!?