Active Domestic Speakers

[honmanm]

Why are they praises for amps like naim and rega which are full of harmonics. Ditto most tube amps, some of which measure terribly esp. Prima Luna’s. Am I missing something here?

You might be missing the fact that harmonic distortion in electromechanical transducers (speakers, cartridges, analogue tape records) is an order of magnitude higher than that of a reasonably well designed amplifier. Somewhere A.S. posted some scary numbers on the performance of drivers, and links to a comparative study of the tape recorders which were used in the pre-digital recording era.

So as long as distortion from the amplifier is (apparently) less than about 1% it get buried in all the other distortion - and apparently our ears only object when distortion gets to something like 3% THD.

Sorry to be so anecdotal about this - just wanted to make a quick note as we are now going really OT. Alan's comments about distortion (which I've almost certainly misrepresented) should be in other threads of the "ask the designer" section - though the site's search facility is either broken or I'm struggling to drive it!

BTW have you had the chance to compare the Yamaha amp against the low-end Naims and Regas?

In the UK, Japanese amps are under-rated and are often a good second-hand option. However thanks to a Naim-nut friend I've had plenty of opportunities to listen to Naim gear and while I don't understand the Love of Naim I can't say that the amps ever sounded offensive (BTW a lot of the Naim "house sound" comes from their use of tantalum capacitors in certain parts of the amplifier circuit - build the same amp with different parts and the sound is more conventional).

[kittykat]

Hi honmanm

You have a good point, which if I’ve read correctly, means that speakers are the weakest link and i totally agree with you. But this does not detract from the fact that there are by varying degrees differences between amplifiers on a test bench. Even though I always say “there are no differences between amps”, it’ll probably be more accurate to say “there are few perceivable differences between decent amps”.

My belief, and it’d be great to have a discussion on, is that the sum of parts of design (and choice of parts like you mentioned) does contribute to a sonic signature (which would invariably be reflected in measurements). I have heard the Naims in isolation and thought I heard they sounded “different”. They sound thicker, richer and has a “driving” beat. Im sure this is further from what it was meant to be, than say an amp which measured “cleaner”. Whether all this comes out in a blind comparison is another issue.

Having said all this, i am convinced I am not hearing the SHL5’s as it was meant to be (or to its full potential), cause ive fallen foul, and currently using a KT88 amp. It has its limitations and weaknesses.

[EricW]

As far as I know, there are two main theoretical advantages to an active speaker:

1) improved efficiency, and
2) improved sound quality

Now, I would think #1 is definitely important if you're doing sound reinforcement (i.e. operating a PA system), perhaps important if you're running at tremendous loudness levels in a recording studio. In a normal domestic environment, I would think it's a complete non-issue, given that any decent combination of amp and passive speaker will likely play, with ease, as louder as or louder than you want to or should be playing anyway.

That leaves #2. In theory, I imagine there's potential for better sound, if the system is well designed, and all the pieces truly optimized to work with each other. But just because something's potentially better doesn't mean that that potential is realized, or even likely to be realized, and your system is now far more finicky and less flexible.

The best systems I've heard (with passive speakers) are so shockingly good (M40.1 included in that number) that I really can't imagine wanting anything better at home. I suspect there's far more to be gained by optimizing the room environment (and perhaps the source) than worrying about going completely active.

[Labarum]

An active loudspeaker has the crossover at line level - before the power amplifiers.

Line level crossovers introduce less distortion, especially around the critical crossover frequencies.

The power amplifiers are immediately connected to the drivers, which gives them better control.

The amp and speakers can be designed of a piece, and with good engineers that offers advantage.

But, comparing the best passives with available domestic actives, who knows? Has anyone done a survey?

Distortion seem to be

1. Driver distortions.

2. Crossover distortions.

3. Enclosures distortions (resonances)

4. Room effects.

How are these distortions to be ranked? If the crossover distortion is small compared to the others then clearly it matters not.

[hifi_dave]

I've not done a survey but I've certainly had a lot of active speakers through my hands over the years and can't really see the attraction other than the theoretical.

Early days we had the Likes of Linn, Linn/Naim, Arc/Naim, Arc/Nytech combos and yes, active did always sound better than passive but that was because the amplifier count went up significantly and/or was entirely different.

Later days I have had demo speakers which run passive or active by plugging in the mains. With these, we could always get the passive sounding better than active by using better amps to those internal to the speakers. Where the actives scored was that they were good value with the amps on board and were very robust.

See also post no.2

[Thanos]

Hi,
I'll have to fully agree with my friend Dave, I think he knows very well the subject, after all he's a dealer, having seen/heard hundreds of examples. And, some more things I believe we should severly reconsider:
It is very hard, costly and highly inconvenient to repair, fix, maintain, move, accomodate (need double extra A/C sockets) -say it as you wish- two boxes with electronics included. And much more difficult to resale and update if decided... Not to mention also that ageing is usually different between electronics and speakers, so I am restricted with posible replacements.
Just some extra thoughts,
Regards,
Thanos

[DSRANCE]

There's good and bad in all forms of speaker implentation in my opinion. Active should be better if the amps are properly designed for the drivers they're used with (and built in), and the amps don't have to be exposed to the damping and filling in an enclosure either (bigger ATC actives have the amps in a partitioned off section of the cabinet).

I read elsewhere tonight that there's no such thing as cabinet induced colouration, as John Bowers (B&W) supposedly, in the 70's, researched using an accelerometer and found that unless you use cardboard or similar, the box resonances are 60db below the music signal and, therefore, lower than the driver or crossover distortions. I don't know about this, but I know what kind of box, ported or otherwise, I enjoy listening to...


It's all too much. I reckon I need a holiday :)

[JoeHutch]

Whatever the design theory, surely the only thing that matters to the end user is whether we like way the speakers look and the sound they make. Few things annoy me more than being told that if I don't like the sound of a system it's because my ears need 'educating' to appreciate the benefits of a supposedly superior component or replay system.

[DSRANCE]

Few things annoy me more than being told that if I don't like the sound of a system it's because my ears need 'educating' to appreciate the benefits of a supposedly superior component or replay system.

I couldn't agree more and I very sincerely apologise for forcing this kind of rubbish on people all those years ago when I didn't know better.... Actually, I don't think I was the worst out there either........

[markus sauer]

I read elsewhere tonight that there's no such thing as cabinet induced colouration, as John Bowers (B&W) supposedly, in the 70's, researched using an accelerometer and found that unless you use cardboard or similar, the box resonances are 60db below the music signal and, therefore, lower than the driver or crossover distortions.

That other site doesn't seem to be a reliable source of information. To put it as mildly as possible.

[Labarum]

I must admit, Dave, to hear that said in the other place left me utterly amazed.

So cabinet distortions are below the threshold of audibility, but passive crossover distortions are not?

mmm

(Emailing now from Cyprus hanging out of a window pinching broadband from a neighbour!
Anyone want a Tangerine? If I leaned a bit further I could pick one from the tree!)

[Stephen PG]

That other site doesn't seem to be a reliable source of information. To put it as mildly as possible.

No kidding?

Perhaps if a little honesty had been employed and the place clearly marked as a manufacturer forum from the start, but trying to pass it off as an 'independent' forum doesn't do a great deal for the integrity of those concerned.

[honmanm]

I read elsewhere tonight that there's no such thing as cabinet induced colouration... the box resonances are 60db below the music signal and, therefore, lower than the driver or crossover distortions.

Box resonances are measureable - have a look at the last plot (cumulative spectral decay) on this page for example.

One could quibble over the accuracy of the measurements, but they do tend to correlate with audible colourations. Compare the LS3/5a measurement to the P3 chart in the previous link.

Do Harbeth do their own measurements of this nature and would they be willing to share any?

[Dougal]

I'd be interested to know if anyone has found out any more details about this - did John Bowers come to the right conclusion?

{Moderator's comment: please remind us .... what was his question?}

[Dougal]

Sorry, to clarify, I'm picking up on DSRANCE's comment:

"I read elsewhere tonight that there's no such thing as cabinet induced colouration, as John Bowers (B&W) supposedly, in the 70's, researched using an accelerometer and found that unless you use cardboard or similar, the box resonances are 60db below the music signal and, therefore, lower than the driver or crossover distortions. I don't know about this, but I know what kind of box, ported or otherwise, I enjoy listening to..."

{Moderator's comments: I think Harwood's papers on this are the definitive word.}

[DSRANCE]

I'm sure Dudley's papers say something different, but I was originally being a little sarcastic perhaps? High Q resonances are definitely what we don't want, as I understand from experience, but the "BBC" way of tuning the main panel noises as far away from the midband as possible is the way I keep returning to and I find it preferable to my ears.

The thing for me about active speakers is that you can't just take any old speaker, sling out the passive crossover and replace it with an active amp pack. In my view, the drive units and cabinets must be sorted out first and *this* is why I'd love to see some active smaller Harbeth models in years to come, as they work so well in passive form.

[audisp]

I read elsewhere tonight that there's no such thing as cabinet induced colouration, as John Bowers (B&W) supposedly, in the 70's, researched using an accelerometer and found that unless you use cardboard or similar, the box resonances are 60db below the music signal and, therefore, lower than the driver or crossover distortions.

Don't know about that quote but B&W braces their cabinet using their "Matrix" precisely to reduce cabinet-induced colouration. So I doubt it's accurate?

{Moderator's comment: Doubtless JB was quoting Dudley Harwood's BBC research work which we have covered here before and is detailed in the BBC Research Dept. LS3/5a design report (as an example). You can extend the thinking to say that any signal buried 60dB under the wanted signal is inaudible I guess which makes a mockery of chasing 100dB+ s/n ratios in electronics etc..}