Why would anyone use an AB amp for a tweeter?

[boxerfan88]

Woah, it seems my comment kicked up a little bit of a storm.

While there are valid technical reasons to use a Class AB for a tweeter over a Class D, those that said the reason was "marketing" are certainly not wrong: -

Designing loudspeakers for sale involves assessing the best technical solution, but also the solution that will result in the product being attractive to the most people. Due to the history of the development of the different classes of amplifier, there are implicit biases (or expectations) in the market for each of these types.
Class A has always been as the highest quality, but most reasonable amplifiers were Class AB due to their similar sound quality but much better thermal performance. As others have said, when Class D amplifiers were introduced, they were only for high power/high efficiency applications where sound quality was secondary (or tertiary). Are these "tiers" still applicable today, where we have amplifiers like the AHB2 and LA90 and EIGENTAKT modules? Not really, but in this case, the market technology expectations move slower than the actual technological progress.

Thanks for sharing your perspective and the back story.

Personally I think choosing AB for the tweeter was a good design decision.

 

[tmtomh]

My favorite thing about what KEF says about the LS60:

It is all reasoned in their LS60 Wireless White Paper chapter 3.4.1 Amplification, exemplary from it:

The HF section has very different power statistics. Only
low continuous output is required but the amplifier
must have the ability to deliver sudden large peaks.
[...] Class AB amplifiers have a much
lower efficiency than Class D, but the thermal power
generated is reasonably low under these signal
conditions. Class AB has the added advantage of wide
HF bandwidth and of not requiring an output filter.

These bolded bits are my favorite part of this. We have some members here who routinely extol the virtues of AB over D, are stridently demanding that amps must be able to delivery their max rated power continuously at the highest treble frequencies, and are excoriating Class D amps for not having the thermal capacity to do so.

And here we have KEF explaining that they can use AB amps for the higher frequencies because they don't need to deliver high continuous output at those frequencies, which makes AB possible because AB amps would have.... thermal issues if they had to produce high continuous output.

Love it!

 

[KSTR]

And here we have KEF explaining that they can use AB amps for the higher frequencies because they don't need to deliver high continuous output at those frequencies, which makes AB possible because AB amps would have.... thermal issues if they had to produce high continuous output.

Yep. In an active speaker of the better kind you tend to have a limiter based on thermal modelling in the tweeter path, to thermally protect it. Many tweeters, like smaller AMT's, can't take any more than ~10Watts of continuous dissipation but can happily be driven to 100Watts or even 300W++ for short peaks (300Watts still only means sqrt(30)=5.5 times output voltage which is not overkill in terms of required headroom -- for unprocessed raw signals in the studio at least).
Hence, the amplifier also doesn't need to be thermally overengineered.

 

[torgeirs]

Totally agree on this
Using class A/B the bom can omit the output filter. That money can be used at the transistors and low distortion drivers + passive component instead. And it can be taylored for HF. Input cap can be small for instance
Now the amp can have high rail voltage. And the power supply of the class D amps can be used.
As mentioned high rails and A/B usually requires big heat sinks.
But with low crest factor it is not longer the case.
Guess class D amps that is HF spesific will come, but a class A/B one is much simpler to make with good spec for HF. (If not heatmanagement is the issue anymore)
To paraphrase Reagan: It`s the bandwith,stupid ;-)

 

[somebodyelse]

Guess class D amps that is HF spesific will come

Like the NC100HF?

 

[PGansz]

It's also interesting to note how SPL limited this model is, and how frequently it goes on sale for 33% off.

Marketing and price points were definitely at play.

Question for the group - I seem to keep correlating ultra high damping factors to perceived "thin" sound. Is this a thing? If so, maybe also explains the decision a bit?

 

[dfuller]

We've really only seen good, high fidelity class D amps come around in the last decade or so. Prior to that they were... not good. Filter:driver interactions causing FR weirdness, high HF distortion, intermodulation from the (lowish) switching frequency into the audible range, etc etc were all valid concerns and why it wasn't (and isn't) uncommon to see AB for HF and D for LF. Can you get good Class D? Sure thing. But that's a recent thing, especially at an affordable (for manufacturer BOM levels) pricetag. FWIW: I'd keep in mind that Neumann use TDA7293s - $5 or less per, in large quantities - in their older models (120A, 310, 420) and similar cost Class D chip amps are (or at least were) for the most part pretty horrid. 7293s on the other hand are really quite clean.

 

[Marc v E]

It's also interesting to note how SPL limited this model is, and how frequently it goes on sale for 33% off.

Marketing and price points were definitely at play.

Question for the group - I seem to keep correlating ultra high damping factors to perceived "thin" sound. Is this a thing? If so, maybe also explains the decision a bit?

Well....I just heard them 1 hour ago.

I would rate them among the best, regardless of price. Spl for music within a reasonably sized room, say 5 x 5 meters should be no problem. I think my ears sooner give up than the speakers do. Edm sounds fine on the ls60.

For thx level explosions I would recommend a suitable sub, though.

Regarding thin perceived sound imo and ime, is more to do with a neutral, revealing speaker. Our ears somehow correlate distortion with loud sound and more bass. Without distortion it can get ear deafeningly loud before I realise it.

And what struck me too, is that poor recordings can give me a perception of thin, bland sound. Genre or content doesn't matter.

 

[Purité Audio]

Well....I just heard them 1 hour ago.

I would rate them among the best, regardless of price. Spl for music within a reasonably sized room, say 5 x 5 meters should be no problem. I think my ears sooner give up than the speakers do. Edm sounds fine on the ls60.

For thx level explosions I would recommend a suitable sub, though.

Regarding thin perceived sound imo and ime, is more to do with a neutral, revealing speaker. Our ears somehow correlate distortion with loud sound and more bass. Without distortion it can get ear deafeningly loud before I realise it.

And what struck me too, is that poor recordings can give me a perception of thin, bland sound. Genre or content doesn't matter.

If you can I would add sub/s , and of course they are easily integrated via the KEF app,
Keith

 

[torgeirs]

Like the NC100HF?

Yes function wise. Ncore are super high gain bandwith analog design. Then it doesn’t matter if it is D or A/B
So thinking more of a integrated chip design like TI or Merus.
That is designed for 50k bandwith and 40 dB global feedback. That is designed to work with post filter feedback at HF

 

[torgeirs]

In the other end of the spectrum the post filter feedback design of cheap 200 watt class D designs are excellent. I have just bought one for 20 to 200 hz duty.
Abundant gain bandwith for 200 hz duty. Actually also nearly up to 3k
A little irritating they design them to start the harmonic rise at 1 kHz and not 3 kHz

 

[torgeirs]

Active design amps should be noise and distortion optimized for 3 bandwiths
20-300 hz
100 - 3 kHz
500 to 50 kHz