advertisement


Class D Amps - will they be the future.

Have been thinking about Class D amplification. Why for all its qualities and eco friendliness has it not become the norm. Since were all moving closer and closer to a full on digital invasion is there any reason why we stick to analogue, is class D just not there yet or is it just a technology that doesn’t work as well. ..... Are there any British made units?

.. so is Class D not used because were so entwined in the industry, its production process and its current form that a wholesale change would effectively re-invent it? or is Class D just in its infant days and as of yet, needs to find its feet?

Hi r-tee,

Ignoring the "digital vs analogue" debate for the moment, I think there are a number of purely business reasons why Class D has not become "universal"...

In the majority of cases, the "technology" of Class D is "bought-in" rather then developed in-house. Examples of hifi manufacturers that have adopted this "bought-in" approach using B&O's ICEpower modules are:

  • Bel Canto
  • Jeff Rowland Design Group
  • PS Audio
  • Rotel
  • Wyred-for-Sound

For these manufacturers there has to have been a conscious "strategic" decision taken to reduce in-house R&D into amplifier design and to adopt third party technologies. Most of these manufacturers do some bespoke peripheral tweaking to the overall circuitry - leaving the modules intact - in order to achieve some improvement over the basic module performance.

Other manufacturers may well have a view that their past R&D investment has yet to yield an adequate return for them to simply abandon their in-house (and analogue) designs in favour of "bought-in" digital designs. Also, some manufacturers have built a reputation around their own proprietary circuit designs - Krell and Pass Labs spring to mind - and are unwilling to throw away that element of "differentiation" that sets them apart from the herd.

So, in this scenario we have "financial inertia" coming into play coupled with the desire to easily differentiate offerings from the masses.

Another factor lies in the gross profit margins possible when using in-house designs versus using licensed designs from third parties. This is probably most valid in the 50-100 wpc integrated amplifier segment where competition is fierce and pricing is dictated by positioning rather than branding. This segment is also a popular target segment for UK amplifier manufacturers such as Arcam, Creek, etc. For these manufacturers to suddenly switch from their own analogue designs that have evolved over time, to a digital technology sourced from a third party, would pose some problems due to write-off of R&D (as mentioned above), loss of "house sound" and profit margin erosion due to having to pay for someone else's R&D.

To my knowledge, there are very few British amplifier manufacturers employing third party digital amplifier technology (Meridian is one that comes to mind).

There are probably other business-related issues, but these two should be sufficient to illustrate the point that there are factors other than sound quality that influence the adoption of digital amplifier technology.

On the issue of sound quality, there appears to be a polarisation of opinion on this forum (and many others) as to the sonic merits of digital amplifiers - with avid supporters on one hand and vicious detractors on the other hand (and with very few in the middle...).

Personally, I am very impressed with the ICEpower technology as implemented by Jeff Rowland and, after >3 years of living with a pair of 250 watt (8ohm load) monoblocks, would not consider a move back to a non-digital offering.

So, I honestly believe that digital amplification - as exemplified by B&O's ICEpower - has "come of age", but universal adoption is probably a very long way off (and for a number of reasons - not the least of which is the polarisation of opinion in the potential markets).

My $0.02 worth...
 
Hi r-tee,

Ignoring the "digital vs analogue" debate for the moment, I think there are a number of purely business reasons why Class D has not become "universal"...

In the majority of cases, the "technology" of Class D is "bought-in" rather then developed in-house. Examples of hifi manufacturers that have adopted this "bought-in" approach using B&O's ICEpower modules are:

  • Bel Canto
  • Jeff Rowland Design Group
  • PS Audio
  • Rotel
  • Wyred-for-Sound

For these manufacturers there has to have been a conscious "strategic" decision taken to reduce in-house R&D into amplifier design and to adopt third party technologies. Most of these manufacturers do some bespoke peripheral tweaking to the overall circuitry - leaving the modules intact - in order to achieve some improvement over the basic module performance.

Other manufacturers may well have a view that their past R&D investment has yet to yield an adequate return for them to simply abandon their in-house (and analogue) designs in favour of "bought-in" digital designs. Also, some manufacturers have built a reputation around their own proprietary circuit designs - Krell and Pass Labs spring to mind - and are unwilling to throw away that element of "differentiation" that sets them apart from the herd.

So, in this scenario we have "financial inertia" coming into play coupled with the desire to easily differentiate offerings from the masses.

Another factor lies in the gross profit margins possible when using in-house designs versus using licensed designs from third parties. This is probably most valid in the 50-100 wpc integrated amplifier segment where competition is fierce and pricing is dictated by positioning rather than branding. This segment is also a popular target segment for UK amplifier manufacturers such as Arcam, Creek, etc. For these manufacturers to suddenly switch from their own analogue designs that have evolved over time, to a digital technology sourced from a third party, would pose some problems due to write-off of R&D (as mentioned above), loss of "house sound" and profit margin erosion due to having to pay for someone else's R&D.

To my knowledge, there are very few British amplifier manufacturers employing third party digital amplifier technology (Meridian is one that comes to mind).

There are probably other business-related issues, but these two should be sufficient to illustrate the point that there are factors other than sound quality that influence the adoption of digital amplifier technology.

On the issue of sound quality, there appears to be a polarisation of opinion on this forum (and many others) as to the sonic merits of digital amplifiers - with avid supporters on one hand and vicious detractors on the other hand (and with very few in the middle...).

Personally, I am very impressed with the ICEpower technology as implemented by Jeff Rowland and, after >3 years of living with a pair of 250 watt (8ohm load) monoblocks, would not consider a move back to a non-digital offering.

So, I honestly believe that digital amplification - as exemplified by B&O's ICEpower - has "come of age", but universal adoption is probably a very long way off (and for a number of reasons - not the least of which is the polarisation of opinion in the potential markets).

My $0.02 worth...

Excellent post Sir and sums it up very well... Interesting thread and a very civil one too... I'm not sure if they're still using the technology, as I think they were either in financial difficulty or were taken over, but certainly Tripath produced Class D chipsets used by non other than Naim in the NVi. Whether they still do or not, I don't know. I must away to my bed, but will catch up with this tomorrow...nite all.... :)

Jon..
 
I used to own a Lyngdorf SDAI 2175 and it was a fine integrated amplifier but lacked emotion and was a bit boring to listen to for longer periods of time IMO. The sound was rather flat but yet powerful. The bas certainly wasn't lacking and it had a really good grip on my ribbon speakers. I've also read that a lot of people prefer ICE Power/PWM amps in their subwoofers. But when it comes to their main systems they rather want solid state or tube amps.
I believe that digital amps are the right step forward and they have great potential if used and implemented right. Plus it's "green".

The SDAI 2175 isn't fully digital (it's semi digital) and it doesn't use ICE Power technology. The TDAI 2200 is fully digital, the same with Millennium (which costs around £9-10.000 for the newest MK IV).
 
yes I agree very good posts.... so is full digital amplification along with full solid state storage going to be the ultimate setup in the very near future. No moving parts, low running costs, simple build, no or little heat. Ultimately creating a sound beyond the conventions of todays digital to analogue and analogue to digital systems?
 
There are pros and cons of the so-called full digital approach; using a DAC and a circuit like say the Hypex one may turn out to be better in the long run; I think the jury is out on this one.

There are some hybrid approaches as well; for example the pro-amps made by LabGruppen use an analogue pathway surrounded by a class D circuit to give near class D efficiency, with analogue precision.

To quote Yogi Berra, in turn quoting Neils Bohr:“It’s tough to make predictions, especially about the future.”

Quite why an American baseball coach was quoting a Danish quantum physicist is beyond me.
 
I have the Acoustic Reality version of the largist ICE one. I bought the ar's when sterling was strong against the euro .In fact they sounded sh1t out of the box but got better.I ended buying a isotek titan ( on a whim whilst drunk) to run them. This really helped and now i'm very happy with them.
 
I was never that keen on the ICE amps, to me they lacked any life, in a back to back comparison to the Lyngdorf gear, im not sure why though.

I rather like the Lyngdorf/Tact stuff with the right speakers it sounded fantastic (ie large full range designs or 2+2)

Its a shame Lyngdorf didnt carry on development and release the monoblocks and all the other stuff. It all ended up in the Steinway System it seems (which pissed a lot of Lyngdorf owners off)
 
I always thought the reason class D wasn't more widely used was becuase it still suffers from higher distortion than class A/B designs?

The way I understood it was:

Class A/B = low distortion but also low efficiency

Class D = high efficiency but also high distortion

Lefty
 
Never heard any lyngdorf kit but I've heard the steinway piano. That was true analogue :)
lefty, I thought it was the other way Round?

I was Reading last months HFC with the lyngdorf millenium IV review which is what prompted this thread
 
No, class D can have very low distortion, especially at low frequencies.

It is hard to design and make.

There are technical problems; depending on which circuit is used you can have:
1) Ultrasonic switching noise radiated/conducted down speaker cables, may cause interference
2) Output impedance rises at high audio frequencies, so that response is not flat into low impedance loads
3) Some designs have distortion problems at higher frequencies
4) Some design are fragile, and fail if the output is shorted
5) Bandwidth is limited; if you want to use your ribbon tweeter to send signals to horseshoe bats at 110kHz, few off the shelf class D amps are any use. Opinions differ if this matters for audio!

3 and 4 are solvable.

1 is a bigger problem, but good designs manage to control output switching noise fairly well.

2 is a problem for some designs (say ICEpower) but not for others (Hypex)

If 5 is important is a matter of opinion.
 
I always thought the reason class D wasn't more widely used was becuase it still suffers from higher distortion than class A/B designs?

The way I understood it was:

Class A/B = low distortion but also low efficiency

Class D = high efficiency but also high distortion

Lefty

Lefty have you ever heard a Lyngdorf Millennium or TDA 2200? Distort they do not, their not perfect mind, getting the right speakers is paramount.

Have a look at the french fellas system a little bit down the page, id have loved to have tried the lyngdorf gear with horns, but i dont have the space.

http://www.stereotimes.com/showreportces05page2.shtml
 
It is hard to design and make.


5) Bandwidth is limited; if you want to use your ribbon tweeter to send signals to horseshoe bats at 110kHz, few off the shelf class D amps are any use. Opinions differ if this matters for audio!

Yes, they are hard to make, ours took over 3 years to get them to the standard they are now at and the designer spent many a night burning the midnight oil.

Ours have very good bandwidth, he felt it was important and it's been worthwhile.
 
I was never that keen on the ICE amps, to me they lacked any life, in a back to back comparison to the Lyngdorf gear, im not sure why though.

As johnhunt wrote:

In fact they sounded sh1t out of the box but got better.I ended buying a isotek titan ( on a whim whilst drunk) to run them. This really helped and now i'm very happy with them.

I found the same phenomenon when first installing the Jeff Rowland Model 201s - it took some 200+ hours before they "came on song". Initially, I was somewhat underwhelmed and had begun to doubt the wisdom of my selection, when - out of the blue - everything just clicked into place and they sounded great.

Like johnhunt, I added a mains filter at a later stage and this made a significant difference - lowering the noise floor and allowing more low level detail to peep through. The area where this was most apparent was in reproducing spatial cues which had, previously, been buried in the noise.

So muz, I would hazard a guess that your audition must have been to new units not powered via a mains filter. If the opportunity presents itself, you should try to audition some ICEpower amps that are well run-in and run via a mains filter - I think you'll be pleasantly surprised.

Lefty wrote:

I always thought the reason class D wasn't more widely used was becuase it still suffers from higher distortion than class A/B designs?

I would honestly like to know where this belief originated. If distortion figures are to be the yardstick by which amplifiers are judged, then all valve amp afficionados better ditch them quick! :D

The only disadvantage of owning Class D amps is during a cold winter when they don't generate enough heat to warm the building... ;)

Apart from that minor quibble, they are actually rather fine!!!
:cool:
 
I think after you've lived with class D for a while you realise something is missing, they are somehow just to clean, maybe it is the rolled off frequency response, when you go back to normal SS amps you miss what the class D does!
 
Depends on the module used.

For Hypex, it is about 50KHz, largely independent of load impedance.

For ICEpower, it varies - into 8Ohms it is about 80kHz, and falls in proportion as the load impedance falls - probably not ideal for Quad ESL57s!

These two have much of the class D market between them.
 
So muz, I would hazard a guess that your audition must have been to new units not powered via a mains filter. If the opportunity presents itself, you should try to audition some ICEpower amps that are well run-in and run via a mains filter - I think you'll be pleasantly surprised.


Well I only had a few weeks with them, they were NHT power 2's made by a company in Finland. They were connected to a dedicated mains spur.
Im sure all ICE amps dont sound the same, however i preferred the older Bel cantos (Tripath) to the newer ICE ones so maybe its just me.
 
There are technical problems; depending on which circuit is used you can have:
1...
To which I'd add a subtle issue: some Class D amps can end up with very high DC potentials available on the loudspeaker terminals due to the way they work - up to 80v above ground at idle* has been reported. I'm not sure I'd want such a thing in a domestic situation with small children around without more care taken than is usually the case.
 


advertisement


Back
Top