The Truth about Modern Class D


All my amps right now are Class D. ICEpower in the living room, and NAD D 3020 in the bedroom.

I’ve had several audiophiles come to my home and not one has ever said "Oh, that sounds like Class D."

Having said this, if I could afford them AND had the room, I’d be tempted to switch for a pair of Ayre monoblocks or Conrad Johnson Premiere 12s and very little else.

I’m not religious about Class D. They sound great for me, low power, easy to hide, but if a lot of cash and the need to upgrade ever hits me, I could be persuaded.

The point: Good modern Class D amps just sound like really good amplifiers, with the usual speaker/source matching issues.

You don’t have to go that route, but it’s time we shrugged off the myths and descriptions of Class D that come right out of the 1980’s.
erik_squires
 @abasia 
good question, and one I have asked earlier when I was thinking the same thing ...

to to avoid some unnecessary quibbles, we probably should have established some parameters for Class D in terms of quality r/t:
1. Price of the class d amp, and 
2: perhaps overall system cost, assuming some reasonable sense of equality (price, performance) b/w the main components in the system 

Examples:
1.  E.g., for the price of the Class D amp alone, at what Price point would the GENERAL price break be?  b/w a good Class D and one that the unsatisfactory Class D amp. 
$1000 ?   $2000? $3k? And so on and so forth. 
E.g., “even”George has said he finds Class D acceptable at the 1.5 MHz switching speed. How much are those SE-1s?

—> Same question for everyone else. There will be a variety of price points given,, reflecting individual tastes. 

B/c right now both sides are making blanket statements about Class D. And both are wrong. But both are right. It’s the specifics guys. Cuz right now we are both just hitting our heads against the wall. 

In the words of a prior college professor, “it’s the specifics stupid”.
guidocorona
I remember no more than one year ago you advocating that the day of class D would come once designers adopted Gallium Nitrite transistors switching at 1.5Mhz....

1graber2
E.g., “even”George has said he finds Class D acceptable at the 1.5 MHz switching speed. How much are those SE-1s?



What I said was G, they can’t yet because the mainstream manufacturers (Motorola, Hitachi, Fairchild STMicroelectronics Texas Instruments ect ect) haven’t started manufacturing them yet for maybe $2 each, because GAN is invented and made by a small (relative to the above) development company (EPC-Co) , they on sell the licence to major manufacturers to make them.
(EPC-Co) also invented the power Mosfet in the 60’s which the major manufacturers use under licence from them to make. Yes Technics has the GAN probably direct from EPC in limited supply, and you can bet they’re not $2 each, hence the $30k price tag on the SE-R1, anlong with limited tooling costs.

I’ve always stated that the 1.5gHz switching frequency used in $30k Technics SE-R1 was a massive stepping stone in the right direction, instead of the 400-600mHz used today by others. Because at 1.5gHz the low order output filter can remove "almost" all the switching noise without any effects down to around 5khz into the audio band which happens with all other Class-D’s now.

As Cyril Hammer of Soulution amps also says, we need to get even higher, he’s at 5gHz I’ve always thought 3gHz is fine, and when I see that on the way with future technology, it’s time for me to start advertising my boat anchors.

Cheer George

George,

You are so yesterday.....he he. Check this link out.....no, they are not mosfets as Mouser has no category for them at present.....download the data sheet.....these are $5.50 each and anyone can buy them when they arrive in Nov. They also have higher current ones for $8 each due in Oct.

https://www.mouser.com/ProductDetail/GaN-Systems/GS61004B-E01-MR?qs=sGAEpiMZZMvplms98TlKYzJ7xUKby2GKgG%2funvqgEAPyX8Mt10R2gA%3d%3d

What so funny is that you only need a couple of these per channel to make multi-hundred watt amps. These are only slightly more expensive than the mosfets everyone is using now.

The revolution is happening. Get rid of those heavy boat anchors before they are worthless!

Class D Output filters are load dependent. From TI:
The LC filter response also varies with speaker load impedance. The load impedance determines the damping ratio of the output LC filter and is classified as overdamped, critically damped, or underdamped. It is also important to understand the speaker load impedance variations for the application and select the L and C values that suit the expected load variations. Ideally, the LC filter value is selected for a critically damped, flat passband, and phase response. Two considerations when selecting components for the second-order low-pass filter is the cutoff frequency and Q factor or damping ratio.

Like any other amplifier, Class D will interact with non-linear speaker impedances. There are multiple output filter topologies and each will interact differently with different loads.

Stating that 1.5GHz will solve all the problems is as silly as saying a 10MHz Class AB bandwidth has to sound better than 100KHz.

As far as GaN devices, there have been countless ’type’ as the new sliced bread for the entire history of electronics, be it tube, transformer, transistor, FET, IC, etc.

ALL electronics involve a compromise and ALL electronics interact with their source and load.

Specific recommendations border on risible.

You might have a point George.... I just did a quick Google search and came up with this article....


https://phys.org/news/2018-07-path-high-performance-transistors.html

If indeed Gallium Nitrite transistors may be applied to push into the Ghz switching frequencies for 5G transmitters, they will need to be eventually supplied by mainstream/major fabs... Perhaps we will see more high speed GaN transistors before long.


BTW, I did not know that Soulution is already using GaN technology.


G.