Digital Amps

[Tomorrow]

Can someone please summarize the benefits and disadvantages of digital amplifiers? I'm clueless when ICE power and such are discussed. A primer for dummies like me would be appreciated. Links would be great, too.

T'anx.

 

[mtrycrafts]

Can someone please summarize the benefits and disadvantages of digital amplifiers? I'm clueless when ICE power and such are discussed. A primer for dummies like me would be appreciated. Links would be great, too.

T'anx.

Digital amps are very efficient compared to other amps. Gobs of power for the weight, when you need it, to its design limits. Certainly some are designed cheaply, higher distortions, than others.

http://www.audioholics.com/education/amplifier-technology/the-truth-about-digital-class-d-amplifiers

http://users.ece.gatech.edu/~mleach/ece4435/f01/ClassD2.pdf

http://en.wikipedia.org/wiki/Electronic_amplifier

http://electronicdesign.com/Articles/Index.cfm?AD=1&ArticleID=7533

 

[mike c]

here are the benefits for us regular consumers:
-cheaper because there are pretty much no heatsinks
-should be cheaper because there are a lot less copper for the amps
-runs cool
-very small
-a lot less demand on your electrical circuit

 

[Nomo]

here are the benefits for us regular consumers:
-cheaper because there are pretty much no heatsinks
-should be cheaper because there are a lot less copper for the amps
-runs cool
-very small
-a lot less demand on your electrical circuit

This is something I've been wondering as well. I actually thought that this was something every one knew about and I didnt want to embarrass myself by asking a stupid question. Kudos to Tomorrow for asking. 'cause he knows lots more than me. So now I feel better.
OK.
Then please allow me to ask the stupid question.
Why are we killing ourselves lifting 40 pound receivers and 100 pound amps around if digital is that much more efficient?

 

[Pyrrho]

This is something I've been wondering as well. I actually thought that this was something every one knew about and I didnt want to embarrass myself by asking a stupid question. Kudos to Tomorrow for asking. 'cause he knows lots more than me. So now I feel better.
OK.
Then please allow me to ask the stupid question.
Why are we killing ourselves lifting 40 pound receivers and 100 pound amps around if digital is that much more efficient?

Many people are prejudiced against digital amplifiers, and so manufacturers, being in business to make money, make what they think will sell for the most profit. It is extremely common for people to believe that the more a component weighs, the better it is. This applies to pretty much any type of component.

In defense of those who dislike digital amplifiers, some have been made that are not good. But that is true of pretty much every kind of amplifier; it is always possible to screw something up.

Anyway, being relatively new, the Luddites always are going to claim it isn't as ...[insert your favorite subjective words here; e.g., "non-fatiguing", "smooth", "creamy", delicious", etc.].

Anyway, from a buyer's standpoint, you can only buy what others sell, and then it may be that you will end up with a conventional design as your best choice for your particular application.

 

[mike c]

probably for the same reason some people carry giant LP's over "digital" compact discs



but seriously, the digital amps of today are pretty much expensive as compared to the regular analog amps. that's why i said SHOULD be cheaper, but not YET cheaper today.

also, there also just a few brands using these digital amps, and even fewer manufacturers actually producing the ICE power modules. and if you consider that it is B&O producing the most number of modules, you'll understand why they're so expensive. though b&o has some very nice modules available.

http://www.icepower.bang-olufsen.com/

 

[Seth=L]

It's not exactly correct to call them digital amplifiers though. Class D is just a tag label, like Class A or Class A/B. The "D" doesn't stand for "Digital" it's a marketing gimmick. In the late 1990's it became a big deal to have something that was "Digital" that trend is still going on today. To audiophiles the Class D amplifier is considered part of the trendy, cheesy, home theater in a box classification.

Any well designed Class D, Class A, Class A/B, Class H, Class Whatever amplifier will be transparent and leave no mark on the sound of your system. Their are some misconceptions that Class D amplifiers can't handle complex loads, but this is not true.

The weight factor only loosely applies to like amplifier topologies. You must also not forget other weight factors, such as heat sinks, case, extra parts, or other additional weight (such as brass feet or other unnecessarily added weight).

There is nothing wrong with Class D amplifiers, they are very cost effective from just about every point of view. They are light, efficient, and are not very expensive to manufacture.

 

[mike c]

i think the ep2500 is class H. but I get your point.

 

[MDS]

The 'D' doesn't stand for digital necessarily but it is at least somewhat correct to call them digital amplifiers. They use either PAM (pulse amplitude modulation) or PWM (pulse width modulation) switching at very high frequencies (100 kHz or higher).

 

[Seth=L]

i think the ep2500 is class H. but I get your point.

You are correct, my bad. The EP2500 is Class H. The A500 and EP1500 are both Class A/B. I thought the EP1500 was the same class as the EP2500 but the manual to the EP1500/EP2500 indicates otherwise. Good catch Mike.

Class H is a more efficient derivative of Class A/B.

 

[fredk]

Another reason digital amps arn't mainstream is that there are currently very few people who can design them.

Apparently none of the stuff engineers learn in the analog world applies to the digital world.

I wonder if there are limits to how much power you can push through a digital amp? I would think that the folks that use pro amps would be interested in the potential power savings.

I wonder if the recent improvements in the world of subs is an indicator that digital amp design is maturing a little?

Fred

 

[Seth=L]

Digital amplifiers in subwoofers need to do one thing, have high output at various impedances, typically lower ones. The distortion could be very high and you wouldn't hear it. THD factors on subwoofer amplifiers are typically just under 10%.

 

[fredk]

Hmm... Wishfull thinking on my part I guess. I like the idea of efficient amps.

Fred

 

[Seth=L]

Hmm... Wishfull thinking on my part I guess. I like the idea of efficient amps.

Fred

Oh, don't get me wrong...

I am not saying there aren't accurate Class D amplifiers, just that typically the subwoofer amplifiers (even the ones that aren't Class D) are higher in THD. They make these sacrifices for reasons, to cut cost and give the amplifier an edge in overall power. Since you can't hear the distortion at that low of a frequency, what's there to fret about?

 

[bandphan]

Another reason digital amps arn't mainstream is that there are currently very few people who can design them.

Apparently none of the stuff engineers learn in the analog world applies to the digital world.

I wonder if there are limits to how much power you can push through a digital amp? I would think that the folks that use pro amps would be interested in the potential power savings.

I wonder if the recent improvements in the world of subs is an indicator that digital amp design is maturing a little?

Fred

read mds post, they are switching amps, and have been made for a very long time btw. From what ive read in the past is that while they are effiecent in making power, there is power that is lost in transfer, so recoverery has been a challange. Velo has been using them for the sub amps b4 2k.

 

[fmw]

To answer the OP's question directly, there is no such thing as a digital amp. Audio amplifiers don't amplify in the digital domain. They do what they do in the analog domain. Amplifying in the digital domain wouldn't require an amplifier. It would require a computer program. The same holds true for any class of audio amplifier. The term digital amp is a marketing statement, not a statement of how the amplfier is designed.

Amplifers are designed to generate feedback into the output stage because the feedback reduces distortion. Hi fidelity was born when this was discovered in the 1930's. How that feedback is applied determines the "class" of an amplifier. When the feedback is applied continuously you have what is called a class A amplifier. All the other classes of audio amps modify this type of feedback or transistor biasing in order to improve efficiency. Class A amplifiers are quite inefficient but have the lowest measurable distortion. The other classes of amplifiers trade efficiency for higher measureable distortion. Notice that I didn't say audible distortion. I said measurable distortion.

I won't get into all the details and classes. You can find them in Wikipedia or some other place on the net. I'll just deal with the two mentioned here. Class D are switching amps. They simply turn off the output biasing when the input level reaches a certain point and turns it on when the input increases past that point. The biasing is switched on and off and hence the term switching amp. They have lower distortion than amplifiers without any biasing but more than other types that are considered high fidelity amplifiers.

Class H amps modulate the internal voltage of the amplifier rails against the input signal so the potential power output and biasing actually changes as the input signal changes. In effect, the "class" of the amplifier changes as needed. These are very efficient, high tech amps that can be designed with inaudible distortion.

 

[mike c]

yes, pretty much all the articles about class d starts with : "the d in class d does not stand for digital". but for the sake of ease of communication ... lets call class d digital or if you prefer, mickey mouse.

 

[bandphan]

From what ive read in the past is that while they are effiecent in making power, there is power that is lost in transfer, so recoverery has been a challange.

just found some notes i made from a mini ces, and i want to retract the recovery portion. Sorry i cant even parrot myself correctly.

 

[OttoMatic]

How that feedback is applied determines the "class" of an amplifier. When the feedback is applied continuously you have what is called a class A amplifier.

I don't think the feedback applied to an amplifier determines its "class." It's the biasing of the amplifier that determines what "class" it is. In a class A amp, it's biased continuously on; in a class B amp, it's biased on only half the time. Class A or class B can be implemented with a single transistor (and some resistors), and has absolutely nothing to do with feedback. Other amplifier classes follow, and can be reduces to a trivial case with a few electronic components.

I believe that each of these classes could be elaborated upon to implement feedback (either positive or negative), but feedback does NOT define the class.

 

[fmw]

I don't think the feedback applied to an amplifier determines its "class." It's the biasing of the amplifier that determines what "class" it is. In a class A amp, it's biased continuously on; in a class B amp, it's biased on only half the time. Class A or class B can be implemented with a single transistor (and some resistors), and has absolutely nothing to do with feedback. Other amplifier classes follow, and can be reduces to a trivial case with a few electronic components.

I believe that each of these classes could be elaborated upon to implement feedback (either positive or negative), but feedback does NOT define the class.

Biasing is the process of delivering feedback to the output stage.