Sign in with
Sign up | Sign in
Your question

Continued existence of Class-AB amps?

Last response: in Home Audio
Share
Anonymous
August 28, 2005 10:26:09 PM

Archived from groups: rec.audio.tech (More info?)

I would like to know what people consider the state of the art to be in
the (hopefully) rapid obsolescence of Class-AB power amps. Given all of
the apparent benefits of switching power amps, I am surprised that
there are so few PA (pro audio, public address, whatever) amps with
switching design. Any reasons? Is a Class-AB still cheaper to make,
even with the big heat sink and fan?
Anonymous
August 29, 2005 8:38:44 AM

Archived from groups: rec.audio.tech (More info?)

Pardon me for my ignorance, but what's a "switching power amp"? Are you
referring to the power supply? If so, does that have much bearing on
class A/B vs other classes? Are you referring instead to class D power
amps? Those are kinda switching I guess. I'm apparently not up on the
nomenclature.

On Sun, 28 Aug 2005 elephantcelebes@yahoo.com wrote:

> I would like to know what people consider the state of the art to be in
> the (hopefully) rapid obsolescence of Class-AB power amps. Given all of
> the apparent benefits of switching power amps, I am surprised that
> there are so few PA (pro audio, public address, whatever) amps with
> switching design. Any reasons? Is a Class-AB still cheaper to make,
> even with the big heat sink and fan?
>
>
Anonymous
August 29, 2005 8:38:45 AM

Archived from groups: rec.audio.tech (More info?)

There are some amps out there that are based on the idea that the power
supply needs only to be a roughly filtered, high current DC source. Then
this power is switched on and off very fast to create a PWM (pulse width
modulation) signal. The PWM is controlled by the audio signal. This is not
like a traditional amp that uses a tube or a transistor as a valve that lets
out a varying voltage. This thing lets out all or nothing. But it switches
on and off so fast that the accumulated power over time _IS_ the signal.
With just a little bit of filtering to get rid of the RF, it works very
well. The nice thing about a system like this is that there is no need for
an AC power transformer at all. If AC is directly rectified and filtered
with some big caps. That's all you need. One would think that if all of the
heavy copper was eliminated, that power amps would be very cheap, but of
course, we all still have to pay for designs, name brands and hype.

~James. :o )


"MZ" <mark@mdz.no-ip.org> wrote in message
news:20050829005121.Q839@mdz.no-ip.org...
> Pardon me for my ignorance, but what's a "switching power amp"? Are you
> referring to the power supply? If so, does that have much bearing on
> class A/B vs other classes? Are you referring instead to class D power
> amps? Those are kinda switching I guess. I'm apparently not up on the
> nomenclature.
>
> On Sun, 28 Aug 2005 elephantcelebes@yahoo.com wrote:
>
> > I would like to know what people consider the state of the art to be in
> > the (hopefully) rapid obsolescence of Class-AB power amps. Given all of
> > the apparent benefits of switching power amps, I am surprised that
> > there are so few PA (pro audio, public address, whatever) amps with
> > switching design. Any reasons? Is a Class-AB still cheaper to make,
> > even with the big heat sink and fan?
> >
> >
Anonymous
August 29, 2005 9:44:04 AM

Archived from groups: rec.audio.tech (More info?)

In <1125278769.792161.243670@g47g2000cwa.googlegroups.com>, on 08/28/05

at 06:26 PM, elephantcelebes@yahoo.com said:

>I would like to know what people consider the state of the art to be
>in the (hopefully) rapid obsolescence of Class-AB power amps. Given
>all of the apparent benefits of switching power amps, I am surprised
>that there are so few PA (pro audio, public address, whatever) amps
>with switching design. Any reasons? Is a Class-AB still cheaper to
>make, even with the big heat sink and fan?

Basically, we have existing designs that are working well enough and we
have designers who learned their craft a while ago. For small volume
production (less than hundreds of thousands of units) it is hard to
justify the retraining, tooling, and inventory costs of changing over.
Most of the parts needed for Class-AB designs are commodity items. We
are not quite at that point for Class-D.

One must retrain all of the production and service personel, not just
the engineers. Also, there is a not so obvious cost of stocking a
separate set of repair parts.

Another issue is customer resistance to new technology. Class-D
amplifiers have some "baggage" in the marketplace.

---

Consider the current state of automobile technology. There have been
alternative technologies for decades, but none of them can gather any
traction.

-----------------------------------------------------------
spam: uce@ftc.gov
wordgame:123(abc):<14 9 20 5 2 9 18 4 at 22 15 9 3 5 14 5 20 dot 3 15
13> (Barry Mann)
[sorry about the puzzle, spammers are ruining my mailbox]
-----------------------------------------------------------
Anonymous
August 29, 2005 10:01:24 AM

Archived from groups: rec.audio.tech (More info?)

<elephantcelebes@yahoo.com> wrote in message
news:1125278769.792161.243670@g47g2000cwa.googlegroups.com

> I would like to know what people consider the state of
> the art to be in the (hopefully) rapid obsolescence of
> Class-AB power amps. Given all of the apparent benefits
> of switching power amps, I am surprised that there are so
> few PA (pro audio, public address, whatever) amps with
> switching design. Any reasons? Is a Class-AB still
> cheaper to make, even with the big heat sink and fan?

The price performance advantages of class AB amps in certain
power categories, remains.

Switchmode amps are dependent on special ICs, high-speed
output devices, and in some cases complex analog power-level
output filters. They all cost money and the filters take up
space.

There is a third option - class H, that you don't seem to be
considering.

Switching amps aren't all gravy from a performance
standpoint.

Switchmode amps that are dependent on complex output filters
have problems with rising source impedance at high
frequencies, which results in an amp that is more sensitive
to the impedance curve of the speaker load.

Some switchmode amps are ugly-sounding when they clip. The
obvious response should be to simply use a larger amp, but
in some applications, clipping is not religiously avoided,
and so switchmode amps may have bad reputations.

I think the end game for power amps is apparent if you
consider the ubiquitous switchmode PC power supply. When PCs
first came out about 25 years, analog power supplies may
have been cheaper but they probably would have been larger
and run hotter. Due to continued development and immense
volume production over 25 years, the economic issues are
totally moot.
Anonymous
August 29, 2005 10:31:40 AM

Archived from groups: rec.audio.tech (More info?)

"Barry Mann" <zzzz@zzzz.zzz> wrote in message
news:4312dee7$1$avgroveq$mr2ice@wcnews.cyberonic.com

> Consider the current state of automobile technology.
> There have been alternative technologies for decades, but
> none of them can gather any traction.

Consider the confluence of automotive and audio. The OEM
equipment avoids switchmode because of EMI concerns. The
after-market is whole-hog into switchmode power supplies for
all but the most basic power levels, and moving quickly into
switchmode amps.

I don't know if the automotive switchmode amps have
independent switchmode power supplies or not.
Anonymous
August 29, 2005 11:07:25 AM

Archived from groups: rec.audio.tech (More info?)

Is Class-H not a switching topology?

Thanks for the useful comments. I was just wondering what the rough
edges were that still needed to be smoothed.
Anonymous
August 29, 2005 12:44:35 PM

Archived from groups: rec.audio.tech (More info?)

"MZ" <mark@mdz.no-ip.org> wrote in message
news:20050829005121.Q839@mdz.no-ip.org...
> Pardon me for my ignorance, but what's a "switching power amp"? Are
you
> referring to the power supply? If so, does that have much bearing
on
> class A/B vs other classes? Are you referring instead to class D
power
> amps? Those are kinda switching I guess. I'm apparently not up on
the
> nomenclature.
>
> On Sun, 28 Aug 2005 elephantcelebes@yahoo.com wrote:
>
> > I would like to know what people consider the state of the art to
be in
> > the (hopefully) rapid obsolescence of Class-AB power amps. Given
all of
> > the apparent benefits of switching power amps, I am surprised that
> > there are so few PA (pro audio, public address, whatever) amps
with
> > switching design. Any reasons? Is a Class-AB still cheaper to
make,
> > even with the big heat sink and fan?
> >
> >

Have a look at
http://stereos.about.com/od/homestereotechnologies/a/am....
htm
which reviews the different types and their effects.


--
Woody

harrogate2 at ntlworld dot com
Anonymous
August 29, 2005 2:02:10 PM

Archived from groups: rec.audio.tech (More info?)

Arny Krueger wrote:
> "Barry Mann" <zzzz@zzzz.zzz> wrote in message
> news:4312dee7$1$avgroveq$mr2ice@wcnews.cyberonic.com
>
>
>>Consider the current state of automobile technology.
>>There have been alternative technologies for decades, but
>>none of them can gather any traction.
>
>
> Consider the confluence of automotive and audio. The OEM
> equipment avoids switchmode because of EMI concerns. The
> after-market is whole-hog into switchmode power supplies for
> all but the most basic power levels, and moving quickly into
> switchmode amps.
>
> I don't know if the automotive switchmode amps have
> independent switchmode power supplies or not.
>
>
Speaking of automobile audio, I think BOSE car audio use this type of
technology, or a related one. They turn straight DC into high frequency
(100 MHz) AC to power their systems. Doing so enables their system to
generate more power with much greater efficiency. Interesting stuff. I
always scratch my head, though, at the possible side effects, like EMI
from the rapidly switching supplies, as mentioned before, and I gather
there may be other effects as well.

CD
Anonymous
August 29, 2005 2:03:53 PM

Archived from groups: rec.audio.tech (More info?)

"Codifus" <codifus@optonline.net> wrote in message
news:FzEQe.3613$vu5.2732@fe11.lga
> Arny Krueger wrote:
>> "Barry Mann" <zzzz@zzzz.zzz> wrote in message
>> news:4312dee7$1$avgroveq$mr2ice@wcnews.cyberonic.com
>>
>>
>>> Consider the current state of automobile technology.
>>> There have been alternative technologies for decades,
>>> but none of them can gather any traction.
>>
>>
>> Consider the confluence of automotive and audio. The OEM
>> equipment avoids switchmode because of EMI concerns. The
>> after-market is whole-hog into switchmode power supplies
>> for all but the most basic power levels, and moving
>> quickly into switchmode amps.
>>
>> I don't know if the automotive switchmode amps have
>> independent switchmode power supplies or not.

> Speaking of automobile audio, I think BOSE car audio use
> this type of technology, or a related one. They turn
> straight DC into high frequency (100 MHz) AC to power
> their systems. Doing so enables their system to generate
> more power with much greater efficiency. Interesting
> stuff. I always scratch my head, though, at the possible
> side effects, like EMI from the rapidly switching
> supplies, as mentioned before, and I gather there may be
> other effects as well.

Bose was a very early adopter of switchmode audio for
automotive. I believe their equipment works at more like 100
KHz than 100 MHz.
Anonymous
August 29, 2005 2:19:09 PM

Archived from groups: rec.audio.tech (More info?)

<elephantcelebes@yahoo.com> wrote in message
news:1125324445.774832.71900@f14g2000cwb.googlegroups.com

> Is Class-H not a switching topology?

Sort of. The output devices don't switch, but there are
switching devices modulating the power supply for the output
devices.

http://www.screensound.gov.au/glossary.nsf/Pages/Amplif...

> Thanks for the useful comments. I was just wondering what
> the rough edges were that still needed to be smoothed.
Anonymous
August 29, 2005 3:22:58 PM

Archived from groups: rec.audio.tech (More info?)

In <1125324445.774832.71900@f14g2000cwb.googlegroups.com>, on 08/29/05
at 07:07 AM, elephantcelebes@yahoo.com said:

>Is Class-H not a switching topology?

>Thanks for the useful comments. I was just wondering what the rough
>edges were that still needed to be smoothed.

Class-D is not new. I've been waiting for it to take over the world
since the 1970's. Despite the marketing hype, the audio industry is
slow at embracing new technology.

For a while it was thought that the EMI problem could not be solved for
high powered devices, but 100's of millions of cheap 200+ Watt PC
switch mode power supplies blew a hole in that assumption.

Another sticking point has been linearzing the chopper. (the circuit
that converts the analog input signal to pulses of one sort or another)
In industrial power applications distortion components that are 60dB
down are usually considered to be insignificant. In high end audio
terms, such a device would be considered crude. Now, with the advent of
cheap, high speed integrated circuits, we can afford a much more
complex circuit.

Over the next few months I think we will see more and more traditional
audio manufacturers bring out a Class-D model.

---

My original an analogy evidently didn't work very well. I was thinking
car engines, not car audio. Our current gasoline engines are very crude
in many ways, but we have been unable to develop a viable replacement
technology. (Yes, I know there are and have been some very interesting
demonstration products, but they have had an insignificant market
share.)

-----------------------------------------------------------
spam: uce@ftc.gov
wordgame:123(abc):<14 9 20 5 2 9 18 4 at 22 15 9 3 5 14 5 20 dot 3 15
13> (Barry Mann)
[sorry about the puzzle, spammers are ruining my mailbox]
-----------------------------------------------------------
Anonymous
August 29, 2005 4:13:02 PM

Archived from groups: rec.audio.tech (More info?)

James Lehman wrote:

> There are some amps out there that are based on the idea that the power
> supply needs only to be a roughly filtered, high current DC source. Then
> this power is switched on and off very fast to create a PWM (pulse width
> modulation) signal. The PWM is controlled by the audio signal. This is not
> like a traditional amp that uses a tube or a transistor as a valve that lets
> out a varying voltage. This thing lets out all or nothing. But it switches
> on and off so fast that the accumulated power over time _IS_ the signal.
> With just a little bit of filtering to get rid of the RF, it works very
> well. The nice thing about a system like this is that there is no need for
> an AC power transformer at all. If AC is directly rectified and filtered
> with some big caps. That's all you need. One would think that if all of the
> heavy copper was eliminated, that power amps would be very cheap, but of
> course, we all still have to pay for designs, name brands and hype.

You have reversible 2-prong mains plugs in your part of the world?
Your system ground will be connected to neutral or live, both bad ideas.

Either a big mains transformer or a switch-mode psu is still required.

--
Eiron

I have no spirit to play with you; your dearth of judgment renders you
tedious - Ben Jonson.
Anonymous
August 29, 2005 4:13:03 PM

Archived from groups: rec.audio.tech (More info?)

"Eiron" <e1ron@hotmail.com> wrote in message
news:3ng8vdF17s4vU1@individual.net

> Either a big mains transformer or a switch-mode psu is
> still required.

Agreed.

Of course, running transformers at high frequencies shrinks
their size and weight magnificently.

The cost, size and weight of a power amp is largely based on
the transformers, main filter caps, and heat sinks.

Switchmode operations shrinks them all, but at a cost in
complexity and high-tech semiconductors. The latter two
influences have a long history of improving dramatically.
Anonymous
August 29, 2005 5:06:55 PM

Archived from groups: rec.audio.tech (More info?)

In article <4_KdnVLDvqj8a4_eRVn-tw@comcast.com>,
"Arny Krueger" <arnyk@hotpop.com> wrote:

> Switchmode operations shrinks them all, but at a cost in
> complexity and high-tech semiconductors. The latter two
> influences have a long history of improving dramatically.

Did you mean to say latter, rather than latter two?
Anonymous
August 29, 2005 6:49:14 PM

Archived from groups: rec.audio.tech (More info?)

In the US we have 3 prong 110VAC, Hot, Nutral and Earth, that is usually
single phase 220VAC with a ground split into 2 110 circuits inside the house
breaker box. Things like electric clothes dryers and electric oven-stove are
220. What's the big deal? Just make an amp that takes 220VAC. If it was done
right, you'd have almost 2KW of power into 8 ohms.

James. :o )


"Eiron" <e1ron@hotmail.com> wrote in message
news:3ng8vdF17s4vU1@individual.net...
> James Lehman wrote:
>
> > There are some amps out there that are based on the idea that the power
> > supply needs only to be a roughly filtered, high current DC source. Then
> > this power is switched on and off very fast to create a PWM (pulse width
> > modulation) signal. The PWM is controlled by the audio signal. This is
not
> > like a traditional amp that uses a tube or a transistor as a valve that
lets
> > out a varying voltage. This thing lets out all or nothing. But it
switches
> > on and off so fast that the accumulated power over time _IS_ the signal.
> > With just a little bit of filtering to get rid of the RF, it works very
> > well. The nice thing about a system like this is that there is no need
for
> > an AC power transformer at all. If AC is directly rectified and filtered
> > with some big caps. That's all you need. One would think that if all of
the
> > heavy copper was eliminated, that power amps would be very cheap, but of
> > course, we all still have to pay for designs, name brands and hype.
>
> You have reversible 2-prong mains plugs in your part of the world?
> Your system ground will be connected to neutral or live, both bad ideas.
>
> Either a big mains transformer or a switch-mode psu is still required.
>
> --
> Eiron
>
> I have no spirit to play with you; your dearth of judgment renders you
> tedious - Ben Jonson.
Anonymous
August 29, 2005 7:16:54 PM

Archived from groups: rec.audio.tech (More info?)

"James Lehman" <james[remove]@akrobiz.com> wrote in message
news:%OIQe.31281$Tf5.20171@newsread1.mlpsca01.us.to.verio.net
> In the US we have 3 prong 110VAC, Hot, Nutral and Earth,
> that is usually single phase 220VAC with a ground split
> into 2 110 circuits inside the house breaker box. Things
> like electric clothes dryers and electric oven-stove are
> 220. What's the big deal? Just make an amp that takes
> 220VAC. If it was done right, you'd have almost 2KW of
> power into 8 ohms.

It's done on the pro audio circuit, all the time, for
example:

http://www.qscaudio.com/products/amps/powerlight/powerl...

PL 6.0II 1150 wpc@8 ohms 20-20K < 0.1% THD
Anonymous
August 29, 2005 8:28:45 PM

Archived from groups: rec.audio.tech (More info?)

On 29 Aug 2005 07:07:25 -0700, elephantcelebes@yahoo.com wrote:

>Is Class-H not a switching topology?

In class G, a low Vcc1 power supply is coupled to a standard class AB amp,
and a large Vcc2 kicks in during peaks, driving the same amp or a larger one
coupled to the same output load. Class H adds a tracking system to the power
supply, thus somewhat locking it to the input signal and varying Vcc
according to program demands.

Not exactly switching.
Anonymous
August 29, 2005 8:56:01 PM

Archived from groups: rec.audio.tech (More info?)

"hoarse with no name" <no@2spam.com> wrote in message
news:no-CA4099.13065329082005@cnews.newsguy.com
> In article <4_KdnVLDvqj8a4_eRVn-tw@comcast.com>,
> "Arny Krueger" <arnyk@hotpop.com> wrote:
>
>> Switchmode operations shrinks them all, but at a cost in
>> complexity and high-tech semiconductors. The latter two
>> influences have a long history of improving dramatically.
>
> Did you mean to say latter, rather than latter two?

No, I meant that both apparent complexity for a given
function, and semiconductors have a long history of
improving dramatically. And, often its the latter that
enables the former. Higher level of semiconductor
integration reduce apparent complexity by taking a
basketfull of parts and integratnig them into one small
package with a few leads coming out if it and few if any
external parts.
Anonymous
August 29, 2005 10:14:08 PM

Archived from groups: rec.audio.tech (More info?)

Thanks to both of you. And I think there may be another "class" as
well. Looking up the patents belonging to Tripath, they claim a hybrid
where a Class AB operates on the lowest and highest sighal levels, and
Class D elsewhere. The same output transistors operate in both modes --
it is only the driver circuit that changes over. If anybody cares,
this is patent US6297697. This, combined with dithered modulation
frequency, may be what they refer to as "Class T."
Anonymous
August 30, 2005 3:26:17 AM

Archived from groups: rec.audio.tech (More info?)

On Mon, 29 Aug 2005 15:16:54 -0400, "Arny Krueger" <arnyk@hotpop.com> wrote:

>It's done on the pro audio circuit, all the time, for
>example:

The QSC amps you mention use a switching power supply and can work on 115 V
mains.
Anonymous
August 30, 2005 3:26:18 AM

Archived from groups: rec.audio.tech (More info?)

"François Yves Le Gal" <flegal@aingeal.com> wrote in
message news:n8v6h1djdqa27figvtavr92khjlrcsr4vb@4ax.com
> On Mon, 29 Aug 2005 15:16:54 -0400, "Arny Krueger"
> <arnyk@hotpop.com> wrote:
>
>> It's done on the pro audio circuit, all the time, for
>> example:

>> QSC Powerlight 6.0II 1150 wpc@8 ohms 20-20K < 0.1% THD

> The QSC amps you mention use a switching power supply

Agreed. They are also Class G.

> and can work on 115 V mains.

Normal 120 volt power circuits in the US are circuit-breaker
protected @ either 15 or 20 amps.

The PL 6.0 120v Current Consumption(1/8 Power Pink Noise @
4 ohms) is specified to be 16 amps.

It would seem to be difficult to fully exploit a QSC
Powerlight 6.0 on a normal US 120 volt circuit.

I've observed them used for SR at a live concert, running
off of 230-240 volt circuits.
Anonymous
September 3, 2005 2:26:25 AM

Archived from groups: rec.audio.tech (More info?)

In article <--SdnUN-sI__647eRVn-1g@comcast.com>,
"Arny Krueger" <arnyk@hotpop.com> wrote:

> No, I meant that both apparent complexity for a given
> function, and semiconductors have a long history of
> improving dramatically. And, often its the latter that
> enables the former. Higher level of semiconductor
> integration reduce apparent complexity by taking a
> basketfull of parts and integratnig them into one small
> package with a few leads coming out if it and few if any
> external parts.

I do not understand why this is not entirely an improvement in chip
tech. I do not understand how complexity can be subject to terms such as
improved or worsened.
!