[geezer[_2_]]

I saw a review for a JBL car amp that was a class 1, BCA design. It
output a ridiculous # of watts

I thought I was passingly familiar with most amp types. [class A, A/
B, d, h, etc....

Does the balanced current thing work, or sound any good?

-glenn

[Paul Stamler]

"geezer" wrote in message
...
I saw a review for a JBL car amp that was a class 1, BCA design. It
output a ridiculous # of watts

I thought I was passingly familiar with most amp types. [class A, A/
B, d, h, etc....

Does the balanced current thing work, or sound any good?

If the phrase means what I think it means, yes, it works; it's called
bridging in the rest of the world. Whether the JBL amp sounds good, I
haven't the faintest idea. In a car, who could tell?

Peace,
Paul

[David Gravereaux]

Paul Stamler wrote:
"geezer" wrote in message
...
I saw a review for a JBL car amp that was a class 1, BCA design. It
output a ridiculous # of watts

I thought I was passingly familiar with most amp types. [class A, A/
B, d, h, etc....

Does the balanced current thing work, or sound any good?

If the phrase means what I think it means, yes, it works; it's called
bridging in the rest of the world. Whether the JBL amp sounds good, I
haven't the faintest idea. In a car, who could tell?

Peace,
Paul


Also called bi-polar loading, too.

[geezer[_2_]]

No, it's not bridging, I know what that is. Apparently it's class I
(interleave), not Class 1.

According to the review http://www.jbl.com/car/featured/spon...0GTireview.pdf
it was developed by Crown (for the K series amps? I'm guessing that
because the k series were fairly high power, but had no fans, or even
external heat sinks. That would co-incide with the claimed 85%
efficiency at 4 ohms.)

Anyway, just trying to figure our what the basic design approach is,
and if it actually sounds like anything.

-glenn



On May 12, 12:35*am, "Paul Stamler" wrote:
"geezer" wrote in message

...

I saw a review for a JBL car amp that was a class 1, BCA design. It
output a ridiculous # of watts

I thought I was passingly familiar with most amp types. *[class A, A/
B, d, h, etc....

Does the balanced current thing work, or sound any good?

If the phrase means what I think it means, yes, it works; it's called
bridging in the rest of the world. Whether the JBL amp sounds good, I
haven't the faintest idea. In a car, who could tell?

Peace,
Paul

[Eeyore]

geezer wrote:

No, it's not bridging, I know what that is. Apparently it's class I
(interleave), not Class 1.

According to the review http://www.jbl.com/car/featured/spon...0GTireview.pdf

..... it has 'snazzy blue neon tubes' !

Shame that neon only actually glows orange.

Graham

[geezer[_2_]]

Gee, perhaps many of you have missed my point entirely. I wasn't
asking about the AUTOMOTIVE part of the design, as I mostly listen to
AM radio when I'm driving.

My question was about the design principles of the Class I amplifier,
developed by Crown. Can anybody tell me whether this is a sound
operating principle?

One of the design parameters, if you look at the Crown CTS 2000, is
that it generates the same wattage into virtually any impedance load.
The CTS 2000 is rated rated at 1000 watts, into either 8, 4, or 2 ohm
loads. This is certainly quite different from other amplifier designs,
and seems somewhat counter-intuitive, since you'd expect watage to
decrease and the speaker impedance increases, or vice versa. Since
this isn't happening with this design, what is? and why? are there
any negative consequences of the design?

-glenn



On May 12, 6:18*am, Eeyore
wrote:
geezer wrote:
No, it's not bridging, I know what that is. Apparently it's class I
(interleave), not Class 1.

According to the reviewhttp://www.jbl.com/car/featured/sponsors/hot_news/JBLA6000GTireview.pdf

.... it has 'snazzy blue neon tubes' !

Shame that neon only actually glows orange.

Graham

[John Williamson]

geezer wrote:
Gee, perhaps many of you have missed my point entirely. I wasn't
asking about the AUTOMOTIVE part of the design, as I mostly listen to
AM radio when I'm driving.

My question was about the design principles of the Class I amplifier,
developed by Crown. Can anybody tell me whether this is a sound
operating principle?

One of the design parameters, if you look at the Crown CTS 2000, is
that it generates the same wattage into virtually any impedance load.
The CTS 2000 is rated rated at 1000 watts, into either 8, 4, or 2 ohm
loads. This is certainly quite different from other amplifier designs,
and seems somewhat counter-intuitive, since you'd expect watage to
decrease and the speaker impedance increases, or vice versa. Since
this isn't happening with this design, what is? and why? are there
any negative consequences of the design?

-glenn



On May 12, 6:18 am, Eeyore
wrote:
geezer wrote:
No, it's not bridging, I know what that is. Apparently it's class I
(interleave), not Class 1.
According to the reviewhttp://www.jbl.com/car/featured/sponsors/hot_news/JBLA6000GTireview.pdf
.... it has 'snazzy blue neon tubes' !

Shame that neon only actually glows orange.

Graham

It seems to be a variant on large number of paralleled switching output
stages driving a single output transformer. There's a paper on the
design principle he-

http://forums.klipsch.com/forums/sto...0Amplifier.pdf

Page 4 has a sort of schematic sketch of how it works.

Not exactly light either, it arrived in a case weighing 145 pounds
(nearly 70 kilograms), apparently.

The one quoted in the review is a bass amplifier for automotive
applications, with nothing above 2KHz showing on the response curve.

--
Tciao for Now!

John.

[Scott Dorsey]

geezer wrote:
No, it's not bridging, I know what that is. Apparently it's class I
(interleave), not Class 1.

Anything other than class A, B, AB, and C is something made up by the marketing
department. Everyone more or less knows what "Class D" means these days.

There are a lot of class B amplifiers with switching supplies out there. The
marketing department makes up a silly name. There are some class B amplifiers
with trick rail-switching supplies (Carver loved these), and the marketing
department makes up a silly name. There are a bunch of class D amplifiers
out there that use a PWM oscillator going into a class C amplifier stage
going into a power integrator stage, and the marketing department all wants
to call them something else to distinguish them.

According to the review http://www.jbl.com/car/featured/sponsors/hot_news/JB=
LA6000GTireview.pdf
it was developed by Crown (for the K series amps? I'm guessing that
because the k series were fairly high power, but had no fans, or even
external heat sinks. That would co-incide with the claimed 85%
efficiency at 4 ohms.)

What you need to know is that car amplifiers do not need to follow the
FTC laws for advertising power, so car amplifiers always have outrageously
high power level ratings that bear no resemblance to the actual amplifier
porformance. I have seen some "thousand watt" car amplifiers that were
more like fifty watts when measured by the FTC method.

Anyway, just trying to figure our what the basic design approach is,
and if it actually sounds like anything.

For the most part, I am very suspicious of trick amplifiers. The whole point
of class D operation is to get more output power with less input power.
That's great for a car application and maybe for a subwoofer, but for the most
part I deal with applications where we already have way more power than we
ever need.
--scott

Another hilarious thing... the Chinese have been buying up all sorts of
traditional American trademarks, and consequently we have been seeing a lot
of cheap consumer crap showing up under names like Bell and Howell, RCA,
and the like. They even got Dual and Wharfdale. But what I find very funny
is that they are now selling car stereos under the "Phase Linear" name.
200W heads (actually closer to 5W) with CD players for $50. I asked the
guy at the auto parts store if they caught fire but he didn't know.
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Scott Dorsey]

geezer wrote:
My question was about the design principles of the Class I amplifier,
developed by Crown. Can anybody tell me whether this is a sound
operating principle?

I do not believe that switching amplifiers are a good idea for general
applications, no. You get a lot of efficiency but you give up a lot
in the process. For an inefficient subwoofer, sure.

One of the design parameters, if you look at the Crown CTS 2000, is
that it generates the same wattage into virtually any impedance load.
The CTS 2000 is rated rated at 1000 watts, into either 8, 4, or 2 ohm
loads. This is certainly quite different from other amplifier designs,
and seems somewhat counter-intuitive, since you'd expect watage to
decrease and the speaker impedance increases, or vice versa. Since
this isn't happening with this design, what is? and why?

Because it's run by a cheap switching power supply which has a high
output impedance. The power rating is being limited by the ability of
the power supply to deliver current and not by the size of the output
stage.

are there
any negative consequences of the design?

I have never heard a switching amplifier which didn't sound bad. Tripath
is working on it, but I wouldn't consider any existing switching amplifiers
for anything approaching a critical monitoring application.

I did the A-B between a K-1 and my fifty-year-old Citation II amp, and the
Citation II sounded hands down better than the K-1. Sure, the K-1 could put
out about 50 times the power, but what good is power if it doesn't sound good?
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Arny Krueger]

"geezer" wrote in message


I saw a review for a JBL car amp that was a class 1, BCA
design. It output a ridiculous # of watts

Google coughed up this technical paper, first on the retrieval list:

www.crownaudio.com/pdf/amps/137234.pdf

What is there in this paper that needs further explanation?

I thought I was passingly familiar with most amp types.
[class A, A/ B, d, h, etc....

Does the balanced current thing work, or sound any good?

I seriously doubt that in this day and age, Crown would field an amplifier
that simply failed to work or would unconditionally sound bad, if in good
working condition. I say this somewhat advisedly, having some inside
information about the reliability of some of their parent company's OEM
power amps for automotive use.

The paper cited above outlines the basic technology, which is based on a
pair of PWM switchmode power amplifiers.

[Don Pearce]

Arny Krueger wrote:
"geezer" wrote in message


I saw a review for a JBL car amp that was a class 1, BCA
design. It output a ridiculous # of watts

Google coughed up this technical paper, first on the retrieval list:

www.crownaudio.com/pdf/amps/137234.pdf

What is there in this paper that needs further explanation?


Good paper, but it would have been even better as a compare/contrast
against Class D, rather than A & B.

d

[Arny Krueger]

"Don Pearce" wrote in message
et

Arny Krueger wrote:

"geezer" wrote in message


I saw a review for a JBL car amp that was a class 1, BCA
design. It output a ridiculous # of watts

Google coughed up this technical paper, first on the
retrieval list:

www.crownaudio.com/pdf/amps/137234.pdf


Here is a paper that makes specific performance claims:

http://www.jbl.com/car/featured/spon...0GTireview.pdf

We thus have the model number - A6000G


Apparently, this innovation is not limited to auto sound, but is also
implemented in the

JBL VP7212MDP stage monitor:

http://www.jblpro.com/installedsound...vp7212MDP.html


What is there in this paper that needs further
explanation?

Good paper, but it would have been even better as a
compare/contrast against Class D, rather than A & B.

I read the description of Class I as being a variation on Class D.

I see a minor variation on PWM that attempts to reduce electrical losses and
heating due to simultaneous conduction.

One interesting claim - that these amps recapture energy that is temporarily
stored in reactive loads like loudspeakers.

[Don Pearce]

On Mon, 12 May 2008 12:42:52 -0400, "Arny Krueger"
wrote:

Good paper, but it would have been even better as a
compare/contrast against Class D, rather than A & B.

I read the description of Class I as being a variation on Class D.

I see a minor variation on PWM that attempts to reduce electrical losses and
heating due to simultaneous conduction.


Yes, this is the important stuff, and exactly why a comparison between
the two would have been useful.

One interesting claim - that these amps recapture energy that is temporarily
stored in reactive loads like loudspeakers.


Clearly silly. Any audio amplifier - however implemented - is
ultimately a voltage source. And here at last we find a use for that
orphan parameter - damping factor. It is the ratio of the output
impedance to the nominal speaker impedance; in other words the degree
of mismatch. If we assume 1000 as a reasonable minimum figure, no
amplifier will absorb more than a thousandth of the power the speaker
is attempting to reflect at it.

If this amplifier is somehow "recapturing" that energy, what on earth
is it doing with it, how is it recapturing it and where is it putting
it?

d
--
Pearce Consulting
http://www.pearce.uk.com

[geezer[_2_]]

On May 12, 8:45*am, (Scott Dorsey) wrote:

Because it's run by a cheap switching power supply which has a high
output impedance. *The power rating is being limited by the ability of
the power supply to deliver current and not by the size of the output
stage.

Ok, the Crown CTS2000 is not cheap, by any stretch - it sells for
something like $2,300. I've had a QSC Powerlight 1.8 for maybe 10
years, and it delivers increasing wattage as load impedance drops (400
into 8 ohms, 650 into 4, etc.) The CTS2000 is a straight 1000 into
either 2, 4, or 8, and that's driven full bandwidth 20-20k, with 0.35%
thd, so it's not exactly a made up number a la car audio gear mostly
is. And the damping factor is a gi-normous 3,000 up to 100hz.
http://www.crownaudio.com/pdf/amps/135009.pdf


are there
any negative consequences of the design?

I have never heard a switching amplifier which didn't sound bad. *Tripath
is working on it, but I wouldn't consider any existing switching amplifiers
for anything approaching a critical monitoring application.

I've had the Powerlight for a while, and never thoughrt anything was
amiss with the sound, at least in PA applications. Interestingly, I
once compared a Mexican built Ross MTA1200 (the ultimate cheapo
switching amp), vs a McCormack amplifier on the Vandersteen II-Cs.
(Both amps driven by a McCormack pre) All four of us, including the
owner of the Vandys, thought the Ross sounded a lot better in the mids
& highs, which none of us expected (shocked would a better
description) We also compared to a Soundcraftsmen PCR model. The Ross
was the hands-down winner of the 3. So I don't think one can indict
switching designs, per se.


I did the A-B between a K-1 and my fifty-year-old Citation II amp, and the
Citation II sounded hands down better than the K-1. *Sure, the K-1 could put
out about 50 times the power, but what good is power if it doesn't sound good?

Is your 50 year old a tube design? Couldn't "euphonic" distortion,
could it? :-)



--scott
--
"C'est un Nagra. *C'est suisse, et tres, tres precis."

[geezer[_2_]]

On May 12, 9:46*am, "Arny Krueger" wrote:
"geezer" wrote in message

Google coughed up this technical paper, first on the retrieval list:

www.crownaudio.com/pdf/amps/137234.pdf

What is there in this paper that needs further explanation?

A lot, if you're basically a dummy like me. Guess I'm a little
unclear on the pulse width modulation thingie. Also can't figure why
other switch-mode amps seem to have a somewhat inverse relationship
with impedance load and wattage, but the CTS2000 has a constant
wattage into any impedance between 2 and 8 ohms. Guess I'm just
missing something

I seriously doubt that in this day and age, Crown would field an amplifier
that simply failed to work or would unconditionally sound bad,

That's what occurred to me. Crown makes decent enough stuff, & has
been around forever. I assume their going to put a totally BS
product..... OTOH, who would have predicted "New Coke" ?

-glenn

[geezer[_2_]]

That's what occurred to me. Crown makes decent enough stuff, & has
been around forever. *I assume their going to put a totally BS
product..... * OTOH, who would have predicted "New Coke" ?

-glenn

SHOULD have written:

"I assume they're NOT going to put a totally BS
product..... OTOH, who would have predicted "New Coke" ?"

That's what I get for posting in between phone calls during lunch.

-glenn

[Scott Dorsey]

geezer wrote:
On May 12, 8:45=A0am, (Scott Dorsey) wrote:

Because it's run by a cheap switching power supply which has a high
output impedance. =A0The power rating is being limited by the ability of
the power supply to deliver current and not by the size of the output
stage.

Ok, the Crown CTS2000 is not cheap, by any stretch - it sells for
something like $2,300.

That doesn't keep it from being cheap. You can sell cheap crap at any
price point.

I've had a QSC Powerlight 1.8 for maybe 10
years, and it delivers increasing wattage as load impedance drops (400
into 8 ohms, 650 into 4, etc.)

Right. If an amplifier is a zero ohm source, it will driver twice as much
power into 4 ohms that it does into 8 ohms, and twice as much into 2 ohms
as it does into 4 ohms.

The reason that your amplifier delivers only 650 ohms into 4 watts instead
of 800 is that the output stage and power supply have some series impedance
to them. The bottleneck in the system that limits the output power is the
series impedance of the supply and the output stage.

The CTS2000 is a straight 1000 into
either 2, 4, or 8, and that's driven full bandwidth 20-20k, with 0.35%
thd, so it's not exactly a made up number a la car audio gear mostly
is.

Right. It has a switching supply.... and no huge supply capacitors. The
power isn't limited by the series resistance of the output stage any more,
it's limited by the input side of the switching supply. By using a switching
supply, the "power bottleneck" has been moved from one part of the system
to another. And the place it's been moved to is a place where the power is
independant of the output load.

And the damping factor is a gi-normous 3,000 up to 100hz.
http://www.crownaudio.com/pdf/amps/135009.pdf

Yes, this is a consequence of having a low output impedance. The only problem
with these amps is that they sound awful.

I've had the Powerlight for a while, and never thoughrt anything was
amiss with the sound, at least in PA applications. Interestingly, I
once compared a Mexican built Ross MTA1200 (the ultimate cheapo
switching amp), vs a McCormack amplifier on the Vandersteen II-Cs.
(Both amps driven by a McCormack pre) All four of us, including the
owner of the Vandys, thought the Ross sounded a lot better in the mids
& highs, which none of us expected (shocked would a better
description) We also compared to a Soundcraftsmen PCR model. The Ross
was the hands-down winner of the 3. So I don't think one can indict
switching designs, per se.

I remain skeptical, still. I haven't tried the Ross, but I have tried
some Stewarts, the K-1, and some other amps out there and not been so
pleased.

I did the A-B between a K-1 and my fifty-year-old Citation II amp, and the=

Citation II sounded hands down better than the K-1. =A0Sure, the K-1 could=
put
out about 50 times the power, but what good is power if it doesn't sound g=
ood?

Is your 50 year old a tube design? Couldn't "euphonic" distortion,
could it? :-)

It's a tube design, and the distortion is outrageously low for the tube
era... but most of it comes from the output transformer. Because it is
transformer-coupled, THIS amplifier has the power bottleneck there. The
thing is flat within 1 dB from 2 Hz to 40 KHz driving 2W into a load,
but at full rated 60W power it's +/- 3dB from 20Hz to 20 KHz only... the
transformer saturation limits the low frequencies.

The tube amp ALSO can deliver the same rated power into any load, because
the transformer has different taps for 4, 8, and 16 ohm loads. The saturation
level of the transformer core is at the same power level no matter which
tap you use.

The output transformer really is a severe bottleneck, and the big advantage
of solid-state amps was to do away with them. It limits power and is the
main source of distortion in the path.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Arny Krueger]

"Don Pearce" wrote in message

On Mon, 12 May 2008 12:42:52 -0400, "Arny Krueger"
wrote:

One interesting claim - that these amps recapture energy
that is temporarily stored in reactive loads like
loudspeakers.

Clearly silly. Any audio amplifier - however implemented
- is ultimately a voltage source.

Agreed.

And here at last we
find a use for that orphan parameter - damping factor. It
is the ratio of the output impedance to the nominal
speaker impedance; in other words the degree of
mismatch.

I don't think so. I'm under the impression that damping factor is the ratio
of source impedance to load impedance. Speakers and amplifiers are not
impedance-matched systems like transmitters and antennas. Therefore, source
impedance and load impedance are usually vastly different.

If we assume 1000 as a reasonable minimum
figure, no amplifier will absorb more than a thousandth
of the power the speaker is attempting to reflect at it.

I question your analysis. There will be times during an output cycle where
the load will be dumping current back into the amplifier. The amplifier
either absorbs this current, or the output voltage of the amplifier will be
greater than the output voltage which corresponds to the input signal. If
the amplifier provides a low source impdance, the voltage across the load
will be 99.9%+ as determined by the input signal.

It seems possible that virtually all of the current that the load dumps will
be absorbed by the amplifier.

If the amplifier's output devices are only partially conducting, then much
of the power that the load is dumping will be dissipated in the output
devices. That is why SOA was such a big issue for traditional Class A-C
power amplifier designs.

If the amplifier is a switchmode amplifier then the output devices will be
conducting very heavily, and very little power will be dissipated in the
output devices.

If this amplifier is somehow "recapturing" that energy,
what on earth is it doing with it, how is it recapturing
it and where is it putting it?

As always, the law of conservation of energy will assert itself, so the
excess energy stored in the load will have to go *some place*. Thus, it will
be absorbed by any lossy components of the load, any amplifier output
filtering network, the amplifier output devices, and the amplifier's power
supply.

In a low-loss system, most of the energy will end up back in the amplifier's
power supply.

[David Gravereaux]

Scott Dorsey wrote:
....
I have never heard a switching amplifier which didn't sound bad.

Ditto. Although I would love to use this MOSFET class A/B design I have
in a product, which sounds amazingly good, I can't due to thermal
requirements. I've tried numerous high efficiency modules made by
IcePower and PowerOne, and I've come the same conclusion. They all
sound like crap. Mostly an unnatural crispyness to the top-end and lack
of punch in the bottom.

[Don Pearce]

On Mon, 12 May 2008 13:30:56 -0400, "Arny Krueger"
wrote:

"Don Pearce" wrote in message

On Mon, 12 May 2008 12:42:52 -0400, "Arny Krueger"
wrote:

One interesting claim - that these amps recapture energy
that is temporarily stored in reactive loads like
loudspeakers.

Clearly silly. Any audio amplifier - however implemented
- is ultimately a voltage source.

Agreed.

And here at last we
find a use for that orphan parameter - damping factor. It
is the ratio of the output impedance to the nominal
speaker impedance; in other words the degree of
mismatch.

I don't think so. I'm under the impression that damping factor is the ratio
of source impedance to load impedance. Speakers and amplifiers are not
impedance-matched systems like transmitters and antennas. Therefore, source
impedance and load impedance are usually vastly different.


Exactly (although I had worded it back to front). And the rest of your
para is simply rephrasing what I said. we agree - there is no
matching, and the damping factor is a measure of the degree of
mismatch.

If we assume 1000 as a reasonable minimum
figure, no amplifier will absorb more than a thousandth
of the power the speaker is attempting to reflect at it.

I question your analysis. There will be times during an output cycle where
the load will be dumping current back into the amplifier. The amplifier
either absorbs this current, or the output voltage of the amplifier will be
greater than the output voltage which corresponds to the input signal. If
the amplifier provides a low source impdance, the voltage across the load
will be 99.9%+ as determined by the input signal.


Negative feedback ensures that the output impedance is close to zero
at all points in the voltage range. The resistance of the speaker is
in series with this at all frequencies, so there is a potential
divider which drops the potential transfer from speaker to amplifier
by a ratio equal to the damping factor.

It seems possible that virtually all of the current that the load dumps will
be absorbed by the amplifier.

The power - and hence the energy of 99.9% gets dissipated in the
speaker resistance.

If the amplifier's output devices are only partially conducting, then much
of the power that the load is dumping will be dissipated in the output
devices. That is why SOA was such a big issue for traditional Class A-C
power amplifier designs.

If the amplifier is a switchmode amplifier then the output devices will be
conducting very heavily, and very little power will be dissipated in the
output devices.

If this amplifier is somehow "recapturing" that energy,
what on earth is it doing with it, how is it recapturing
it and where is it putting it?

As always, the law of conservation of energy will assert itself, so the
excess energy stored in the load will have to go *some place*. Thus, it will
be absorbed by any lossy components of the load, any amplifier output
filtering network, the amplifier output devices, and the amplifier's power
supply.

In a low-loss system, most of the energy will end up back in the amplifier's
power supply.

There is no mechanism for it to get there. Negative feedback makes the
speaker's connection point far too stiff to allow anything to be
transferred.

d
--
Pearce Consulting
http://www.pearce.uk.com

[Scott Dorsey]

David Gravereaux wrote:
Scott Dorsey wrote:
...
I have never heard a switching amplifier which didn't sound bad.

Ditto. Although I would love to use this MOSFET class A/B design I have
in a product, which sounds amazingly good, I can't due to thermal
requirements. I've tried numerous high efficiency modules made by
IcePower and PowerOne, and I've come the same conclusion. They all
sound like crap. Mostly an unnatural crispyness to the top-end and lack
of punch in the bottom.

Look at the Tripath modules. They seem to be the best of the set. Also
look at a more aggressive integration stage... don't worry if your top end
response sags a bit on paper... it will sound better.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[geezer[_2_]]

On May 12, 1:27*pm, (Scott Dorsey) wrote:


That doesn't keep it from being cheap. *You can sell cheap crap at any
price point.

SNIP

Yes, this is a consequence of having a low output impedance. *The only problem
with these amps is that they sound awful.


OK, so the premise is that Crown is building an amp which sells for a
couple grand, which has crappy components and sounds like crap.
At the same time they're building amps for much less money, with
better power supplies and which presumably sound fine. I'm thinking
maybe the xs or xls series.

Somehow, the premise doesn't quite make sense in a free market. How,
or why, would Crown make an inferior product for more money?

I guess I'm just missing something.

-glenn

[Scott Dorsey]

geezer wrote:
On May 12, 1:27=A0pm, (Scott Dorsey) wrote:

OK, so the premise is that Crown is building an amp which sells for a
couple grand, which has crappy components and sounds like crap.

No, no, I didn't say it had crappy components. I just said it sounded like
crap, because of the design considerations that went into it.

It's designed for power and light weight, and when you want these things,
you have to give something up.

At the same time they're building amps for much less money, with
better power supplies and which presumably sound fine. I'm thinking
maybe the xs or xls series.

Those also have trick power supplies, but they are different kinds of
trick power supplies and they are conventional amplifiers. If you look
at dollars per watt, though, you will find they are more expensive than
the class D amps.

Somehow, the premise doesn't quite make sense in a free market. How,
or why, would Crown make an inferior product for more money?

Because that is what the market is demanding. The manufacturers make
what they can sell.

Halfler made great sounding amps, but they had to shut their doors because
they couldn't make money doing it.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Arny Krueger]

"geezer" wrote in message

On May 12, 9:46 am, "Arny Krueger"
wrote:
"geezer" wrote in message

Google coughed up this technical paper, first on the
retrieval list:

www.crownaudio.com/pdf/amps/137234.pdf

What is there in this paper that needs further
explanation?

Guess I'm a
little unclear on the pulse width modulation thingie.

This might help:

http://en.wikipedia.org/wiki/Pulse-width_modulation

Also can't figure why other switch-mode amps seem to have
a somewhat inverse relationship with impedance load and
wattage, but the CTS2000 has a constant wattage into any
impedance between 2 and 8 ohms. Guess I'm just missing
something

Probably the simplist way to think about the relationship between power and
impedance is to think about a power amp like it is a power supply.

If you have a power supply that puts out the same voltage no matter what the
load, every time you cut the load impedance in half, the amount of current
is going to double, and so is the power delivered to the load. This is what
we call a constant voltage source.

If you have a power supply that puts out the same current no matter what the
load, every time you cut the load impedance in half, the voltage dropped
across the load is going to be cut by half, and so is the power delivered to
the load. This is what we call a constant current source.

The usual kinds of real-world power supplies fall someplace in between these
two limits: constant voltage or constant current.

The most common situation is where the power amp is more like a constant
voltage source than a constant current source. That means that the power
delivered to the load goes up as the load impedance goes down, but it does
not quite double with every halving of the load impedance.

Power amplifiers with output transformers are the next big notch down - the
power output is the same as the load impedance varies over a certain range
like 4-16 ohms. That's because the transformer transforms the load to
appear to be the same, despite the fact that it is actually being changed.
The transformer action changes as you pick different taps on the windings.

We almost never see audio power amps that act like constant current sources,
so I won't waste time talking about them.


I seriously doubt that in this day and age, Crown would
field an amplifier that simply failed to work or would
unconditionally sound bad,

That's what occurred to me. Crown makes decent enough
stuff, & has been around forever. I assume their going
to put a totally BS product..... OTOH, who would have
predicted "New Coke" ?

Despite the fact that Crown is a division of Harman International, I would
never expect them to pull some total BS like "New Coke". OTOH, New Coke is
the only kind of Coke I drink when I drink Coke. ;-)

-glenn

[Chris Hornbeck]

On Mon, 12 May 2008 13:30:56 -0400, "Arny Krueger"
wrote:

I question your analysis. There will be times during an output cycle where
the load will be dumping current back into the amplifier. The amplifier
either absorbs this current, or the output voltage of the amplifier will be
greater than the output voltage which corresponds to the input signal. If
the amplifier provides a low source impdance, the voltage across the load
will be 99.9%+ as determined by the input signal.

It seems possible that virtually all of the current that the load dumps will
be absorbed by the amplifier.

If the amplifier's output devices are only partially conducting, then much
of the power that the load is dumping will be dissipated in the output
devices. That is why SOA was such a big issue for traditional Class A-C
power amplifier designs.

If the amplifier is a switchmode amplifier then the output devices will be
conducting very heavily, and very little power will be dissipated in the
output devices.

What bugs me is the issue of common-mode current. Apparently
the output inductors are a much bigger part of the design than
is obvious in an otherwise interesting description.

Maybe the reflected power is a related issue. Wish I were
smart enough to say how.


If this amplifier is somehow "recapturing" that energy,
what on earth is it doing with it, how is it recapturing
it and where is it putting it?

As always, the law of conservation of energy will assert itself, so the
excess energy stored in the load will have to go *some place*. Thus, it will
be absorbed by any lossy components of the load, any amplifier output
filtering network, the amplifier output devices, and the amplifier's power
supply.

In a low-loss system, most of the energy will end up back in the amplifier's
power supply.

Well, yeah, a linear amplifier must dissipate reflected power.
Is that all they were saying?


Much thanks, as always,

Chris Hornbeck
"I have a gift for enraging people, but if I ever bore you,
it'll be with a knife." -Louise Brooks

[Arny Krueger]

"Chris Hornbeck" wrote in
message
On Mon, 12 May 2008 13:30:56 -0400, "Arny Krueger"
wrote:

I question your analysis. There will be times during an
output cycle where the load will be dumping current back
into the amplifier. The amplifier either absorbs this
current, or the output voltage of the amplifier will be
greater than the output voltage which corresponds to the
input signal. If the amplifier provides a low source
impdance, the voltage across the load will be 99.9%+ as
determined by the input signal.

It seems possible that virtually all of the current that
the load dumps will be absorbed by the amplifier.

If the amplifier's output devices are only partially
conducting, then much of the power that the load is
dumping will be dissipated in the output devices. That
is why SOA was such a big issue for traditional Class
A-C power amplifier designs.

If the amplifier is a switchmode amplifier then the
output devices will be conducting very heavily, and very
little power will be dissipated in the output devices.

What bugs me is the issue of common-mode current.

I think that "common-mode current" is similar to what we used to call
"simultaneous conduction" in p-p linear power amps, back in the old days
when SS devices weren't really fast enough.

The issue is back because switchmode amps are running at frequencies that
are about 10 times higher than the usual audio range, and efficiency is a
bigger issue.

Apparently the output inductors are a much bigger part of the design
than is obvious in an otherwise interesting description.

Switchmode power amps have this thing about inherently having big HF signals
inside the box. They must not be allowed to get out of the box, especially
in automotive applications. Also, there is FCC part 15.

Maybe the reflected power is a related issue. Wish I were
smart enough to say how.

The power is not so much reflected as phase-shifted.

If this amplifier is somehow "recapturing" that energy,
what on earth is it doing with it, how is it recapturing
it and where is it putting it?

As always, the law of conservation of energy will assert
itself, so the excess energy stored in the load will
have to go *some place*. Thus, it will be absorbed by
any lossy components of the load, any amplifier output
filtering network, the amplifier output devices, and the
amplifier's power supply.

In a low-loss system, most of the energy will end up
back in the amplifier's power supply.

Well, yeah, a linear amplifier must dissipate reflected
power.

Hence all the dark science related to SOA.

Is that all they were saying?

I think so... ;-)

[Chris Hornbeck]

On Mon, 12 May 2008 19:28:27 -0400, "Arny Krueger"
wrote:

What bugs me is the issue of common-mode current.

I think that "common-mode current" is similar to what we used to call
"simultaneous conduction" in p-p linear power amps, back in the old days
when SS devices weren't really fast enough.

Yeah, that's what I mean. "Common-mode conduction", there're
a variety of poorly performing names for the gig.

What strikes me about the "BCA" paper is that huge "common-
mode" currents seem to be built in. Zero signal level is
with both series'd switches *ON* (or off). Obviously,
something interesting is missing from my first glance.


Maybe the reflected power is a related issue. Wish I were
smart enough to say how.

The power is not so much reflected as phase-shifted.

Yeah, but it's still I^2 R, and the paper seems to imply
something special, something not possible in a linear
(conventional; you know what I mean) amplifier.

They seem to be implying a "re-use" of the reflected
power, like an electric car reuses braking energy.


If this amplifier is somehow "recapturing" that energy,
what on earth is it doing with it, how is it recapturing
it and where is it putting it?

As always, the law of conservation of energy will assert
itself, so the excess energy stored in the load will
have to go *some place*. Thus, it will be absorbed by
any lossy components of the load, any amplifier output
filtering network, the amplifier output devices, and the
amplifier's power supply.

In a low-loss system, most of the energy will end up
back in the amplifier's power supply.

Well, yeah, a linear amplifier must dissipate reflected
power.

Hence all the dark science related to SOA.

Is that all they were saying?

I think so... ;-)

Major bummer. At my level of ignorance, the common-mode
current puzzle and the reflected power puzzle could
have somehow been rolled together into something really
cool and interesting. And brought to you by the magic
of shocking the output inductors. It's so Tesla it ain't
funny.

Much thanks, as always,

Chris Hornbeck
"I have a gift for enraging people, but if I ever bore you,
it'll be with a knife." -Louise Brooks

[geezer[_2_]]

On May 12, 7:04*pm, (Scott Dorsey) wrote:

No, no, I didn't say it had crappy components. *I just said it sounded like
crap, because of the design considerations that went into it.

Well, you did say it had a cheap power supply and sounded crappy



Halfler made great sounding amps, but they had to shut their doors because
they couldn't make money doing it.

Yes they did, I put one of the Trans-novas in an install I did about 7
years ago.
I also like some of the old Adcoms real well, just put a gfa-555 on my
home system. replaced an NAD 216thx.

I thought Hafler folded because of incompetent management at Rockwell
Fosgate. Reminded me of when MTX bought Soundcraftsmen.

BTW, did you ever listen to any of their PCR units? I used to have one
of the little shoebox thingies.

-glenn

-glenn

[Scott Dorsey]

geezer wrote:
On May 12, 7:04=A0pm, (Scott Dorsey) wrote:

No, no, I didn't say it had crappy components. =A0I just said it sounded l=
ike
crap, because of the design considerations that went into it.

Well, you did say it had a cheap power supply and sounded crappy

If it had a linear power supply with conventional reservoir caps, it would
cost ten times as much. The whole point of this amplifier is to give you
very high efficiency for a very low price per watt.

Halfler made great sounding amps, but they had to shut their doors because=

they couldn't make money doing it.

Yes they did, I put one of the Trans-novas in an install I did about 7
years ago.

Yes. You'll notice they aren't making those amplifiers any more, though.
They did sound great, though.

I also like some of the old Adcoms real well, just put a gfa-555 on my
home system. replaced an NAD 216thx.

That is also an excellent conventional amplifier.

I thought Hafler folded because of incompetent management at Rockwell
Fosgate. Reminded me of when MTX bought Soundcraftsmen.

Could be, and it wouldn't be the first time. Still, class D amplifier
designs and conventional amplifier designs with switching supplies are
the hot thing on the market today, because they have vast amounts of
power without a lot of weight, heat, or money.

BTW, did you ever listen to any of their PCR units? I used to have one
of the little shoebox thingies.

I have not played with them at all.

But... for an example here.... my old Citation II has about 500 watts
idle power, and because it's nearly into class A, the input power does
not change much with signal. So, if you're playing something into some
monitors and driving about 10 watts, then the total system has about 2%
efficiency. This was about typical for the era.

If you look at a big Krell today, it's a little bit better, but not a
whole lot better. A Krell wil a similar power rating will have slightly
better efficiency but 5% is optimistic. There is a reason for all those
big heatsinks because the other 95% of the power is being dissipated as
heat.

Look into that Adcom, though, and you'll see something between 30% and
50% efficiency. Nearly half of the input power is coming out of the
output. You pay for that efficiency; it's not as linear as the Krell,
but it's still pretty damn good.

Now, take a look at one of those Class D amplifiers. You'll see efficiency
numbers in excess of 95%. The output stage is operating as a big switch,
so it's very efficient, and the 5% loss of power is mostly in RF being
eaten up by the integrator stage that turns the pulse train back into
baseband audio. This is what the market is demanding right now.

On the small end of things, people want their iPod to run for weeks on
a single battery. On the large end of things, people want to pump a
couple kilowatts into a subwoofer with something that plugs into a 20A
single-phase outlet. Efficiency is king for a lot of applications today,
and Crown, just like everyone else, is building amplifiers to fit that
market.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[geezer[_2_]]

On May 12, 7:28*pm, "Arny Krueger" wrote:

Hence all the dark science related to SOA.



OK, I'll bite. What are we talking about here? Safe Operation Area?
What dark science?

-glenn

[Scott Dorsey]

In article ,
geezer wrote:
On May 12, 7:28=A0pm, "Arny Krueger" wrote:

Hence all the dark science related to SOA.

OK, I'll bite. What are we talking about here? Safe Operation Area?
What dark science?

There is a whole lot of handwaving involved in figuring out safe operating
limits on output stages... in the case of Class B stages, the actual heat
produced is a function of the waveform the amplifier is handling, which
means you either can design for worst-case or expected-case or something
in-between.

There is a little propaganda available at:
http://powerelectronics.com/thermal_...ta-sheet-1107/
in an article that is pretty well-written and explains a lot about why
traditional design involves a little bit of cut and try engineering here.
--scott

--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[geezer[_2_]]

On May 12, 8:58*pm, (Scott Dorsey) wrote:

If it had a linear power supply with conventional reservoir caps, it would
cost ten times as much. * The whole point of this amplifier is to give you
very high efficiency for a very low price per watt.



Now, take a look at one of those Class D amplifiers. *You'll see efficiency
numbers in excess of 95%. *The output stage is operating as a big switch,
so it's very efficient, and the 5% loss of power is mostly in RF being
eaten up by the integrator stage that turns the pulse train back into
baseband audio. * This is what the market is demanding right now.

On the small end of things, people want their iPod to run for weeks on
a single battery. *On the large end of things, people want to pump a
couple kilowatts into a subwoofer with something that plugs into a 20A
single-phase outlet. *Efficiency is king for a lot of applications today,
and Crown, just like everyone else, is building amplifiers to fit that
market.

If I understand your broader point correctly, you generally don't like
amps with switching power supplies, regardless of the amplifier
class.
And you really dislike class D variant amps because the integrator
stage screws up the sound.

Are there _any_ more conventional A/B amps with SMPS that you think
are ok?

-glenn

[Scott Dorsey]

geezer wrote:

If I understand your broader point correctly, you generally don't like
amps with switching power supplies, regardless of the amplifier
class.

In general no. I have actually heard one amplifier with a switching supply
that sounded good, but it also had some heavy reserve capacitance on the
output of the supply.

One of the things about the switching supply is that, since the ripple filter
now only needs to filter out very high frequencies, you can get away with
tiny capacitors instead of a big 60 Hz filter bank. This may be part of the
issue because it reduces the amount of current you can deliver on peaks.'

And you really dislike class D variant amps because the integrator
stage screws up the sound.

No, the integrator stage is just part of the system, you can't really
look at it in parts. The whole system sounds bad. I don't know why
it sounds bad, but the reason why I don't like it is because every
Class D amplifier I have heard sounds like someone is jamming knives
into my ears.

Are there _any_ more conventional A/B amps with SMPS that you think
are ok?

Chevin Research has made a couple that weren't bad. In general, though,
I think trick power supplies are never a good idea. Why settle for "okay"
when there are so many great-sounding inexpensive amplifiers? Unless you
are in an application where you absolutely need the efficiency.

SMPS are also a major pain in the neck to repair when they fail.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Arny Krueger]

"geezer" wrote in message


On May 12, 7:28 pm, "Arny Krueger"
wrote:

Hence all the dark science related to SOA.

OK, I'll bite. What are we talking about here? Safe
Operation Area?

Yes.

What dark science?

If you have a good handle on designing SOA protection circuits, could you
share it?

[geezer[_2_]]

On May 13, 4:53*pm, "Arny Krueger" wrote:
"geezer" wrote in message



On May 12, 7:28 pm, "Arny Krueger"
wrote:
Hence all the dark science related to SOA.
OK, I'll bite. What are we talking about here? *Safe
Operation Area?

Yes.

What dark science?

If you have a good handle on designing SOA protection circuits, could you
share it?

Hell, no. I don't know nuthin'... I just wanted to know what you
were talking about, specifically.

So what are you talking about, specifically?

-glenn

[geezer[_2_]]

Ok, I could only access the first two pages for some reason.
Since the issue appears to thermal runaway, isn't this pretty much
resolved by using MOSFETs as the output devices?

And if you're building prototypes of a particular design using
transistors that would melt down in this way, wouldn't you just
purposely fry a few prototypes on your way to finding the limits of
that particular design? That doesn't seem too much like dark science,
not that I'm an EE or anything close.

-glenn


On May 13, 3:25*pm, (Scott Dorsey) wrote:

There is a whole lot of handwaving involved in figuring out safe operating
limits on output stages... in the case of Class B stages, the actual heat
produced is a function of the waveform the amplifier is handling, which
means you either can design for worst-case or expected-case or something
in-between.

There is a little propaganda available at:http://powerelectronics.com/thermal_...management_sim...
in an article that is pretty well-written and explains a lot about why
traditional design involves a little bit of cut and try engineering here.
--scott

--
"C'est un Nagra. *C'est suisse, et tres, tres precis."

[Scott Dorsey]

geezer wrote:
Ok, I could only access the first two pages for some reason.
Since the issue appears to thermal runaway, isn't this pretty much
resolved by using MOSFETs as the output devices?

It's changed. Using MOSFETs doesn't eliminate the problem, but it does
make the design a whole lot easier and it does prevent the thing from
failing in a horrible and catastrophic "Phase Linear" manner.

Thing is, MOSFET output stages sound different than bipolar ones. Whether
this is a good or bad thing depends on taste.

And if you're building prototypes of a particular design using
transistors that would melt down in this way, wouldn't you just
purposely fry a few prototypes on your way to finding the limits of
that particular design? That doesn't seem too much like dark science,
not that I'm an EE or anything close.

That's called "cut and try" engineering, and it used to be the only way
to make absolutely sure. Still, no matter how well you design the thing,
some customer will find a way to abuse it that you never thought of.

For the most part, this is good enough, unless you're trying to design
the whole thing on a hybrid or monolithic chip where it costs thousands
of dollars to cut a prototype, or if you are trying to push the limits
of the technology.

But to be honest, all of the problems come from trying to push the limits
of the technology... but that's what you have to do to sell a product.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Arny Krueger]

"geezer" wrote in message

On May 13, 4:53 pm, "Arny Krueger"
wrote:
"geezer" wrote in message



On May 12, 7:28 pm, "Arny Krueger"
wrote:
Hence all the dark science related to SOA.
OK, I'll bite. What are we talking about here? Safe
Operation Area?

Yes.

What dark science?

If you have a good handle on designing SOA protection
circuits, could you share it?

Hell, no. I don't know nuthin'... I just wanted to
know what you were talking about, specifically.

So what are you talking about, specifically?

I'm just saying that the design of SOA protection circuits still has some
elements of black art in them.

[William Sommerwerck]

I did the A-B between a K-1 and my fifty-year-old Citation II amp,
and the Citation II sounded hands down better than the K-1. Sure,
the K-1 could put out about 50 times the power, but what good
is power if it doesn't sound good?

The K-1 sounds so bad, that you don't even have to compare with anything
else to know how bad it is.

[William Sommerwerck]

Somehow, the premise doesn't quite make sense in a free market.
How, or why, would Crown make an inferior product for more money?

Stupidity. Particularly in the marketing department.

Not all Crown amplifiers are made and/or designed by Crown. The K-1,
unfortunately, is.