[Eric Desrochers]

Hello! I'd like to get opinions about the pros and cons of regulated
PWM supplies in car amps, in the real world.

Thanks!

--
Eric (Dero) Desrochers
http://homepage.mac.com/dero72

Hiroshima 45, Tchernobyl 86, Windows 95

[Matt Ion]

Eric Desrochers wrote:
Hello! I'd like to get opinions about the pros and cons of regulated
PWM supplies in car amps, in the real world.

The one major drawback would be the additional size, as well as probably heat
generatred, as they cram enough stuff into most HUs as it is. I can't think of
any REALISTIC benefits that would out-weigh that.

[KU40]

well the obvious benefit is constant power output, regardless of input
voltage. Disadvantages include no gain of power at higher voltages and
the amount of current drawn increases when voltage drops. So it's kind
of a circular problem at that point. voltage drops because current
draw is too much, but the amp sucks more current because the voltage is
dropping.


--
KU40

[Kevin McMurtrie]

In article ,
(Eric Desrochers) wrote:

Hello! I'd like to get opinions about the pros and cons of regulated
PWM supplies in car amps, in the real world.

Thanks!

They're all regulated or controlled to some degree, but not solid like a
preamp or computer supply.

With a properly balanced amplifier circuit, ripples and dips in the
power rails don't cause any problems. It's more efficient to drain the
amp rail capacitors a bit rather than try to pass enormous surges
through the whole system.

Despite not holding a constant voltage, audio power supplies are
designed with control over the output. Home systems sometimes have weak
magnetic coupling in the power transformer so that the power rails can
droop on low impedance loads without frying parts. Car amps will adjust
the voltage based on current draw and temperature to avoid overheating.

You'll not likely find a high power car amp that doesn't have a PWM
power supply. It's the simplest and most effective design these days.
Chips handle the complexities like regulation feedback, avoiding
transformer saturation and bias, bootstrapping, soft-start, thermal
overload, current overload, and MOSFET driving.

[Eric Desrochers]

Kevin McMurtrie wrote:

They're all regulated or controlled to some degree, but not solid like a
preamp or computer supply.

With a properly balanced amplifier circuit, ripples and dips in the
power rails don't cause any problems. It's more efficient to drain the
amp rail capacitors a bit rather than try to pass enormous surges
through the whole system.

Despite not holding a constant voltage, audio power supplies are
designed with control over the output. Home systems sometimes have weak
magnetic coupling in the power transformer so that the power rails can
droop on low impedance loads without frying parts. Car amps will adjust
the voltage based on current draw and temperature to avoid overheating.

You'll not likely find a high power car amp that doesn't have a PWM
power supply. It's the simplest and most effective design these days.
Chips handle the complexities like regulation feedback, avoiding
transformer saturation and bias, bootstrapping, soft-start, thermal
overload, current overload, and MOSFET driving.

Thanks for the info, all! It goes with what I read elsewhere and with
my own knowledge of electronic (which I studied). So an amp that is
fully regulated (models by xtant, JL Audio and Alto mobile comes to
mind) will hold its power rating better despite varying supply voltage.
Since audio power output is often quoted at an high supply voltage of
14.4 volts, a regulated model might keep its promise better at a more
realistic 13.8 volts, for example. Or 12.2 volts...

Now a more acute question : do you think a regulated supply will affect
the overall amp efficiency compared with a similar, unregulated model?

FInally, a rave and a rant!

I think the new CEA 2006 standardisation have cleared A LOT of fluff and
clutter in the spec sheets of equipment and I APPLAUDE those responsible
for its creation and the manufacturer who apply them!

OTOH, every manufacturer of class-D or D-variant amps like to talk about
the great efficiency of their amps but fail short of stating the actual
figure in %. Few reviewers actually post those figures either. I find
this info to be more relevant than say the damping factor (which is
already an order of magnitude higher than what is needed in most case)
and it's virtually impossible to find...

If you happen to have numbers for Premier PRS-D2000T or Kenwood
KAC-9109D...

Thanks!

--
Eric (Dero) Desrochers
http://homepage.mac.com/dero72

Hiroshima 45, Tchernobyl 86, Windows 95

[MOSFET]

Now a more acute question : do you think a regulated supply will affect
the overall amp efficiency compared with a similar, unregulated model?

Ooh..... Good question. I don't think I've ever heard that asked here
before (how does the efficiency of regulated amps compare with unregulated
amps).

My hunch (and this is a totally uneducated guess) is that efficiency would
tend to be the same as both are Class AB in operation (which ultimately
determines efficiency), it is only the power supply that is regulated.

MOSFET

[Eric Desrochers]

MOSFET wrote:

Now a more acute question : do you think a regulated supply will affect
the overall amp efficiency compared with a similar, unregulated model?

Ooh..... Good question. I don't think I've ever heard that asked here
before (how does the efficiency of regulated amps compare with unregulated
amps).

Ha! I wouldn't ask to this fine group if the info was readily available
on the web!

My hunch (and this is a totally uneducated guess) is that efficiency would
tend to be the same as both are Class AB in operation (which ultimately
determines efficiency), it is only the power supply that is regulated.

MOSFET

Yeah, in the case of the class AB, some more or less efficient supply
might get lost in the grossly innefficient actual amplification
circuits. But in the case of a class D amp, a few more % efficient PSU
becomes more important. In case you haven't realized yet, i'm an
efficiency freak!

Regards,

--
Eric (Dero) Desrochers
http://homepage.mac.com/dero72

Hiroshima 45, Tchernobyl 86, Windows 95

[Kevin McMurtrie]

In article ,
(Eric Desrochers) wrote:

Kevin McMurtrie wrote:

They're all regulated or controlled to some degree, but not solid like a
preamp or computer supply.

With a properly balanced amplifier circuit, ripples and dips in the
power rails don't cause any problems. It's more efficient to drain the
amp rail capacitors a bit rather than try to pass enormous surges
through the whole system.

Despite not holding a constant voltage, audio power supplies are
designed with control over the output. Home systems sometimes have weak
magnetic coupling in the power transformer so that the power rails can
droop on low impedance loads without frying parts. Car amps will adjust
the voltage based on current draw and temperature to avoid overheating.

You'll not likely find a high power car amp that doesn't have a PWM
power supply. It's the simplest and most effective design these days.
Chips handle the complexities like regulation feedback, avoiding
transformer saturation and bias, bootstrapping, soft-start, thermal
overload, current overload, and MOSFET driving.

Thanks for the info, all! It goes with what I read elsewhere and with
my own knowledge of electronic (which I studied). So an amp that is
fully regulated (models by xtant, JL Audio and Alto mobile comes to
mind) will hold its power rating better despite varying supply voltage.
Since audio power output is often quoted at an high supply voltage of
14.4 volts, a regulated model might keep its promise better at a more
realistic 13.8 volts, for example. Or 12.2 volts...

Now a more acute question : do you think a regulated supply will affect
the overall amp efficiency compared with a similar, unregulated model?

That would totally be dependent on the circuit. You can't generalize an
answer for that because it depends on what the limiting factors are in a
design.



FInally, a rave and a rant!

I think the new CEA 2006 standardisation have cleared A LOT of fluff and
clutter in the spec sheets of equipment and I APPLAUDE those responsible
for its creation and the manufacturer who apply them!

Aw, aren't you going to miss all of those 600W amplifiers that use 5-pin
amplifier chips running off 12V? Hell, let's use a 4 pin amp chip.
That bootstrap circuit costs another 85 cents.


OTOH, every manufacturer of class-D or D-variant amps like to talk about
the great efficiency of their amps but fail short of stating the actual
figure in %. Few reviewers actually post those figures either. I find
this info to be more relevant than say the damping factor (which is
already an order of magnitude higher than what is needed in most case)
and it's virtually impossible to find...

If you happen to have numbers for Premier PRS-D2000T or Kenwood
KAC-9109D...

Thanks!

[GregS]

In article , "MOSFET" wrote:
Now a more acute question : do you think a regulated supply will affect
the overall amp efficiency compared with a similar, unregulated model?

Ooh..... Good question. I don't think I've ever heard that asked here
before (how does the efficiency of regulated amps compare with unregulated
amps).

My hunch (and this is a totally uneducated guess) is that efficiency would
tend to be the same as both are Class AB in operation (which ultimately
determines efficiency), it is only the power supply that is regulated.

With an unregulated supply, the rails are higher than whats needed, so there
is too much power wasted in heat, equals less efficiency.

greg

[MOSFET]

With an unregulated supply, the rails are higher than whats needed, so
there
is too much power wasted in heat, equals less efficiency.

greg

Interesting. And I am not an EE so I know NOTHING about how an amp actually
operates. But just so I'm clear, you are seeing that with a regulated Class
AB amp, the rails are NOT higher than needed? I just want to be sure we are
comparing apples to apples.

MOSFET

[MOSFET]

Interesting. And I am not an EE so I know NOTHING about how an amp
actually
operates. But just so I'm clear, you are seeing that with a regulated
Class
AB amp, the rails are NOT higher than needed? I just want to be sure we
are
comparing apples to apples.

MOSFET

I meant "saying" not "seeing".

The reason I ask is that I thought with ALL Class AB amps the rail voltage
was ALWAYS higher than what was needed (regardless of regulated vs.
unregulated). This, of course, is the KEY thing that separates more
efficient amp designs (Class D, Class T, etc.).

MOSFET

[GregS]

In article , "MOSFET" wrote:
With an unregulated supply, the rails are higher than whats needed, so
there
is too much power wasted in heat, equals less efficiency.

greg

Interesting. And I am not an EE so I know NOTHING about how an amp actually
operates. But just so I'm clear, you are seeing that with a regulated Class
AB amp, the rails are NOT higher than needed? I just want to be sure we are
comparing apples to apples.

The rails are always higher than needed. When you speak of regulation, the rails
don't have to be smart, or follow along to a certain extent, with the music. The rails
could be regulated to a certain value and stay there, but that would be stupid
to do when you can control the rail so it for example, stays 5-10 volts above the
music waveform. Some systems have done this with two or three discrete
rail voltages, have been mostly home amplifiers. The more voltage differential
between the rail and the music waveform, just produces more I times E wasted
power across the transistors in the amplifier stage. The regulated power supply
is an infant form of switching amplifier, where the power supply is basicall acting also
as the amplifier stage directly, so that 5-10 voltage differential does not exist at
all, except for the voltage across the mosfets which is pretty small. We eliminated the transistors
and went to use mosfets because of their fast switching speeds.

I had one of the first amplifiers that had transistor switching power supply, It use a slow swiching
speed campared to the mosfets.

greg

[Kevin McMurtrie]

In article ,
"MOSFET" wrote:

Interesting. And I am not an EE so I know NOTHING about how an amp
actually
operates. But just so I'm clear, you are seeing that with a regulated
Class
AB amp, the rails are NOT higher than needed? I just want to be sure we
are
comparing apples to apples.

MOSFET

I meant "saying" not "seeing".

The reason I ask is that I thought with ALL Class AB amps the rail voltage
was ALWAYS higher than what was needed (regardless of regulated vs.
unregulated). This, of course, is the KEY thing that separates more
efficient amp designs (Class D, Class T, etc.).

MOSFET

A switching power supply can't raise the rails in the 1/80000 second
that could be needed. Switching power supplies that track the output
always risk clipping during a sudden noise.

Class D amps chop the rails into a fractional signal that's filtered
using a nearly lossless LC filter. The negative feedback is loop is
really tricky because the amp's gain is proportional to the rail voltage
and the chopped output lags the input. It's why most Class D amps are
subwoofer grade only.

Class H and variants are a Class AB amp with a Class B amp pushing its
power rails outward as needed through capacitor coupling. It's more
efficient but the high current pulsing in the power rails can cause
minor audio quality problems.

[MOSFET]

Thank you for your thorough explanation. I really appreciate it!

MOSFET

A switching power supply can't raise the rails in the 1/80000 second
that could be needed. Switching power supplies that track the output
always risk clipping during a sudden noise.

Class D amps chop the rails into a fractional signal that's filtered
using a nearly lossless LC filter. The negative feedback is loop is
really tricky because the amp's gain is proportional to the rail voltage
and the chopped output lags the input. It's why most Class D amps are
subwoofer grade only.

Class H and variants are a Class AB amp with a Class B amp pushing its
power rails outward as needed through capacitor coupling. It's more
efficient but the high current pulsing in the power rails can cause
minor audio quality problems.

[Kevin McMurtrie]

In article
,
Kevin McMurtrie wrote:

In article ,
"MOSFET" wrote:

Interesting. And I am not an EE so I know NOTHING about how an amp
actually
operates. But just so I'm clear, you are seeing that with a regulated
Class
AB amp, the rails are NOT higher than needed? I just want to be sure we
are
comparing apples to apples.

MOSFET

I meant "saying" not "seeing".

The reason I ask is that I thought with ALL Class AB amps the rail voltage
was ALWAYS higher than what was needed (regardless of regulated vs.
unregulated). This, of course, is the KEY thing that separates more
efficient amp designs (Class D, Class T, etc.).

MOSFET

A switching power supply can't raise the rails in the 1/80000 second
that could be needed. Switching power supplies that track the output
always risk clipping during a sudden noise.

Class D amps chop the rails into a fractional signal that's filtered
using a nearly lossless LC filter. The negative feedback is loop is
really tricky because the amp's gain is proportional to the rail voltage
and the chopped output lags the input. It's why most Class D amps are
subwoofer grade only.

Class H and variants are a Class AB amp with a Class B amp pushing its
power rails outward as needed through capacitor coupling. It's more
efficient but the high current pulsing in the power rails can cause
minor audio quality problems.

Actually, there are two kinds of Class H. The compact version uses
diodes and capacitors and a Class B amp to raise the rails, more like a
Class G. Another version has one Class B or D amp driving the common
line of an isolated power supply for another Class AB amp. The second
has good audio quality but it requires one extra power supply per
channel and it's only efficient if the first amp is Class D.


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