00%>Yahoo! is not affiliated with the authors of this page or responsible for its content.
Design Considerations for High-Output Portable Amplifiers 
Overview of Application Requirements
Since the late 1960s, touring sound has led the professional
audio industry in the development and application of new
technologies. This, coupled with the high profile of major
tours and their sound system providers, makes touring an
important proving ground for advances in sound reinforce-
ment technology.
The Show Must Go On, Efficiency is a Must, but Its
All About the Sound
Ruggedness and reliability, coolness under fire, watts per kilo,
initial costs and long-term operating costs; all of this matters
to the tour operator, but audio performance comes first with
the artist, whether on stage or behind the console. As ticket
prices continue to rise, the expectations of audiences rise
along with them. It takes more and more of an experience
to pry the ticket-buyer out of their home theater or away
from the internet. Income from concert tickets and merchan-
dise is becoming increasingly important to artists and the
music industry.
Total fidelity requires headroom, flat frequency response,
good full-range damping, controlled and non-destructive clip-
ping when pushed, and freedom from any kind of premature
breakup. In addition, signal alignment due to digital process-
ing, including any delays in the amplifiers, must now be
taken into account.
To summarize, concert amplifiers must be light but extremely
rugged, precise but powerful, high quality but cost-effective,
and more recently, must work equally well in digital and
analog systems.
Further Considerations
Rental companies demand flexibility. Inventory only earns
money when its on the road. The more a given piece of
equipment can do, the less time it spends in the warehouse.
So an amp that can drive subwoofers in bridged mono, 12
inch or 15 inch woofers in stereo, bi-amplified floor wedges
and even high-frequency drivers, is a better investment than
a one-trick pony.
Compatibility is also a key factor. Many sound systems use
one type of amplifier for all four frequency bands. This may
require adjustable input sensitivity to optimize the amp for its
designated role. Other systems use different power levels for
different frequency bands. When updating a system, owner/
operators of this type of sound system may prefer to replace
one frequency band at a time, requiring any new amps to
play well with the rest of the system.
Signal delay, adaptability to digital processors, remote moni-
toring and control, full protection against hazards and ac-
cidents, freedom from radio emissions, rack depth, adequate
mechanical strength, sufficient cooling these and many
other factors must be thoroughly tested before any piece of
equipment is ready to take to the road.
Installed systems too are increasingly expected to provide
concert sound performance. In fact, the primary difference
between many installed systems and concert systems is that
the amp racks of the former lack casters.
Next-Generation Power Amplification for
Portable Live Sound: Requirements
Challenges facing the designer of new power
amplifiers include:
Reducing size and weight.
Increasing output power; low-frequency amplifiers should
deliver 1000 -1500 watts per 8 ohm driver to take full ad-
vantage of the latest cone driver capabilities. This requires a
rating of at least 4000 watts per channel into 2 ohm loads.
Avoiding signal latency so that phase alignment can be as
precise and repeatable as possible; this is also a requirement
for backward compatibility with most existing amplifiers.
Providing reasonable adjustments and indicators that let op-
erators fit the amps into various system architectures, often
on short notice, without specialized equipment.
WHITE PAPER
Design Considerations for High-Output
Portable Amplifiers
Written by: Patrick H. Quilter
Date: April 30, 2007 
PFC vs. Fixed-Voltage Power Supplies
One feature not mentioned above is Power Factor Correc-
tion. In a typical active PFC circuit, an additional boost con-
verter is inserted in front of a standard switch mode supply
to pre-regulate its input while optimizing power draw from
the AC line. If sufficiently well-designed, this may also permit
the power supply to work on a wide range of AC voltages.
Unfortunately, this approach requires additional semicon-
ductor switches and control electronics and the additional
circuitry also reduces overall efficiency of the power supply.
However, its increasingly rare to transport the P.A. all over
the world. Usually there are acceptable rigs located wherever
large concerts are being promoted. Moreover, touring amps
are not connected directly to the wall socket, but to an AC
distribution system which may have the ability to supply the
correct voltage to the power amp racks. The bottom line is
that PFC can theoretically reduce peak AC current require-
ments, but we can do this with greater reliability and lower
cost by simply improving overall efficiency.
Upper Class, Lower Class:
Which Operating Class?
Clearly, amplifier efficiency is very important in concert sound
applications. However, audio performance is also critical.
Most modern amplifiers use switch mode power supplies for
weight reduction while still using various types of linear output
sections for high audio quality. Recently, switch mode tech-
nology (various forms of Class D) has been applied to high
power amplifiers, most successfully for low-frequency applica-
tions. Lower power midrange and high-frequency amplifiers
can still use Class H linear outputs to optimize audio perfor-
mance with acceptable efficiency.
So Lets Go to Class!
The earliest high power professional amplifiers used linear
Class B operation. The concept is simple and the perfor-
mance can be excellent. Positive and negative power transis-
tors are connected to positive and negative DC rails. When
idling, both transistors are basically off, so idle temperature
is low. A relatively simple phase splitter causes the positive
transistor to reduce resistance when we want to move the
speaker cone forward, and the negative transistor reduces
resistance to move the cone backward. Only one transistor is
working at any one time, so they never oppose each other,
thus reducing unwanted losses. However, losses within each
transistor are quite high when delivering typical, partial output
powers, since the devices are operating as variable resistors.
(Figure ) On music, losses can greatly exceed the average
power delivered to the speaker. Efficiency is very low, and
therefore it is difficult to scale this approach up to very high
power levels.
Figure  - Class B Amplifier with Losses
Line Level Input
Phase
Splitter
Class B
Positive
Output
Class B
Negative
Output
Amplified Class B Output
Positive Rail
Negative Rail
Shaded Areas
Show Relative
Losses
Speaker
Load 
We can reduce these losses significantly, especially at aver-
age music levels, by using a Class H system. This technique
adds one or more additional sets of power supply rails at
intermediate voltages. (Figure ) Special steering circuits
connect the output transistors to the nearest available rail,
thus greatly reducing the average voltage drop across the
output stage, which greatly reduces its resistive loss. It takes
considerable experience to do this without sacrificing audio
performance, but QSC has successfully used this technique
to double or triple the power capacity of a given platform.
However, heat sink size and voltage limitations of available
linear devices still limit the maximum safe power to about
2000 watts per channel in a 2RU chassis.
The fundamental problem with all linear designs is that ther-
mal losses are inherent to the technique, regardless of how
much one is willing to spend on the circuitry. To make further
progress, we must consider a fundamentally different way of
controlling the power we deliver to the load. Designers have
long understood that the only way to eliminate power losses
is to eliminate the resistance; an ideal semiconductor should
operate either fully on (as close to zero resistance as pos-
sible) or fully off (which avoids all current flow and losses).
We must drive the device as a switch instead of a resistor
and vary the power by controlling the percentage of time it
spends in the on state. This is the essence of switch mode
operation, and it is theoretically capable of nearly lossless op-
eration. This mode of operation was recognized early in the
history of amplifiers, and was assigned the letter D hence
the commonly used term, Class D amplification.
Figure  - Class H Amplifier with Losses
Line Level Input
Phase
Splitter
Class H
Positive
Output
Class H
Negative
Output
Amplified Class H Output
Positive Mid Rail
Speaker
Load
Positive Full Rail
Negative Mid Rail
Negative Full Rail
Shaded Areas
Show Relative
Losses 
Basic Elements of a Class D Amplifier
The basic principle of Class D is to operate the switches at
a frequency much higher than the audio range so that the re-
sulting pulses can be filtered and smoothed out leaving only
the desired audio signal. This takes three major processing
blocks plus