FAQ  •  Register   •  Gallery •  Login

It is currently Thu Oct 30, 2014 11:06 pm

Magnetic Field Amp explained?

<<

RichP714

SILVER-7t

Posts: 1814

Joined: Fri Oct 12, 2007 2:12 pm

Post Mon Nov 05, 2007 11:16 am

Magnetic Field Amp explained?

<<

TNRabbit

User avatar

(pair of) SILVER-7 TUBE AMPs

Posts: 13338

Joined: Sun Jan 14, 2007 11:53 am

Location: TN Native Languishing in VA

Post Mon Nov 05, 2007 11:29 am

Here's the text in case the link goes bust (sorry for the errors):

Meeting Feature - The Carver M-400 Power Amplifier
The story of the magnetic amplifier begins just over two years ago, shortly before the January,
1978 CES. Bob Carver had just been ousted from Phase Linear in a boardroom coup. He
went to the CES, staying with an acquaintance from ESS because the room he had reserved was
unavailable to him. He was sitting outside the hotel, feeling very sad, when a manufacturer's
rep named Les Davis introduced himself. Having heard Carver's story, Davis offered encouragement
and support, assuring Carver that he could make a strong comeback within a year. The
timetable was slightly optimistic but the prediction has proved correct.
"I knew I had to have a really great idea, " said Carver, "or two, maybe ..." (much laughter).
The first idea was inspired by a clock he saw at Tiffany's in the shape of a lustrous golden
cube, which gave him the notion of trying to build a high-powered amplifier in a similar package.
The final result is amazingly small and light, contains three printed circuit boards, uses no exotic
parts, has a transformer made of ordinary materials (albeit of unusual design) which is
about the size of the transformer in a conventional 20-Watt amp, contains no heat sink other than
the chassis, and offers for $349 almost identical performance to the original Phase Linear 700,
which cost $800 ten years ago. It is conservatively rated at 200 Watts per channel continuous
power into 8 Ohms, and will put out over 300 Watts into both 4 and 2 Ohms. Unlike the original
Phase Linear, it will deliver lots of power into an electrostatic loudspeaker. It can be bridged
to operate as an extremely powerful monaural amplifier without additional hardware. Its dynamic
headroom is about 2 dB.
The Power Supply
The unusual power/weight and power/cost ratios of the magnetic amplifier are the result of
two things: an efficient power supply and an efficient output stage. First, we will discuss the
power supply.
- 9-
Figure 1 shows the power supply circuit of a conventional power amplifier, consisting of a
power transformer, a bridge rectifier, and two filter capacitors. The peak value of the rectified
voltage determines the voltage of the supply. In its quiescent state, i.e., when no power is
being drawn, no current flows through the rectifier. When the amplifier starts feeding power to
the load, current is drained from the capacitors and their voltage falls slightly. The rectifier
then conducts during the portion of the wave in which the voltage is higher than the voltage on the
capacitors. Because the voltage on the capacitors never drops by more than about 20%, even at
maximum output, the portion of the 60 Hz waveform during which conduction takes place is quite
small. In this design, all the energy is stored in the electrical field of the capacitors. The
power transformer has to be big, because at full power it is required to provide 120 short bursts
per second of very high current, but its capabilities are underused because the conduction angle
is small.
Figure 2 shows one of Carver's early attempts to build a more efficient amplifier, which he
used to introduce the concept of the magnetic power supply. Superficially, the elements of the
circuit are the same as in the conventional circuit: a transformer, a rectifier, and a filter capacitor.
Here, however, there is a triac in series with the primary of the transformer. A triac
is an electronic switch, controllable by an external voltage; it turns on and off during each half
cycle of the 60 Hz wave. This intermittent operation causes the transformer to behave like the
ignition coil in a car, as a field coil rather than as a transformer in the usual sense. When the
triac conducts, current flows in the primary, and a magnetic field builds up. The triac is turned
off as the voltage crosses zero, which is the point at which the current, and hence the magnetic
field, is at its peak. The resulting collapse of the field produces a very high voltage in the secondary,
which is then rectified to charge the filter caps. By properly controlling the triac, the
conduction angle can be made much larger than in a conventional design, increasing the efficiency
substantially, and allowing the use of a smaller transformer to produce the same voltage and
current output. This is a "magnetic" power supply, because it first stores energy in the magnetic
field around the primary coil, then dumps the energy into the capacitors.
Figure 3 shows the final version of the magnetic power supply. The power transformer,
now called the field coil, is approximately one fifteenth the size and weight of the transformer in
the original Phase Linear 700, and puts out nearly the same power. "It's not so much that this
one's unbelievably small;" Carver says, "it's that a conventional transformer is unbelievably
large." In this design, the field coil is chosen to have lots of leakage inductance, which in the
figure is drawn separately from the secondary. There is a connection from the filter capacitors
to the triac which causes the triac to turn off before the zero crossing. As before, when the triac
turns on, current flows and a magnetic field builds up in the field coil. In this version, there
is a capacitor in the secondary before the rectifier. This capacitor combines with the leakage
inductance of the field coil to form a resonant circuit. When the triac turns off, the secondary
circuit, instead of producing a single large pulse, goes into oscillation at a frequency of around
600 Hz, and its stored energy is then transferred through the bridge rectifier into the filter capacitors.
Because the frequency at which this takes place is relatively high, the caps can be
smaller' and still down adequate job of filtering. When heavy power demands drain the filter
caps; the triac stays on longer and the conduction angle increases.
Another clever scheme that helps make the most efficient use of the power supply is that the
two channels of the amp are operated out of phase, so that the "+" loudspeaker terminal of one
channel is at ground potential. The heaviest demands fall on the power supply during passages
with heavy bass, and in most recordings the lowest bass is made to be monaural to allow easier
cutting and tracking of the disc. When the same signal is fed to both channels, one will swing
positive while the other goes negative. In this way the positive and negative halves of the power
supply each take care of only one channel at any instant; in a normal power amp both channels
would be drawing current from half the supply while the other polarity lies idle.
The presence in the primary of the triac, which controls the power supply in response to an
externally applied voltage, gives the designer a chance to use protection circuitry that is effective
without compromising the amplifier's performance. Carver assumes, no doubt correctly,
that the low price of the unit will attract many customers who have never before owned a highpower
amplifier, and who are unaware of the dangers. Accordingly he designed the M-400 to resist
the most deliberate attempts to blow it up. The amp will shut down if either output voltage,
output current, or their product becomes excessive. The VI limiting does not take effect until
-10-
Fig. 1. Conventional power supply
Fig. 3. M-400 power supply
Fig. 2. Early attempt at more efficient power supply
Fig. 4. Power supply (P) and signal (S) waveforms
Conduction angle
a = No load
b = Full load (15% voltage sag)
Conduction angle
a = No load
b = Full load
(b) Carver M-400
- 1 1-
(a) Conventional amplifier
the current reaches 15 Amps/channel, and will allow for a load line with an imaginary:real ratio
of 2:1 (i.e., it will drive typically reactive speaker impedances). Too high a temperature will
shut it down, as will the presence of more than a few millivolts of DC in the output. There is a
circuit which, in effect, estimates the voice coil temperature of the speakers and prevents them
from overheating. This function is frequency-dependent, so that it protects midranges and tweeters
more stringently than woofers. It doesn't sense the very highest frequencies, though, because
that would cause the amp to shut down during the FTC preconditioning routine; instead it
guesses at the power being delivered to the tweeter by looking at the midrange. It is therefore
possible to burn out a tweeter by feeding it high frequency pure tones. This thermal protection
for the speakers is provided by a circuit that integrates the voltage over the previous two minutes
or so and shuts things down if the average power gets too high.
The final, and perhaps the most interesting, protection feature is one which shuts down the
power supply in response to large amounts of out-of-phase infrasonic material. According to
Carver, this circuit does not give trouble on actual music, but it renders the system immune to
dropping the stylus with the gain up, because that accident produces a huge out-of-phase thump.
The Output Stage (or, Throw Away That Heat Sink!)
The diagram in Figure 3 shows a single 80-Volt power supply. In reality there are three
separate supplies in the M-400, each with its own bridge rectifier. The others put out 50 and 25
Volts, and there are switching transistors which provide intermediate values of 65 and 37.5
Volts. In all, there are five different voltages to which the B+ and B- connections can be made,
and the amp contains an electronic switch which Carver calls a commutator that chooses among
the values according to the demands of the input waveform. The difference between this mode of
operation and the usual one can be seen in Figure 4. The behavior of the conventional circuit is
shown in Figure 4a. When power is demanded from the amp, the B+ falls slightly, and the ripple
increases somewhat. As more power is drawn, the signal and the B+ get closer together, and
when they touch, the amplifier clips. In Figure 4b we see the M-400 responding to a high level,
low frequency signal: the power supply voltages track the signal up and down. The most important
feature of this diagram is that the vertical distance between the power supply and the signal
voltages, which determines the amount of power that is dissipated in the transistor, is large in
the conventional amp, and small (never exceeding about 15 Volts) in the M-400. The output devices
are operated in non-switching class B mode, which prevents crossover notch distortion but
usually makes an amp run hot. Here, however, the dissipation is low, particularly at moderate
power levels where conventional amps dissipate the most power in their transistors. The M-400
is 33% more efficient than a typical conventional unit at full power, and about 300% more efficient
at one-third power.
Similar schemes for obtaining both high peak power and low quiescent dissipation have been
used by Hitachi ("class G") and Soundcraftsmen ("class H"). The Carver design falls somewhere
between these simpler forms and the so-called "smart power supply" amps in its degree of approximation
of the signal by the B+. The extra components needed for the more sophisticated designs
increase the complexity and cost of the circuit, but, as Carver says, silicon is cheap compared
to copper and iron, and silicon can be made to have a certain rudimentary form of intelligence.
The M-400 exemplifies the substitution of intelligence for brute force. In response to a
question about the feasibility of smaller amps being made in this way, Carver pointed out that the
savings are greatest in a high-powered amp, because that is where the most copper and iron can
be eliminated. The power supply of an ordinary 30-Watt amp is cheap enough that trimming some
copper and iron from it would not save enough to pay for the added complexity of the Carver design.
There is, however, a larger magnetic amplifier on the drawing board. A member of the
audience observed that the high power to weight ratio of the M-400 seemed to make it ideal for a
traveling rock band. Carver's reply is that the present model is optimized for home hi-fi use,
and that its protection circuitry would be a source of constant irritation and puzzlement to rockand-
roll sound men. He is designing a professional model, with higher rated power, to meet
this need. (Member of audience: "Since the professional model will have a higher rated power
than this one, how do you propose to keep audiophiles from buying it?" Carver: "I don't know.
Maybe I'll make it very ugly." Audience: "That never stopped them before ...")
- 12-
Carver considered using a high frequency switching power supply before he thought of the
present design. He shied away from this approach for several reasons: complexity, cost, unreliability,
and interference. A 20 kHz supply uses high frequency switching transistors which,
in order to be fast enough to do the job, have to have very thin base and emitter layers. These
thinly deposited layers can be broken through easily by relatively small overvoltages, so the
failure rate is high. The transients generated in a 2 kiloWatt, 20 kiloHertz switcher are tremendous,
necessitating a massive shield held together by screws one-quarter inch apart. As the
projected retail price of the unit climbed past $1,800 with no end in sight, Carver realized that
he was on the wrong track.
Carver offered the supposition that many engineers in the audience must be a bit disappointed
to find no exotic or mysterious parts inside the M-400. But he pointed out that the unit had to be
made of ordinary pieces, and not very many of them, if it was to gain the rapid acceptance that
he wanted for it. The conventional designs against which it is competing have undergone decades
of refinement, and the new approach had to have obvious advantages from the very beginning.
The M-400 has an unusual shape, an unusual name, unusually light weight, and the alluring veil
of mystery over its design. With this presentation, the veil was lifted, to reveal a design of
such obvious merit that we were moved to wonder why it hadn't been done before. Carver said
that he had heard that question a lot, and that he didn't know the answer.
After a brief intermission, during which an M-400 supplied some loud music at the front of
the room, Carver answered questions from the audience about the Sonic Hologram preamp (reports
are coming in that it works with all kinds of speakers, even the Bose 901, although omnidirectional
speakers as a class are the worst; a less expensive unit containing the preamp and
the hologram, but not the other features, will be out in about a year). A BAS member asked
about building his own magnetic amplifier and was told that the circuit diagram will be available,
but that Carver saves enough on parts by buying them in large quantities that the builder would
probably realize no cost advantage by doing it himself. Finally, he was asked if he had any other
bright ideas in the works. He replied that he did have a third one, and when someone from the
audience asked, "Is it bigger than a breadbox?" someone else remarked, "It used to be."
-- Brad Meyer and Peter Mitchell
First Impressions of the M-400
Al Foster ran some bench tests on a prototype of the M-400 the day of the meeting. The
perils of testing prototypes are many, but a brief look at one of the first production models has
revealed no big differences. With both channels driving resistive loads, the prototype M-400
put out 276 Watts (continuous) into 8 Ohms, 361 Watts into 4 Ohms, 312 Watts into 2 Ohms, and
202 Watts into 1.1 Ohms. In the bridged (mono) mode, using a tone burst consisting of a single
cycle of 1 kHz with a duty cycle of about 0.5%, peak power readings of 756 Watts into 1.1 Ohms
and 1,024 Watts into 4 Ohms were obtained. At 12.5 Watts into an 8 Ohm resistive load, the
distortion (THD) was 0.05% at 1 kHz and 0.047% at 20 kHz.
A brief listening comparison was made at my house between another Carver prototype and a
pair of Audionics CC-2s operating in bridged mode, driving a pair of Snell Type A speakers,
which have an impedance of roughly 4 Ohms. I played a tape recording of a thunderclap, which
has a very brief, high-intensity peak with its greatest energy in the 125 and 160 Hz one-thirdoctave
bands, and measured peak sound pressure levels. The peak is so brief that it is easy to
light the clipping indicators on any amp I have tried without the sound being uncomfortably loud.
In this test, the Carver was from 0 to 1 dB louder than the Audionicses (Audionices?) which, considering
the 60% cost saving, is very good indeed. The CC-2s have a mechanical hum which is
annoying at night when the room is very quiet; the Carver had the same problem, compounded by
the presence of higher harmonics because of the switching in the power supply. Carver is aware
of this problem, and has apparently fixed it in the production version.
There was only one piece of musical material on which the two amplifiers seemed to behave
very differently: the M&K organ record entitled "The Power and the Glory," Volume 2. This
disc has substantial amounts of energy below 20 Hz cut into it, and because it was made with
spaced omnidirectional microphones much of the deep bass is out of phase. This caused the
M-400's 0 dB LED to light at levels 10 dB or so lower than the CC-2's, and what was either clipping
or power supply limiting could be heard at sound levels that were only moderately high. It
- 1 3-
must be said, however, that the extent of the low bass on this record, and its randomness of
phase, are positively freakish, and that on normal musical material the Carver could play louder
than I wanted to listen.
The man who brought over the Carver and I did some listening comparisons with music, using
the archaic molasses-slow A-B technique, with only a cursory attempt at level matching. We
agreed that the amps sounded very similar, but not exactly alike. He preferred the CC-2s very
slightly over the Carver, while I liked the Carver a bit better. Our impressions of the difference
were similar as well. The CC-2s sounded warmer (if you preferred them) or more veiled (if you
didn't), while the Carver sounded more transparent (if you preferred it) or slightly strident (if
you didn't). Frequency response measurements, taken while the speakers were attached, showed
differences of about 0.3 dB whose character coincides rather well with our informal subjective
judgments. Because the Carver is a prototype, the measurements say more about the correspondence
between subjective impressions and measurement than they do about the performance
of the M-400, so the graphs in Figure 5 are not labeled by name; you must read the text to identify
them.
-- Brad Meyer
Postscript: Bob Carver has provided an explanation for the behavior of the M-400 on the M & K
organ record. It seems that the prototype amplifiers have signal-sensing circuitry which responds
only to the positive half of the waveform. When the bass is monaural this is no problem,
because the demands on the two halves of the supply are symmetrical. With out-of-phase bass
the negative half encounters a wide voltage swing without any large signal in the positive half, so
the triac never gets the message and the required voltage increase does not take place. The amp
therefore clips much sooner than it otherwise would. Production units sense both sides of the
supply; Al Foster has tested one of these with two oscillators, one for each channel, running at
slightly different low frequencies, and reports no loss in power output. -- EBM
+1dB
20 40 80 160 315
630
1250
Frequency, Hz
2500 5000
Fig. 5. Frequency response of Carver M-400 and Audionics CC-2
10,000 20,000
- 14-
-1dB
0
The Boston Audio Society does not endorse or criticize products, dealers, or services. Opinions
expressed herein reflect the views of their authors and are for the information of the members.
TNRabbit
Image

From FrankieD's lips to your ears: Sunfire - a quiet box of endless power.

Sunfire TG-IV/400~7 Amp
Carver SD/A-360 CDP
Benchmark DAC-1
Sony SACD/DVD-A
Active bi-amp: Ashly XR-1001 & 2 Rane PEQ-15s
Main: HotRodded AL-IIIs
Sub: Klipsch RT-12d
Center: Sunfire CRS-3c
Surround: Sunfire CRS-3 (x 2)

OconeeOrange wrote:"Gary likes to play it 'loud' as do I. His system begs you turn it up until you die"

RIP WIlliam B. Dibble, 1948-2012. I'll miss you my friend.
<<

bob p

User avatar

SILVER-7t

Posts: 1340

Joined: Wed Jul 18, 2007 10:40 pm

Location: Chicago

Post Mon Nov 05, 2007 2:36 pm

Funny that you would mention the Boston Audio Society. They had some pretty interesting interactions with Bob Carver. They did a review of Sonic Holography and pointed out some things they didn't' like.

Middle of page 8.

http://www.bostonaudiosociety.org/pdf/b ... 3-7912.pdf

I guess Bob didn't like the review. Page 3.

http://www.bostonaudiosociety.org/pdf/b ... -8001b.pdf
Carver Preamps: C-1, C-2, BillD's JVD-modded C-4000, C-9
Carver Amplifiers: BillD's C-500, M-1.5t (4) PM-1.5 (4) M-500t (2)
Repair/Restoration/Upgrade expert for all of these components.
<<

TNRabbit

User avatar

(pair of) SILVER-7 TUBE AMPs

Posts: 13338

Joined: Sun Jan 14, 2007 11:53 am

Location: TN Native Languishing in VA

Post Mon Nov 05, 2007 2:55 pm

bob p wrote:Funny that you would mention the Boston Audio Society. They had some pretty interesting interactions with Bob Carver. They did a review of Sonic Holography and pointed out some things they didn't' like.

Middle of page 8.

http://www.bostonaudiosociety.org/pdf/b ... 3-7912.pdf

I guess Bob didn't like the review. Page 3.

http://www.bostonaudiosociety.org/pdf/b ... -8001b.pdf


That was some interesting reading....did you continue to read past page 3 in the latter article? A couple of the other folks blasted the reviewer as well for making it "personal"...
TNRabbit
Image

From FrankieD's lips to your ears: Sunfire - a quiet box of endless power.

Sunfire TG-IV/400~7 Amp
Carver SD/A-360 CDP
Benchmark DAC-1
Sony SACD/DVD-A
Active bi-amp: Ashly XR-1001 & 2 Rane PEQ-15s
Main: HotRodded AL-IIIs
Sub: Klipsch RT-12d
Center: Sunfire CRS-3c
Surround: Sunfire CRS-3 (x 2)

OconeeOrange wrote:"Gary likes to play it 'loud' as do I. His system begs you turn it up until you die"

RIP WIlliam B. Dibble, 1948-2012. I'll miss you my friend.
<<

bob p

User avatar

SILVER-7t

Posts: 1340

Joined: Wed Jul 18, 2007 10:40 pm

Location: Chicago

Post Mon Nov 05, 2007 3:42 pm

Yes, I am familiar with the entire article. Regardless of whether the review was personal or not, I can't say that that the reviewer was wrong in his assessment of the noise problems in what was essentially a prototype piece of gear. As Carver Forum member Jvandyketexas has already pointed out, there are some significant noise issues in the production C-4000 circuit. He's developed mods to fix them.

What I thought was most interesting was Bob Carver's response to the reviewer:

Bob Carver wrote:ouch! That was hurtful. You're wrong, Brad. A fully-functioning, properly set up Sonic Hologram is spectacular.


Ouch, that hurts, you're wrong??? I hate to say it on a Bob Carver fan site, but that sounded kind of childish to me.
Carver Preamps: C-1, C-2, BillD's JVD-modded C-4000, C-9
Carver Amplifiers: BillD's C-500, M-1.5t (4) PM-1.5 (4) M-500t (2)
Repair/Restoration/Upgrade expert for all of these components.
<<

angelod307

SILVER-7 TUBE AMP

Posts: 2724

Joined: Sat Jun 06, 2009 10:56 pm

Location: manning, sc

Post Thu Sep 09, 2010 12:15 am

Re: Located in Marietta Georgia

pang55215 wrote:SPAM [-X



why post this?
CarverFest: I just go for the arts and crafts
<<

BillD

User avatar

(pair of) SILVER-7 TUBE AMPs

Posts: 7295

Joined: Wed Jun 06, 2007 8:29 pm

Location: The west's most mid-western town, Scottsdale, Arizona

Post Thu Sep 09, 2010 10:03 am

Re: Magnetic Field Amp explained?

It's spam
It should sound like it isn't there!
There is a difference between hearing and listening...
Making life enjoyable through expensive electronics.
_________________
Carver: C-4000 & C-1 preamps, PSC-60 preamp/tuner, TX-11a tuner, M-400 (2), C-500, M-500, M-500t, M-500t Mk.II, A-500x, AL-III loudspeakers (2 pr.)
Sunfire:Theater Grand III processor, Ultimate Receiver, Cinema Grand Signature 400 ~ seven, True Subwoofer Mk. II, D-10 Subwoofer

Return to Bob Carver, the Myth, the Legend

Who is online

Users browsing this forum: No registered users and 1 guest

Powered by phpBB © 2011 phpBB Group.