• Check out the Annual AK Secret Santa exchange Program, signup before November 23rd in the AK Charities & Christmas forum!

Echo:: I have a technical question for you.

R

rca2000

Super Member
I am posting this, as maybe it may be on the mind of some other AK'ers, also.

If you can, I would appreciate a detailed explanation of how a magnetic field amplifier operates. I THINK I understand it, (I have figured out things on my own, most people I talk to do not know, or care, how those amps work).

From what I have surmised:: A basic , direct-coupled amplifier, with more transistors, than the sink size in the amp would normally dictate, is used in these carver amps. The "magic", (if that is the word for it), seems to be the power supply. Small caps, that are MUCH too small for the rails(in a normal amp, anyhow), and a "transformer", that is connected, through a triac, to the power line. The transformer has an impedance MUCH too low to operate directly from the line, and so, it is just "barely " on, when there is no, or a low, signal. This is accomplished, by sensing how hard the amp is being driven, and then, modulating the power transformer drive, by that same amount. The b+ and b- rails are very low, just enough to turn on the output stage, until a demand is given, Then, the rails are increased, up and down, as the load varies. By doing this, the dissipated heat in the amp and tranny is low, or so it is claimed.

Is this right, and if so, isn't this pretty much the same as a "class H", amp? And, why the name "magnetic field amp"? There are no magnetic devices(like chokes or transformers )in the actual amp sections after all, just a specially driven "transformer", (in some ways, the power supply is more like a switcher, but it operates at much lower frequencies, than most switchers I have worked with.)

I also have a real feeling, that while this will work with music, the whole power supply, and amp, would probably "melt down", with a true sine wave tone, of say, 100hz., after a couple of minutes.
 
Register to hide this ad
Sounds like you have it down much better than I. Never opened one up, and never even heard one, but in years past I remember reading tech data on the amp, and I do remember thinking that is sounded a lot like a Class H-type amp, but with many more available 'steps'.

Maybe ilimzn will give us his thoughts.
 
rca2000 said:
...If you can, I would appreciate a detailed explanation of how a magnetic field amplifier operates... From what I have surmised:: A basic , direct-coupled amplifier, with more transistors, than the sink size in the amp would normally dictate, is used in these carver amps. The "magic", (if that is the word for it), seems to be the power supply. Small caps, that are MUCH too small for the rails(in a normal amp, anyhow), and a "transformer", that is connected, through a triac, to the power line. The transformer has an impedance MUCH too low to operate directly from the line, and so, it is just "barely " on, when there is no, or a low, signal. This is accomplished, by sensing how hard the amp is being driven, and then, modulating the power transformer drive, by that same amount. The b+ and b- rails are very low, just enough to turn on the output stage, until a demand is given, Then, the rails are increased, up and down, as the load varies. By doing this, the dissipated heat in the amp and tranny is low, or so it is claimed.
Is this right, and if so, isn't this pretty much the same as a "class H", amp?
Click to expand...

Yes, it is a class H amp. Actually, carver had several class h designs of which the 'Magnetic' power amp is the first.

rca2000 said:
And, why the name "magnetic field amp"? There are no magnetic devices(like chokes or transformers )in the actual amp sections after all, just a specially driven "transformer", (in some ways, the power supply is more like a switcher, but it operates at much lower frequencies, than most switchers I have worked with.)
Click to expand...

Although I have never seen one of these inside, I have seen one later incarnation of the technology. i am told there is now one more.
From what i can glean from the available literature, the power supply uses what one could most accurately describe as a flyback transformer.

I am not completely sure it uses a triac, although it is possible to do what it's doing with a fast gate-turn-off triac as a driver.
In essence, unlike a traditional transformer that has a soft magnetic characteristic and has (in theory) infinite inductance, the 'transofrmer' inside the carver amp is actually a magneticaly coupled pair of coils.

Basic electrics tells us that a coil tends to keep a constant current through it. It's inductance is a measure how good it is at that. If you put a constant voltage across a coil, current in it will rise inearly, at a rate inversely proportional to the inductance and proportional to the voltage. So, 1V across a 1H coil gives you a coil current that rises by 1A every 1s.

Now immagine what happens if at the point where the current reaches 1A we suddenly remove the voltage from the coil altogether, and substitute this with an open circuit: the coil will try to keep it's current constant - in other words, the voltage across it's terminals will theoretically increase to infinity as it tries to drive 1A into an open circuit! In practise, since inductors are not ideal, the voltage increases to a finite level, though very high.

If the same inductor has another winding, then that winding will see a voltage proportional to the first times ratio of windings.

AFAIK the MF amplifier uses this principle to modulate the voltage of it's power rails. The 'transformer' is not a true transformer, rather an inductor, and it is essentially used for energy storage - one side gets 'pumped' with appropriately gated mains, the other is used to 'pick off' a secondary voltage from the inductive 'flyback'. The longer the primary is on, the larger the flyback pulse and the higehr the output voltage.

Now, based on some of the literature (which is very scant as the principle was patented by Carver), the output circuit of the power supply is a sort of resonant circuit, which is why it is possible to store enough energy even with relatively small smoothing caps.

Hopefully this makes some sense - I am definitely not sure of the details and I may be quite wrong in interpreting the facts from the varius pseudo-sales-pitch blurb, so I don't really KNOW for sure, and, to be honest, neitehr have I met anyone who does.

rca2000 said:
I also have a real feeling, that while this will work with music, the whole power supply, and amp, would probably "melt down", with a true sine wave tone, of say, 100hz., after a couple of minutes.
Click to expand...

based on the reason why I had the next generation amp in for a fix, I would tend top agree. The amp uses many power transistors to provide the required safe operating area for them, but does not support long term full power use.
 
I have fixed a couple of Carvers over the years. I have the service manual for a PM175. I could post it if yall are interested. They do use a triac to control the how much of the sinewave is used. As I recall it is used to control the on time for each cycle. At power up the triac is fired for a fraction of a cycle which allows the low voltage rails to come up. All of the voltage rails are monitored by one op amp which then controls when the triac fires. This also gives an ideal way of having speaker protection without the need for relays in the signal path. If any faults exist the triac doesn't fire. I thought it was a well designed amp. As for running a 100 hz sinewave, no problem. When it come to pro sound gear, like whats used at concerts, Carver is well respected.
 
So VERY different than the mag-amps we used to drive the control-rod motors for the nuclear reactors I worked on in the Navy. The ones we used were truly magnetic amplifiers. A few transistors for control functions, but everything else was coils and transformers of one sort or another. But, IIRC, they weren't made by Carver. And weighed a few hundred pounds, each. And, with 280,000hp on tap, we didn't really care about having to store energy. When more was needed the throttle was opened and away we went.
 
Last edited:
Matt said:
I have fixed a couple of Carvers over the years. I have the service manual for a PM175. I could post it if yall are interested. They do use a triac to control the how much of the sinewave is used. As I recall it is used to control the on time for each cycle. At power up the triac is fired for a fraction of a cycle which allows the low voltage rails to come up. All of the voltage rails are monitored by one op amp which then controls when the triac fires. This also gives an ideal way of having speaker protection without the need for relays in the signal path. If any faults exist the triac doesn't fire. I thought it was a well designed amp. As for running a 100 hz sinewave, no problem. When it come to pro sound gear, like whats used at concerts, Carver is well respected.
Click to expand...

You would be surprised how low the long term RMS power is for an audio signal, even for hard rock ;), though if the amps are used just for the bass, it can get substantially higher. As I explained in the 'amplifier class' thread, class H amps exploit the fact that a class B amp efficiency can approach 70% just before clipping. In theory, that alowes the class H amp to have slightly less than half the heatsink size compared to a comparable standard class B amp. The amp I fixed was one of the Carver cubes, and this one, sadly did not strike me as adequatly heatsinked. Hopefully the older ones were.

Also, I don't think the power transformer was a regular transformer. Chopping the since with a triac would rather quickly overheat a regular transformer due to the high order harmonics in the waveform increasing the iron losses drastically unless it's a special construction, and IIRC it is. Carefully dimensioning the filter caps is also mandatory, as is sapplying a stray field shorting shield on the transformer.

I would love to see the power supply circuit, AFAIK the amp itself is nothing new. Do you have access to a scanner?
 
Yamaha B-2 said:
So VERY different than the mag-amps we used to drive the control-rod motors for the nuclear reactors I worked on in the Navy. The ones we used were truly magnetic amplifiers. A few transistors for control functions, but everything else was coils and transformers of one sort or another. But, IIRC, they weren't made by Carver. And weighed a few hundred pounds, each. And, with 280,000hp on tap, we didn't really care about having to store energy. When more was needed the throttle was opened and away we went.
Click to expand...

Ah, yes, but if I remember right, that is something completely different. This is an interesting and fairly old technique (reportedly one of the first uses was the gyro servo used in the WWII German rocket 'superweapon', the V2 - so definitely discovered before Carver was even born ;) ) which uses a special winding across a magnetic shorting part in a transformer to modulate the transformer current. Many subvariants are possible, some have direct mechanical control. A similar thing is also used in some electric welders, lead-acid battery chargers, etc.
 
Yes. Fairly crude. But extremely reliable. More modern stuff was all SCR and today is most likely power MOSFET, but not back then (talk about getting old!). These amps carried loads of over 100 amps with no problem. Quite industrial.

Can't tell you more. If I do I'll have to kill you.
 
I dont have a scanner, but I borrowed a digital camera. In order to make out the schematic I had to take two pics. This is just the p/s. In the center of the first pic are four diodes, and in the center of those is an opto-isolator. An op-amp monitors the voltage rails and increases the current through this opto-isolator increasing the on time of each half of the sinewave.
 
Sure like to see the rest of amp circuit. My guess is the opamp keeps the voltage across the output tranny at close to and still away from saturation. So far it is an adjustable power supply by using a triac to control the duty cycle of the 60 Hz.
 
The voltage rails remain fixed. But there are three B+/B- rails. One is at 21V, 44V and 71V. As the drive signal increases the collector voltage is brought up to 44V and then 71V. The schematic is a scanned copy i got off Ebay. Supposedly scanned at 1200 DPI, still barely readable. Anyways heres pics of one channel most of the control circuitry. Pic 855 shows the control circuitry that drives the P/S. The pin that I wrote "grey" is what drives the opto isolator in the P/S. Pic 854 show how the 71V rail gets turned on. Pic 852 is the output stage. In the upper/lower right corner are the transistors that swtitch the 44V rail on.
 
There are two transistors and another opamp driving the control opamp. Any chance of getting a view on where those get thier signal? :D
They are part of the power supply loop.
 
If you are referring to pic 855 the lower transistor is part of the protection circuit for high freqs. As the lower circuit is engaged the circuit above it drives the protect led. I think the opamp is looking for DC on the output. If you print the pics and can line up all 3 pics you can trace the signals for the one channel. I will include another pic of the whole schematic for ref, since it is unreadable.
 
You must log in or register to reply here.
Back
Top Bottom