Calculating Wattage of Class A/B Amplifiers

[BrettMendes]

From my understanding, a class A amplifier will double its wattage output with each doubling of impedance. I.E. if it puts out 100 watts at 8 ohms it will produce 200 watts at 4 ohms. Is there a similar formula for class B, A/B, and D amplifiers? I ask because some amplifiers, specifically the Marantz MM7055 and MM8077, can surely drive a 4 ohm load, but only post ratings at 8 ohms.

Thanks,
Brett

 

[ImcLoud]

Marantz are only good down to 6 ohm, I wouldn't use 4 ohm speakers with them, especially when there are amps out there for less money that are rated for 4 ohms.. Also it doesn't always double when you go from 8 to 4 or 4 to 2, just like it doesn't always double when you bridge, you just need to rely on the manufacturers specs. Marantz usually is marked 6 ohm. I try to always buy equipment that is 4 ohm stable just so when I am swapping speakers I don't have to worry about it, sure them amps will run a 4 ohm load, but for how long? and at what sound quality cost?

 

[ski2xblack]

I think you're getting your power supplies confused with the output stage. Doubling down really doesn't happen even with the amps with big regulated power supplies (although they get the closest to meeting that ideal) and it adds a lot of cost to the amp. Think Krell, Accuphase, and other very expensive kit. The bulk of gear available uses unregulated power supplies and won't double down, and if the manufacturer claims it does, they may be deflating the 8 ohm power ratings in order to make that claim. It's a confusing subject, particularly if you're a hobbyist and not an engineer, but there are lots of articles and threads on the topic here at Audioholics.

Now, as far as your specific questions on the Marantz gear, I don't know why they don't publish 4 ohm ratings, nor if your speakers would tax them.

Do you happen to have an impedance/phase plot for your speakers? That would help determine if they're easy to drive, or amp-suckers. Chances are good that they are not amp-suckers, and that one of those Marantz models would do just fine.

 

[fuzz092888]

Marantz doesn't rate their amplifiers down to 4 ohms because the amps aren't rated down to 2 ohms, or put another way, speaker manufacturers will rate their speakers at 4 ohms, but the speakers may have dips down to 2 ohms or lower which would severely tax the amp. This is why they only rate the amplifier down to 8 or 6 ohms. It's a legal thing to avoid responsibility for unreliable speaker ratings.

Also, there is no formula and not all class A amps double down on top of ski said.

 

[Swerd]

Doubling down really doesn't happen even with the amps with big regulated power supplies (although they get the closest to meeting that ideal) and it adds a lot of cost to the amp.

The bulk of gear available uses unregulated power supplies…

Can you or anyone explain, in plain English for non-EE types, what a regulated power supply is, and what it means for an audio amplifier?

The only thing I know about audio amplifiers with regulated power supplies are these two links:

avahifi - The Brand New Synergy Power Amplifiers

[url]http://www.enjoythemusic.com/magazine/equipment/0912/audio_by_van_alstine_synergy_450.htm
[/URL]

 

[fuzz092888]

Can you or anyone explain, in plain English for non-EE types, what a regulated power supply is, and what it means for an audio amplifier?

The only thing I know about audio amplifiers with regulated power supplies are these two links:

avahifi - The Brand New Synergy Power Amplifiers

Audio By Van Alstine (AVA) Synergy 450 Power Amplifier Blockbuster performance for only $1999! Review By Anthony Nicosia

Try these, they mostly made sense and I think it might be what you're looking for.

http://www.extron.com/download/files/whitepaper/powersupp_reg.pdf

Solid State Power Amplifier Supply Part 1

Audio amplifier power supply design - Part 1: Power supply types & transformer considerations

 

[BrettMendes]

I think you're getting your power supplies confused with the output stage. Doubling down really doesn't happen even with the amps with big regulated power supplies (although they get the closest to meeting that ideal) and it adds a lot of cost to the amp. Think Krell, Accuphase, and other very expensive kit. The bulk of gear available uses unregulated power supplies and won't double down, and if the manufacturer claims it does, they may be deflating the 8 ohm power ratings in order to make that claim. It's a confusing subject, particularly if you're a hobbyist and not an engineer, but there are lots of articles and threads on the topic here at Audioholics.

Now, as far as your specific questions on the Marantz gear, I don't know why they don't publish 4 ohm ratings, nor if your speakers would tax them.

Do you happen to have an impedance/phase plot for your speakers? That would help determine if they're easy to drive, or amp-suckers. Chances are good that they are not amp-suckers, and that one of those Marantz models would do just fine.

Here's the scenario: I have a set of B&W CM5s and a that I intend to eventually use as rears with the CM9s as my FR and FL, and the CMC2 as my center. Currently, I have no trouble driving my bookshelves with my Marantz SR6006 and am very pleased with the way they sound. I'm just looking towards the future. I got the "doubling down" info from a review of the Gryphon Colosseum, which I obviously will never own, but assumed it was true of all class A amplifiers and thus I was assuming there would be a similar formula for other classes. I don't have a phase plot for the CM9s but deduced from the Audioholics review of the CM8s that they would likely be 4 ohm speakers in reality. Further searches through the forums seem to support this theory as most people characterize them as a very difficult load.

Truly, I'm a long ways from being able to purchase any of this kit, but am trying to figure out what I should be saving up for. I'm mostly drawn to Marantz because of the aesthetic and the fact that I can get the gear at a discount, but if it's not going to do the speakers justice then it's a wash, you know?

By the way, Hello from Denver!

 

[ski2xblack]

Can you or anyone explain, in plain English for non-EE types, what a regulated power supply is, and what it means for an audio amplifier?

I think the confusion may just be a semantic thing. Functionally, a linear regulated supply means the amp can be made to operate as a perfect voltage source, i.e. double down.

Thanks for the links, fuzz. I like the eetimes one the best, straight and to the point.

 

[BrettMendes]

Try these, they mostly made sense and I think it might be what you're looking for.

http://www.extron.com/download/files/whitepaper/powersupp_reg.pdf

Solid State Power Amplifier Supply Part 1

Audio amplifier power supply design - Part 1: Power supply types & transformer considerations

Very lucid articles. Thank you much.

 

[Irvrobinson]

Can you or anyone explain, in plain English for non-EE types, what a regulated power supply is, and what it means for an audio amplifier?

For audio amplifiers, a regulated power supply means that there is active circuitry regulating the voltage levels on the power rails for the output stage transistors, meaning that it holds the voltage at a certain level. This technique virtually guarantees that the amplifier actually "doubles down", so long as the mains power is sufficient, and the rail voltage is chosen so that the rest of the power supply is sufficient for the current required by the lower impedance levels.

 

[3db]

I think the confusion may just be a semantic thing. Functionally, a linear regulated supply means the amp can be made to operate as a perfect voltage source, i.e. double down.

Thanks for the links, fuzz. I like the eetimes one the best, straight and to the point.

The term regulated power supply simply means that there is some feedback mechanism employed to maintain either a constant voltage or current within the intended design limits. Nothing more.

 

[ski2xblack]

Greetings from Longmont, Brett.

I would suggest doing what ImcLoud suggests, also mentioned in fuzz's links by the way, and err on the side of caution, get something with reliable low impedance drive capability, and with enough power that you're nowhere near clipping. I'm not knowledgable regarding all those B&Ws you have, but some of their models are notorious power hogs.

 

[Swerd]

I have a set of B&W CM5s and a that I intend to eventually use as rears with the CM9s as my FR and FL, and the CMC2 as my center. Currently, I have no trouble driving my bookshelves with my Marantz SR6006 and am very pleased with the way they sound. I'm just looking towards the future…

I wouldn't worry about it. Although the the CM9s may be a 4 ohm impedance speaker, I don't think they qualify as very difficult to drive. If and when you get the CM9s, try them with your Marantz receiver. If it becomes a problem, the overload protection circuits will shut it down.

The Marantz SR6006 does have preamplifier output jacks that allow you to add an external amplifier while using the receiver as a preamp. For example, you could get a 2-channel external amp of your choice (forget the double-down Class A business, it's largely marketing nonsense) to drive the CM9s, while using the Marantz to drive the center and rear speakers.

I'm sorry I highjacked your thread before you got answers to your original question.

 

[TLS Guy]

From my understanding, a class A amplifier will double its wattage output with each doubling of impedance. I.E. if it puts out 100 watts at 8 ohms it will produce 200 watts at 4 ohms. Is there a similar formula for class B, A/B, and D amplifiers? I ask because some amplifiers, specifically the Marantz MM7055 and MM8077, can surely drive a 4 ohm load, but only post ratings at 8 ohms.

Thanks,
Brett

Whether an amp is Class A or A/B or any other class, has nothing do with current output into the load.

The ability to increase current without dropping the supply voltage and therefore power as load impedance drops is a function of power supply design and how they are used.

Class A biasing means high quiescent current and inefficiency and tends to a shorter life for the output devices, since they are always "on" and therefore hot. This is a significant price to pay to eliminate all crossover distortion. The best bets are the feed forward designs of people like Peter Walker and Nelson Pass, which in essence have a small very good low powered class A amp to correct the A/B output stage. Peter Walker called it current dumping. It can be mathematically proven that the performance is class A without the huge disadvantages of biasing a powerful output stage to class A. These types of circuits perform well and have proved to be very durable and reliable. The patents have expired and this approach is not used nearly enough.

Swerd an unregulated power supply has a transformer, rectifier and smoothing caps, and served us well for years.

Regulated power supplies have chip circuits,in addition to the above components.

When fed an input voltage in a specified range of voltages higher than the output, these devices help keep a constant voltage at the output over a wide range of input voltages and load current. However the transformer and power supply caps have to be adequately sized, as the regulator must see in input voltage at least at its minimum specified. This is always above the output voltage. A regulator can not boost voltage.

This is how your laptop power supply will work at 120 volts US and 240 volts UK, without you having to think about it, or throw a switch. If it was an unregulated power supply and you plugged it in in the UK, you would fry you computer in a puff of smoke instantly.

 

[BrettMendes]

Whether an amp is Class A or A/B or any other class, has nothing do with current output into the load.

The ability to increase current without dropping the supply voltage and therefore power as load impedance drops is a function of power supply design and how they are used.

Class A biasing means high quiescent current and inefficiency and tends to a shorter life for the output devices, since they are always "on" and therefore hot. This is a significant price to pay to eliminate all crossover distortion. The best bets are the feed forward designs of people like Peter Walker and Nelson Pass, which in essence have a small very good low powered class A amp to correct the A/B output stage. Peter Walker called it current dumping. It can be mathematically proven that the performance is class A without the huge disadvantages of biasing a powerful output stage to class A. These types of circuits perform well and have proved to be very durable and reliable. The patents have expired and this approach is not used nearly enough.

Swerd an unregulated power supply has a transformer, rectifier and smoothing caps, and served us well for years.

Regulated power supplies have chip circuits,in addition to the above components.

When fed an input voltage in a specified range of voltages higher than the output, these devices help keep a constant voltage at the output over a wide range of input voltages and load current. However the transformer and power supply caps have to be adequately sized, as the regulator must see in input voltage at least at its minimum specified. This is always above the output voltage. A regulator can not boost voltage.

This is how your laptop power supply will work at 120 volts US and 240 volts UK, without you having to think about it, or throw a switch. If it was an unregulated power supply and you plugged it in in the UK, you would fry you computer in a puff of smoke instantly.

Are there companies still using Walker and Pass's design? That sounds very interesting and worth reading about.

 

[TLS Guy]

Are there companies still using Walker and Pass's design? That sounds very interesting and worth reading about.

Yes, the Quad QSP is a current dumper.

Nelson Pass are still making amps.

 

[3db]

Whether an amp is Class A or A/B or any other class, has nothing do with current output into the load.

The ability to increase current without dropping the supply voltage and therefore power as load impedance drops is a function of power supply design and how they are used.

Class A biasing means high quiescent current and inefficiency and tends to a shorter life for the output devices, since they are always "on" and therefore hot. This is a significant price to pay to eliminate all crossover distortion. The best bets are the feed forward designs of people like Peter Walker and Nelson Pass, which in essence have a small very good low powered class A amp to correct the A/B output stage. Peter Walker called it current dumping. It can be mathematically proven that the performance is class A without the huge disadvantages of biasing a powerful output stage to class A. These types of circuits perform well and have proved to be very durable and reliable. The patents have expired and this approach is not used nearly enough.

Swerd an unregulated power supply has a transformer, rectifier and smoothing caps, and served us well for years.

Regulated power supplies have chip circuits,in addition to the above components.

When fed an input voltage in a specified range of voltages higher than the output, these devices help keep a constant voltage at the output over a wide range of input voltages and load current. However the transformer and power supply caps have to be adequately sized, as the regulator must see in input voltage at least at its minimum specified. This is always above the output voltage. A regulator can not boost voltage.

This is how your laptop power supply will work at 120 volts US and 240 volts UK, without you having to think about it, or throw a switch. If it was an unregulated power supply and you plugged it in in the UK, you would fry you computer in a puff of smoke instantly.

Like I said, A feedback mechansim is employed to maintain either a constant current or voltage. That mechanism is usually doen by active components like transistors, back when I hand built a power supply them or now by ICs...having many many transistors.

 

[slipperybidness]

For the record, there are advantages and dis-advantages of regulated vs un-regulated power supplies.

From the reading that I have done, I tend more towards the camp of un-regulated supplies.

 

[Swerd]

Thanks everyone for your comments about regulated power supplies.

TLS Guy, thanks especially for reminding us that "a transformer and power supply caps have to be adequately sized… This is always above the output voltage. A regulator can not boost voltage". Conservation of energy and matter is a law of physics that applies everywhere, including audio.

I also appreciate the fact that unregulated power supplies have served us well for years in audio.

But there is a reason why I asked about this. A number of years ago, as an experiment, Frank Van Alstine built a completely balanced audio system.

… It required a four-channel DAC analog filter and audio circuit set, a four-channel preamp, and a four-channel power amp (or two stereo power amps) maintaining balanced operation from input to output, summing the signal at the speakers. In theory, this canceled all common mode distortion in the system, and provided audibly superior musical operation, depending upon how much distortion was originally there to cancel. Of course it essentially doubled the cost of the audio system.

Although this was an experiment, it had positive results. The audible improvement in this fully balanced system alerted us that there must be distortion from components in the audio chain that we had not yet addressed.

This sent us on a serious effort to discover remaining weak links in the audio chain we produce and fix them at a reasonable cost, if possible. The result was significantly improved performance for several of our products. When these improved components were tried again in a fully balanced system, they resulted in only minimal improvements. The whole exercise sure was helpful in putting up red flags that we were not as good as we had thought we were and helping us to make things better.

One of the results of this past experiment is the active regulated power supplies in his present line of amplifiers. Frank Van Alstine claims it helps his amplifiers perform with an audibly lower noise floor.

What was once costly, and of questionable necessity, can become worthwhile if it can be done at a reasonable cost. I don't know if regulated power supplies will become standard in the future, but here's at least one example that it can be done at perhaps not unreasonable cost, with audible results.

 

[ImcLoud]

Oh boy... I swear you guys can complicate peanut butter and jelly....
Just buy an amp that says its 4 ohm capable if you plan on running 4 ohm speakers, problem solved, lol....

PS I know PB & J is complicated, first you have the PB, chunky or smooth, Organic, natural, or "regular", Commercial or Private label. Then the Jelly, jelly or jam or preserves which one? Old school grape, or Blackberry, Strawberry, Peach, Raspberry, or a tropical mix of a bunch of berries, or maybe something exotic like Cactus jelly, mango banana, or Papaya who knows, so much to think about and we didn't even consider the bread yet??? Oh no, old fashioned american, sliced Italian, Rye, Marble, Banana, french, toasted raisin, maybe it would be good on some cold french toast? This turned out to be very confusing, but at least that is all we have to choose, but wait Hre comes mister Fluff? WTF do we add fluff and what does Nutella have to do with this, does that go in there too???? Someone help me, Which goes an first, when I just had PB and jelly, it was easy, jelly on one side PB on the other, or do we mix them in a bowl then spread them on?????? Ill never figure this out...

But seriously I have figured it out, Koezes PB {I prefer crunchy}, with Mackays Scottish 3 berry, on lightly toasted Arnolds Potato bread, NO FLUFF, NO CHOC CHIPS, NO NUTELLA...