What are the most powerful Pure Class A Amplifiers ever made?

mep said:

Francisco-I'm simply talking about the cost of retubing them and keeping them running assuming all of the output tubes live a nice long life.

Click to expand...

The cost of retubing both channels is about $1800.00. That is not so bad considering how much power we get vs what our competition charges for their tubes. We warrant the tubes for a year.

Also the tubes hold up quite well in that amp since the effective output impedance is lower. It puts out its maximum power into 2.3 ohms.

Another way of putting this is that with any tube amp, if you operate the amp into a higher impedance, the tubes will run cooler as more power is dissipated into the load rather than the output section itself. Since the tubes are thus running cooler and dissipating less power, they last longer. Having more power tubes is a lot like running a smaller amp into a higher impedance as far as the tubes are concerned. With regards to the MA-3, even if the amp is driving a 4 ohm load the tubes run pretty cool- cooler than they do in any of our other amps. We don't know exactly how long the tubes last, but we do know they hold up better than in our other amps. We've seen over 10,000 hours in them if the amp is set up right. So the MA-3 is a lot easier to live with that one would expect on first blush.

Not mentioned yet but worthy of note is that bias functions are automatic, and the amplifier has its own AC power regulation- the line can drop to 95 volts and the amp will still make full power.

We first introduced this amp about 10 years ago, it remains the world's most powerful OTL, and one of the most powerful class A tube amps made; if you include the fact that it is also a triode power amp the list gets a lot shorter.

One distinct advantage it has is without an output transformer it is very fast- it rivals the fastest transistor amps in that regard. So from 1Hz it can run right up to 100KHz at full power within less than 0.5db. I don't know of another tube amp with this sort of power than can do that. As a result, it does not sound ponderous; musically it is as nimble as it gets.

Atmasphere said:

The cost of retubing both channels is about $1800.00. That is not so bad considering how much power we get vs what our competition charges for their tubes. We warrant the tubes for a year.

Also the tubes hold up quite well in that amp since the effective output impedance is lower. It puts out its maximum power into 2.3 ohms.

Another way of putting this is that with any tube amp, if you operate the amp into a higher impedance, the tubes will run cooler as more power is dissipated into the load rather than the output section itself. Since the tubes are thus running cooler and dissipating less power, they last longer. Having more power tubes is a lot like running a smaller amp into a higher impedance as far as the tubes are concerned. With regards to the MA-3, even if the amp is driving a 4 ohm load the tubes run pretty cool- cooler than they do in any of our other amps. We don't know exactly how long the tubes last, but we do know they hold up better than in our other amps. We've seen over 10,000 hours in them if the amp is set up right. So the MA-3 is a lot easier to live with that one would expect on first blush.

Not mentioned yet but worthy of note is that bias functions are automatic, and the amplifier has its own AC power regulation- the line can drop to 95 volts and the amp will still make full power.

We first introduced this amp about 10 years ago, it remains the world's most powerful OTL, and one of the most powerful class A tube amps made; if you include the fact that it is also a triode power amp the list gets a lot shorter.

One distinct advantage it has is without an output transformer it is very fast- it rivals the fastest transistor amps in that regard. So from 1Hz it can run right up to 100KHz at full power within less than 0.5db. I don't know of another tube amp with this sort of power than can do that. As a result, it does not sound ponderous; musically it is as nimble as it gets.

Click to expand...

Ralph,

Why the 12 x 6sn7 versus the 6 of the MA2? Are you running the triodes in parallel?
BTW, the lifetime of the 6sn7 is similar to that of the 6as7?

Not being a tube guy, my choice would be this one, old but still one of my favorites.

Krell - - - KRS-200 - - - Krell Reference Standard - - -



https://sites.google.com/site/jovlshifisite/Home/krell

Atmasphere said:

The cost of retubing both channels is about $1800.00. That is not so bad considering how much power we get vs what our competition charges for their tubes. We warrant the tubes for a year.

Also the tubes hold up quite well in that amp since the effective output impedance is lower. It puts out its maximum power into 2.3 ohms.

Another way of putting this is that with any tube amp, if you operate the amp into a higher impedance, the tubes will run cooler as more power is dissipated into the load rather than the output section itself. Since the tubes are thus running cooler and dissipating less power, they last longer. Having more power tubes is a lot like running a smaller amp into a higher impedance as far as the tubes are concerned. With regards to the MA-3, even if the amp is driving a 4 ohm load the tubes run pretty cool- cooler than they do in any of our other amps. We don't know exactly how long the tubes last, but we do know they hold up better than in our other amps. We've seen over 10,000 hours in them if the amp is set up right. So the MA-3 is a lot easier to live with that one would expect on first blush.

Not mentioned yet but worthy of note is that bias functions are automatic, and the amplifier has its own AC power regulation- the line can drop to 95 volts and the amp will still make full power.

We first introduced this amp about 10 years ago, it remains the world's most powerful OTL, and one of the most powerful class A tube amps made; if you include the fact that it is also a triode power amp the list gets a lot shorter.

One distinct advantage it has is without an output transformer it is very fast- it rivals the fastest transistor amps in that regard. So from 1Hz it can run right up to 100KHz at full power within less than 0.5db. I don't know of another tube amp with this sort of power than can do that. As a result, it does not sound ponderous; musically it is as nimble as it gets.

Click to expand...

$1,800.00 is a bargain if you compare to these, $10,000.00

T-161O = $2,500.00 match pair.

microstrip said:

Ralph,

Why the 12 x 6sn7 versus the 6 of the MA2? Are you running the triodes in parallel?
BTW, the lifetime of the 6sn7 is similar to that of the 6as7?

Click to expand...

Yes, there are more sections in parallel in the voltage amplifier and driver section. The 6SN7s hold up very well. Never been able to tell how long they last, but 10,000 hours is commonplace.

Any heard this one?

http://amaudio.it/page242.html

AM Audio...pure Class A 100 watts...but 1000 watts into 8ohms and a beast at 275lbs. I know Wizard is a big fan.

Also,

http://amaudio.it/page241.html

Class A 300 watts?

Anyone know where the Krell KMA-400 fits in here? I had 4 of the KMA-160's driving my big Apogee ribbons and you'd be sweating by the end of a listening session!

Lee

Honest specification of Class A Watts

Cautionary note. Some Class AB amplifiers (NOT Nelson's) give a "watts of Class A power" spec that is highly misleading, because they don't specify the impedance. It turns out that you can maximize the "watts of class A power" spec without changing the amplifier simply by choosing to use a way high load impedance like 50 ohms to measure (or calculate) it at. But what you are more likely interested in is an honest number into some meaningful speaker load, such as 4 or 8 ohms. Nelson Pass does in fact give you this honest Class A power for 8 ohms with his amplifiers. However I do not believe even his class A power doubles down into lower impedances. Read on.

I use this formula to calculate watts of Class A power given Idle Bias I and load resistance R: 2RII. If this seems unfamiliar (formula for power is IIR normally) it's because the current delivered into a load will be twice the quiescent idle bias, hence 2RII (that means two times R times I squared).

So it would seem you could increase the class A power simply by increasing the R! But you run into a limit were IR (which is V, voltage, following ohms law) is the maximum output voltage of the amplifier. So you can't just keep increasing R forever, there is some particular R for any amplifier where you will get the maximum Class A power, and it will scale linearly down below that (aka "halve down"). For most Class AB amplifiers this R value is considerably higher than 8 ohms.

The Mondial/Klipsch number for the Aragon 8008 BB which I use was inflated in this way… They claimed 26W Class A per channel. My unit was biased significantly higher, so I was getting 30.8W Class A. But I would only get that at a load of 68.7 ohms. At 8 ohms, I was only getting 3.58A W Class A, or 1.79W Class A into 4 ohms.

I discuss this all here: http://audioinvestigations.blogspot.com/2011/11/class-power-from-class-ab-amplifiers.html

My amp was dissipating 180W at idle, btw. I need to re-measure, too, because I added 4 large brass Heavyfoot's under my amp, and it runs way cooler now, suggesting the idle may have lowered too, so I may be getting less class A power now, but I think it sounds better anyway and it runs at about 110 degrees F rather than 135 which had me seriously worried about longevity of the unobtainium output transistors.

Now you can see the stark tradeoff given a certain amount of transformer iron, or heatsink area. With a given amount of metal, you can have low rail voltage (like Levinson ML-2) and get large class A power into 8,4,2 ohms, but that's it no more, or Class AB which gives you far more max Class AB power, but smaller class A power, sometimes surprisingly small. IIRC the ML-2 stays in Class A until 2 ohms (25W at 8, 50W at 4, and 100W at 2, all class A).

I think Nelson Pass has done a nice tradeoff with the XA amplifiers, say 30W class A and 150W push-pull Class AB for the XA 30.5. Most 150W Class AB amps will have far less than 30W Class A power into 8 ohms. But unlike Levinson ML-2, the Class A power won't be doubling into lower impedances (even though the Class AB power may increase for somewhat lower impedances).

But the ML-2 design does nicely match electrostats like Quads ESL-57 and ESL-63, for which impedance falls greatly at high frequencies and yet you'd like to be sure to stay in Class A to control a highly reactive load, and you don't need that much max power (voltage) into 8 ohms anyway, in fact you really don't want too much voltage into 8 ohms.

ML-2's would not work for my Acoustat 1+1's though, which need much more power. Probably the best approach would be direct drive amplifier. Acoustat did earlier use a direct drive amplifier. I believe it may have had some reliability issues, a friend of mine said the tubes were being driven past a few of their spec limits. That same friend has designed a direct drive tube amp for ESL-63's, class A, maximizing the linearity of the tubes, and operating everything within spec.

OK, I found the numbers for the Pass Labs XA 30.5 here (and others too):

https://www.passdiy.com/project/articles/leaving-class-a

Now he says the XA30.5 has 1.3A idle bias. Following the formula 2IIR that I previously described for Class A power, we have:

2*1.3*1.3*8 or 27W at 8 ohms

2*1.3*1.3*4 or 13.5W at 4 ohms

Now Nelson himself on the same page describes the Class A power as 57W peak. Well that is the same as 28.5W RMS (technically it's 28.5W average power, RMS applies only to voltage, except audiophiles have created a term-of-art RMS so that when we say RMS we really mean average). Average power is half peak power (for sine waves).

Thanks! Very interesting to read...thanks for taking the time to post.

I am quite certain the Krell MRA is the most powerful solid state amp of all time. A test in the German magazine Audiophile shows measurements where it can produce up to 2KW in class A. They are 1KW at 8Ohm and doubles that all the way down to 16KW at 0.5Ohm. They have a 16KVA transformer and a capacitor bank of 3.3 Farad. They use 202 transistors. All pr channel.

Tube amp I think it must be TRL 800s. They are stable at below 1 Ohm at 800W class A push-pull triode.

Mvh
Roysen

Roysen said:

I am quite certain the Krell MRA is the most powerful solid state amp of all time. A test in the German magazine Audiophile shows measurements where it can produce up to 2KW in class A. They are 1KW at 8Ohm and doubles that all the way down to 16KW at 0.5Ohm. They have a 16KVA transformer and a capacitor bank of 3.3 Farad. They use 202 transistors. All pr channel.

Tube amp I think it must be TRL 800s. They are stable at below 1 Ohm at 800W class A push-pull triode.

Mvh
Roysen

Click to expand...

oh...my...word...no wonder they weigh over 650 pounds each monoblock. 3.3 Farads is 9x the capacitance of a Gryphon Colosseum dual-mono amp (which weighs 176 lbs) and is a beast in my experience. Even the Mephisto monos are 1 Farad (500,000 micro-farads each monoblock). Jeez.

RBFC said:

Anyone know where the Krell KMA-400 fits in here? I had 4 of the KMA-160's driving my big Apogee ribbons and you'd be sweating by the end of a listening session!

Lee

Click to expand...

Seems to me the Krells had a sliding bias system, so at idle they made a relatively low amount of heat as they were also lightly biased. But as the amplifier was asked to make more power, the bias was also increased to maintain the output devices in the A range.

We have used a similar technique in our amplifiers, although we do it to prevent the amplifier from roasting the power tubes due to AC line voltage variation.

With regards to charlesp210's comments: There has been some controversy of whether an OTL can be a class A amplifier as well. The fact of the matter is that it can. However, just as we see in solid state amplifiers above, OTLs are load-dependent in this regard. For example our M-60 is a class A2 amplifier into 8 ohms but into 4 ohms it is a class AB2 amplifier. A similar phenomena exists with regular transformer-coupled tube amps as well- IOW the load on the output devices is part of the formula whether tube or solid state.

Atmasphere said:

Seems to me the Krells had a sliding bias system, so at idle they made a relatively low amount of heat as they were also lightly biased. But as the amplifier was asked to make more power, the bias was also increased to maintain the output devices in the A range.

Click to expand...

My 4 KMA-160s would heat the listening room when idling. Not kidding. Krell's new amplifiers are advertised as using a sliding bias technique, but not sure about the older ones. Note the heat fins on these 160s from a net photo I found:



Lee

That sliding bias system has been around since sometime in the 90s.

Atmasphere said:

That sliding bias system has been around since sometime in the 90s.

Click to expand...

Krell has also used sliding bias designs in some of their previous amplifiers. However they claim that their latest amps use a newly developed bias system that monitors and actively changes the output device's bias in real time to maintain class A operation. I believe that their previous designs tracked the input signal and then adjusted the output stage bias accordingly. We'll have to wait until they are available for audition to determine just how successful their new designs are.

I had a pair of Krell KMA 160s and they ran very warm. They were a little too "sweet" for my taste and did not have the balls to drive my MBLs. Somewhere on WBF I posted a list of all the heavy metal amps I've tried that survived the MBL challenge and were to my taste... it's a very short list. Krell KMA 400s and AVM MA8s.

I should add that the MBL challenge dates back to the late 80s when I heard the Black Banana Club's famous sound system. The music actually entered your body and was not even close to deafening.

^^ The MBLs are not that hard to drive! You do need some power, but 200 watts seems to be plenty. It's be a different story if the speaker were outside though...

Atmasphere said:

^^ The MBLs are not that hard to drive! You do need some power, but 200 watts seems to be plenty. It's be a different story if the speaker were outside though...

Click to expand...

Ralph,

Not to be a pia but how many MBLs have you owned? Also, perhaps it's my bad... I should qualify it: MBL 101Es. The 116 can be driven on 100 watts.

I was referring to the 101es. I have a friend in town who had them for a number of years, and customers also. Not owned any myself.