that is certainly helpful and might alleviate the situation. How does it look in the bass? Are there any peaks?
Atma-Sphere Class D Mono blocks
- Thread starter Chorus
- Start date
There are other advantages to using Ralph’s class-D amps besides the sound quality being superior to many tube amps. I was a tube amplifier only person ever since the SET revolution in the mid 90’s. That is until this class-D amp arrived. Now I actually listen to and enjoy more music than I did when I was using tube amplifiers.
First off these class-D amps improved in SQ over about the first 400 to 500 hours of operation. Biggest difference after the first few hundred hours then more subtle gains until around the 500 hour mark at which time the amplifiers were fully settled in. I’m assuming this was the time it took for the power supply capacitors to become fully formed.
They only draw about 5mA of mains current at idle. I leave them powered on 24/7. Which is not practical nor advisable with tube amps. My system is ready to go and all warmed up all the time. There is no waiting period for the magic to happen. It’s there all the time.
One of my favorite things to do is wake up before sunrise, pour a cup of coffee, sit in my listening chair and listen to some of my favorite musicians such as Bill Frisell at a low volume for an hour or so as the sun rises. Similar in overall volume to an acoustic nylon string guitar being played from about twelve feet away. Very pleasant.
I’m using these amps with 99dB efficient horn speakers so the amps are putting out only a few tens of milliwatts at this volume. These amps sound the same when putting out a few milliwatts as they do putting out several tens of Watts. The first Watt is sweet and the rest follows!
For the front end, I purposely chose a server and DAC combination to provide the most detailed, musical sound possible. My IT setup, server and USB cable combination delivers great fidelity to the DAC. The DAC uses a transformer for the I/V stage, a single direct coupled 10k hour rated triode for the voltage amplification stage and a transformer balanced floating output. This drives 15 feet of balanced interconnect into remote controlled balanced attenuators using relays and Texas Components resistors located at the input of each mono-block amplifier. I don’t need a preamplifier or amplifier to change the sound that I am getting out of the front-end equipment. Quite the contrary. I want the amps to pass that goodness along to the speakers without editorialization. Which these class-D amps do better than any other amplifier technology that I have ever heard.
Then there is the constant moving target of tubes as they age. You may get a set of tubes that you love the sound of. But if you use your amps as regularly as I do, the sound will noticeably change over time and not for the better. After a certain amount of use it’s time to re-tube your amps after which they will sound different again and the moving target of aging tubes repeats. There is also the cost of retubing your amps with high quality tubes if you can find them. Some people prefer fiddling with that stuff but after decades of chasing that tail, I prefer spending my time listening to music now.
Below is a picture of Harvey Rosenberg who started New York Audio Labs in the early 80’s and me who started Ambient Recording Co. in 1990 in Stamford, CT. This picture was taken in the late 90’s when we had the system setup for several months at Ambient. The speakers at that point had the crossovers removed, were filled with sand and weighed over 600Lbs each. As can be seen we were not overly concerned about the quality of the rack which was made from plywood and 2x4’s. We just needed something large enough to fit all the amps and power supplies. The entire system was wired unbalanced with Alpha-Core silver foil interconnects. It wasn’t the quietest system ever, but even my dear friend Allen Wright (RIP) of Vacuum State Electronics said it was far and above the best sounding system he had ever heard. For all of us it was the new goal in sound reproduction.
Harvey had been receiving SET amplifiers from Japan to review on an almost weekly basis. We had access to everything from 2 Watt to 20 Watt SET amplifiers. Many based on classic Western Electric designs at our disposal to try with the 99dB efficient Westminsters. We also had his New York Audio Labs “Futterman” OTL’s and Moscode amps as well as a modified Marantz 8B with outboard regulated power supplies.
After going thru all those iterations and trying many different combinations of the above on both the woofer and tweeter with active crossovers, we finally settled on Atma-sphere M60 OTL’s modified with outboard tube regulated power supplies in a bi-amp configuration as seen in the picture. It was an all tube signal path including the active crossover, which also had tube regulated supplies. This was at a time when true NOS tubes were readily available. We could cherry pick thru them to find the lowest noise, non-microphonic tubes with tightly matched halves. Try that today.
To my dismay Harvey passed unexpectedly from a heart attack at the age of 59. Three weeks before 9/11/2001. He left a different planet than we live on now. He always had good timing.
Both he and Allen were close friends who opened my mind to what could be accomplished in hi-end audio. I also have great respect for Ralph Karsten of Atma-sphere who’s decades of experience in tube amplifier design has led to this breakthrough in class-D amplification.
First off these class-D amps improved in SQ over about the first 400 to 500 hours of operation. Biggest difference after the first few hundred hours then more subtle gains until around the 500 hour mark at which time the amplifiers were fully settled in. I’m assuming this was the time it took for the power supply capacitors to become fully formed.
They only draw about 5mA of mains current at idle. I leave them powered on 24/7. Which is not practical nor advisable with tube amps. My system is ready to go and all warmed up all the time. There is no waiting period for the magic to happen. It’s there all the time.
One of my favorite things to do is wake up before sunrise, pour a cup of coffee, sit in my listening chair and listen to some of my favorite musicians such as Bill Frisell at a low volume for an hour or so as the sun rises. Similar in overall volume to an acoustic nylon string guitar being played from about twelve feet away. Very pleasant.
I’m using these amps with 99dB efficient horn speakers so the amps are putting out only a few tens of milliwatts at this volume. These amps sound the same when putting out a few milliwatts as they do putting out several tens of Watts. The first Watt is sweet and the rest follows!
For the front end, I purposely chose a server and DAC combination to provide the most detailed, musical sound possible. My IT setup, server and USB cable combination delivers great fidelity to the DAC. The DAC uses a transformer for the I/V stage, a single direct coupled 10k hour rated triode for the voltage amplification stage and a transformer balanced floating output. This drives 15 feet of balanced interconnect into remote controlled balanced attenuators using relays and Texas Components resistors located at the input of each mono-block amplifier. I don’t need a preamplifier or amplifier to change the sound that I am getting out of the front-end equipment. Quite the contrary. I want the amps to pass that goodness along to the speakers without editorialization. Which these class-D amps do better than any other amplifier technology that I have ever heard.
Then there is the constant moving target of tubes as they age. You may get a set of tubes that you love the sound of. But if you use your amps as regularly as I do, the sound will noticeably change over time and not for the better. After a certain amount of use it’s time to re-tube your amps after which they will sound different again and the moving target of aging tubes repeats. There is also the cost of retubing your amps with high quality tubes if you can find them. Some people prefer fiddling with that stuff but after decades of chasing that tail, I prefer spending my time listening to music now.
Below is a picture of Harvey Rosenberg who started New York Audio Labs in the early 80’s and me who started Ambient Recording Co. in 1990 in Stamford, CT. This picture was taken in the late 90’s when we had the system setup for several months at Ambient. The speakers at that point had the crossovers removed, were filled with sand and weighed over 600Lbs each. As can be seen we were not overly concerned about the quality of the rack which was made from plywood and 2x4’s. We just needed something large enough to fit all the amps and power supplies. The entire system was wired unbalanced with Alpha-Core silver foil interconnects. It wasn’t the quietest system ever, but even my dear friend Allen Wright (RIP) of Vacuum State Electronics said it was far and above the best sounding system he had ever heard. For all of us it was the new goal in sound reproduction.
Harvey had been receiving SET amplifiers from Japan to review on an almost weekly basis. We had access to everything from 2 Watt to 20 Watt SET amplifiers. Many based on classic Western Electric designs at our disposal to try with the 99dB efficient Westminsters. We also had his New York Audio Labs “Futterman” OTL’s and Moscode amps as well as a modified Marantz 8B with outboard regulated power supplies.
After going thru all those iterations and trying many different combinations of the above on both the woofer and tweeter with active crossovers, we finally settled on Atma-sphere M60 OTL’s modified with outboard tube regulated power supplies in a bi-amp configuration as seen in the picture. It was an all tube signal path including the active crossover, which also had tube regulated supplies. This was at a time when true NOS tubes were readily available. We could cherry pick thru them to find the lowest noise, non-microphonic tubes with tightly matched halves. Try that today.
To my dismay Harvey passed unexpectedly from a heart attack at the age of 59. Three weeks before 9/11/2001. He left a different planet than we live on now. He always had good timing.
Both he and Allen were close friends who opened my mind to what could be accomplished in hi-end audio. I also have great respect for Ralph Karsten of Atma-sphere who’s decades of experience in tube amplifier design has led to this breakthrough in class-D amplification.
Well said, this is one of the reasons I stopped with tubes. I want my system to sound as close to the same over a long range of time. And listening to a beautiful set of NOS tubes slowly die took a lot of the fun out of it for me.
You think solid-state dos noy age, albeit at a much slower rate? As that is always the question/ A candle that burns brightest extinguishes more quickly.
Thankfully, my Class D is not candle-powered. If it lasts for 10 years without deterioration, that's good with me as I'm likely to have changed it before then. Class D is still evolving (now more slowly, unlike other amp types that matured decades ago), so I'm now enjoying an amp that betters my previous SETs and I'm looking forward to even greater things in the future.
Ca n they coexist, or will they fight it ut to the death in a cage match?
The maker of a class A OTL tube amp hails the virtues of his class D amp which is the antithesis of his current line. Does one have to die, or can they peacefully coexist?
The maker of a class A OTL tube amp hails the virtues of his class D amp which is the antithesis of his current line. Does one have to die, or can they peacefully coexist?
I know plenty of people having to recap SS electronics at some time. Of course, the same applies to tube.
I'm not sure about the transistor. Does it burn out over time. What I have heard is when you loose a transistor, its not at all a simple swap. Its a big deal to get matched and they may not be made any more.
I.wonder the total life span of a SS vs tube amp before you have to throw it out. How many years do either run before the cost to refurbish is beyond getting new.
Preamps may be a little different. Especially with tubes. About 6 month ago I tried a fresh pair of Telefunken E88CC in my First Spund MK11SI. They sounded exactly the same as the original pair I have had in for 5 years or so. I have 4 pair as backup so I'm feeling pretty good my preamp.
I'm not sure about the transistor. Does it burn out over time. What I have heard is when you loose a transistor, its not at all a simple swap. Its a big deal to get matched and they may not be made any more.
I.wonder the total life span of a SS vs tube amp before you have to throw it out. How many years do either run before the cost to refurbish is beyond getting new.
Preamps may be a little different. Especially with tubes. About 6 month ago I tried a fresh pair of Telefunken E88CC in my First Spund MK11SI. They sounded exactly the same as the original pair I have had in for 5 years or so. I have 4 pair as backup so I'm feeling pretty good my preamp.
They do. They sound really similar. I do think though that tube power amps are currently on borrowed time- why bother with them when you can get sound just as good and not worry about tube replacement?? The war in Ukraine certainly isn't helping...
Transistors can corrode over time. Anyone who has renovated older solid state gear knows that! The industry has improved that issue over the decades- it appears that newer semiconductors do last longer since going to surface mount. It certainly helps that there is less heat in class D circuits. Class D seems to have the advantage that if the amp is well built, it will always be worth renovation (which will be likely about 30-35 years down the road; filter caps last longer when there's no heat). So I don't see that as a problem.
"They do. They sound really similar. I do think though that tube power amps are currently on borrowed time- why bother with them when you can get sound just as good and not worry about tube replacement?? The war in Ukraine certainly isn't helping..."
Atmasphere
Thats breaking news!
Atmasphere
Thats breaking news!
Not really breaking news. The coming of age of Class D means that other types are already on borrowed time, although this was probably not true 5 years ago. It will be even more true in 5 years time when it will be really difficult for tube and Class D builders to sell their costly kit in the face of obviously superior and much more cost-effective Class D kit. It's enlightening that at least one well-respected tube amp builder has seen the writing on the wall. Hopefully others will follow.
I've been watching class D for over 20 years. I realized then that is was a rising star and like most technologies is on a Sigmoid cost/development curve. I realized about 6 years ago that we are very nearly at the top of that curve so it was obvious it was time to sort things out. IMO any audio manufacturer that hasn't sorted out class D will get left behind.
How much of the sound of a class D amp is attributed to the GaNFET output transistors, Exclusive Atma-Sphere class D module, case and/or other components.
Where does the upgrade path start? As in, what is still being assessed to improve upon what we hear from these amps. Or how will you make them sound as natural as possible and up the power to say 300 or more watts.
Do you think in a blind test a group of people could tell the difference between monoblock amps like Boulder/Gryphon/Dartzeel? What would they hear different after living with them for a month or so.
Do you add in some color to keep an amp from being sterile. To flat and perfect?
I know I am asking a lot here. Its just such a hurdle to go from the accepted "Audiophile" world of high performance (whatever that means) with atmospheric pricing, to a new topology that is suppose to be as good or better for $6K or so.
I have been wrestling with this idea that maybe I like some degree of distortion or frequency bumps. I am not sure if some amount of deviation from a perfect amp is required to create a more natural sound???? Sure that sounds stupid on the surface. At times I feel I have heard a better violin or piano in my system, but other parts of the whole were wrong enough to make me move on. I notice this more dramatically when I used a horn speaker where the violin and brass instruments were fantastic, but the Harpsicord was buried. When I changed to a Coax with a paper driver and only a horn high, the harpsicord came out and stands as it should, but the violin and brass don't have quite the alive and up front I was hearing before.
Where does the upgrade path start? As in, what is still being assessed to improve upon what we hear from these amps. Or how will you make them sound as natural as possible and up the power to say 300 or more watts.
Do you think in a blind test a group of people could tell the difference between monoblock amps like Boulder/Gryphon/Dartzeel? What would they hear different after living with them for a month or so.
Do you add in some color to keep an amp from being sterile. To flat and perfect?
I know I am asking a lot here. Its just such a hurdle to go from the accepted "Audiophile" world of high performance (whatever that means) with atmospheric pricing, to a new topology that is suppose to be as good or better for $6K or so.
I have been wrestling with this idea that maybe I like some degree of distortion or frequency bumps. I am not sure if some amount of deviation from a perfect amp is required to create a more natural sound???? Sure that sounds stupid on the surface. At times I feel I have heard a better violin or piano in my system, but other parts of the whole were wrong enough to make me move on. I notice this more dramatically when I used a horn speaker where the violin and brass instruments were fantastic, but the Harpsicord was buried. When I changed to a Coax with a paper driver and only a horn high, the harpsicord came out and stands as it should, but the violin and brass don't have quite the alive and up front I was hearing before.
I think the simple answer is just to arrange a home trial of a couple of top-drawer Class D amps. Sadly, I've not yet had the opportunity to hear the Atmosphere monos, but I do use an amp featuring the Purifi Eigentakt technology. This is considered by most as the best out there and certainly I've not heard better.
There are plenty of cheap Purifi amps listed on Ebay, etc that are often garage-built using the Evaluation Board, but I'd suggest that the much better option is to go for a big-brand offering where they build under license after maximising performance in conjunction with Purifi. The amps that are probably top of the pile are NAD's M23 power amp or its sister M33 all-in-one, or the T & A A200 power amp or the M200 mono amps. I'm hoping the Atmosphere may out-perform these amps, but I'll wait for their integrated before jumping ship from the NAD M33.
Tubes were declared 'obsolete' in the 1960s when germanium transistors were really the only game in town. Obviously the declaration was premature. Then silicon came in, but we still had the problem of 'turn it up and it sounds bright and harsh'. The measurement/numbers guys spent 50 years trying to convince us that solid state was more natural. Obviously brightness and harshness is every bit a coloration and IME more dastardly than that of the warmth of tubes caused by the 2nd and 3rd harmonics.
So tubes failed to go away which is because they weren't obsolete. This is economics not engineering! The tubes were clearly doing something that you couldn't get from solid state. I've explained elsewhere why a class D amp finally has the ability to get around the harshness and brightness of solid state. I've explained how it does that. I've explained how the ear perceives amplifier differences and why a class D can sound like a tube amp or a cheap solid state amp depending on how its designed.
I'm not feeling like repeating myself at this point; but tomorrow is another day if you want to hear about it. In a nutshell there are now class D amps that have the neutral and musical character that I've always associated with the best tube amps but no worries about replacing tubes.
The sound of the module has to do with its topology (and there are a lot of differences there), how much feedback it runs, how the feedback is implemented and how well its able to not have noise from parasitic inductances. GaNFETs have the advantage of really reducing stray inductances in the output since they literally have no leads. Because of their low gate capacitance they are easier to drive as well so the risetime and shutoff slopes are a lot steeper. So if you're careful a very low noise module can be created. If you read the early parts of this thread you'll see that we didn't know if we were passing the EU standards for noise so were reluctant to talk about export. But as it turned out when we were finally able to get testing it was no worries.
I can't speak to your Boulder/Gryphon/Darzeel question as I've not spent enough time with those amps to know anything about them. But I can say that if their distortion profiles are different they will sound different too.
We don't add 'color'. It worked out in our design (and prototypes) that the things that cause distortion in the amp tend to create lower ordered harmonics, much like a tube amp, but at a much lower level. Because it is at a lower level, on some systems its might sound a bit leaner. But it won't be harsh or bright, owing to less higher ordered harmonics (and those that are there are being masked in the same way this happens in a tube amp). But the goal has been and continues to be to get the amp as neutral as possible. One thing that has really been heartening in this whole thing has been that when we've done comparisons with our tube amps (on the right speakers, as they are load sensitive) that its obvious that we were on the right track all this time.
IMO one of the bigger sins a system can have is to be bright and harsh. I've experienced systems that actually inflict pain and in the words of Earl Root (RIP), 'Pain sucks dammit!' So I don't give a bright and harsh system much credence. I like 'smooth and detailed' instead.
WRT to a 'new topology' class D was first proposed in the 1950s and the first class D amps for the home were offered in the 1960s. IOW they have been around as long as conventional solid state. But its taken a while to really work out how to get them to sound right.
{Text omitted}
"I'm not feeling like repeating myself at this point; but tomorrow is another day if you want to hear about it. In a nutshell there are now class D amps that have the neutral and musical character that I've always associated with the best tube amps but no worries about replacing tubes."
You are a teacher and learning comes through repetition. there is no need be remedial, however. I have followed your post since Audiogon. I have read the white papers on your website. I also have read what you have said about Class D amps. I said they are a lateral move. I gave my choice of best sound at this year's CAF to an exhibit that featured a class D monoblocks. I have not heard your amp; thus, I have no opinion of its 'sonic character.
In fact, my presence here is because your apparent future commitment to Class D amplification is a 180-degree pivot to your all-out tube designs. You have to admit that is fuel for debate.
"I'm not feeling like repeating myself at this point; but tomorrow is another day if you want to hear about it. In a nutshell there are now class D amps that have the neutral and musical character that I've always associated with the best tube amps but no worries about replacing tubes."
You are a teacher and learning comes through repetition. there is no need be remedial, however. I have followed your post since Audiogon. I have read the white papers on your website. I also have read what you have said about Class D amps. I said they are a lateral move. I gave my choice of best sound at this year's CAF to an exhibit that featured a class D monoblocks. I have not heard your amp; thus, I have no opinion of its 'sonic character.
In fact, my presence here is because your apparent future commitment to Class D amplification is a 180-degree pivot to your all-out tube designs. You have to admit that is fuel for debate.
^^ I'm sure it is. And I'm open to it.
I had to reread this thread to see if I had made some of the explanations here; turned out I'd done very little. When I got done, it was about time to head home. I'm not sure if I should try to collect all the different posts I've made or just write up a paper trying to cover all the various aspects of this topic and simple provide links to it.
I had to reread this thread to see if I had made some of the explanations here; turned out I'd done very little. When I got done, it was about time to head home. I'm not sure if I should try to collect all the different posts I've made or just write up a paper trying to cover all the various aspects of this topic and simple provide links to it.
Just to show you I was paying attention professor.
" It is self-oscillating, similar in spirit to Bruno Putzeys' designs. Unlike his stuff of course we're using GaNFETs. They are so fast that the deadtime used is no longer about the on and off switching times of the transistors- they can easily switch at 60MHz, at least from what I've seen. I can't speak for other designs but in our design the distortion sources are in the encoding scheme and the deadtime used in the output section (which is mostly used for the choke in the output filter) and tend to result in lower ordered harmonics rather than higher ordered. We attribute this to the amp's smooth character; of course, its considerably lower distortion than any tube amp."
Similar threads
