|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
71.53.138.121
it's all about the new Ganfet tech
Edits: 07/07/23Follow Ups:
Orchard Audio has their own proprietary ganfet modules. They are running an amp tour. Listen to the amp free for a week then ship it to the next person on the list. Sign up on their website.
I participated and ended up buying the Stereo Ultra amp. Love it. The tube and class A amps have been sitting.
Thanks,
barondla
here it is again
The use of Gallium Nitride (GaN) semiconductors has been around since the 1990's. GaN FETs have gained popularity in recent years in power devices like laptop battery chargers as they can be made even smaller than just a couple years ago and produce even less heat. I have one that's 1/2 the size of a deck of cards that can charge my laptop in a couple hours.Amps that use GaN FETs are still Class D amps. Not all Class D amps are equal even those using the same technology as we have experienced over the years. For example the ICEpower Peachtree Class D amp that I had sounded terrible. Same for the ICEpower Bel Canto S300. My Wyred4Sound ICEpower monoblocks were very good while the PS Audio were excellent. My Hypex nCore amps didn't sound so great. My present Purifi Class D amp is also excellent.
I'm sure we will experience terrible sounding GaN FET amps just like we have with other Class D amps.
The point being, GaN FET is not a game changer in Class D amps - not yet anyway. I'm taking a wait and see attitude. Waiting for more capable designers to implement them in their products. And then we'll be able to weed out the bad ones from the good.
Edits: 07/07/23 07/07/23
The link took me to an AoL sign in page and this is 2023 so I don't know what's the story Story.
But, here is an interview with Bruno Putzeys where he is asked about GaN FETs for class-D (at 17:50). If I may paraphrase his answer Math beats GaN!
Edits: 07/07/23
Sounds like Bruno has a new high power version about to come out that will knock the NCx500 off the high SINAD thrown and finally an alternative power supply option.
.
2022/03/30 Historical Records CENSORED
Edits: 07/09/23 07/09/23 07/09/23
Bruno left Hypex in 2014 to start Wii and later, Purifi.
Fixed, he is Bruno. He talks about a new Purifi high power amp coming out that is closer to 130 dB SINAD than 120 db SINAD in the last section of the video posted by 13th Duke of Wymbourne.
.
2022/03/30 Historical Records CENSORED
Edits: 07/09/23 07/09/23
No surprise there.
Time will tell.
"Reality cannot exist because it cannot keep up with the lies on the Internet."
nt
He uses them in his model and is not who I would consider a "lesser" designer.
Ralph has posted here that he has designed his own modulator so nothing 'lesser' there. Other products are, basically, the same modulator with silicon FETs replaced by GaN FETs.
and read it a bit more slowly. :)
I think it likely Ralph knows what is important for class-D as he is not re-using an off-the-shelf chip set. However, Putzey has been doing this sort of work for decades so Ralph is relative newbie ..... TBD.
Edits: 07/07/23
In fact the amps you linked to have some pretty average specs as Class D goes but it was probably intentional to deliberately introduce some "tube like" distortion in that brand.
And they used a linear power supply? Why?
"And they used a linear power supply? Why?"Linear vs switch mode?
since when have SM power supplies been considered any good?
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
Edits: 07/07/23 07/07/23
One of the quietest linestages and preamps in the world use SMPS. But not any off the shelf SMPS or "wall wart". Requiring a heavy linear power supply is just old school audiophile myth.
nt
Open up your mind, in pours the trash. - Meat Puppets, 1987
That's a whole thread in itself :)
There is a company making a power amp from Putzey's Eigentakt modules and they add their own 'audiophile' zero-feedback buffer at the input and use an over sized 'linear' PSU too. That buffer has no supply rejection so any supply noise goes straight through the amplifier. Maybe if they'd designed the input buffer properly they wouldn't need such a PSU.
Of course 'linear' power supplies are anything but linear - full-wave rectification and only sampling the mains on the voltage peaks.
"full-wave rectification and only sampling the mains on the voltage peaks."Yes, for capacitor input filters.
Make it a critical inductance choke input filter and the current to the filter from the power transformer is constant. The diodes each (in turn) stay on for their full half cycle. Charging currents are only a tiny bit larger than the load current and there are no peaks at all. Ripple is low and a sinusoidal.
Don't switch mode power supplies generate a lot of RF? That is not something I want around my audio signal.
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
Edits: 07/07/23
I can't hear RF :) and it is easier to shield RF than 60Hz.
I don't know anything about choke input filters, they sound interesting. Is the input choke prohibitively large/expensive?
Choke filters have been around for several decades in tube electronics. They're no longer popular but -some- tube amp/preamp manufacturers still use them. The choke can be at the input or 'middle' of a filter network.See the link below in response to E-Stat a few weeks ago.
Chokes vary in size and weight depending on their construction materials, voltage rating and their inductance in Henry (H). These inductors are generally called "chokes" in filtering applications including power supplies at 50/60-Hz. Inductor components can use an iron core or air core depending on the application. Inductors can also form a tuned circuit in RF applications. At extremely high frequencies even the tiny bit of 'stray' inductance and 'stray' capacitance in the PCB traces need to be considered in the design layout.
Edits: 07/08/23 07/08/23 07/08/23 07/08/23
In fact the amps you linked to have some pretty average specs as Class D goes...
Yes. And?
And they used a linear power supply? Why?
Ralph has already answered your question .
- If the GaN FET has better specs why would one use them if the end result ended in worse specs than current MOSFET designs?
- Did not say WHY they use a linear power supply - just that they do.
he, like other designers is not *spec* driven.
The distortion profile of an amplifier is determined by more than simply the output devices employed.
He made choices that met his goals(?). Why are you asking me?
Obviously not spec driven so why choose a device with better specs if you're not going to take advantage of it? I don't know. You don't know. I can guess why.
Why use a linear power supply? I can guess why on that too.
It comes down to marketing.
Found on the webpage:
"The power supply employs a toroidal power transformer and has over-current protection to keep the amplifier safe in case of a short at the speaker terminals."
Hmmm. Maybe something related to what took out yours!
"....short at the speaker terminals."
Nope. My speakers are fine, and no shorts at the terminals or cables. Nothing in the system changed from the last time I played it the day before and the day the NAD blew up.
If I were going that route I'd have to give it serious consideration despite what Putzey has to say. Especially as he seems to be bad-mouthing someone else's choice of technology.
Always a red flag in my view.
ESPECIALLY among audio designers and manufacturers where we have seen WAY to much of it over the years.
Known Ralph Karsten for what seems like forever so would give him thoughts on the subject fair consideration plus...
Bet him unlikely to bad-mouth someone else's design choice.
"Reality cannot exist because it cannot keep up with the lies on the Internet."
Bruno's design philosophy is to use as much feedback as possible - and then add more. So, in that sense, it is not about designer gurus (his words) 'voicing' components. Just design them transparent without pandering to audiophile sensibilities. My heart likes to think that designers impart some magic to special components but my head thinks this unlikely.
Putzey wasn't bad mouthing anyone. He was merely pointing out that there's a LOT MORE to implementing a high quality Class D amp than choosing one component part (GaN FET) over another (MOSFET). Just like choosing the supposedly 'better' KT120 tube didn't magically make those tube amps 'better'.
I prefer a tube power triode.
So did I until I home tested a wide range of SS amps and chose a Class D one as equal to or better sound than my SET amps that I'd been very happy with for the previous 15 years. No regrets for switching - now I listen to far more hours of music than ever I did using valve amps.PS - Just ordered Ralph's Class D monos
Edits: 07/08/23
What SET amps did you use for 15 years?
A number of them. Carissa 845 and PX-25 from Art Audio, Consonance Cyber 845 monos, a poor 300B offering from Audio Note and others. Also OTL from Graaf. Why do you ask?
Which one did you prefer?
PX-25 for small scale music, female vocal, etc. Carissa or Cyber 845 for full blooded music. Used to power 102 dB horn speakers.
You probably would have had better luck if you Bi-Amped with the PX-25 dedicated to the horn and a Separate Bass Amp...... PX-25 did not have enough power for 102db speakers. It would have sounded glorious on just the horns, and a class D amp for the Bass would have freed up the PX-25 for everything you needed.
Edits: 07/09/23
I believe Ralph's Class D amps are 'voiced' to sound a bit more like tubes vs some other Class D amps.I would be very interested in how they sound compared to your NAD M33.
On a related note does the M33 have a DAC output that can drive the Atma-Sphere Class D monoblocks? Or do you have another digital source that can drive your M33 and the Atma-Spheres? The point being, the only way to truly get a feel for how the amp section in your M33 compares to the standalone Atma-Sphere amps is to use the same source.
BTW I have a NAD C298 that sounds surprisingly good - until it died last night ;-(. But up to that point I was very impressed.
Edits: 07/08/23
I've talked to Ralph extensively about his Class Ds and have a pair. Ralph did not voice them to sound like tubes. He does not voice his tube amps to sound like tubes. He designs his stuff to his own technical specs (which are the technical parameters he thinks makes the most revealing and neutral sound) and then dials them in a bit to even more neutral. He prefers the technical specs of his GAN Class Ds to the specs of his tube amps. That is really what matters to him.
Edits: 07/12/23
The M33 has a preamp out, although it is single ended. The AP amps accept either SE or balanced.
So I can use the M33 as streamer / DAC / pre and the signal can then go to either the M33's own Purifi amp or to the AP amps. In fact I have an amp switcher that will allow direct comparison without changing cables each time.
Ralph suggests a 400 or so hour burn in before optimal sound quality. Hopefully I'll notice a worthwhile improvement over the M33, although I have my doubts as the M33 is so very good!
I', planning an upgrade to the front end by changing the M33 with the new M66 when it's released. Again I'm hoping for a worthwhile SQ improvement. It certainly has a better DAC (9038 vs 9028) a better volume control and more comprehensive Dirac Bass Control, although I'm unlikely to use that feature if it's anything like Dirac Live or RoomPerfect.
I'll post impressions is a week or two
I concur that the amps need at least 400 hours for the bass to fully come in and balance out.
Really hard to dispute such a listening experience!
For more than a decade I have been listening to a very highly rated OTL Tube amp/Tube Preamp multi-thousand dollar combination or Musical Fidelity Class A amp, and then along came $80 Aiyima A07 and boy, hard to argue with the sound I hear from it!
I tried the Peachtree GaNFET amp. It sounded pretty good, but to me, wasn't better than the Devialet Expert Pro 250 I was using, so I let it go.
I have a lot of respect for Ralph, believe he's a straight shooter. If he comes up with a higher powered amp, I would definitely look into it.
Been fooling around with tube amps for decades. Love their sound, but to be honest, the Devialet Expert Pro is a better sonic option. I'll keep the ones I have, and occasionally listen to them, but the Devialet is my primary setup. If anything, the longer I have the Devialet, the more I realize just how good it actually performs.
" Don't look back. Someone may be gaining on you"
Sctchel Paige
Interesting - I had the Devialet 200 for a while, but didn't like it much. It couldn't match the SETs I was using at the time and the socketry at the back was a disaster area. I guess that Devialet have improved their Class D over the years as I would certainly rate the NAD M33 way above that early French amp!
The Devialet Analog Digital Hybrid (ADH) is NOT a class D amp per se. The voltage is Class A, and Class D is strictly for current. The Class D section does not has the output filter all Class D amps have, because it's not needed.My speakers use SEAS Excel drivers, which are very revealing. The Expert Pro 1000 sounds amazing with the speakers. The sound is even better when using the Chord M-Scaler as a front end processor. It's the cleanest and most dynamic set of electronics I've heard by far, more refined than any Class D amp I've heard.
Might be why there are so many purchasing options. You know what they say, ask 12 audiophiles are asked what they hear, and there are 13 different opinions offered.
" Don't look back. Someone may be gaining on you"
Sctchel Paige
Edits: 07/08/23 07/08/23 07/08/23
There is often confusion as to whether an amp is true Class D or something else that's rather similar! I'm sure that "strictly speaking" you are correct, though many consider these amps to be so near Class D as to be easily mistaken for Class D.
However, having said that, I'm a fan of Class D when implemented properly. After home demo'ing a dozen good ss amps a while ago, I chose a Class D one above the As and ABs and others that could be confused for Class D. Chosen carefully the best Class D can now match or better just about anything - and at a lower price and their cost will be soon be saved (compared with valve or Class A) in the lower electricity bill they generate.
The current crop of Class D amps are very good. The earlier Class D offerings weren't very good, but they have improved over time a lot. I had a pair of Primare Class D amps in mono that were quite nice sounding.I also would much rather wrestle with a pair of 25 pound integrated amps that output 1000 watts than the 100 lb. amps I used to wrestle with.
Kind of missed the boat on the lower cost :-)
I did get mine at a big discount, but it was still $$$. I do really like having the DAC, streamer, preamp and power amp all in one box. Saves a lot of shekels on separate boxes and cables as well. So, it kind of works out.
" Don't look back. Someone may be gaining on you"
Sctchel Paige
Edits: 07/08/23
Which Peachtree GaNFET amp? They had one that only accepted a digital input. I think their more recent ones offer more traditional analog inputs.
I have an older Peachtree iDecco Class AB integrated that sounds wonderful. I tried their newer Class D Nova 150 a couple years ago and I thought it was less than average as Class D amps go.
The Peachtree amp was the GaN400. Not a bad sounding amp, but not in the same class sonically as the Devialet.
Been wanting to try an speaker amp to take advantage of the Chord Hugo TT2 upscaling sonics. Have a headphone amp where the advantages of the TT2 are obvious, but so far, every speaker amp I've tried falls short compared to the Devialet setup. Have a couple of tube amps that sound nice with the Chord for small scale music, but runs out of stem with large symphonic music.
" Don't look back. Someone may be gaining on you"
Sctchel Paige
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: