New Generation Of Class D Amplifiers for 2010 - AVS Forum
Forum Jump: 
Reply
 
Thread Tools
post #1 of 74 Old 08-25-2010, 06:01 PM - Thread Starter
 
HiFiFun's Avatar
 
Join Date: Nov 2008
Location: USA
Posts: 2,209
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 17
Up until now I stayed away from Class D amplifiers because of their considerable EMI emissions which can seriously degrade sound quality. However there appears to be some breakthrough in the latest Class D designs. One is in the new Samsung M700 receiver and the other is from a new line of amplifiers from Extron:
http://www.extron.com/download/files...eries_revA.pdf
http://www.extron.com/company/articl...d=amplifiersad
HiFiFun is offline  
Sponsored Links
Advertisement
 
post #2 of 74 Old 08-25-2010, 06:19 PM
AVS Addicted Member
 
MichaelJHuman's Avatar
 
Join Date: Nov 2006
Location: Colorado Springs, CO
Posts: 18,756
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 164 Post(s)
Liked: 99
There's quite a few class D amp designs out there. Out of curiosity, how do you know that the audio quality has been affected on various class D amps? Have you tested them personally or seen tests you would like to share? I have not seen specifics.

Also, how do you know these new amp designs are better?

"But this one goes up to 11"
MichaelJHuman is online now  
post #3 of 74 Old 08-25-2010, 06:27 PM - Thread Starter
 
HiFiFun's Avatar
 
Join Date: Nov 2008
Location: USA
Posts: 2,209
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 17
The largely unacknowledged downside of all digital electronics is the harmonics of the of high frequency square waves wrecking havoc into the surrounding analog circuitry. The worst spewing case is with the high power square waves in the switching power supply or output stage, as in Class D/PWM amplifiers.

Another unexpected corner-case is the true local dimming LED powered displays with thousands of low powered LED switching in the nanosecond rate (combine them and it becomes high-power). For this reason plasma sounds better. Ha!
HiFiFun is offline  
post #4 of 74 Old 08-25-2010, 06:35 PM
AVS Addicted Member
 
MichaelJHuman's Avatar
 
Join Date: Nov 2006
Location: Colorado Springs, CO
Posts: 18,756
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 164 Post(s)
Liked: 99
I understand what you are saying, but where's the specifics? Do you have bench test output of specific class D amps showing these issues?

"But this one goes up to 11"
MichaelJHuman is online now  
post #5 of 74 Old 08-26-2010, 06:09 AM - Thread Starter
 
HiFiFun's Avatar
 
Join Date: Nov 2008
Location: USA
Posts: 2,209
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 17
I've read your other thread and admire your quest to find out what the real truth is about amplifiers. The answer is, like the movie, its complicated and can changed quickly, as opinions can become invalidated in a few months.

Please read the complete pdf document posted above from Extron as there is engineering charts and white papers. In general manufactures do not talk of shortcomings until after they have a product to sell which solves the problem. That is certainly the case here!

Making meaningful RF measurements outside of gear is very difficult and takes specialized gear and facilities. This is FCC certification, which is why many manufactures throw in a separate common mode choke as a band-aid to pass. Even Samsung is not above this in their $5K TVs.

Now what about making measurements inside of gear? Forget it. Engineering courses teach this stuff and its quite complicated, as a grounding always involves compromises and is frequency dependant.
One way is to use resonant mode switching power supplies which use a sine wave which is much easier to filter than square waves.
The point here is Extron and Samsung technology use some new techniques which is noticable sonically. The Samsung M700 is chalked full of digital signals yet is super clean. Realize that Samsung has the deep pockets and engineering skill to develop superior products, but almost everything is kept confidential. So learning anything is difficult. They even rub off the chips inside...

Note: Class D static thd measurements are poorer than with pure analog amplifiers
HiFiFun is offline  
post #6 of 74 Old 08-26-2010, 06:15 AM
AVS Addicted Member
 
penngray's Avatar
 
Join Date: Feb 2006
Location: Florida
Posts: 26,779
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 29
Good specs if they are remotely accurate. I have been looking for a really good S/N ratio design for waveguide designs.

Extron is a solid audio/video product company. I own 3 of their video and audio switches that are used for full house distribution. I suspect the price tag to be higher then normal though, they generally sell to business and not to homes.

It is not "open-minded" to reject knowledge - Bob Lee
penngray is offline  
post #7 of 74 Old 08-26-2010, 06:47 AM
AVS Special Member
 
mastermaybe's Avatar
 
Join Date: Sep 2007
Posts: 7,156
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 52 Post(s)
Liked: 155
I'm certainly not an amplifier designer, nor do I play one on tv, but I took a "gamble" on this: http://www.audioholics.com/reviews/p...plifier-review , which I believe runs A/B then G during extremely high output.

Now, I'm not entirely sure how/if "G" parallels "D" in operation/ouput, but I'll tell you that I simply cannot discern a difference in SQ over my other/past Class A, A/B amps, even at insane (to me 100+) SPLs.

I'm always interested to know the "audible" outcomes of these benefits. I would never deny that they measure "better", however.

Very much anxious to learn if these are capable of such (audible) distinctions.


James

Actual phone call (see pic to left):

 

Tech (responding to laughter): "I'm sorry sir, did I miss something?"

Me: "Yeah, a case of Diet Mountain Dew walking across my living room."

mastermaybe is offline  
post #8 of 74 Old 08-26-2010, 07:18 AM
AVS Special Member
 
mrjktcvs's Avatar
 
Join Date: Jun 2005
Posts: 1,432
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 14
Unfortunately, it will take at least 2 units for a home theater application.
mrjktcvs is offline  
post #9 of 74 Old 08-26-2010, 10:37 AM - Thread Starter
 
HiFiFun's Avatar
 
Join Date: Nov 2008
Location: USA
Posts: 2,209
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 17
Here is the heart of the Samsung 700:

Photo courtesy of CowaJawa
Note Samsung now rubs the chip markings off.
HiFiFun is offline  
post #10 of 74 Old 08-26-2010, 10:47 AM - Thread Starter
 
HiFiFun's Avatar
 
Join Date: Nov 2008
Location: USA
Posts: 2,209
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 17
Pulsus Dominates international market in digital-to-digital amplifying chip technology
Quote:
The quality of pounding and screeching sounds is vital when it comes to domestic entertainment.
In today's digital world, even a dull horror flick can make viewers shriek if their home theater system is good enough.It's all music to the ears of Oh Jong-hoon, chief executive officer of Pulsus Technology, which supplies groundbreaking semiconductors that enhance the sound of electronic products to companies worldwide.

Oh's keen interest in audio systems led him to commercial uses in 1999, when he was teaching physics at Pohang University of Science and Technology.

The professor invited audio professionals and students to work on a unique digital pulse modulation amplifier technology.

A year later, Pulsus developed the world's first digital-to-digital amplifying chip (DDC), which transforms digital signals from a CD or a DVD into another digital form for a speaker system's amplifier. In conventional amplifiers, digital signals are changed into analog before being amplified, thus causing static noise and distortion.

More importantly for Pulsus customers, DDC dramatically increases energy efficiency and enables the size reduction of products in which they are used, such as home theater in-a-box, digital TVs, set-top boxes, mini-components and mobile communication devices.

Pulsus chips now dominate domestic and international markets in digital audio processors for full digital amplifiers.

The venture's sales last year reached 6 billion won ($50 million)."Our competitors in the DDC market include Texas Instruments and ST Microelectronics.

They are world's third and fourth-largest makers of fabless semiconductors - companies that design chips and outsource the production of silicon wafers - after Qualcomm and Intel.

We plan to develop a variety of chips other than those for digital amplifiers and aim to become one of the top 20 by the year 2010," said Oh."This means our sales would have to exceed $200 million by then. Our goal this year is to double the sales to $100 million," said the professor-turned-entrepreneur.

The company's major investors include global players Carlyle Group and Softbank.

Pulsus keeps them informed through a corporate governance system that meets foreign standards. "Everything, including the employees' salaries, is reported to the board of directors in English after each quarter.

Five out of the six board members are from outside the company," explained Oh. Unlike many small ventures where management information is rarely open to the public, Pulsus has reached global standards in terms of transparency, according to Wesley Koo, senior associate of Softbank Korea.

Pulsus DDC is now used by LG Electronics Corp., Samsung Electronics Co., Sony Corp., Kenwood, RCA, Toshiba and Taekwang. Original equipment manufacturers such as Europe's Thompson, America's Apex and Singapore's Creative Labs have also applied them in this year's new products.

The venture operates branches in Hong Kong and Shenzhen hoping to seize the Chinese market because nowadays most customer electronics products are manufactured in China once sales growth matures.

"Although our overseas distributors in China, Japan, Britain and Singapore were quite helpful, it was usually the original equipment makers themselves who found out about Pulsus Technology and proposed deals before distributors could even reach them," said Oh.

Another area that differentiates Pulsus from others is its devotion to technical support.
"When we sell our products, we don't just hand them the chips.
We provide a total solution.
We give them a half-assembled set showing how the chips should be placed inside the equipment.
We work together with the OEMs to complete the final equipment and provide technical support even after the equipment has been sold to customers," said Oh. Pulsus is currently preparing to launch a new product for mobile communication devices.

"Our aim is to become the Qualcomm of Korea," said Oh."
-------
So there is NO D/A conversion in the Samsung 700 receiver!! A huge step forward in technology. Impressive engineering that makes an audible improvement over everything, else just as I observed. SFB?

Does Extron license Pulsus technology? I guess yes

http://www.pulsus.co.kr/news/news_vi...keyfield=&key=
HiFiFun is offline  
post #11 of 74 Old 08-26-2010, 10:51 AM
AVS Special Member
 
M Code's Avatar
 
Join Date: Mar 2003
Location: Joshua Tree, CA
Posts: 9,855
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 88 Post(s)
Liked: 121
Quote:
Originally Posted by HiFiFun View Post

The largely unacknowledged downside of all digital electronics is the harmonics of the of high frequency square waves wrecking havoc into the surrounding analog circuitry. The worst spewing case is with the high power square waves in the switching power supply or output stage, as in Class D/PWM amplifiers.

Some of your information is dated..
It is true that Class D solutions have a big challenge to remove/filter out the high frequency harmonics but here alot of progress has been accomplished...
Also note that the switching power supplies can generate distortion components as well, thats why certain Class D solution can use a conventional linear power supply. Also a linear power supply is lower cost as well..

Just my $0.00..
M Code is online now  
post #12 of 74 Old 08-26-2010, 11:09 AM
AVS Special Member
 
M Code's Avatar
 
Join Date: Mar 2003
Location: Joshua Tree, CA
Posts: 9,855
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 88 Post(s)
Liked: 121
Quote:
Originally Posted by HiFiFun View Post

So there is NO D/A conversion in the Samsung 700 receiver!! A huge step forward in technology. Impressive engineering that makes an audible improvement over everything, else just as I observed. SFB?

Again your information is dated..
The 1st 7.1 AVR without D/A conversion was the Harman/Kardon DPR2005 introduced 6 years ago in 2005...
Its Class D solution was provided by D2Audio which is now part of Intersil, also it is interesting to note that the original D2Audio solution passed THX certification as well..
Their amplifier module could output 125W x 7 continuously, all channels driven.

Note that the Samsung AVRs that use the Pulsus solution are very innovative products, however their marketing challenges are:
1. Samsung brand is not established or accepted well in the quality audio circles
2. Like in their Samsung Blu-ray players they did not get the industry's HDMI/HDCP certification for use with HDMI repeaters, and hence they have multiple HDMI inter-operability issues

IMHO..
If they fail to recognize and address the above, they face major challenges for market acceptance in the AVR category..

Just my $0.00..
M Code is online now  
post #13 of 74 Old 08-26-2010, 01:38 PM
emf
Senior Member
 
emf's Avatar
 
Join Date: May 2008
Posts: 200
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
You are right that Samsung has a marketing challenge. I see their receivers on the shelf, but I ignore them. They seem very much a computer monitor name. Other than your post, I have never seen positive comment about their audio quality. Pioneer has been implementing amps patented by Bang&Oluffson, and for many of us, the name makes it OK; although you may have doubts as to ultimate audio quality, you don't doubt that at the very least Pioneer makes the amp modules OK.

Crown - an excellent pro audio name - has come up with a series of feather-weight amps. I don't need a kilo-watt amp that weighs in at 8.6 lbs, but it provokes curiosity. That's a lot of watts per pound.

Other than the world of pro audio, the new generation of class D amps gets motivation from a need to cram 7 amplifiers into a moderate size and weight box at a very moderate price. It's not really "new," because my JVC D702 receiver has a small form factor and weighs in at 17 lbs, and it has a tag that says 2005.
emf is offline  
post #14 of 74 Old 08-26-2010, 02:35 PM
AVS Addicted Member
 
Bill Mac's Avatar
 
Join Date: Nov 2003
Posts: 11,465
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 79 Post(s)
Liked: 423
Quote:
Originally Posted by HiFiFun View Post

For this reason plasma sounds better. Ha!

Do you mean that loud buzzing noise that emits from many Samsung plasmas.

Bill

My SACD collection, watch it grow and my wallet shrink ;-).

 

Denon 4311 (in preamp mode), Parasound 2100, Boston Acoustics A7200 amp, Oppo BDP-103, Consonance CD120, Panasonic TC-P60GT50 plasma, Panamax 5100EX, Salk Song Towers, Song Center, ADS 300C (surrounds) and two Rythmik F12SEs.
Bill Mac is online now  
post #15 of 74 Old 08-26-2010, 03:46 PM
Advanced Member
 
xianthax's Avatar
 
Join Date: Oct 2008
Posts: 513
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 12
Quote:
Originally Posted by HiFiFun View Post

So there is NO D/A conversion in the Samsung 700 receiver!! A huge step forward in technology.

This is just marketing hype. The output is obviously analog, the only difference here is they are not converting to analog (actually they still kinda are) to send the signal to the class D amplifier input section, rather the D to A conversion occurs only in the conversion performed by the class D amp. A class D amp is itself a DAC by design.

Most class D amps take an analog input and generate PWM signals to drive the class D power stage, this topology just takes the digital signal and converts it to the PWM drive signal directly.

This isn't new, for example the TI TAS5518 has been around since 2004, its an 8 channel digital audio to PWM drive IC like the pulsus PS9830B. Ti calls this hookup "PurePath" and they currently make 27 different class D power stages that accept PWM input, and a whole bunch of conversion IC's as well as several DSP's that output PWM directly thus only requiring the class D power stage externally. I don't know how long this has been around but i remember seeing such parts pre 2004 as well. This PWM conversion is also exactly what the pulsus PS9830B does (the part pictured above).

Incidently, the digital audio to PWM conversion can also be looked at as a D to A conversion, in fact if all you wanted was a line level analog audio signal, you can take the PWM signal, basically just low pass filter it, and poof, theres your analog audio signal. More care is also required to ensure there is no noise pick up in the PWM signal, because it runs at the same frequency as the high power output its easier for noise to transfer to this signal than one in the audio spectrum, although the effect of the noise picked up is likely lower.

Generally speaking the analog signal traces between a conventional DAC and the input to an Analog class D front end are very short and usually differential, If at all well protected the noise pick up should be well below the noise floor. Modern class D amps have much, much higher switching frequencies than older designs (which is actually the major source of improvement in THD as low frequency is indicative of long dead times). So noise impact in the audio spectrum is virtually non existent.

The danger is often in high frequency pick up in the reconstruction filter at the output of the DAC. These filters with almost always have a op amp or 2 which have limited GBP, high frequency transients can cause funkiness, even oscillation, in the filter. It shouldn't be an issue in a production design as such effects are accounted for in the circuit design and PCB layout.

The real reason to move to this topology is cost. High end burr-brown DAC's are really expensive. Digital audio to PWM parts are very cheap, as are class D power stages (really just matched high current switching transistors, usually in an H-bridge, with dead time control circuitry). Class D amps also produce much less heat resulting is lower heat sinking requirements.

The added cost to Class D designs is the need for low pass filtering of the output of the power stage. Due to the high power this requires passive filters, which means high power shielded inductors, which aren't cheap or small. They also need to be carefully designed to avoid introducing distortion, to avoid saturation during current spikes. Interestingly, with some coils, you can actually hear the music being played from the coil itself, the current is high enough to cause the coil to vibrate and in turn vibrate whatever it's attached to (usually a PCB or shielding material) turning it into a speaker.

The filter also needs to be very robust, because even small high frequency transients sent down a long speaker cable can introduce are EMI emissions from the cable itself (or an unshielded driver at the end). The length of the speaker wire is the real reason companies often have to employ EMI chokes to pass FCC part 15. Since the requirement is to test in an "expected configuration" i would bet they choose the shortest possible speaker cables and use shielded speakers to further ease the FCC testing.

For instance, a product i designed some time back employed class D amps, only 18W per channel, but the system drove 20 channels across long cable distances with high impedance unshielded drivers. Chokes ended up being required on the cables to pass FCC.

The EMI emissions inside the receiver are mostly caused by the inductors used to the filter the class D output, using quality shielded inductors mostly solves this issue. A poorly designed power supply that can't handle the high power switching can also cause emissions but this can and should be designed for.

I do consider moving to this type of design to be a good thing, its cheaper and removes unnecessary conversion. I highly doubt you'll see audible changes in performance, unless the analog input class D system you compare to was very poorly designed or an older design with just barely high enough switching frequency to get by (indicative of long dead times).
xianthax is offline  
post #16 of 74 Old 08-26-2010, 05:57 PM - Thread Starter
 
HiFiFun's Avatar
 
Join Date: Nov 2008
Location: USA
Posts: 2,209
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 17
Nice explanations but wrong conclusion, anyone with a good system should be able to hear the difference. Think - why is it called a crystal amplifier?

Its great when members make a "probably can't hear it the difference", but have never listened to it.
Is the problem is that we have long suffered with mediocre to terrible sound quality?

The reactuion to this 2010 Samsung gear (the Oppo guys ran me out of town for suggesting the Samsung 6800 Blu-ray is better than the 83) is very negative as everyone with a vested interest tries to tear this superior new gear down, by casting doubt. The oldest technique since the beginning of mankind.

Does not anyone have the self confidence in their ears and eyes to give an accurate review? I guess we will need to wait a few months for Home Theater Magazine to post a review and once again claim (an inside joke to last year best Blu-ray player) they had no desire to go back to their conventional, more expensive gear.

Samsung must be doing some type of mathematical transform from the (PCM) time domain to the frequency/PWM domain. Sweet.

In any event there is a neutral sweetness, articulation and inner detail to the music I've never heard before. Take drums, everyone hears the main note but what about the harmonics? Its a new experience in amplification

There is still one flaw (and ONLY with 3D after the firmware upgrade) in that the receiver must get very busy processing double-rate pass-thru data and making the audio transforms, as commands from the remote are only infrequently accepted.

M-Code,
Traditional Class D is old news but this is another animal altogether. You have read my posts in other threads raving over the Samsung 700 for the past month (I bought four of them), yet only now you warm up to it when presented with convincing evidence. Welcome aboard, but please go buy one and start listening.
HiFiFun is offline  
post #17 of 74 Old 08-26-2010, 06:07 PM
AVS Addicted Member
 
MichaelJHuman's Avatar
 
Join Date: Nov 2006
Location: Colorado Springs, CO
Posts: 18,756
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 164 Post(s)
Liked: 99
Self mentioned that "digital" class D has issues because it's hard to use feedback to correct errors. Certainly someone could have found ways to deal with that issue.

Anyway, it does not seem like digital class D is considered to be better by all just because...

"But this one goes up to 11"
MichaelJHuman is online now  
post #18 of 74 Old 08-26-2010, 06:19 PM - Thread Starter
 
HiFiFun's Avatar
 
Join Date: Nov 2008
Location: USA
Posts: 2,209
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 17
Quote:
Originally Posted by MichaelJHuman View Post

Self mentioned that "digital" class D has issues because it's hard to use feedback to correct errors. Certainly someone could have found ways to deal with that issue.

Anyway, it does not seem like digital class D is considered to be better by all just because...

One can talk endlessly what car to buy, but one 15 minute test drive can change everything.
HiFiFun is offline  
post #19 of 74 Old 08-26-2010, 06:34 PM
AVS Special Member
 
M Code's Avatar
 
Join Date: Mar 2003
Location: Joshua Tree, CA
Posts: 9,855
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 88 Post(s)
Liked: 121
Quote:
Originally Posted by HiFiFun View Post

M-Code,
Traditional Class D is old news but this is another animal altogether. You have read my posts in other threads raving over the Samsung 700 for the past month (I bought four of them), yet only now you warm up to it when presented with convincing evidence. Welcome aboard, but please go buy one and start listening.

Don't need to warm up or buy 1, we know the Samsung lead engineers and AVR platform well....
As I mentioned previously in other threads this platform has serious HDMI interoperability issues, note that this info comes from selling dealers not from some unsubstantiated source. However this situation is very similar to their Blu-ray products, Samsung chooses to bypass industry certification for HDMI/HDCP to save $ and ship sooner.

Check out the Blu-ray section of this forum and you can find more details for this subject. Samsung does test their HDMI products internally, however their testing process is very abbreviated compared to the more extensive industry certification process. If the Blu-ray product is connected directly to an HDMI display there are fewer issues, the problems arise when connected to an AVR thats uses an HDMI repeater circuit. If Samsung fixes these issues, the AVR platform will be much stronger. Note that they just released a major S/W update addressing many of the reported issues for their AVRs.

Bottom line..
Samsung is an incredibly sucessful company in certain CE categories including video display, cell phone and semi-conductor biz...
And they are pushing very hard to be #1, but frequently fall short here.

Just my $0.00..
M Code is online now  
post #20 of 74 Old 08-26-2010, 06:38 PM
AVS Addicted Member
 
MichaelJHuman's Avatar
 
Join Date: Nov 2006
Location: Colorado Springs, CO
Posts: 18,756
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 164 Post(s)
Liked: 99
Quote:
Originally Posted by HiFiFun View Post

One can talk endlessly what car to buy, but one 15 minute test drive can change everything.

I wish it was that simple for audio electronics. In my experience, they sound so much alike, 15 mins of listening won't tell you anything useful.

"But this one goes up to 11"
MichaelJHuman is online now  
post #21 of 74 Old 08-26-2010, 07:12 PM
Advanced Member
 
xianthax's Avatar
 
Join Date: Oct 2008
Posts: 513
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 12
Quote:
Originally Posted by HiFiFun View Post

Nice explanations but wrong conclusion, anyone with a good system should be able to hear the difference. Think - why is it called a crystal amplifier?

Post actual measurements or i won't believe you because your claim doesn't make scientific sense.

The primary source of distortion in a class D amplifier is not in PWM quantization error or linearity, but rather in the power stage. the primary source of THD by an order of magnitude or more is excessive dead time. Secondary important sources are finite Rds in the H-Bridge MOSFETs and manufacturing tolerances for matching components on each leg of the bridge. All of these are issues with the power stage, not with what the input signal is.

As michael said, by not feeding the amplifier an analog input you actually make it more difficult to remove the distortions caused by these issues as you remove an easy way to implement negative feedback in the system. If your gaining PWM input but its forcing to go open loop, its most likely a performance loss, not a win.

Its not at all easy to implement feedback correction with a PWM input but it can be done. For instance using a topology with direct digital feedback to mitigate the distortion issues with class D amps will go miles further to improve the performance of such amps than messing with the input stage will.

Quote:
Originally Posted by HiFiFun View Post

Samsung must be doing some type of mathematical transform from the (PCM) time domain to the frequency/PWM domain. Sweet.

Nothing special and this occurs within the pulsus IC, samsung has nothing to do with it, this is a well known algorithm, for instance those little $1.00 greeting cards that play a message or music when you open them store a PCM audio signals and modulate a PWM output to make the sound. The implementation is a simple quantization against a triangular waveform (visually). Implementing in reality just requires a count up/down per sample and a timer.

Quote:
Originally Posted by HiFiFun View Post

Does not anyone have the self confidence in their ears and eyes to give an accurate review? I guess we will need to wait a few months for Home Theater Magazine to post a review and once again claim (an inside joke to last year best Blu-ray player) they had no desire to go back to their conventional, more expensive gear.

In any event there is a neutral sweetness, articulation and inner detail to the music I've never heard before. Take drums, everyone hears the main note but what about the harmonics? Its a new experience in amplification

Not at all, human visual and sound processing isn't very reliable. Placebo effect, chemical balance changes, ear wax, etc, etc.

Double blind or it didn't happen
xianthax is offline  
post #22 of 74 Old 08-26-2010, 07:41 PM - Thread Starter
 
HiFiFun's Avatar
 
Join Date: Nov 2008
Location: USA
Posts: 2,209
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 17
I took my new Droid X cell phone to work and only one engineer out of four wanted to compare it to their i-Phone 3&4 side-by-side after seeing the Droid X.
Human nature you suppose?
HiFiFun is offline  
post #23 of 74 Old 08-27-2010, 12:33 AM
Newbie
 
Join Date: May 2007
Posts: 6
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 3 Post(s)
Liked: 10
Anybody have any amplifier information on the new Marantz NR1601 and NR1501 AVRs? The form factor, output specs, and weight hint at Class D.
energizerfellow is offline  
post #24 of 74 Old 08-27-2010, 01:19 AM
Advanced Member
 
welwynnick's Avatar
 
Join Date: Jul 2005
Location: Welwyn, Herts, UK
Posts: 976
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 13
Quote:
Originally Posted by xianthax View Post

This is just marketing hype. The output is obviously analog, the only difference here is they are not converting to analog (actually they still kinda are) to send the signal to the class D amplifier input section, rather the D to A conversion occurs only in the conversion performed by the class D amp. A class D amp is itself a DAC by design.

This is an interesting philosophical argument (to some of us ). I've been wrestling with this one for a while, and I think I have an answer - roll your sleeves up! A pure class D amplifier with no analogue intermediate stage can be thought of as EITHER:

A high-power DAC with no amplifier, OR:

A digital power amplifier with no DAC.

In isolation I don't think that class D amps necessarily make very good amplifiers, nor do they make very good DACs, but dropping one stage must eliminate some degradation. Either way you look at it, eliminating a conversion or amplification stage is desirable, but which one is it? Both points of view seem valid. I think the answer lies with the sampling and reconstruction rates.

With digital audio analogue signals are sampled at say, 44kHz, and analogue audio is reconstructed by the DAC at the same sample rates. Of course, modern DACs use up-sampled audio and convert at 192kHz, but I don't think that's a fundamental difference. Each amplitude sample is valid at each 2.3ms interval, or whatever, and the D to A conversion is performed at those intervals. That's what a DAC does, it just doesn't do it at speaker power.

Class D amplifiers don't reconstruct analogue audio at digital audio sampling frequencies - they do it at much higher frequencies. Class D works in the switching domain, not the audio domain. When a class D amp reconstructs analogue audio, the switching interval is much shorter, so there are many switching cycles within each digital audio cycle. It's the switching cycles that are LC low-pass filtered to reconstruct the analogue signal. But the actual transition from the digital to analogue domain, where the waveform is re constructed, is performed in the switching domain.

Therefore I would volunteer that an all-digital amplifier shouldn't be considered as a high-power DAC with no amplifier, but is better thought of as a digital power amplifier with no DAC.

Nick
welwynnick is offline  
post #25 of 74 Old 08-27-2010, 01:46 AM
AVS Addicted Member
 
MichaelJHuman's Avatar
 
Join Date: Nov 2006
Location: Colorado Springs, CO
Posts: 18,756
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 164 Post(s)
Liked: 99
You could maybe think of a digital class D amp as akin to a one bit DAC though (delta-sigma.) Like how SACDs work.

It's all mental ************ in the end...I think.

The bottom line is whether it sounds good. I have insufficient experience with any digital amp to answer that.

If a class D amp sounded 95% as good as my Emotiva XPA-3 for not much more money, I would happily trade for the lighter amp But so far, no one is making standalone class D amps with the power and price of Emotiva class AB amps.

It's interesting, you can read stuff from various people I would consider experts, and even they don't agree on how to design an amp. As I recall, John Curl thinks a heavy class A bias is important, and Douglas Self thinks class A bias is a questionable practice (due to the issues resulting at the switchover point from A to B IIRC.) Anyway, you can find many examples where experts disagree. So I guess ultimately you are going to have to listen for yourself (which is in itself fallible, IMO

You would think you could measure distortion, and that would be sufficient. But there's some evidence that a simple THD measurement does not tell the whole story.

"But this one goes up to 11"
MichaelJHuman is online now  
post #26 of 74 Old 08-27-2010, 04:28 AM - Thread Starter
 
HiFiFun's Avatar
 
Join Date: Nov 2008
Location: USA
Posts: 2,209
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 17
Pulsus Patents Excerpts

Abstract: An efficient method of amplifying a digital signal with reduced electromagnetic interference is disclosed. The invention applies an average amplitude detector to the input signal to control the clock frequency of a pseudo-random number generator so that the clock signal decreases when the average amplitude of the signal increases. The input signal and the pseudo-random number generator are compared to generate a control signal for a switching power amplifier followed by a low-pass filter. When the average amplitude of the input signal falls below a pre-determine threshold, the switching amplifier is placed in a sleep mode.

Or stated another way:
A power amplifier with random switching, comprising: a) an amplitude detector, the amplitude detector having an input and an output, the input connected to a digital signal; b) a variable clock, the variable clock having an input connected to the output of the amplitude detector, and an output; c) a pseudo-random number generator, the pseudo-random number generator having an input connected to the output of the variable clock and an output; d) a comparator, the comparator having a first input, a second input, and an output, the second input connected to the output of the pseudo-random number generator, and the first input connected to the digital signal; e) a power amplifier, the power amplifier having an input connected to the output of the comparator, and a control input connected to the output of the amplitude detector, and an output; f) a filter, the filter having an input connected to the output of the amplifier, and an output; and g) a transducer, the transducer having an input connected to the output of the filter.
http://www.patentgenius.com/patent/7400194.html
With the construction of the present invention, the use of the paired inductor in the digital audio amplifier may minimize space and heat generation within the digital audio amplifier, thereby decreasing the costs as well.
... a DDC (digital to digital converter) so that a PCM-type digital audio signal is converted into a PWM-type digital audio signal.(read no D/A converter)
http://www.patentgenius.com/patent/6867646.html
The shape of the core establishes a high degree of self-shielding in order to minimize radiated emissions which may affect adjacent elements, especially at higher frequencies and drive levels. The area occupied by the transformer or inductor device of the present invention on the PCB is small and the volumetric efficiency is high in order to achieve high circuit densities. Openings in the core allow the coil windings to be terminated easily on a bobbin prior to core assembly, although some wires may extend beyond the core. In addition, openings in the core may be mounted in a preferred orientation, i.e. facing the PCB where use of a ground plane can most easily minimize undesirable effects such as radiated emissions or susceptibility to EMI and cross-talk.
http://www.patentgenius.com/patent/7135949.html

Now that you have the technical knowledge, it would be a disservice to lump Pulsus amplifier technology in with other inferior class D designs, as there is considerable (and audible) refinement offered here.
Did Pioneer state they had to work for two years to get an ICE amplifier to sound acceptable in their $2.2K receivers?
Pulsus advanced technology described herein has overcome these sonic limitations, and is ready for world-wide use, and without fine tuning. (Interestingly Toshiba appears to be the only major Japanese company on board).
Will traditional "high-end" manufactures will have a hard time differentiating their products? Perhaps they can put it into a large heavy case and advertise in Stereophile!
The rest of us should consider the $400 7.1 channel Samsung and enjoy the benefits over virtually every other receiver or amplifier in the under 150 watt range.
HiFiFun is offline  
post #27 of 74 Old 08-27-2010, 06:22 AM
AVS Special Member
 
smurraybhm's Avatar
 
Join Date: Jul 2007
Location: Birmingham, Alabama
Posts: 1,568
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 102 Post(s)
Liked: 381
Do you ever read the crap you post all over this forum? I think its great you are happy with your Samsung receiver (really) but calling it the best receiver or amp in the 150 watt range based on no REAL data other than what you think and some marketing info - thanks for making me laugh so early in the morning. May, might and other like words are not what you use to start a fact based claim with - is it love when your girlfriend says "I may love you?" Maybe, maybe not. I think I'd wait to buy the ring until she was sure.
Next you're going to throw up a stat on the LCD thread talking about the display that has a 8 million contrast ratio - manufacturer provided of course. I don't know of anyone who paid $2.2k for the Pioneer Elite receiver - unless it was an SC-09. Nice shot at plasma - I believe its Samsung that is well known for making buzzers. One more thing since you mentioned blu-ray players, how about that great firmware update Samsung put out in the past week? Enjoy your receiver, I believe I will stick with outdated technology that the others use like Denon, Onkyo, Pioneer, etc. for reasons other than marketing hype - first one being proven performance.
smurraybhm is offline  
post #28 of 74 Old 08-27-2010, 06:28 AM
AVS Addicted Member
 
MichaelJHuman's Avatar
 
Join Date: Nov 2006
Location: Colorado Springs, CO
Posts: 18,756
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 164 Post(s)
Liked: 99
Most of that description seemed to match a digitally controlled class D amp, which already exist. I don't see how that description make the other class D designs "inferior"!!

I would need way more info to determine that

"But this one goes up to 11"
MichaelJHuman is online now  
post #29 of 74 Old 08-27-2010, 07:45 AM
Advanced Member
 
Mr. Brownstone's Avatar
 
Join Date: Nov 2005
Location: Virginia
Posts: 719
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 12
I used to have reservations about Class D. I think everyone did. That partially explains why they were largely restricted to subwoofer applications for so long. But Class D technology has advanced.

I've owned a Class D power amplifier for almost a year after three different Class AB amps (and a few integrated amps and receivers before that). If there's been some significant degradation in sound quality, damned if it's apparent to me.

Class D seems highly refined now, and I wonder if it will eventually come to dominate the market, as the advantages in size, weight and temperature relative to output are absolutely enticing. I would think the major impediment now is licensing fees to Class D specialists such as B&O ICEpower.

There will always be an "audiophile" niche that resists Class D based on theory rather than practical results. But, hell, there're still people stuck on Class A.
Mr. Brownstone is offline  
post #30 of 74 Old 08-27-2010, 08:12 AM
AVS Special Member
 
gurkey's Avatar
 
Join Date: Aug 2007
Posts: 2,749
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 22
There must be technical reasons, why most of the earlier suspects, who had class D amps or similar in their repertoire, like Panasonics etc. reverted back to analog amplification.
Especially taken into consideration, that analog technology is normally more expensive than digital technology, considering current and future channel and feature count.

It is suspicious, that there are only such few offers, looking at the economics (weight, cost, casings etc.) of building digital amps. Especially the major players in the multimedia field would have gone there already for obvious reasons if not ... (fill in the blanks)
gurkey is offline  
Reply Receivers, Amps, and Processors

User Tag List

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off