Starting to have second thoughts on Class D amps - Page 4 - AVS Forum
Forum Jump: 
Reply
 
Thread Tools
post #91 of 148 Old 01-30-2014, 02:38 AM
Senior Member
 
Antripodean's Avatar
 
Join Date: Apr 2009
Location: The Antripodes (OZ)
Posts: 328
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 12
According to the patent Lab Gruppen is a linear audio power amp using bipolar transistors. It's not Class D. It's Class H for want of a precise classification due to the tracking SMPS.

Paskal9 explains it well. In addition, patents can be had for many different reasons, perhaps just for the marketing side of things.

explore the music
Antripodean is offline  
Sponsored Links
Advertisement
 
post #92 of 148 Old 01-30-2014, 05:51 AM
Senior Member
 
splotten's Avatar
 
Join Date: May 2009
Posts: 468
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 84 Post(s)
Liked: 52
If anybody is interested in LG service manuals a few can be downloaded form here. I haven't looked at them yet but there might be useful info available. You can download a few manuals for free every day without registering or paying.

http://elektrotanya.com/?q=showresult&what=lab%20gruppen&kategoria=&kat2=all
splotten is offline  
post #93 of 148 Old 01-30-2014, 05:59 AM
AVS Special Member
 
J_Palmer_Cass's Avatar
 
Join Date: May 2005
Posts: 6,463
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 271 Post(s)
Liked: 102
Quote:
Originally Posted by bossobass View Post

Quote:
Originally Posted by Antripodean View Post

Hi Bosso

LG calls it Class TD. To quote from their site " The flagship of Lab.gruppen’s touring amplifier line, the FP 14000 produces a staggering total output power of 14000 W – an unprecedented achievement in a 2U frame weighing only 12 kg. The forceful high-current output stage is built around patented Class TD® technology, a breakthrough topology that approaches the efficiency of Class D while retaining the favored sound quality of proven Class B designs. At the same time, Lab.gruppen’s Regulated Switch Mode Power Supply (R.SMPS) has been updated to provide more sustained high power during extended bursts of low frequency content."

Or if I quote for the smaller FP range "Lab.gruppen’s patented Class TD output stage is incorporated in the fP 6400 and fP 3400. Combining the artifact-free sonic performance of Class A/B with the efficiency of Class D, this technology guarantees exceptional sonic transparency along with optimal use of energy."

The tracking power supply is a separate thing while the amp topology is another. What exactly is it if its not a variation on Class A/B?

Hi A,

Marketing departments are not the place to find out how amplifiers work. smile.gif

Class B/AB supplies full voltage all the time. Class H tracks the output and uses a low v rail for the most part until higher v is needed, then it switches to the high v rail (it may employ more than 2 rails, but the improvement with more rails is that of diminishing returns). Therefore, if implemented properly, class H improves efficiency over class B/AB amps.



That describes a Class G amplifier. I own that type of amplifier, and it is classified as Class G in the marketing literature. Class AB with two power rails.



Quote:
Originally Posted by bossobass View Post


Lab does not use multiple rails. Class TD uses class D. tracks the output, supplying Xv + 6V headroom, thus multiple rails are not required. Class TD improves efficiency by providing only the v +6v needed and has more in the tank for transient spikes than class B/AB or H. Unlike Class D, class TD does not require a LPF, as it uses the AB output stage as the filter. Lab also claims that because of that, pumping is not a problem and high frequency amplification is more stable/higher quality than class D.

In any case, class TD is not class H with a made up name, as members have claimed in this and other threads.



That describes a Class H amplifier. Class AB amplifier with a tracking power supply.
J_Palmer_Cass is offline  
post #94 of 148 Old 01-30-2014, 06:10 AM
AVS Special Member
 
J_Palmer_Cass's Avatar
 
Join Date: May 2005
Posts: 6,463
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 271 Post(s)
Liked: 102
Quote:
Originally Posted by bossobass View Post

Quote:
Originally Posted by Antripodean View Post

Hi Bosso

LG calls it Class TD. To quote from their site " The flagship of Lab.gruppen’s touring amplifier line, the FP 14000 produces a staggering total output power of 14000 W – an unprecedented achievement in a 2U frame weighing only 12 kg. The forceful high-current output stage is built around patented Class TD® technology, a breakthrough topology that approaches the efficiency of Class D while retaining the favored sound quality of proven Class B designs. At the same time, Lab.gruppen’s Regulated Switch Mode Power Supply (R.SMPS) has been updated to provide more sustained high power during extended bursts of low frequency content."

Or if I quote for the smaller FP range "Lab.gruppen’s patented Class TD output stage is incorporated in the fP 6400 and fP 3400. Combining the artifact-free sonic performance of Class A/B with the efficiency of Class D, this technology guarantees exceptional sonic transparency along with optimal use of energy."

The tracking power supply is a separate thing while the amp topology is another. What exactly is it if its not a variation on Class A/B?

Hi A,

Marketing departments are not the place to find out how amplifiers work. smile.gif

Class B/AB supplies full voltage all the time. Class H tracks the output and uses a low v rail for the most part until higher v is needed, then it switches to the high v rail (it may employ more than 2 rails, but the improvement with more rails is that of diminishing returns). Therefore, if implemented properly, class H improves efficiency over class B/AB amps.

Lab does not use multiple rails. Class TD uses class D. tracks the output, supplying Xv + 6V headroom, thus multiple rails are not required. Class TD improves efficiency by providing only the v +6v needed and has more in the tank for transient spikes than class B/AB or H. Unlike Class D, class TD does not require a LPF, as it uses the AB output stage as the filter. Lab also claims that because of that, pumping is not a problem and high frequency amplification is more stable/higher quality than class D.

In any case, class TD is not class H with a made up name, as members have claimed in this and other threads.



Where did you dig up your Class AB amplifier description used in Class TD? LG says that they use a Class B amplifier with a tracking power supply.!


"Patented Class TD topology – a masterpiece of design – utilizes a sonically high quality Class B stage for the audio signal path passing directly to the speakers, while at the same time a high frequency tracking rail voltage provides the high efficiency characteristic of a Class D topology. This unique combination produces a flat frequency response into reactive loads, and does not color the sound in any way."



http://labgruppen.com/products/c_series/c/c_884/class_td_psu/
J_Palmer_Cass is offline  
post #95 of 148 Old 01-30-2014, 07:33 AM
AVS Special Member
 
Trepidati0n's Avatar
 
Join Date: Nov 2003
Posts: 1,113
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 16 Post(s)
Liked: 92
Quote:
Originally Posted by J_Palmer_Cass View Post

Where did you dig up your Class AB amplifier description used in Class TD? LG says that they use a Class B amplifier with a tracking power supply.!


"Patented Class TD topology – a masterpiece of design – utilizes a sonically high quality Class B stage for the audio signal path passing directly to the speakers, while at the same time a high frequency tracking rail voltage provides the high efficiency characteristic of a Class D topology. This unique combination produces a flat frequency response into reactive loads, and does not color the sound in any way."



http://labgruppen.com/products/c_series/c/c_884/class_td_psu/

For better or for worse AB and push-pull B can get interchanged quite easily since the only fundamental physical difference is a bias current which is only relevant for small signal (power). Where we typically use class AB and push-pull it looks almost entirely "class B" (e.g. large signal). We get pragmatic or dogmatic on it but from core topology view they are the same damn thing. If I said AB, I apologize since I keep forgetting to post here you need to be incontrovertibly correct and then back it up with a dozen PhD papers which have been blessed by Danley and the blood of a rare Chinese chicken raised in Brazil and slaughtered in a Greek temple was offered to Mohamed. tongue.gif

So yes, the gruppen amplifier probably does have a "push-pull class B" output. However, it is entirely possible because of the shear amount of silicon there is "some bias" that is always present. via leakage (yes, silicon leaks). It is entirely possible that a class B amp looks class AB to a small speaker if the scaling is quite disparate.

I must be guilty because people say I am guilty because they chose to call me guilty because they refuse to see the truth. Much easier to be part of the mob..
Trepidati0n is offline  
post #96 of 148 Old 01-30-2014, 07:43 AM
AVS Special Member
 
Join Date: Mar 2003
Posts: 4,509
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 10 Post(s)
Liked: 609
Class G and class H are mostly interchangeable terms. There is a slight difference and that difference usually becomes a matter of opinion in these fruitless discussions.

Class H tracks the output signal and switches the supply rails accordingly, the # of rails being irrelevant, multiple rails and the method of tracking being the crux. The difference between this class and TD is that TD actually tracks the output where H/G jumps from the low V rail to the high V rail in one step.

Class TD tracks the output but does not switch the supply rails. There are no tiers in class TD. The rails track at output V +6V headroom.

Lab is not class H. Lab is not class G. If you disagree, state your case specifically other than just saying "that's a class H amplifier".
bossobass is offline  
post #97 of 148 Old 01-30-2014, 11:16 AM
AVS Special Member
 
J_Palmer_Cass's Avatar
 
Join Date: May 2005
Posts: 6,463
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 271 Post(s)
Liked: 102
Quote:
Originally Posted by bossobass View Post

Class G and class H are mostly interchangeable terms. There is a slight difference and that difference usually becomes a matter of opinion in these fruitless discussions.

Class H tracks the output signal and switches the supply rails accordingly, the # of rails being irrelevant, multiple rails and the method of tracking being the crux. The difference between this class and TD is that TD actually tracks the output where H/G jumps from the low V rail to the high V rail in one step.

Class TD tracks the output but does not switch the supply rails. There are no tiers in class TD. The rails track at output V +6V headroom.

Lab is not class H. Lab is not class G. If you disagree, state your case specifically other than just saying "that's a class H amplifier".



Semantics and BS come into play here. There is no official Class TD amplifier topology. For that matter, some people claim that there is no official Class G or H amplifier topology.

Wiki and the link below say that Class G amplifiers use a rail switching power supply, and Class H amplifiers use a "voltage tracking" power supply. Aside from the power supply, then you can apply whatever class amplifier that you want to with AB being typical.

LG makes up their own name for an amplifier that uses a "voltage tracking" power supply and a Class B amplifier and names it TD. Some of the LG amplifiers are straight class D amplifier that "emulates Class TD" ( a made up LG class). Not sure what that means exactly either.


http://www.apsdc.com/Amplifiers%20Topologies.htm



LG Class D

http://labgruppen.com/products/c_series/c/c_20_8x/unique_class_td_psu/
J_Palmer_Cass is offline  
post #98 of 148 Old 01-30-2014, 11:32 AM
AVS Special Member
 
desertdome's Avatar
 
Join Date: Aug 2004
Location: Elkhorn, NE
Posts: 1,784
Mentioned: 8 Post(s)
Tagged: 0 Thread(s)
Quoted: 90 Post(s)
Liked: 271
Tommy O'Brien, who designed my amps, has an article on ProSoundWeb explaining Class G & Class H:

Riding The (Moving) Rails: Detailing Class-G And Class-H Amplifier Topologies
LTD02 likes this.
desertdome is offline  
post #99 of 148 Old 01-30-2014, 12:14 PM
AVS Addicted Member
 
noah katz's Avatar
 
Join Date: Apr 1999
Location: Mountain View, CA USA
Posts: 20,628
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 300 Post(s)
Liked: 167
Class D efficiency is the result of the output device being either on or off and not in between where it would bve dropping voltage and generating I^2*R losses.

So how can it approach anywhere near class D efficiency with class B output stages?
Quote:
Originally Posted by J_Palmer_Cass View Post

"Patented Class TD topology – a masterpiece of design – utilizes a sonically high quality Class B stage for the audio signal path passing directly to the speakers, while at the same time a high frequency tracking rail voltage provides the high efficiency characteristic of a Class D topology. This unique combination produces a flat frequency response into reactive loads, and does not color the sound in any way."

http://labgruppen.com/products/c_series/c/c_884/class_td_psu/

Noah
noah katz is offline  
post #100 of 148 Old 01-30-2014, 12:31 PM
AVS Special Member
 
desertdome's Avatar
 
Join Date: Aug 2004
Location: Elkhorn, NE
Posts: 1,784
Mentioned: 8 Post(s)
Tagged: 0 Thread(s)
Quoted: 90 Post(s)
Liked: 271
Quote:
Originally Posted by noah katz View Post

Class D efficiency is the result of the output device being either on or off and not in between where it would bve dropping voltage and generating I^2*R losses.

So how can it approach anywhere near class D efficiency with class B output stages?
Lap Gruppen explains it on their paper on the theory of class TD and their Technical Note on Class TD amplification.
desertdome is offline  
post #101 of 148 Old 01-30-2014, 12:43 PM
AVS Special Member
 
J_Palmer_Cass's Avatar
 
Join Date: May 2005
Posts: 6,463
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 271 Post(s)
Liked: 102
Quote:
Originally Posted by noah katz View Post

Class D efficiency is the result of the output device being either on or off and not in between where it would bve dropping voltage and generating I^2*R losses.

So how can it approach anywhere near class D efficiency with class B output stages?
Quote:
Originally Posted by J_Palmer_Cass View Post

"Patented Class TD topology – a masterpiece of design – utilizes a sonically high quality Class B stage for the audio signal path passing directly to the speakers, while at the same time a high frequency tracking rail voltage provides the high efficiency characteristic of a Class D topology. This unique combination produces a flat frequency response into reactive loads, and does not color the sound in any way."

http://labgruppen.com/products/c_series/c/c_884/class_td_psu/




LG in Class TD uses a class D amplifier to provide / regulate power to the power supply circuit.

The audio amplifier itself if not Class D, but Class AB.
J_Palmer_Cass is offline  
post #102 of 148 Old 01-30-2014, 01:14 PM
AVS Addicted Member
 
noah katz's Avatar
 
Join Date: Apr 1999
Location: Mountain View, CA USA
Posts: 20,628
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 300 Post(s)
Liked: 167
Quote:
Originally Posted by J_Palmer_Cass View Post

The audio amplifier itself if not Class D, but Class AB.

Yes, ergo my question.

Noah
noah katz is offline  
post #103 of 148 Old 01-30-2014, 06:47 PM
AVS Special Member
 
Join Date: Mar 2003
Posts: 4,509
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 10 Post(s)
Liked: 609
Quote:
Originally Posted by Trepidati0n View Post

I apologize since I keep forgetting to post here you need to be incontrovertibly correct and then back it up with a dozen PhD papers which have been blessed by Danley and the blood of a rare Chinese chicken raised in Brazil and slaughtered in a Greek temple was offered to Mohamed. tongue.gif.

No sh!t. People have casually said on many occasions that the clones are class H and no one calls BS. But, state the obvious, that it isn't class H and have to be stripped searched and interrogated for pages. But, yes, this is the norm around here to get a point across for the benefit of those who rarely post but want to know.

Quote:
Originally Posted by J_Palmer_Cass View Post

Semantics and BS come into play here. There is no official Class TD amplifier topology. For that matter, some people claim that there is no official Class G or H amplifier topology.

Wiki and the link below say that Class G amplifiers use a rail switching power supply, and Class H amplifiers use a "voltage tracking" power supply. Aside from the power supply, then you can apply whatever class amplifier that you want to with AB being typical.

LG makes up their own name for an amplifier that uses a "voltage tracking" power supply and a Class B amplifier and names it TD. Some of the LG amplifiers are straight class D amplifier that "emulates Class TD" ( a made up LG class). Not sure what that means exactly either.

The bottom line is that the amp is not class H. The rest of the discussion is, as is usually the case, irrelevant.

Quote:
Originally Posted by desertdome View Post

Tommy O'Brien, who designed my amps, has an article on ProSoundWeb explaining Class G & Class H:

Riding The (Moving) Rails: Detailing Class-G And Class-H Amplifier Topologies

Thanks DD.

The 2nd posted link shows the difference between the class TD and H, D, etc., in simple diagram form. Should make it clear to the usual suspects.
_______________________________________________________

The confusion comes from the typical poster on these forums who basically knows little about amplifiers but loves "Wiki" and "Google". Wiki has no information about class TD and LG isn't so stupid as to try to spell out its patent in marketing brochures, so Google will be of no help either when it comes to a drive-by version of amplifier topology.

Another probable source of confusion is that LG calls their power supply a class D amplifier, or, in other technical writings, more specifically, a Tracking Class D Voltage Provider, which it technically is. There's nothing new about the idea of using a class D amplifier as a power supply. Bob Carver calls his similar design "Tracking Downconverter". It tracks the input, but the efficiency increase comes from the same principal as the LG design. They both also have the ability to deliver extremely high current as voltage requirement drops (as the impedance of the load being driven drops).

I have tried to mention this before but some objected that it was BS or whatever. In any case, a typical amp has a power supply that offers its full voltage all the time which high voltage is lost in heat through the transistors that are receiving that full voltage when there is low or zero input, which is the case with music program and especially the case with a sealed, L/T'd HT subwoofer. They also are less able to provide high burst energy for the transients typically found in music program.

Class H simply uses multiple supply rails, one for low V situations and one for high demand situations. It switches between the rails accordingly, saving energy lost to heat = increase in efficiency vs a standard linear amp.

The LG (and Carver and BASH, etc.) amps arrange for the single rail power supply to 'track' either the input (Carver) or the output (LG) signal and deliver only the voltage needed. Therefore, less energy lost to heat = higher efficiency than class H. The 6V headroom is cited because the supply is never at zero V. When in input is zero, the supply is 6V and it stays at +6V to max voltage.

Both of them received patents which have run out and, to my knowledge, have not been renewed. Neither of them is class H.

LG's test results show efficiency equal to class D, especially at 1/10 to 1/3 power. Competing amp companies have show their own test results that supposedly prove that LG's efficiency claim is not true. The difference in results relates directly to the type of test being used. LG uses something like 33ms on/66ms off input cycle (in an attempt to approximate music program, and other amp companies have other varying ratios of on/off times for the same reasons) and the 'dis-provers' use steady state sine wave input into a resistive load.

It should be a common sense deduction that the savings against energy loss comes from the tracking supply offering only the voltage required for music program, as a %, will change when the input is steady state at maximum output.

Class TD does not equal some class D amps in efficiency but for the purpose of HT amplification it definitely does and with higher SQ in my opinion and by my own test results.
MemX likes this.
bossobass is offline  
post #104 of 148 Old 01-30-2014, 07:25 PM
AVS Addicted Member
 
LTD02's Avatar
 
Join Date: Oct 2003
Posts: 16,744
Mentioned: 19 Post(s)
Tagged: 0 Thread(s)
Quoted: 502 Post(s)
Liked: 1125
"No sh!t. People have casually said on many occasions that the clones are class H and no one calls BS."

maybe part of the confusion is that there isn't a real clear definition at work here. the amp designer dd linked up indicated that a class h is distinct from a class g in that the second rail is not switched, but modulated, i.e. it follows the signal. that would make the clone's behavior quite similar in performance and efficiency to class h in the region where the second rail kicks in. one of lab's documents that has the various efficiencies shows that once the second rail kicks in, the efficiency of td and h are pretty much the same.

personally, I don't really see why it matters if the second rail is switched (g) partially modulated (h) or fully modulated (td) . seems like there are much bigger fish to fry than the last few percentage points of efficiency.

edit for examples:

class G:


class H:

Listen. It's All Good.
LTD02 is offline  
post #105 of 148 Old 01-31-2014, 03:55 AM
AVS Addicted Member
 
LTD02's Avatar
 
Join Date: Oct 2003
Posts: 16,744
Mentioned: 19 Post(s)
Tagged: 0 Thread(s)
Quoted: 502 Post(s)
Liked: 1125
LTD02 is offline  
post #106 of 148 Old 01-31-2014, 04:35 AM
AVS Special Member
 
Trepidati0n's Avatar
 
Join Date: Nov 2003
Posts: 1,113
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 16 Post(s)
Liked: 92
Quote:
Originally Posted by LTD02 View Post

personally, I don't really see why it matters if the second rail is switched (g) partially modulated (h) or fully modulated (td) . seems like there are much bigger fish to fry than the last few percentage points of efficiency.

Silicon up until about 15 years ago sort of sucked in the power electronics industry. I can remember for quite a while going into work and thinking "hrm, wonder what time today I'll hear a bang" and now it is "hrm, its been a few months since the last bang". And the word "bang" to me is when a 1 MW converter with a few mJ of energy decides that it wants to let its smoke out. But more importantly, it wasn't until the past 10 years when silicon sort of became "cheap". Before then my managers would beat on me relentlessly to "use less devices" since that was the very much the dominant cost of the product. Today it is still a lot, but it isn't so much to where I can't margin a bit more. So yeah, going from G to H was a big freaking deal for a while when either managing cost or power density of an amplifier.

As for efficiency. It matters. Efficiency for amplifiers is THEE # cost sliding variable. There is a measurable and verifiable financial benefit for every watt you don't burn up as heat in the box. In a competitive market..it matters.

I must be guilty because people say I am guilty because they chose to call me guilty because they refuse to see the truth. Much easier to be part of the mob..
Trepidati0n is offline  
post #107 of 148 Old 01-31-2014, 08:34 AM
Senior Member
 
WVSyd's Avatar
 
Join Date: Feb 2013
Posts: 333
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 42
Quote:
Nice cool.gif
Some important characteristics of amplifiers can be observed:
Maximum efficiency is NOT at low power.
The difference in efficiency narrows after 100 watts ( approximately ) for these 1000 watt amplifiers ( a 10db power difference )
Consider the crest factor of music - the most efficiency is obtained on peaks.

"Beware of Salesmen: They are the modern Svengali, immune to Science and Reality"
WVSyd is offline  
post #108 of 148 Old 01-31-2014, 09:24 AM
Advanced Member
 
MBentz's Avatar
 
Join Date: Jun 2006
Location: Chicago
Posts: 946
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 20 Post(s)
Liked: 41
Quote:
Originally Posted by splotten View Post

Thank for commenting.

It is true that 200w is not that much louder than 50w, but then you might as well have said that a 200w class A/B is not that much louder than a 50w class A/B, or that a 200w class A/B is not much louder than 50w class D... tongue.gif

Lets see what MBentz has to say about a watt not being a watt.

Quote:
Originally Posted by splotten View Post


This does not make any sense. What does it matter if its class D or A/B if you have equal voltages?

I've been writing on my phone lately so I haven't been going into all the details...

The short story is that the clipping behavior is very different. Class D will hard clip right up against the rail voltage (and then whatever AC signals on the rail come right through the output). The A/B amplifier will have more of a soft compression behavior before hard clipping.

The Class D amplifier can also pass a sine wave with P-P voltages nearly hitting the rail all day long. Your A/B amplifier will probably burn up doing that (depending on how the A/B amplifier's specs are created). The A/B amplifiers I measured several years ago could momentarily deliver output voltages much higher than what you'd calculate from the long term steady state output. CEA standard 490-A requires output power to measured after 5 minutes driving a load with less than 1% THD.

I am very much in the camp that a "watt is a watt", but the reality is that a lot of people don't realize how much they clip their amplifiers all the time and to put it simply: the clipping characteristic of your A/B is more pleasing than Class D. And even if you're not clipping, the amplifier linearity starts to approximate the clipping sound long before you're hitting up against the rails - basically it's part of the sonic signature of the amplifier at lower output levels. Btw, I'm not talking about sonic behavior at 1W for a 1000W amplifier here.


Btw, I'm also in the camp that class D should be the superior amplifier topology. The problem I see is that most class D designs get carried way with how simple/small/cheap they can make the amplifier and be comparable to other much more expensive amplifiers. They also typically employ a switch-mode power supply that don't got the oomph of some of your other better power supply designs....and it's not that a switching supply can't be good - it's "too expensive" to roll one that would actually play nice with the dynamic demands of audio.


I also wanted to mention that the mixing products that I'm referring to typically won't show up in a THD plot - but that doesn't mean they aren't very audible. Anyone that thinks a "power supply is a power supply" is just flat out ignorant of the complexities involved with power supply design - especially as it relates to audio circuits. The idea of a DC voltage is just a pleasant simplification to make other calculations much simpler. The reality is that power supplies are complicated control systems using high powered RF signals that attempt to emulate a DC voltage. The reality is that the power supply is moving all over the place all of the time and that interaction with audio circuits is a very complicated signal processing problem. I think people would be surprised how much of a difference in sound can be achieved by simply changing the "AC behavior" of a power supply. And it's not just an audible thing either - you'll see it in the measurements when you know what to look for.

Who here has actually put an oscilloscope on the rail of their amplifier while playing music? It's anything but DC and all that crazy crap is getting dumped into the audio - relying on the amplifier's feedback to servo it out. Everyone is using different kinds of feedback and biasing techniques so the net result is always different - and I would argue that difference is why you need a more powerful Class D amplifier to have less artifacts at lower output levels.


Then once you take crest factor of the audio signal into account, then it gets all sorts of more crazy. That feedback loop of your SMPS can only track so fast and will be lagging the output demand - so then your output stage is doing extra work to try to keep the output voltage where it needs to be. So ya, in the steady state long term output demand your SMPS will eventually get back up to its desired regulation voltage so that the output stage can keep the sine wave where it needs to be to measure the really good long term output power. But try to instantaneously change the output power by 100x (20dB crest factor) and that SMPS's output is gonna fall flat on its face - relying only on the output capacitance, which ironically has to be lower than other power supply topologies to maintain stability and be "responsive enough".



All that said, "50W class D should be the same as 50W class A/B" - but the reality is that nobody is building a class D or class A/B that actually matches exactly the ideal performance expected of those topologies (or any other topology on the planet). There is always some level of non-ideal about the system: often related to the power supply architecture and that is why I say you'll want a little more power from the switching designs so you can avoid their nasty transient behavior.

-Mike Bentz
~It's all about compromise~
MBentz is offline  
post #109 of 148 Old 01-31-2014, 09:38 AM
Advanced Member
 
bwaslo's Avatar
 
Join Date: Jul 2006
Posts: 825
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 28 Post(s)
Liked: 78
Quote:

Am I missing something? Doesn't that photo show the class TD yellow curve at 30% efficiency for 0 watts out? How does that work?

DIY Synergy horn spreadsheet http://libinst.com/SynergyCalc/
XSim -free crossover designer and simulator http://libinst.com/Xsim/XSimSetup.exe
bwaslo is offline  
post #110 of 148 Old 01-31-2014, 09:42 AM
Senior Member
 
WVSyd's Avatar
 
Join Date: Feb 2013
Posts: 333
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 42
All amps consume power even at "rest" ( ie no signal is being applied to be amplified )
Often quoted as an idle value of consumption of power .

Google "Quiescent current"

"Beware of Salesmen: They are the modern Svengali, immune to Science and Reality"
WVSyd is offline  
post #111 of 148 Old 01-31-2014, 09:54 AM
AVS Special Member
 
notnyt's Avatar
 
Join Date: Dec 2008
Location: Long Island, NY
Posts: 4,495
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 285 Post(s)
Liked: 461
notnyt is offline  
post #112 of 148 Old 01-31-2014, 09:54 AM
AVS Special Member
 
notnyt's Avatar
 
Join Date: Dec 2008
Location: Long Island, NY
Posts: 4,495
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 285 Post(s)
Liked: 461
Quote:
Originally Posted by bwaslo View Post

Am I missing something? Doesn't that photo show the class TD yellow curve at 30% efficiency for 0 watts out? How does that work?

Technically, it's approaching 0, but not at 0.
notnyt is offline  
post #113 of 148 Old 01-31-2014, 10:12 AM
Senior Member
 
WVSyd's Avatar
 
Join Date: Feb 2013
Posts: 333
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 42
What is the significance of the guy eating popcorn?
Some sort of pratfall.

"Beware of Salesmen: They are the modern Svengali, immune to Science and Reality"
WVSyd is offline  
post #114 of 148 Old 01-31-2014, 11:43 AM - Thread Starter
Advanced Member
 
Swolephile's Avatar
 
Join Date: Sep 2013
Posts: 520
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 12 Post(s)
Liked: 53
Quote:
Originally Posted by notnyt View Post


biggrin.gif
notnyt likes this.
Swolephile is offline  
post #115 of 148 Old 01-31-2014, 11:54 AM
Senior Member
 
WVSyd's Avatar
 
Join Date: Feb 2013
Posts: 333
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 42
OK I think I got it.
A pop culture reference to an ongoing comic foil used by JS.
Considering his viewership is 2.5m in a country of 350M - that's only about 0.7% significance biggrin.gif.
eXa likes this.

"Beware of Salesmen: They are the modern Svengali, immune to Science and Reality"
WVSyd is offline  
post #116 of 148 Old 01-31-2014, 11:59 AM
AVS Special Member
 
notnyt's Avatar
 
Join Date: Dec 2008
Location: Long Island, NY
Posts: 4,495
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 285 Post(s)
Liked: 461
Quote:
Originally Posted by WVSyd View Post

OK I think I got it.
A pop culture reference to an ongoing comic foil used by JS.
Considering his viewership is 2.5m in a country of 350M - that's only about 0.7% significance biggrin.gif.

John Stewart was not the significant part, just the popcorn eating. Are you a robot? =[
notnyt is offline  
post #117 of 148 Old 01-31-2014, 11:59 AM
AVS Special Member
 
Trepidati0n's Avatar
 
Join Date: Nov 2003
Posts: 1,113
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 16 Post(s)
Liked: 92
Quote:
Originally Posted by WVSyd View Post

All amps consume power even at "rest" ( ie no signal is being applied to be amplified )
Often quoted as an idle value of consumption of power .

Google "Quiescent current"

Quote:
Originally Posted by notnyt View Post

Technically, it's approaching 0, but not at 0.

Both of you missed the question. At zero output power efficiency can either be infinite or zero. there is no other answer.

Definition: efficiency = power out / power in

The graph is slightly bonked. My guess is they were trying to approximate the shape of the yellow curve a bit better. But if they put in 0% efficiency @ 0W it wouldn't look like they were "better" than the others. What they are trying to show is even at low output power they are "more efficient" than other types. So they put in a non zero value at 0 power to "approximate" it.

I must be guilty because people say I am guilty because they chose to call me guilty because they refuse to see the truth. Much easier to be part of the mob..
Trepidati0n is offline  
post #118 of 148 Old 01-31-2014, 12:03 PM
AVS Special Member
 
notnyt's Avatar
 
Join Date: Dec 2008
Location: Long Island, NY
Posts: 4,495
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 285 Post(s)
Liked: 461
Quote:
Originally Posted by Trepidati0n View Post



Both of you missed the question. At zero output power efficiency can either be infinite or zero. there is no other answer.

Definition: efficiency = power out / power in

The graph is slightly bonked. My guess is they were trying to approximate the shape of the yellow curve a bit better. But if they put in 0% efficiency @ 0W it wouldn't look like they were "better" than the others. What they are trying to show is even at low output power they are "more efficient" than other types. So they put in a non zero value at 0 power to "approximate" it.
Quote:
Originally Posted by notnyt View Post

Technically, it's approaching 0, but not at 0.

How is that missing the point? That's the point exactly. It can't have a value at 0, they can only graph as it approaches the limit at 0 where it becomes undefined.

They're probably graphing efficiency at like 0.1 watts
notnyt is offline  
post #119 of 148 Old 01-31-2014, 12:06 PM
Senior Member
 
WVSyd's Avatar
 
Join Date: Feb 2013
Posts: 333
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 42
Notnyt:
So because your opaque pop reference falls flat - then your direct insults starts.
How many times does one have to post to be granted such Special Member insult privileges...

"Beware of Salesmen: They are the modern Svengali, immune to Science and Reality"
WVSyd is offline  
post #120 of 148 Old 01-31-2014, 12:24 PM
Senior Member
 
WVSyd's Avatar
 
Join Date: Feb 2013
Posts: 333
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 42
Quote:
Originally Posted by Trepidati0n View Post

,,, At zero output power efficiency can either be infinite or zero. .
I though that was presumed - an initial measurement at the lower mw signal possible.

"Beware of Salesmen: They are the modern Svengali, immune to Science and Reality"
WVSyd is offline  
Reply DIY Speakers and Subs

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