I answered to your pm some minutes ago. Judgement based on these cheap class-d boards isn't that fair i think as you then have to compare them against some "low grade" class-a boards as well. (like TDA2003 or so)
Best option is to get yourself a 3251 or 3255 EVM from TI directly (with 50% discount by using their coupon) via Samples-Store.
Best option is to get yourself a 3251 or 3255 EVM from TI directly (with 50% discount by using their coupon) via Samples-Store.
I read it and responded 
Speaking of the mini board linked above, do you think the miniscule capacitance on that board will be a problem for bass response? I think this might be addressed in the main supply but don't know enough about how these operate to be sure. What are the strategies you use to keep the noise floor on these D amps down?
Speaking of the mini board linked above, do you think the miniscule capacitance on that board will be a problem for bass response? I think this might be addressed in the main supply but don't know enough about how these operate to be sure. What are the strategies you use to keep the noise floor on these D amps down?
I was somewhat skeptical about cheap class D amplifiers as well (I had heard about ICEPower). I was raised with class AB amplifiers and even a class A amplifier (Linsley Hood). The idea of cutting the nice sound into square-waves did not sound appealing.
Recently I noticed a simple class D amplifier board sold for very little. I bought it just to confirm my prejudice that it was not as good as the "real" class A/AB stuff.
Just trying out this simple class D amplifier made me doubt - it actually sounded surprisingly well. I then bough some ten (rather cheap) class D amplifier boards of different make and tried them out. I have never been disappointed with any of them when I consider the price. It does not mean that in blind-test the class AB/A may not be a bit better for audiophile use. But, a priori the class D amplifiers cannot be discarded even for audiophile use.
Class D amplifiers primarily serve commercial production in the sense that they are attractive for portable use (due to better efficiency) and easier to enclose in a structure (like a sound-bar) with only a small heatsink because of reduced heating.
Tripath took class D a giant step ahead and the improvement has not stop.
Another issue is the ability of the ordinary user to notice the difference and his/her willingness to pay considerably more for a marginally better performance. The human physiology sets important limits (also for the output power you "need").
Recently I noticed a simple class D amplifier board sold for very little. I bought it just to confirm my prejudice that it was not as good as the "real" class A/AB stuff.
Just trying out this simple class D amplifier made me doubt - it actually sounded surprisingly well. I then bough some ten (rather cheap) class D amplifier boards of different make and tried them out. I have never been disappointed with any of them when I consider the price. It does not mean that in blind-test the class AB/A may not be a bit better for audiophile use. But, a priori the class D amplifiers cannot be discarded even for audiophile use.
Class D amplifiers primarily serve commercial production in the sense that they are attractive for portable use (due to better efficiency) and easier to enclose in a structure (like a sound-bar) with only a small heatsink because of reduced heating.
Tripath took class D a giant step ahead and the improvement has not stop.
Another issue is the ability of the ordinary user to notice the difference and his/her willingness to pay considerably more for a marginally better performance. The human physiology sets important limits (also for the output power you "need").
DPH
I linked the mini board with the low capacitance I'm talking about several posts up. It looks like less than 1mf of supply of very fast capacitance to me but others with better eyes (or a better photo) can correct me.
FauxFrench
Yes, I feel as you did. Perhaps my mind will change when i receive the mini boards described above that I just ordered.
I linked the mini board with the low capacitance I'm talking about several posts up. It looks like less than 1mf of supply of very fast capacitance to me but others with better eyes (or a better photo) can correct me.
FauxFrench
Yes, I feel as you did. Perhaps my mind will change when i receive the mini boards described above that I just ordered.
Ah, I missed it and thought you were talking about the TI EVM boards. Yeah, local bypass in the ~1000 uF range would be good to have.
Kjeldsen
I've noticed the great effort to make these class D amps hifi
wushuliu
That's a bit harsh isn't it? I'm sure there are many like me who are uncomfortable with the inherent characteristics/limitations of class D amps. I never said they are insurmountable but they are there. The two that come immediately to mind are their propensity for poor linearity above 6khz and their poor noise floor but there are others. People who live in cities have to deal with the usually poor RF filtering and using tried and true tricks like massive feedback is much more complicated with class D. There are many reasons to suspect this newer technology but the same was once said of crossover distortion with class B. If I hated it that much I wouldn't have ordered those totally SMD boards that I linked above. I share interest in that board with both you and doctormord.
I've noticed the great effort to make these class D amps hifi
wushuliu
That's a bit harsh isn't it? I'm sure there are many like me who are uncomfortable with the inherent characteristics/limitations of class D amps. I never said they are insurmountable but they are there. The two that come immediately to mind are their propensity for poor linearity above 6khz and their poor noise floor but there are others. People who live in cities have to deal with the usually poor RF filtering and using tried and true tricks like massive feedback is much more complicated with class D. There are many reasons to suspect this newer technology but the same was once said of crossover distortion with class B. If I hated it that much I wouldn't have ordered those totally SMD boards that I linked above. I share interest in that board with both you and doctormord.
Last edited:
Some of you guys/girls own amplifiers so expensive that I have not even heard about them. No doubt also what you call "well constructed class D amps".
But, for each "well constructed class D amp" sold the toy "cheap class D boards" sell by ten thousands.
The "cheap class D boards" will steadily approach the "well constructed class D amps" in performance, but the "well constructed class D amps" will not increase their sales towards the numbers of the "cheap class D boards".
In conclusion, the monolithic (cheap) class D amplifiers are interesting because they satisfy the needs of the absolute majority of users (ordinary people) and indicate the developments in the market.
The very high performance class D amplifiers are important because they demonstrate what is possible with this technology, but they will never be mainstream in any way.
It seems that ICEPower was (is?) an important source of that sophisticated technology.
So, for us ordinary people the toy amplifiers are interesting because they reflect the trend and give us pleasure from convincing performance though not the ultimate.
Said in other words: it is not because there is a Bentley on the market that there is no need to consider the new Opel Insignia.
Yep, (some) class D amplifiers absolutely qualify to audiophile standards.
But, for each "well constructed class D amp" sold the toy "cheap class D boards" sell by ten thousands.
The "cheap class D boards" will steadily approach the "well constructed class D amps" in performance, but the "well constructed class D amps" will not increase their sales towards the numbers of the "cheap class D boards".
In conclusion, the monolithic (cheap) class D amplifiers are interesting because they satisfy the needs of the absolute majority of users (ordinary people) and indicate the developments in the market.
The very high performance class D amplifiers are important because they demonstrate what is possible with this technology, but they will never be mainstream in any way.
It seems that ICEPower was (is?) an important source of that sophisticated technology.
So, for us ordinary people the toy amplifiers are interesting because they reflect the trend and give us pleasure from convincing performance though not the ultimate.
Said in other words: it is not because there is a Bentley on the market that there is no need to consider the new Opel Insignia.
Yep, (some) class D amplifiers absolutely qualify to audiophile standards.
Last edited:
Ah, that's apples and pears. The TacT Millennium is for a very good part that good because it does not amplify analogue signals, it's directly driven by the internal DAC, and it does not re-sample the signal for different volumes. It cannot reproduce any analogue signals and not even all digital sources. If you'd add the analogue part (or get rid of it on other class D amps), the distance between them would melt like ice in the sun. Possibly still with the tact ahead but that's to be expected, considering the price tag of meager 12000€. No, that's not a zero too much.
Not true IMHO. The digital frontend of the Tact Millenium is not the strongest issue. Yes, you don't need an expensive DAC. But, a very good class D amp with a very good DAC is comparable to Tact Millenium.
Not all Tact and Lyngdorf amps are pure digital.
I have tried them all all types of class D in all price ranges (well not all - but more than most). Class A - cheap and expensive. Class AB cheap and expensive. Tubes only cheap and mid priced.
All topologies can make very very good reproduction. My point is - please do not judge class D based on cheap TI boards from China.
Not all Tact and Lyngdorf amps are pure digital.
I have tried them all
all types of class D in all price ranges (well not all - but more than most). Class A - cheap and expensive. Class AB cheap and expensive. Tubes only cheap and mid priced. All topologies can make very very good reproduction. My point is - please do not judge class D based on cheap TI boards from China.
This is not true. Where do you see that the class d topology is not linear above 6 khz (are you talking frequency or distortion ?), and where do you see the poor noise floor (not specific amplifiers but the topology)
Well, that's exactly what it is at the Tact Millenium, the eliminatin of the analogue source, all it's distortions, influence and tolerances and unlinearities of a lot of components in the input. Add the analogue part and the huge difference shrinks together like nothing. Ofcourse it's still much better but not by worlds anymore. That's not just my opinion, a lot of tests say exactly the same.
That I can fully sign.
I don't see how this article contributes. I can show you Class D measurements that are very very low distortion at all frequencies and very linear up to +100 khz.
I'm not a class d fanatic - actually I'm planning to have all types of amplifiers in my system, to be able to change technology from time to time.
I'm not a class d fanatic - actually I'm planning to have all types of amplifiers in my system, to be able to change technology from time to time.
Any suggestions what would be a good OP amp to replace NE 5532 for this TPA3250？
2 choices：
1# Just for great audio quality
2# If cost is not an issue.
That article is a prototype on how not to write objective articles.
Firstly, they measure some frequency response but don't tell you what it is except the brand of the amplifier (just Axiom Audio Class D). And they don't even tell this in the article itself either, you have to grab that information out of the measurement diagrams.
Secondly, the output filter is clearly heavily mismatched with the load impedance. That is, if they connected one at all, but they don't write anything about that either. Ofcourse that displays a problem of class D amplifiers, but it's that bad, it's pretty obvious the load impedance is much too high for the filter and it pretty much looks like it was intentionally done that way to make class D look bad.
Thirdly, the 2nd measurement is a different amplifier. To make the impression how inferior the amplifier impedance/damping factor is, they use sentences like 'We can hope (without success)..' to describe non-resistive loads - instead of just measuring it themselves. They did not do any distortion or noise measurements.
In any way, these measurements are crude, they didn't tell about any measurement circumstances and -equipment, which is standard. But that got an obvious reason, they seem to have inferior measurement tools, at least the very low sample resolution and frequency steps of the impedance measurement suggest that - there are just 4 different, always exact same values from 10Hz up to 5kHz.
To put the resulting damping factor into perspective, the measurements of the amplifier impedance at 8 Ohm load translate to a DF of 53 at 20kHz, 66 at 15kHz and 100-114 for everything below 9kHz. That's pretty much standard specs also for average class A/B amplifiers, even good for multi-channel amps. Sure, that's not impressive for a supposedly high end amplifier (if there's such a thing in home cinema) But again, like someone already said, it's not 2007 anymore and there were a lot of developments on the class D amplifiers in the last 10 years. And there are also a lot of other very expensive amplifiers which don't have any better specs there either.
To dig up such old and very obviously 'tinted' articles as unalterable facts to prove 'class D is baaad' suggests it might either be ignorance, willful mischief or trolling but I honestly believe it's neither of these and there might be hope for an open ear for some arguments, measurements and facts.
Things change. A previously widely accepted assumption/opinion can lack it's very foundations because of technological advancements over time. And 10 years is a very long time in electronics.