LM3886 chip amp vs Purifi/Niali class D

Jump to Latest
#42
FnpzeuJXkAEVcMt.jpg
 
#46
maty tinman said:
In class D amplification there is the DIY Hypex Nilai500 and the others. It is the only class D I would use to power the tweeter, for quality recordings with physical instruments and vocals without autotune.

Electronic music and others, Purifi is OK or maybe better.
Click to expand...
Based on test you did at home in controlled manner or something else - like youtube etc?

//
 
#48
Those curves seems to originate from a thorough test that Amir from AudioScience made around Purify 1ET400. And they clearly shows residue
of the switching frequency around 300 KHz. But they don´t show any problems with that it modulates down in the audio range.

I don´t remember but I don´t think he measured the FR above 20 KHz. If you do so it will show ringing in the area around 50KHz, and personally I
am sure that this modulates down in the high mid band, and causes the ´ziing´ that many people don´t like in Class D amps.

The problem originate from the 12 dB output filter which isn´t steep enough to filter the switching frequency. I am not sure why it is only 12 db
filter, but in crossovers it is possible to cascade filters to get a higher slope further up, or just go with a higher order filter in the first place.
 
#49
drjlo said:
"I find Class D - even the ones by Purifi/Hypex - to have a bit of harshness"

I think this is big part of the reason GaN amps have been popular more recently, sounding a bit softer. Tons of people have tried traditional class D by now, but many are thinking they sound great except for almost subliminal feeling of coolness/harshness. I have Hypex, GaN, Modulus, as well as AB (and tubes) sitting in my room, and I think the better of them all sound "good enough," with all having far better cost/performance ratio than say 10 years ago, a boon for audiophlles for sure..
Click to expand...
TI TPA3255 based amps are also much softer. I used to run some small TPA 3116 based systems, that sounded very good, but the bigger brother
comes in with no aggressiveness at all.
 
#51
The Chinese are making some interesting LM3886 based amplifiers including what they call scaffolding designs-which means hardwired rather than circuit board.Also some doubled up ones for more power.For sale on Ali Express.
I have owned a 3876T chip amp for years which sounds very good.They are best with 8 ohm speakers rather than 4.
 
#52
maty tinman said:
Youtube and 24/48 FLAC files

View attachment 1192010
Click to expand...
Is this what you played on your rig while doing the evaluation i.e. changing between the different physical amps or did you use these via e.g. headphones and made your conclusion?

Please try to be precise in your communication...

It looks like you used different recordings where the different amps was a part of the system recorded - is this correct?

//
 
#53
I did it on my second audio system, the cheap one, with my modified coaxial KEF Q100, nearfield, (they have sounded even better since two weeks ago after flattening the acoustic phase with rePhase but that is another matter). Sound via YouTube is played by JRiver MC thanks to WDM Driver.
https://wiki.jriver.com/index.php/WDM_Driver

Then I downloaded the FLAC files and same result: Pass Labs XP12 preamp + Hypex Nilai500 was almost always the best combination (at least for the type of music I listen to, for electronics Purifi was a bit better, maybe because of more forceful bass, I do not remember, I was commenting on it in real time, Live chat and in my Twitter account).
 
#54
OK. Judging individual component by replaying UT videos over speakers is as serious as choosing your next car by driving the model in a video game. Sorry, its just not possible and I think you should refrain for making such certain statement about power amps like you just did in the previous post.

Why do you keep repeating this kind of "consumer guidance"?.... https://www.audiosciencereview.com/...ilai500diy-modules.34024/page-19#post-1467312

At least, when you do it next time, please state the circumstance around your recommendations - so that the reder can judge how to treat them.

//
 
  • Like
Reactions: waxx
#57
ArthurDK said:
Some of them have, many does not.
Click to expand...
Actually, all of them do. Maybe not in the form of a global negative feedback loop, but they do employ local negative feedback. Even a common emitter gain stage has negative feedback from the internal emitter resistance (re) of the transistor.

Whether an amp employs a global negative feedback loop is independent of the bias of the output stage. You can have a Class A, B, or AB output with/without global negative feedback.

Tom
 
#58
I read on some other forums, one reason against Class D is the ability to repair them and that you can still find old Class AB amps from years ago that work and can still be repaired.

Not sure what on Class D isn't repairable but a vote towards Class AB maybe.
 
#59
I think some of the ability to repair comes down to the skills of the person performing the repair. If you have good hot air tools it's not that big of a deal to work on SMD circuits, but it's certainly trickier than working on older PTH technology. I would think that most of the challenges around repairing Class D amps come down to parts availability and access to schematics/BOMs. In the older technology you can read the resistor value from the coloured rings and the part numbers of devices are often stamped on the part itself. That's not so easy with SMD tech.

Also, I seem to recall that the OEM pricing on, say, a Hypex NC500 is around $100, so paying someone to repair one doesn't really make sense. Just swap out the module and move on. I don't like generating waste, but I also don't like wasting time. Tradeoffs, tradeoffs...

Tom
 
You must log in or register to reply here.
Top Bottom