1
\$\begingroup\$

I am planing to assemble a 5.1 channel amplifier. I selected the famous TDA2030 as the basic element and is simple to proceed. But I have a confusion, whether to use some other Class D amplifier for better quality.

Please share your experience about these classes of amplifiers in the term of their output quality. I really don't care about the efficiency or the size.

\$\endgroup\$
  • 2
    \$\begingroup\$ Why is the TDA2030 famous? :-) \$\endgroup\$ – Anindo Ghosh Nov 19 '12 at 12:15
  • 1
    \$\begingroup\$ !!!!! sudden replay, that's why StackExchange... Yes definitely it is good in quality, external part count and its cost. But, what about the class-D vs TDA2030..:-) \$\endgroup\$ – Saneesh A T Nov 19 '12 at 12:21
  • 3
    \$\begingroup\$ This depends a lot whether you are catering to audiophools and will tell them what's inside your amp, or just want something that sounds good. \$\endgroup\$ – Olin Lathrop Nov 19 '12 at 14:11
9
\$\begingroup\$

Conceptually, a Class D amplifier is considered by some music purists to be less "perfect" in music rendition than a Class AB. This is because at the end of the day, a Class D amplifier generates digital pulses, which are then low-pass filtered to reconstruct the original sound.

However, this distinction is perhaps more a preconception than human-discernible reality. Modern Class D amplifiers like the TPA3125D2 work at high enough oscillator frequencies (300 KHz) that the low-pass filter node frequency can be well above the ability of the human ear to perceive - 60 KHz or higher frequency low-pass filters have been used in DIY audio amplifier designs.

Class AB too can suffer from distortion, especially if the load impedance is not well matched to the design, or the AB amplifier inherently has poor crossover distortion profiles. This is not so evident in "famous" Class AB chip amps such as the TDA2030 or the LM3886, both of which boast excellent Total Harmonic Distortion figures.

Also, Class AB amplifiers can conceivably exhibit distortion at lower frequencies as well, where the human ear is more sensitive to such distortion. Class D distortion is significantly at the top end of the audio band, and beyond.

A bigger cause for concern around audible distortion in DIY audio amplifiers is power-rail voltage droop under load. While the power supply could be designed to comfortably provide sufficient current to exceed the RMS power rating of the amplifier design, the actual power requirement rises to much more than a hypothetical Sqrt(2) x RMS wattage + overhead. Even 4 x rated RMS wattage might be momentarily breached in practice, such as during big drums and cymbals clashing at once.

If the power supply tops out in those moments, and especially if the voltage regulator's protection circuitry takes time to recover, you have harsh distortion, more discernible than any amplifier chip choice would expose you to.


There is no simple answer AFAIK, but if you are interested in personal recommendations, going with multiple Class D amplifiers has several advantages - separate amplification, pass-band and load tuning for each channel and each band, designed to align with the specific characteristics of the speakers you plan to use for each channel and band. The x.1 subwoofer / woofer amplifier can be given special attention, since it tends to be a current hungry part of the design.

The low thermal wastage i.e. lower non-utilized power overhead allows for simpler or even no heat sinking, and of course there is less load on the power supply evidently.

Knowing what actual power the 5.1 amplifier is being designed to pump out may provide better insights.

\$\endgroup\$
  • \$\begingroup\$ Actually some people still say that real quality audio can only be heard from vacuum tube amplifiers. I am not supporting it but only commenting on preconceptions. \$\endgroup\$ – ExcitingProjects Nov 19 '12 at 12:59
3
\$\begingroup\$

If a competent EE is designing the amplifiers, there is no noticeable difference in audio quality between Class-AB and Class-D.

This was not always the case, however. Class-D is hard to design correctly and even just 5 years ago most people did not do it well. There are still people who claim that Class-D is bad, but my experience is that these people are not well informed (technology has improved) or they have personal biases that are preventing them from being objective (they also need to spend a stupid amount of money on speaker cables).

This does not mean that YOU should do Class-D. Class-D amps are still more difficult than Class-AB amps. This is due to the high current and high speed switching that is required-- which places demands on your power supply, PCB layout, and filtering. "Amateur" type EE's tend to get all three of those things at least slightly wrong. But I am also not saying that you should do Class-AB either, since I do not know what your engineering strengths are. Only you can decide what is best for your situation.

But I will give you this warning: When looking at amp chips, pay very close attention to the details in the datasheets. Most chips are targeted at the low-wattage, low-quality market (like cell phones). And many of the higher-wattage parts are targeted at the car-audio market where THD+N is not their top priority (or even third priority). This is true for both Class-AB and Class-D chips!

I can also say that the best sounding Class-D chips that I have experienced are the ones from Cirrus Logic. For business reasons I would never use those parts, but for a hobbiest they are excellent products. I know the guy who designed them, and he is smart beyond belief. We've talked at length about the design of those chips and the amount of engineering that went into them is staggering.

If I were designing a home theater amp I would go with Class-D-- but I'm weird that way.

\$\endgroup\$
-1
\$\begingroup\$

If you just want to be surrounded by sound with > 0.5% distortion and lots of phase shift in the speakers , it won't matter. If want to experience 3D sound and loose perception that the sound is coming from speakers but rather coming the direction of source as the recording engineer intended, then you want linear phase speakers and < 0.1 % THD at max power.

Class D is an order of magnitude more difficult than Class A-B but cheaper in the end for materials and power supply.

For me, very high quality stereo is far more pleasing than poor quality 5.1 and higher quality mono is even better than stereo. 5.1 is an illusion of good sound but sacrifices so much quality in speakers and amp., you must know the difference to appreciate it.

Analog controlled class D is the way to go in the future.

Your choice depends on cost budget for make or buy time vs $. Experience vs appreciation, short term vs long term needs.

The quality factors not only include power and THD but neutrality and transparency of sound.

\$\endgroup\$

Not the answer you're looking for? Browse other questions tagged or ask your own question.