Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Frank Koenig]

I don't have the time or energy to explain where their flaws may be hiding, but it is not crossover distortion.

I know the feeling. But if you're ever in the mood I'm interested.

Just for grins, I took a look at the data sheet for a small, integrated class-D amp:

http://www.ti.com/lit/ds/symlink/tpa3116d2.pdf?ts=1588647708626

They give extensive distortion curves as a function of frequency and power. There is an uptick in distortion + noise at mid to low frequencies at power levels < ~1W. This could be the noise dominating the measurement. I believe distortion measurements, as made by Audio Precision, differentiate the harmonics from the noise by signal averaging but, now that I think about it, signal averaging may not work so well on the switching hash as it may be correlated with the test signal. It may be just damn hard to measure.

So I don't know, but this feature of the distortion curves may have something to do with amp manufacturers using a lower power limit when speccing distortion. Commercial amp users don't give a hoot about low power anyway since low power means they specced an amp that's bigger than it needs to be.

Enough for now from me. --Frank

[John Roberts {JR}]

I know the feeling. But if you're ever in the mood I'm interested.

Just for grins, I took a look at the data sheet for a small, integrated class-D amp:

http://www.ti.com/lit/ds/symlink/tpa3116d2.pdf?ts=1588647708626

They give extensive distortion curves as a function of frequency and power. There is an uptick in distortion + noise at mid to low frequencies at power levels < ~1W. This could be the noise dominating the measurement. I believe distortion measurements, as made by Audio Precision, differentiate the harmonics from the noise by signal averaging but, now that I think about it, signal averaging may not work so well on the switching hash as it may be correlated with the test signal. It may be just damn hard to measure.

So I don't know, but this feature of the distortion curves may have something to do with amp manufacturers using a lower power limit when speccing distortion. Commercial amp users don't give a hoot about low power anyway since low power means they specced an amp that's bigger than it needs to be.

Enough for now from me. --Frank

I was involved with class D back in the 80s (at Peavey) when they were far from mature and the device technology was not up to the challenge like it is this century. There are some subtle low level distortions but these are pretty much managed in modern chip sets.

My point was that saturated switching amps do not exhibit the classic problems associated with linear audio amps. 

Back when I was product manager for all Peavey's power amps and I had a class D amp that was lucky to spec 1kW... The limitation back then was power switching devices that could handle the voltage/current. Class D amp output devices do not behave well in parallel like typical analog output stages can. So the switching amp industry was at the mercy of power device manufactures waiting for fast, high current, high voltage devices. Now decades later that wait was rewarded (thanks a lot... ).

The modern generation of cost effective switching amps were after my time in the trenches so I have no hands on experience or inside information, but I can read the tea leaves (customer complaints) and they seem pretty serviceable. I can speculate they use standard IC manufacturer chip sets with cookbook application note designs.

My favorite modern class D amp for modest power is a EU product (Hypex) designed by Bruno Putzey (real deal engineer).

For TMI, I use a cute couple watt class D amplifier chip in my drum tuner. This class D IC sips current from my batteries and is about the size of a wood tick who is still hungry. You have to love modern technology (I do). 

JR

[Frank Koenig]

Here's the follow-up. I bought the Powersoft Mezzo 604A, which has 4 channels, a 600W power supply, DSP without FIR and auto-standby. I'm very pleased. It's a nice compact package that I mounted on the wall behind the TV with the business end close to one vertical edge of the TV so that it's accessible. It's super quiet both electrically and acoustically and, of course, sounds just fine. The auto-standby, which has a fixed -60 dB threshold, more-or-less instant-on and an ~30 min. hold, works perfectly. I'm using a miniDSP 2x4hd upstream for FIR filtering and remote-controlled preset selection (bass-boost on/off depending on program).

Thanks for the help, everyone.

--Frank

[ProSoundWeb Community]