You Can Have My Class AB When You Pry It From My Cold Dead Fingers

musichal

poet emeritus
Will the future answer, similar to the MIB roach, "Your terms are acceptable?"

Well, of course, literally (nobody lives forever)... you know what I mean. Is AB looking over its shoulder at D?
 
AFAIK...D is equivalent to AB in terms of sound.

I've only heard one hi-fi class D amp. It was massive.

It was massive because the builder knew it had to be to impress you. That's the only real reason.

All that needs to be big on D amps is the power supply. The primary chip needs a relatively small heat sink. Nothing much else.

I'll compare a class D amp to my (badass) AB amps when I find a company that builds something quality minus the bullshit. In other words, better than the lie-laden Chinese built trash, minus the 80 lbs of CNC aluminum the "high end" companies strap to them to justify all the extra zeros at their price point. Quality chip, clean power, well designed but 'frugally' built chassis.

If I don't see something in the next few months I may look to provide it to the market myself. This isn't hard... Or rather, it doesn't have to be.
 
Last edited:
It was massive because the builder knew it had to be to impress you. That's the only real reason.

All that needs to be big on D amps is the power supply. The primary chip needs a relatively small heat sink. Nothing much else.

I'll compare a class D amp to my (badass) AB amps when I find a company that builds something quality minus the bullshit. In other words, better than the lie-laden Chinese built trash, minus the 80 lbd of CNC aluminum the "high end" companies strap to them to justify all the extra zeros at their price point. Quality chip, clean power, well designed but 'frugally' built chassis.

If I don't see something in the next few months I may look up provide it to the market myself. This isn't hard... Or rather, it doesn't have to be.
Your 'badass' AB amps ?

What do you have besides the Nak PA-7AII ? The Nak is a badass amp for sure, but you have others ?

The class D amp I'm referring to held it's own against a Parasound HCA-3500.
 
Your 'badass' AB amps ?

What do you have besides the Nak PA-7AII ? The Nak is a badass amp for sure, but you have others ?

The class D amp I'm referring to held it's own against a Parasound HCA-3500.

The other two are for sale, so I took them out of signature.

I also have an Adcom GFA-555, a SUMO Andromeda (first gen), and a Pioneer SX-1250.
 
At least your AB amp is likely to have through-hole components and has already stood the test of time and can be repaired relatively easily. Try that with a Class D. They are disposable unless you have serious SMD reworking capability and bespoke parts.
 
At least your AB amp is likely to have through-hole components and has already stood the test of time and can be repaired relatively easily. Try that with a Class D. They are disposable unless you have serious SMD reworking capability and bespoke parts.
Looks like you could replace the caps...what else could possibly wear out ? :D

class d board.jpg
 
At least your AB amp is likely to have through-hole components and has already stood the test of time and can be repaired relatively easily. Try that with a Class D. They are disposable unless you have serious SMD reworking capability and bespoke parts.
You're really comparing vintage to modern construction techniques, not amplifier circuit topologies. A modern mass-produced class AB amp is just as likely to be SMD construction as a modern mass-produced class D amp, and both will be equally disposable and/or annoying to fix.
 
You're really comparing vintage to modern construction techniques, not amplifier circuit topologies.

Obviously.

'Stood the test of time' means they are already pre Pb-free and unlikely SMD construction.

I would also say that the majority of class AB amplifiers used by members of AK would be through-hole, vintage or semi-vintage designs, whereas the Class D designs used by AK members will be almost 100% SMD, lead free, and no fun to repair down the track. In short, they will become landfill when someone shorts a speaker wire at full power or the tin-whiskers start shorting out the SMD components in 10 years.

Mostly, they are cheap-ass Chinese, SMPS powered little bangers being extolled as 'giant killers' by people with not a speck of test equipment to verify their glowing subjective reviews.

The unit pictured above (BigElCat) is a decent looking unit, real transformer, a reasonable amount of PSU filtering and a nice block of aluminium to dissipate the heat.
 
I've toyed with several of the 3116s
Honestly for general listening they're the bargain of the century.
I have several tube amps and a couple of SS amps I prefer.
That was at a very low price point.
I can see where with some development, D might ably swim in the deep end.
It seems inevitable .
 
It was massive because the builder knew it had to be to impress you. That's the only real reason.

All that needs to be big on D amps is the power supply. The primary chip needs a relatively small heat sink. Nothing much else.

I'll compare a class D amp to my (badass) AB amps when I find a company that builds something quality minus the bullshit. In other words, better than the lie-laden Chinese built trash, minus the 80 lbd of CNC aluminum the "high end" companies strap to them to justify all the extra zeros at their price point. Quality chip, clean power, well designed but 'frugally' built chassis.

If I don't see something in the next few months I may look to provide it to the market myself. This isn't hard... Or rather, it doesn't have to be.
I would love to see another supplier out there that did what Emotiva did, but not change as Emo did. Emotiva, once they got up and running good changed their focus to AV stuff. If you have the ability to build and produce a quality Amp at a great price, you will make money. What I found was as Emotiva moved out of audio and into AV, a quality, well built, and most of all, long lasting amp became very expensive. Hence my signature of Pass Labs. I bought it used for 25% of new price, so I feel I did well. And yes, you can have my clas A/AB amp from my cold dead fingers.
 
All that needs to be big on D amps is the power supply.
Even the power supply doesn't need to be physically large. With proper shielding, switch-mode power supplies can be noise and interference free and supply kilowatts of energy in a lightweight package.
The primary chip needs a relatively small heat sink. Nothing much else.
True, because a class D topology is highly efficient. 90% of the energy that comes in through the power cord goes out through the speakers as sound, so only 10% of the input energy needs to be dissipated as heat. Compare that to AB designs, where at best approximately 25% of the input energy must be dissipated as heat.

Class A designs are, by necessity, even less efficient: over 70% of the input energy must be dissipated as heat via the heat sinks. That's why they tend to have massive heat sinks but relatively low output power.

Note that a class D amplifier doesn't have to use chips (integrated circuits), as in the classic "chip amp". Whilst the high efficiency makes it easy to construct class D topologies using integrated circuit technology because integrated circuits and heat don't mix well, class D amplifiers can and have been constructed using discrete components (http://www.diyaudio.com/forums/class-d/286263-pa-discrete-class-amplifier-schematic.html) including tubes (http://www.studio-rts-ing-rampin.it/ampdiva/ampdivapdfdepot/ampdivawhitepaper1en.pdf).
 
Last edited:
Perhaps in solid state gear. I never expect to come across a class D tube amp and I'm not likely to give up my class AB Oldchen KT88-K3 any time soon.

Not sure what my darTZeel NHB-108B copy is.... I suspect it's more class AB rather than pure class A as it really doesn't run terribly warm....

I expect class D may be capable of being every bit as good though I'm yet to hear a class D amp can compete with the better class A and class AB designs. That's not to say it's not possible. I'm pretty sure that this is down to implementation than topology. It's still early days for class D and the best is still ahead of it. I rather expect it will eventually surpass what is possible in class A or AB SS designs.

Does anyone know if class D is even possible for a tube amp design?
 
Still waiting for a switcher to match AB performance at the top. Midrange with my Ncore is great, but not in the same class as the VTLs.
 
Does anyone know if class D is even possible for a tube amp design?
Sure. See my post above yours. :) I linked to this: http://www.studio-rts-ing-rampin.it/ampdiva/ampdivapdfdepot/ampdivawhitepaper1en.pdf

However, there's no reason to use tubes for class D outputs.

Tubes are used in class A and AB output designs because when they clip, they tend to produce more even harmonics than equivalent clipping when using semiconductors. This may sound better than semiconductors under the same circumstances.

In a class D amplifier, the output devices are used as switches -- they're turned on and off rapidly, where the pulse width is proportional to and thus encodes the amplitude of the signal, which is decoded by a passive low-pass filter to drive the speakers -- and so clipping characteristics are not a function of the output device type (tube vs semiconductor) but the circuit topology and the input signal level.

In other words, there aren't any sonic qualities to gain by using tubes in a class D amplifier. Instead, you might lose some quality because tubes generally can't switch state as quickly as semiconductors.
 
True, because a class D topology is highly efficient. 90% of the energy that comes in through the power cord goes out through the speakers as sound, so only 10% of the input energy needs to be dissipated as heat. Compare that to AB designs, where at best approximately 25% of the input energy must be dissipated as heat.
Well, no, not really. Ca. 90% of the power drawn by the Class D amp from the "grid" is indeed presented to the loudspeakers - but virtually all loudspeakers still have efficiencies of a couple percent. Most of the power delivered to the loudspeaker is wasted - as heat. The acoustic power output of a loudspeaker is fairly meager --although one watt of acoustic power corresponds to pretty impressive SPL.

Per Klipsch:
Most speakers are actually very inefficient; only about 1% of the electrical energy sent by an amplifier to a typical home loudspeaker is converted to acoustic energy. The remainder is converted to heat, mostly in the voice coil and magnet assembly. The main reason for this is the difficulty of achieving proper impedance matching between the acoustic impedance of the drive unit and that of the air into which it is radiating. The efficiency of loudspeaker drivers varies with frequency as well. For instance, the output of a woofer driver decreases as the input frequency decreases.

https://www.klipsch.com/education/speaker-sensitivity

I cannot vouch for this 'conversion factor' but it sounds about right.
Be aware that there is a distinction between efficiency and sensitivity. The first is acoustic watts out for acoustic watts in and the other is SPL on axis for a given voltage. They can be converted but only if you know directivity or polar pattern.

1 acoustic watt is 109dB in full space at 1m. If your radiation is hemispheric then this becomes 112dB at 1 m for half space radiation. Most units are even more directional so you need a number for d.i. (directivity index in dB).

With a typical high frequency d.i. of 10dB then a 50% efficient horn (the maximum possible) would be 109 + 10 - 3 = 116.

Scale from there.

David S.
source: http://www.diyaudio.com/forums/multi-way/3973-klipsch-efficiency.html#post2808973

The efficiency of a horn-loaded loudspeaker (e.g., a Klipschorn -- all horn-loaded) is on the order of 10 to 20% (I'm still scouring the web, looking for a precise value).

Not that a more efficient amplifier isn't a laudable development -- heck, that's why push-pull amplifiers supplanted single-ended amplifiers ca. 80 years ago! -- but the benefits only go so far.

In fact, there's a certain irony in using high power, high efficiency (e.g., Class D amplification) to drive a loudspeaker with ca. 80 dB SPL/watt @ 1 meter sensitivity!
 
Last edited:
Well, no, not really. Ca. 90% of the power drawn by the Class D amp from the "grid" is indeed presented to the loudspeakers - but virtually all loudspeakers still have efficiencies of a couple percent. Most of the power delivered to the loudspeaker is wasted - as heat. The acoustic power output of a loudspeaker is fairly meager --although one watt of acoustic power corresponds to pretty impressive SPL.
This is true regardless of amplifier topology.

A class D amplifier is still typically 90+% efficient, class AB at most about 75% efficient, and class A about 30% efficient in terms of power in/out regardless of the power efficiency of the speakers themselves, which is indeed relatively poor.

But you're right, I was being very sloppy by describing it as, "90% of the energy that comes in through the power cord goes out through the speakers as sound".

To be accurate, I should have written, "90% of the energy that comes in through the power cord makes it to the speakers."
 
Back
Top Bottom