[DaKi][er]

Quote:

Originally Posted by Crowbar /img/forum/go_quote.gif Well, unforgiveness is why you added the protection mechanisms.

That, as well as being able to leave it as a fully DC coupled input, so why block DC when most of the time there isnt any? My protection circuit serves many functions

 

[cotdt]

the pass labs amps are a pain to build because of the big heatsinks.

 

[Crowbar]

What? You tool, that applies to any class-A power amp. If anything, the Pass amps tend to be more efficient than regular SE class A since the Aleph current source switches to PP for large signals.

 

[cotdt]

Quote:

Originally Posted by Crowbar /img/forum/go_quote.gif What? You tool, that applies to any class-A power amp. If anything, the Pass amps tend to be more efficient than regular SE class A since the Aleph current source switches to PP for large signals.

My Class D amps output 100W and 400W, and neither of them even needs heatsinks at all! 10W amps that need massive heatsinks is completely impractical.

 

[Crowbar]

Class D... LOL! This is supposed to be a hi-fi forum. When I hear hi-fi sound from a switching amp, I'll eat my shorts.

10 W massive heatsinks? What garbage is this? My Aleph-X are 150 W per channel, and the heatsinks are not massive (especially since I'm using a slow fan in a heat tunnel configuration).

And don't even bring efficiency into the picture. Power usage of the most inefficient amp is nil compared to the 40 killowatt-hours in a gallon of gasoline.

My plasma speakers run on 1800 W and I'm not worried.

Finally, even your class-D amp doesn't help that much, since the actual portion of output energy converted to sound in the average speaker is a few percent--why worry about amplifier efficiency so much, when the best speakers rarely are more than 10% efficient, with 90% of the electric energy being converted to heat in the speaker?

Efficiency in audio is truly a red herring, unless you're making amplifiers for theaters, concert halls, or outdoor systems (or portable battery-powered headphone amplifiers, but that's not the discussion here).

 

[hugz]

Quote:

Originally Posted by cotdt /img/forum/go_quote.gif the pass labs amps are a pain to build because of the big heatsinks.

It's not that big of a deal. have you built a pass labs amp or are you just talking? Because I've built one and all you have to do is screw two parts to the heatsink. Takes like 20 minutes

Quote:

Originally Posted by cotdt /img/forum/go_quote.gif My Class D amps output 100W and 400W, and neither of them even needs heatsinks at all! 10W amps that need massive heatsinks is completely impractical.

You're compairing apples to oranges though. It's not like the pass labs amps use heatsinks because they're poorly designed. They use heatsinks because they're class A amplifiers, and they're specially designed to work really hard to do very little, but do it perfectly. It takes 75w per channels and turns it into 5w of output. Thats what the heatsinks are for.

You say it's "completely impratical", but i cant see how. How is having an amp with large heatsinks any more impractical than one without them? They dont require any maintanance or require me to switch them on before listening. they just do their job passively.

As far as i'm concerned, REQUIRING a 100w or 400w amp is impractical. My speakers are happy with 5w, so it's all very practical and easy. If you have speakers that need 400w amps than you're using more electricty than me just to get the same result (except my result is class A, generally considered the best; whereas yours is class D, generally considered to be not hifi), plus it seriously limits your amplifier choice. I'd be hard pressed to find an amp that wasn't powerful enough for my needs

 

[greyhorse]

If it helps any, I've built a Buschhorn Mk II using the RS-1197, but found high output impedance amplifiers didn't work well with it. I didn't try any SS amps like the F2, but tubed ones like SE 2A3s, and PP EL84s. With the tubed amps the vocals were shrill, and the bass pretty much non-existant. Once I switched to a moderately powered (100Wpc) Class AB SS amp, the sound was much more controlled and full-bodied.

I think everyone agrees that the FE103 is rather thin sounding when used on its own, so thin sounding amps like Gainclones probably won't give you good results. I've built one myself, and didn't see what the fuss was about anyway. In that sense an Aleph-X might be a good choice, but kind of pricy. Why not try some of the more popular Class AB SS amps out there like the SKA? I've recently built ESP Audio's P101, but find it thin sounding, so I don't recommend that.

 

[star882]

Quote:

Originally Posted by Crowbar /img/forum/go_quote.gif Class D... LOL! This is supposed to be a hi-fi forum. When I hear hi-fi sound from a switching amp, I'll eat my shorts. 10 W massive heatsinks? What garbage is this? My Aleph-X are 150 W per channel, and the heatsinks are not massive (especially since I'm using a slow fan in a heat tunnel configuration). And don't even bring efficiency into the picture. Power usage of the most inefficient amp is nil compared to the 40 killowatt-hours in a gallon of gasoline. My plasma speakers run on 1800 W and I'm not worried. Finally, even your class-D amp doesn't help that much, since the actual portion of output energy converted to sound in the average speaker is a few percent--why worry about amplifier efficiency so much, when the best speakers rarely are more than 10% efficient, with 90% of the electric energy being converted to heat in the speaker? Efficiency in audio is truly a red herring, unless you're making amplifiers for theaters, concert halls, or outdoor systems (or portable battery-powered headphone amplifiers, but that's not the discussion here).

What's your source? Digital or analog? Analog sources are disadvantaged slightly on class D since they would be converted to digital and back to analog. Digital sources, however, simply need to be processed into PWM before amplification and conversion to analog in the speakers.
As for efficiency, class D amps stay efficient over a wide range of output powers. Most amplifiers only achieve maximum efficiency at maximum power output. Note that a mere 5w to the speakers is a lot of volume! At that level, a 100w class D amplifier can still achieve efficiency of 80% or above. In contrast, most 100w amplifiers can only achieve maybe 30% or less efficiency. The class D amplifier allows for much less waste while still giving enough power for peaks.

 

[Crowbar]

Quote:

Originally Posted by greyhorse /img/forum/go_quote.gif In that sense an Aleph-X might be a good choice, but kind of pricy. Why not try some of the more popular Class AB SS amps out there like the SKA?

How is it pricey? The most expensive single thing is the power transformer, but I just wound my own, so the two monoblocks cost me about $300 in parts each. For comparison, it cost me $400 to build the Blue Hawaii, which is a headphone amp. (These prices include chasses.) Winding your own is great because you can make sure there's an electrostatic shield between the primary and secondaries, and you can put on additional custom windings. Anyway, you can't compare an Aleph-X to an AKSA amp LOL! If you insist on a class AB, I suggest you go for an ultra-low distortion design like the one by Bob Cordell; this one's THD at even 20 kHz full power output is less than 0.001%: http://www.cordellaudio.com/papers/MOSFET_Power_Amp.pdf
This is an old design, published in the journal of the AES way back in 1984, so with modern devices and 1% resistors you'd get even better performance. The amplifier uses Hawksford error correction in the output stage which reduces the crossover distortion of a class AB by about 30 dB, and error correction I'd recommend for any AB amp as it's very effective in getting most of the way to class A sound; for a class A amp, it would make much less difference.

Quote:

Originally Posted by star882 /img/forum/go_quote.gif What's your source? Digital or analog? Analog sources are disadvantaged slightly on class D since they would be converted to digital and back to analog. Digital sources, however, simply need to be processed into PWM before amplification and conversion to analog in the speakers.

Class D and PWM are not digital. This has been written about extensively. The conversion from digital PCM to PWM is a type of digital-to-analog conversion; the analog in the PWM is the length of each pulse. There are power-DACs, that is, fully digital amplifiers, but class D is not it.

Quote:

The class D amplifier allows for much less waste while still giving enough power for peaks.

But I've made a good point the waste is not significant, since a few hundred Watts even several hours a day is not much compared to, say, the 40 killowatt-hours of energy in a gallon of gasoline. And I see people all the time argue about amplifier efficiency, then go and use class-D to drive low efficiency speakers that need ten or more times the power to produce the same SPL as a high efficiency speaker.

 

[greyhorse]

Quote:

Originally Posted by Crowbar /img/forum/go_quote.gif How is it pricey? The most expensive single thing is the power transformer, but I just wound my own, so the two monoblocks cost me about $300 in parts each.

The OP asked for a beginner's DIY power amplifier project for a set of speakers that cost around $100 per pair to build. Just judging by the simplicity of the Zigmahornet, one can guess he's looking for a similiarly simple power amp to build. Since when did wrapping your own power transformer become part of a beginner's amplifier project? I think even Grey himself admitted that all the parts matching and careful adjusting involved in setting up an Aleph-X can be complicated for the beginner.

In recommending the Simple Killer Amp (SKA), I pointed towards a simple to make SS amplifier with a relatively low output impedance that a lot of people have had success with. Unlike you, I don't consider THD figures that important. I let my ears decide what's good and what isn't. Having built an Aleph-Mini, I don't find Class-A amplification, in and of itself, to be necessarily superior to a Class AB amplifier. I was driving a set of FE207E Bass Reflexes, BTW, so don't tell me the Aleph-mini was underpowered.
http://www.diyhifi.org/amplifierguru/guru_001.htm

 

[Crowbar]

Class AB has the same distortion mechanisms as class A, with the addition of crossover distortion. There is most certainly no distortion that class A has that class AB does not also have (all else being equal). Take a class AB amp and bias the output stage to up to class A, and distortion will go down. Take a class A amp and bias the output stage down to class AB, and distortion will go up. It really is as simple as that.

The Cordell amp I referenced is excellent, since it's not just the THD, but all other distortion measures that are extremely low: crossover distortion, transient intermodulation distortion, phase intermodulation distortion; testing with real reactive loads was done. The only type of distortion that he did not test for is the thermal memory distortion (see http://peufeu.free.fr/audio/memory/). That means that there's essentially no type of audible distortion that could have gotten through.
THD measures are indeed not correlated with perception, in the sense of the specific spectrum of the distortion being more important as the ear masks certain distortions much more strongly. But when you have extremely low distortion measures in all the mentioned measures, that overall covers any possible distortion, because any possible distortion would affect at least some of these metrics. And the numbers for Cordell's amp are already a couple of orders of magnitude lower than the best speakers' distortion.