What does it take to design a class A amplifier

[wettou]

I would like to build a Class A amplifier that provides 500W power at 8 ohm

[cjd]

biiiiiiig heatsinks for one

[Hdale85]

500watt Class A? I would say thats darn near impossible. I mean you can create that kind of Class A power sure but to cool it is going to take something huge and when I say huge I mean HUGE! An Aleph60 requires heatsinks that normally a Class A/B 7 Channel amp would use. So 500 watts is going to be something like the 2 heatsinks the size of a small fridge or maybe even bigger (LxW wise) for just one channel.

[cjd]

I already talked about big heatsinks... :P

Class A/B is the only sensible way to get to 500W. What do you want that much pure class A for?

[Hdale85]

Well if its a sub its pointless thats for sure

[Gir]

Anyone remember those 195,000W amps from NASA I posted? Full class A, cooled with ionized water.

Seriously, though, A/B's the way to go.

[wettou]

I already talked about big heatsinks... :P Class A/B is the only sensible way to get to 500W. What do you want that much pure class A for?

Drive 802D

[Amphiprion]

You will need two 20A lines from the breaker box, one for each amp, with nothing else on it.

An extra air conditioner for the room they are in.

Heatsinks sized like tanks.

A transformer that would give five people lifting it a hernia.

Seriously, class AB is your friend.

[TacoD]

What about class-D? You can use water cooling, will safe you an air conditioning/ impossible heat sinks.

Even the big Pass Labs X600 amplifiers are not completely class A.

[JoshK]

biamping is your friend too. A/B on woofer, class A on mid/tweet.

[Hdale85]

I would look for a class A/B design that runs in Class A for most of its power band because thats where most of your listening is. Honestly depending on what amp you have now if it performs a lot in Class A then you may not notice much of a difference going to a full Class A amp.

[wettou]

I would look for a class A/B design that runs in Class A for most of its power band because that's where most of your listening is. Honestly depending on what amp you have now if it performs a lot in Class A then you may not notice much of a difference going to a full Class A amp.

Classé CA-5200 I know it is design like a classe A/B an ideas about how many watts in class A

[Saurav]

http://www.hometheaterhifi.com/volume_12_4/classe-ca-5200-amplifier-11-2005-part-1.html

Bias adjusts itself depending on the demands, such that about 30% of the output is always in Class A

So that would be about 60W.

[Hdale85]

I doubt you really ever get more then 60watts rms under normal listening conditions. You might hit higher then that on peaks but even thats pushing it.

[Dennis H]

Seems like I recall JonMarsh saying more A in an AB design isn't necessarily better. There's an optimum amount and more or less than that will give you more distortion.

[Hdale85]

Well it depends on the design. But usually a Class A/B is designed to where Class A goes until it starts getting to much distortion and then it switches to B. Being that he's running a Classe I would take a guess that its designed properly.

[A9X]

What about class-D? You can use water cooling, will safe you an air conditioning/ impossible heat sinks.

Even the big Pass Labs X600 amplifiers are not completely class A.

You won't need water cooling for 500W class D amps, even a modest non fan assisted heastsink will do. Some of my earlier experiments with larger Pass amps used water cooling, but I found high efficiency speakers to be more worth the effort and at LF, for subs, the class of the amplifier won't be audible. That's the only place I'll need all that much power.

Re the Pass, thought some might find this interesting.

http://www.passdiy.com/pdf/leaving_class_a.pdf

[TacoD]

You won't need water cooling for 500W class D amps, even a modest non fan assisted heastsink will do. Some of my earlier experiments with larger Pass amps used water cooling, but I found high efficiency speakers to be more worth the effort and at LF, for subs, the class of the amplifier won't be audible. That's the only place I'll need all that much power.

Re the Pass, thought some might find this interesting.
http://www.passdiy.com/pdf/leaving_class_a.pdf

The water cooling part was for the class A amplifiers.

I also tried different configurations and found out that speakers are imho the biggest factor for the sound experience.

[A9X]

The water cooling part was for the class A amplifiers.

Sorry, I misunderstood.

I also tried different configurations and found out that speakers are imho the biggest factor for the sound experience.

I agree. Amps are far, far less of a sonic issue.

[Hdale85]

Well I wouldn't say far far less. I mean I can easily here the difference in amps with my speakers but they are extremely revealing speakers. So yes the speakers will make the biggest difference up to a point.

[A9X]

Well I wouldn't say far far less. I mean I can easily here the difference in amps with my speakers but they are extremely revealing speakers. So yes the speakers will make the biggest difference up to a point.

You're entitled to your opinion and experience of course, but mine does not agree with it. I've built a lot of amps, some my own design, some other peoples and within reason (not clipping, low THD and output Z), the differences are small. Almost no one has picked out changes made blind, and never anyone consistently. I've built a few speakers too, from modified Tannoys, and Khorns to full range front loaded horns and LABsubs.

PS: I meant speaker + room, not just speakers.

[chrismercurio]

So....why build it?

Buy an old Krell FPB600. 600w into 8 ohms, stable into a dead short (almost), and the power doubles everytime the impedance is halved. I sold these in the 90's and they sound pretty good.

http://cls.audiogon.com/cgi-bin/cls.pl?ampstran&1218865612

http://www.stereophile.com/solidpoweramps/829/index.html

[Hdale85]

Well I wasn't the only one that could pick out differences. Of course we were comparing to a Rotel amp and I'm not a real Rotel fan

[littlesaint]

I'm not an EE, but I have built some class A SETs and a class A/B push-pull amp. All tube of course.

My understanding of A/B isn't that it "switches" from one to the other, but that in a class B amp, an output produces half of the signal wave and a complementary output produces the other half (hence the name push-pull). Joining the two outputs together can introduce distortion so each output is setup to produce a little bit of it's other half. This is technically class A operation. How much of the other half being output determines how much of the amp is operating at class A. So it's never really true class A or true class B, but something in between.

[A9X]

I'm not an EE, but I have built some class A SETs and a class A/B push-pull amp. All tube of course.

My understanding of A/B isn't that it "switches" from one to the other, but that in a class B amp, an output produces half of the signal wave and a complementary output produces the other half (hence the name push-pull). Joining the two outputs together can introduce distortion so each output is setup to produce a little bit of it's other half. This is technically class A operation. How much of the other half being output determines how much of the amp is operating at class A. So it's never really true class A or true class B, but something in between.

Did you read Nelson Pass' paper I linked a few posts ago?

[wettou]

Did you read Nelson Pass' paper I linked a few posts ago?

No where is the link?

[Saurav]

In post #17.

[wettou]

In post #17.

link!

[Saurav]

LOL. All you had to do is scroll up to find post 17

But I will copy and paste the link for you:

http://www.passdiy.com/pdf/leaving_class_a.pdf