Compact Single-Ended Class A Build

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi guys! My name is Aaron, and I'm new to this board, but frequent other audio-related sites as well.

Until recently I had been running a pair of Channel Islands Audio D-200 Class D monoblocks. They are superb amps, but I recently bought a CEC AMP53 single-ended class A amp. It sounds truly amazing, and has inspired me to set about building a pair of amps for myself.

I am a Mechanical Engineer, and have some ideas roughed up for the enclosure/chassis, but am looking for some tips on which circuit to use. The plan is to build a pair of monoblocks - single-ended Class A, that put out somewhere between 30 and 50 watts per channel into 8 ohms. Problem is, this forum is daunting, with 50+ pages in some of the threads. What would be a good circuit/build to go with to accomplish this? Simplicity and purity are key...

I also want these to be fairly compact amps, so a small circuit would be preferable. Cooling won't be an issue, I have that taken care of. Also, planning to go with external matching enclosures to hold the power supply and associated circuitry. Elegantly industrial is what I'm going for. Or industrial elegance?

So, with those goals in mind, what can you recommend? Glad to be a part of this community now :)
 
In that case, your only limit would be the size of the heat sinks. Even a relatively complex circuit like the A-75 can be put on a very small board. ( I think the stereo board from AudioXpress is around 4X4). So you could build pretty much any of the Pass labs projects: A75, Mini A, Aleph X, F5, F4.

I really like the A75 and the F5. (never built an aleph X though.)

JJ
 
Most Pass amps are not designed to be very compact.
Some can be made small, if you wish.

But if your requirement number one is COMPACT, you will have better choices:
- Class D
- Gainclones like LM3875, LM3886
Class D amplifiers ( Class D Forum ) can be extremely small.
Because they can do with very small heatsinking.
 
The words "compact" and "class A" aren't normally found together. I'm assuming that when you say you want the amp to be compact, you're speaking of the active circuitry, since you indicate that you're going to have a separate power supply.
A few more-or-less random points:
--Class A circuits need to dissipate a lot of heat from the output devices. This requires heat sinks of some sort and if you want air to carry off the heat, you're going to be depending on fans to keep the size down. The smaller you want it to be, the more air you will have to move and the louder the fans will be. This will be a balancing act.
--Water cooling allows you to move a preposterous amount of heat and do it quietly if it's done properly. I did a thread on this once upon a time wherein I described a system built around a pair of Aleph 2s. The downside to water cooling is that it isn't portable. Other than that, you're pretty much good to go. For what it's worth, I can easily get the active circuitry of an Aleph 2 into a 1U chassis with water cooling. The power supply is another matter. Transformers don't compress easily. Nor do big electrolytics.
--I'd suggest an Aleph 3/Aleph 30 or Aleph 4 (wasn't that the 60W one?)/Aleph 60 depending on how much power you want. They're relatively straightforward and you'll have oodles of people who can guide you over any bumpy spots, as there are scads of those amps in the field. The A-75 is a wonderful piece, but requires more parts and some adjustment. That may or may not suit you. The Aleph-X might be a contender, but also requires some fiddling and seems to conceptually confound some folks.
--Some of Nelson's First Watt spin-offs might be of interest, depending on power available and scalability.
--Amps with separate power supplies are pretty nifty, but can be difficult to carry around. Consider a detachable umbilical. Neutrik makes some suitable plugs. See also Amphenol, Hubble, et. al.

Grey
 
Water cooling crossed my mind, but that get more complicated than I want, and with the amount of power I'm talking about, probably unnecessary . I'll be using a fan, as Nelson mentioned. You can dissipate substantially more heat than with just natural convection, and make things smaller

The next couple days I'm going to sit down with my buddy and discuss how the circuit will be laid out so I can start getting into more of the packaging details.
 
inspiribomb said:
I recently bought a CEC AMP53 single-ended class A amp. It sounds truly amazing, and has inspired me to set about building a pair of amps for myself.




Dear Aaron

You may have arrived at the wrong place due to shameless and misleading advertising: your CEC appears to be neither single ended, nor class A. It may still be a very nice amp.
 
Yikes! A 120W stereo amplifier that idles at a mere 60W for both channels is definitely not class A and definitely not single-ended. People who make claims like that should be forced to eat their own product without benefit of knife, fork, or salt.
And $2400? Ouch!
Imagine how much better a real class A amp would sound...

Grey

EDIT: It occurs to me that if you're using the CEC amp as a yardstick to judge how big, how heavy, and how hot a class A amp might be, you're in for a serious shock. A stereo 120W Aleph would dissipate over 700W at idle (as opposed to 60W) and require much, much more heat sinking and power supply. Looks as though the CEC is something along the lines of mild class AB at best, possibly just class B.
The upside is that an Aleph will sound better...elseways people would have abandoned class A years ago.
 
I see no information to the contrary to the CEC being a single-ended Class A amplifier, I'd like to know where you found your information. Whether or not it is, it is a fine sounding amp. Maybe it running at 50 C with a fan is some indication? Seems pretty warm to me.

I appreciate all the input, and do understand that heat-sinking can't be overdone. I work in the semiconductor industry and often design manual actuators that have to dissipate quite a bit of heat when testing devices.
 
Well, because a Class A amp would per definition have an idle-current that is equal to the maximum signal swing. But, this means, that in can never be more that 50% efficient (in reality it's more like 25%). So a 120W Amp would have an Idle dissipation of at least 240W (theoretical limit) and more or less likely around 450-500W in the real world.

Please, anybody, correct me if I'm wrong.

Brgds
Michael
 
Sadly, the math is inescapable...it's not single-ended and it's not class A. You've been...well...er...you've been lied to. Yes, I saw where the web site makes those claims, but...
Okay, let's take it from the top.
Class A means that both the upper and lower output devices are conducting at all times during the waveform.
Class B means one bank cuts off and leaves the other to handle the entire load for half the waveform. If you want a good visual analogy, think of the little tap-tap-tap metal balls Magneto has going in the X-Men movie. When one hits, it stops completely and the other starts going. The one that's in motion goes out, comes back, and taps the first one, coming to a stop in the process while the first one cycles.
Class AB is between the two. A different analogy: Relay runners...where one starts running to match speed with the fellow carrying the baton, once the baton is passed, the guy who had been carrying the baton slows, stops and gets out of the way. In the same sense, an AB amplifier output stage has some transition period where both banks of transistors carry the signal at the same time. For that time, and for the brief time period only, you can say that the amplifier is operating in class A. However, it is not correct to describe such an amp as class A, just because it is able to put out, say 3 or 5 or maybe even 10W in class A. If it's a 120W amplifier and it's claimed to be class A, then it's got to be biased hard enough that both banks of transistors remain lit at all times during a full 120W into 8 Ohms waveform.
Unfortunately, it's easy to demonstrate that the CEC ain't...
The math:
SQRT(120W/8Ohms)= 3.87 Amperes of current
This indicates rails of approximately 120W/3.87A= 31V
Those are the RMS numbers, allowing for peaks means that you multiply the 31V by 2.828, which gives us a minimum rail voltage of 87.7V total or +43.8V.
Still with me?
Okay, here comes the bad news...
To deliver the 3.87A of output current, you have to bias for half that amount minimum at idle, and thus your idle dissipation would be...
87.7V rails * (1/2 * 3.87A)= 170W for one channel in a perfect amplifier. 340W for two channels.
Compare this to the claimed 60W idle dissipation for the two channels in the CEC amplifier and you'll see a clear discrepancy.
There's no such thing as a perfect amplifier, so you have to allow for real world things like Vbe or Vgs in the output devices, power loss in the emitter or Source resistors, etc. and the dissipation goes up accordingly. In other words, the math gets even more lopsided in the real world. If you need to supply power to a load less than 8 Ohms, then you have to bias even harder still, with the inevitable result that the math tips further still.
As for the "single-ended" claim, that falls out of the class A thing. Single-ended circuits are by nature class A. If the amp isn't class A, then it isn't single-ended. Yes, yes, yes...it's possible to drive a single-ended output until the follower half shuts off, leaving the current source to carry the load alone, but please believe me when I say that you don't want to do so. It ain't a pretty sight.
So, sorry, but the CEC is neither class A, nor single-ended.
Another nail in the coffin is the picture showing the innards of the amplifier from the top. There ain't no blinkin' heat sinks visible. None. Nada. Nothing.
When looking at a class A amp from the top, the heat sinks will be a dominant, can't-miss-it feature. In other words, they're huge. Big. Gargantuan. We're talking a good 50% of the visible real estate of the amplifier. The lack of heat sinks (I'm reduced to assuming that there's something inside or on the bottom...pretty please?...but that's the worst possible place to put heat sinks because there's no air flow. Fans perhaps?) almost makes me wonder if the thing is actually class D, that bing about as far from class A as you can get and still produce music. (Yes, there's a class C--they didn't skip the letter--but it's not suitable for audio use.)
You're a victim of creative advertising. Sorry. I hate that we're the ones to break it to you, 'cause you like the amp and now you're going to hate us for being the bearers of bad tidings, but if you do the math there's no way around it. They're lying.
Guys, I'm at work and I'm out of time, so this post is going in rough. Somebody who's got a second double-check my math, please. I know it isn't class A, but I don't want to mislead him on the actual numbers. Overlook any typos, etc. the usual disclaimers when I'm rushed.

Grey
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.