Hi, I just finished building an Aleph 3. It sounds good and works perfectly. No problems at all. Fired it right up. I built it from scratch with perf boards. I ran simulation and the real deal. Both very close.
At 20-20khz, I got 25 watts into 8 ohms before any noticeable clipping on oscilloscope. 37 watts into 4 ohms (did not double as I expected), and only 18 watts into 2 ohms. My dc supply is right around 25 volts.
I calculated that if I wanted to see 100 watts at 2 ohms, I would need about 10 amps of idle current on this baby, which is about 5 times what it is running now. Probably less though because I calculated from Class A with standard current sources, not the Aleph source which is more efficient. And the mosfets would not handle the current anyway because there are only 4 per channel.
Any thoughts?
At 20-20khz, I got 25 watts into 8 ohms before any noticeable clipping on oscilloscope. 37 watts into 4 ohms (did not double as I expected), and only 18 watts into 2 ohms. My dc supply is right around 25 volts.
I calculated that if I wanted to see 100 watts at 2 ohms, I would need about 10 amps of idle current on this baby, which is about 5 times what it is running now. Probably less though because I calculated from Class A with standard current sources, not the Aleph source which is more efficient. And the mosfets would not handle the current anyway because there are only 4 per channel.
Any thoughts?
Build a bigger Aleph!! I'd recommend biasing it around 6A, and you'd probably need a total of 8 fets/ch to keep them alive, that'll give you around 120W@2R. Just be sure your heatsinks are up to it, you'll be burning about 300W/ch. No longer an Aleph 3; more like a supersized Aleph 5....same sound, more power.
Well the idle current doesn't have anything to do with the maximum output power of the amplifier. The idle current sets the maximum number of class A watts that the amplifier will produce before transitioning into class AB.
Getting 25 watts @ 8 ohms and then 37 watts @ 4 ohms is nothing unusual. The power rails sag a little under the heavier loading, reducing the output voltage swing of the harder load, but the losses within the output become more significant into lower ohm loads too.
Having the output power crumple into 2 ohm loads is nothing unusual but it depends on the amplifier. The Aleph three uses 2 pairs of output devices and should theoretically be able to driver 2 ohm loads from the position of the output devices being able to deliver enough current. Obviously this isn't the only limitation as any protection circuitry will dominate here to keep the output stage alive during a short circuit. Then there's the previous stages ability to drive the output stage into more demanding loads.
Does the Aleph three have any protection circuitry?
Getting 25 watts @ 8 ohms and then 37 watts @ 4 ohms is nothing unusual. The power rails sag a little under the heavier loading, reducing the output voltage swing of the harder load, but the losses within the output become more significant into lower ohm loads too.
Having the output power crumple into 2 ohm loads is nothing unusual but it depends on the amplifier. The Aleph three uses 2 pairs of output devices and should theoretically be able to driver 2 ohm loads from the position of the output devices being able to deliver enough current. Obviously this isn't the only limitation as any protection circuitry will dominate here to keep the output stage alive during a short circuit. Then there's the previous stages ability to drive the output stage into more demanding loads.
Does the Aleph three have any protection circuitry?
Alephs are single ended, there is no transition to AB, it doesn't exist.
There is current limiting in the Alephs. Negative output current is limited to slightly more than the maximum positive output current, typically 2.5 times the bias current. I believe it's just one resistor for this adjustment, it's value will depend on bias settings.
As for the preceding stage driving the output stage, I believe it's just a matter of the input fets properly driving the input capacitance of the output fets in this case.
Of course I could be completely wrong in all cases here.....I just build them, run em on the ragged edge, then blow em up.
Jacco.....Aleph 3.14!!
Hmm, the amplifier is driven by dual rails and uses separate pass devices to control the current flow from each rail. The two have to meet somewhere.
I thought a traditional single ended amplifier used a single supply rail, one pass device, so to speak and an output coupling capacitor, or am I missing something here?
Edit - so it would seem that this is more like a bridged amplifier and a bit like a BTL class D in that two single ended power stages meet in the middle and in the Alephs case presumably cancel out any output offset negating the need for a coupling cap.
Last edited:
Do you live in a cold climate? That's going to be a seriously hot amp. It's going to sit idling about 400w, twice that if you are talking stereo.
Don't let that stop you, of course, it's still neat as all heck!
But the other posters are correct, you will need to build something new. If you have the heatsink for it, just go straight to an Aleph 1.2
Lately i'm more inclined to a 3 Turbo, with common cathode diodes across the 0.47 bunch.
I calls it the 2-stage titty twister. (or aleph 3.5.5)
Pass DIY Addict
Joined 2000
Paid Member
No doubt! When I bias my Aleph-X up to about 9A, it idles in the neighborhood of 400w and requires heatsinking of about 0.06c/w per chassis. That is one big, heavy, and hot chassis! I have three such amps powering the front speaker array for my theater and the whole system requires its own dedicated 30A breaker and a 10-gauge run to the breaker box.
Definitely lots of power! But they sound so very sweet!
What do the diodes do across the .47 resistors? I am not familiar with that.
Thanks everyone for your input.
Just an experiment for my curiosity, see what happens (read the Aleph current source paper)
Mega-amp was joking about a talk on boost pressure some time ago.
Well, I experimented with putting forward biased diodes across the .47 resistors. Putting them in the current source does nothing, in fact, it creates problems. However, putting them in the lower amplifier section actually helps low z performance, however the top half of the sinewave is the first to clip, so I think I need more current. I want a way to do this in a more traditional sense without using the aleph current source. For example, I removed the capacitor/resistor from the negative feedback line going to the transistor that sets up the current level through the current source.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.