Aleph 2 Current/Voltage Load Issues.

[GRollins]

Higher voltage, like higher current, is good. It reduces distortion. Drawing a balance between voltage and current, as in how much of each, is part science, part voodoo. Okay, let's rephrase that as part heat dissipation, part listening.
The rail voltage sets the maximum wattage you can get out of an amp. The bias current tells you how low an impedance you can deliver that wattage into. If you should have something rare, like a 32 ohm speaker, you can deliver a given wattage with lower bias than if you were to want to drive a 4 ohm speaker.
That's the fix I found myself in. I was driving the Magneplanar ribbon tweeters (ca. 2.5 ohms) with a pair of Aleph 2s. Sounds whacky, right? Only, the Aleph 2 current limits into such a low resistance. Below about 5 ohms the Aleph 2's power begins falling instead of rising because it runs out of current.
Enter the Aleph-X. Lower rails, but higher current. Ta-da! Lower device dissipation and decent power output into low impedances. The price being (at 15V rails) somewhat higher distortion.
Your choice.

Grey

[jh6you]

Originally posted by GRollins

Higher voltage, like higher current, is good. ...

Grey

Hmm... well understood. But, I often wonder if the meaning of the "good" is also with respect to the sound quality which is in general DIYers' ultimate target.

[Fred Dieckmann]

The Aleph 2 current limits at 10 amps. Looking at the crossover schematics for the Tympani IV A and MG-IIIa (which I own) gives a tweeter impedance in the range of 3.5 to 4 ohms. This agrees with the speakers being around the 4 ohm value stated in the Magnapan literature. Last but not least, the speaker uses a 2.5 amp fuse for the tweeter!

I squared R (power) for 10 amps and 3.5 ohms is 350 watts! 10 amps times 3.5 ohms is 35 volts, a little hard to get with 15 volts rails, even with 2.5 ohms (don't forget the voltage drop of 10 amps across the source resistors.)

I am not accusing you of breaking any laws.......... well, except for ohms law. You have either smoked a lot of tweeters or are blowing quite a bit smoke up the forum once again. Learn ohms law, learn to measure stuff, and stop pulling everyone's leg. I will be glad to leave you alone when you post something even remotely plausible. I'm afraid this wasn't one of those times...........

http://www.integracoustics.com/MUG/MUG/

http://www.passlabs.com/pdf/aleph/a2man.pdf

[Nelson Pass]

That would be a pretty distorted 10 amps from an Aleph 2

[Fred Dieckmann]

You won't hear much of it while you are stamping out the little pieces of burning aluminum foil on the floor of your listening room..........

*Flaming Maggies

[GRollins]

Fred,
10 amps from an amp biased for 3 amps? Last I heard, you could get double the bias, which would be more like 6 amps for an Aleph 2.
I'm shooting from memory, but I think the Aleph 3 was the only one with enough bias to double its power into a 4 ohm load--and that was just about its limit. All the other models peaked (wattage-wise) at higher impedances.

Grey

EDIT: Oops, forgot, sorry. My tweeters measure a little under 2.5 ohms (2.3, I think, I usually round that off as 2.5), as measured with a Fluke 8060A. Sorry if that doesn't suit you. The tweeters are no longer in the original Tympani panels. They're mounted with the RD-75s in a stand-alone enclosure. I took the mid/tweeter panels off entirely, leaving the woofer panels. My current speaker system consists of the Magneplanar tweeters and Bohlender-Graebner RD-75s in a 7' enclosure with a narrow frontal area; 8" as I recall, the Magneplanar woofer panels, and the subs (six of the original 12" Titanic woofers per channel).

[MikeW]

It was me that was turning up the voltage. I change the values on a mini-A to match an Aleph 5 (+/- 34 volts) and biased it at 1.3 amps. Thanks for the idea Rod. Still only one set of outputs. Next is the Aleph 2. No smoke yet.

[Nelson Pass]

Originally posted by GRollins
Fred,
10 amps from an amp biased for 3 amps? Last I heard, you could get double the bias, which would be more like 6 amps for an Aleph 2.

I recall measuring that figure, and certainly there's nothing
to prevent the circuit from delivering - 10 amps. (distorted, as
I said)

Most of the Alephs peaked at about 6 ohms, and the 2 had a
bias a little larger than 3 amps and a current gain slightly
greater than 50% (also from recollection).

[Fred Dieckmann]

First read the spec.

http://www.passlabs.com/pdf/aleph/a2man.pdf Page 4

Maximum Output 40 volts, 10 amps

Next look at the schematic page 6. Six output devices in parallel each having a 1 ohm source resistor. 10 amps divided by 6 gives 1.66 amps across each of the 1 ohm source resistors. Current limiting is achieved by the NPN transistor Q4 across R14, the 392 ohm resistor that converts the current output from the front end diff pair to the voltage to drive the gates of the 6 output devices. A two resistor voltage divider across one of the output source resistors to the cause the desired current at limiting to be about 0.7 volts needed to cause the NPN Q4 to start conducting. When it conducts it steals the drive current from the 392 ohm resistor causing the voltage across the resistor to decrease. This lowers the gate voltage to the output mosfets and they start conducting less current and we have current limiting. At 10 amps we have 1.66 volts across each 1 ohm source resistor. 1.66 volts /(150+221) gives 4.5 mA. That current through the 150 ohm resistor across the base to emitter of Q4 is 0.67 volts which is about voltage where Q4 conducts and drops the gate voltage causing the outputs to current limit. If I have messed this up too badly I expect Nelson will be along to dope slap me on the back of my head.


"My tweeters measure a little under 2.5 ohms (2.3, I think, I usually round that off as 2.5), as measured with a Fluke 8060A."

The DC resistance is not the same value as the AC impedance (if you measured the DC resistance). I honestly think that 2.5 ohms is really too low for the value of the impedance for these ribbons.

"Sorry if that doesn't suit you."

Makes no difference to me......... believe what ever you what, it's never stopped you before.

[GRollins]

Fred,
I take the phrase 'maximum output' as being more along the lines of peak output, not RMS. If you're talking peak and I'm talking RMS, then of course we're going to come to different conclusions.
Yes, that's the DC resistance. So you're saying that the Magneplanar ribbon tweeters aren't resistive drivers? News to me. Yes, technically even a piece of straight wire has inductance, but the amount is trivial. It takes a coil to add up to anything worth caring about in this application. The aluminum strip in the ribbon is a pretty fair approximation of a straight wire, so I don't see that there's any significant amount of inductance. Capacitance? From what mechanism? I don't see where you're getting a reactive component for the driver itself. Since I no longer use the passive crossover, being quad-amped, I don't worry about what the crossover might have brought to the table.

Grey