mini aleph build questions
 

Hi, I got some Aleph PCBs from BrianGT ages ago, now I am planning the build.

I want to build mini aleph monoblocks using no extra heatsinks. Just by using the cases I have to dissipate the heat. I estimate case to have 2*C/watt thermal resistance per side (4 case sides in total if I build monoblocks).

Is it true mini Aleph only draws about 1A with +/-15v power... 30W of heat per channel max? In that case, amps should reach 50*C max. But if I find they are running too hot, can I simply reduce power supplies to +/-12v instead?

An Aleph with 1A at +/-15V probably will give you about +/13V, and
the maximum power will be at about 6.6 ohms with about 13 watts.

Should work just fine.

If you're running too hot you can reduce either the rail voltage or the bias or both and that will cure the heat problem. (Or get more or bigger heat sinks.) The thing about reducing the rail voltage is that distortion will rise rapidly; I was pushing it a bit by building the circuit with 15V rails. The problem being that the MOSFETs' Gate capacitances increase below roughly 25V or so. That, in turn, makes the MOSFETs harder to drive, particularly at high frequencies. As a direct consequence, distortion increases.
Is it the end of the world? Of course not. It appears that a lot of these amps (and variations thereof) have been built, and people seem to be quite happy with them. It's just a question of being aware of the tradeoffs before you start making changes.
On the other hand, reducing the output bias will also increase distortion a bit. There's no such thing as a free lunch.
Just for the record, the lowest 'optimum' rail voltage for MOSFETs often turns out to be 25V or so. At that point, the Gate capacitance tends to make a dogleg and begins flattening out. Further increases to the rail voltage tend to produce only incremental reductions in Gate capacitance. That's not to say that there aren't other potential benefits, but we're drifting farther and farther away from your original question.

Grey

P.S.: It's no accident that Nelson tends to publish schematics for MOSFET circuits that show rails ca. 20-25V...

I have thought that it is because Papa wants to 100% utilise his 18VAC-secondary transformers he still has in stock.

Am I wrong. . .? ;)

Not drifting too far off topic here. I do have a spare 25V transformer that would be good to make use of. Trouble is by the time it's been rectified and filtered it's going to be at about +/-34v. Do I dare try to use this? If I can reduce bias enough will it work well? Bare with me here I'm not even sure how to reduce bias current.

This is the PCBs I have... I would need to change R27 and R28 to adjust bias current is that right?
http://www.chipamp.com/diyaudio/miniA-amp.jpg

Decrease bias by increasing the value of R27, R28. With a +/-34 rail @ 1A, you have approximately 35watts dissipation on each fet, so I think you're OK there. But you may run into trouble as your heat-sink or lack thereof, if I understand you correctly, will not meet the requirements.

Now ask yourself why Nelson bought transformers with those particular voltages...



The problem with reducing the current to match your heat sinks with that sort of rail is that you'll be so lightly biased that the amp will leave class A at anything over a zillionth of a watt. Not a pretty picture, particularly with an Aleph. Alephs don't do the class B thing so well.
I'd suggest scrounging up some bigger heat sinks...then you won't have to worry about the heat so much.

Grey

The dogleg is at about 5V, but at low supply voltages, the dogleg
starts to be a substantial part of the curve.

You're right, +/- 25V or so for rails is a good compromise value, so
I bought a butt-load* of them.

:cool:



* "butt-load" is a trademark of Dana Kruse

Looking at something like the IRFP240 (Fig. 5), the input capacitance becomes more-or-less flat after you pass the low to mid-20V range. Below that, it's a fairly smooth curve. The FQA19N20C (also, coincidentally, Fig. 5) shows an even clearer flattening at the same voltage range. Curiously, the FQA shows the curve flattening below 1V as well...didn't remember that. Not that running a power MOSFET at 1V is necessarily all that good a game plan.
I'm not saying that you can't get music out and that the music doesn't sound pretty darned good. It's just a question of which compromises you want to make.
On the other hand, you could run a MOSFET at 20A bias at 1V, then cascode for voltage if you were to provide an alternate path for the bias around the cascode device...

Grey

I don't want to start a new thread, so I will ask here:

- At what current shall I match IRF9610?

I have 20pcs of IRF9610 (Vishay) and I have Vgs within 15mV for all of them at 5 to 20mA. Am I just lucky? or shall I use different current?