Aleph Temperature Sweet Spot

[macka]

I know some of us have touched on heatsink size recently, but but does the Aleph (2) have an ideal operating temperature for best sound/distortion?

Mr Pass indicates 25 c rise above ambient for 50 - 55c and after 1 hour will reach best performance subjectively, and measured.

But how important is this? Obviously if it runs too hot say above 65 there may be some reliability issues, but what if you buy hugh heatsinks and it cruises at say 40 c.

Will this matter to best performance?

The curves for the 244/240 IRF show gentle rise through most parameters and perhaps there is a sweet spot.

My other point is that I assume that it is the top running temperature and not rise above ambient that is important.

Guys how about some feedback!

regards

macka

[Nelson Pass]

50 degrees C has been a standard for me for a
long time, not because it necessarily results in
better sound per se, but is the most ideal high
temperature for sale to consumers. At this point
human skin is quite sensitive to small changes in
temperature. At 45 degrees, you can keep your
hand on a sink for quite a long time, and at 50
you cannot.

Aside from assuring that a consumer or his family
will not be injured by the temperature, when I get a
call from wondering if an amp is biased properly, I can
ask the question, "How long can you put your hand on
the sink?" and get the information I need.

At home I have stuff running hotter, say 65 degrees,
which have not yet died, but then again, I don't lay awake
at night about a potential transistor failure.

[GRollins]

This is a question that I have considered. On the face of it, it would seem silly that the temperature would have an effect on the sound quality.
But...
There are so many other things that defy rational explanation (cable directionality is my favorite example--makes no sense at all, logically speaking) that it might...just might be worth investigating. Basically those who choose to use passive heat sinks will have a tough time investigating this because the creation of heat via bias vs. the dissipation via the heat sinks is going to be fixed by the original mechanical design.
But for those who choose water-cooled circuits, the intriguing option of varying the water flow or the efficiency of the heat exchanger (air flow) would allow one to explore temperature questions independent of bias (to a first approximation--I'm ignoring the fact that the bias will vary with the temperature).
It's one of those things that's on my to-do list for about the year 2010. If anyone gets to it ahead of me, let me know how things turn out.

Grey

[macka]

I have been doing the touch test on my heat sinks and your right, 50 c is too hot to hold for very long.

Thankyou for your clever and practical logic.

Was this another Pass innovation?

regards

macka

[Nelson Pass]

Don't lose sight of the fact that it is the chip
temperature that is important, not the heat sink
temperature. They are not the same thing, and
you can fry a transistor on a cold heat sink.

[Andrew Steele]

In some places of Australia, the temprature during extreme days of heat can get to 50 degrees C but not where I live (thank goodness)

[macka]

I appreciate your point on chip temperature.

I had this problem with my first class A diy project .. A JLH which only has one pair of output devices.

I used cheap SiIicon washers from Jaycar.They appeared to shrink after the first use and l cooked the output stage in 5 minutes because the devices had come loose.

I always use Mica washers now, but I undestand there are better quality self greasing washer available.

Do you have a preference for heatsinks with or without flanges?

I expect I will have to allow for any temperature gradient across the flange and heatsink junction as the flange appears hotter than the heatsink.

My current project for the Aleph 5 uses 1 x heatsink rated at 0.25 c /watt / channel. Theyhave twin flanges which makes it easier to mount the devices, this seems the main advantage.

I will refer this point to my heatsink manufacturer Conrad Engineering.

Thankyou for your assistance.

regards

macka

[Nelson Pass]

I always use Mica for TO-3's, as the silicone insulators
shear at the hole edges. Usually we use silicon for the
plastic devices which have a plastic insulating ring on
the mounting hole.

Thermalloy white grease is my favorite for Micas.

[alaskanaudio]

I find that a output stage temperature range of 145-165 degrees F is quite acceptable. I useally test my amps to 205 degrees F.

With proper temperature compensation circuits the output stage currents can be kept at a near constant value from 85 to 185 degrees F fairly easy. Heatsink design plays an important factor and most be considered when building compensation circuits.

All power amplifiers should have some type of temperature compensation. That is unless you believe that this is not important.

John Fassotte
Alaskan Audio

[mrfeedback]

Hello all,
Regarding temperature, I have read comment in the past by Peter Stein, http://www.me-au.com/sound.html , that he temperature regulates (variable speed fan) to 50*C or something because he finds this to be a sonically optimum operating temperature.

"There are so many other things that defy rational explanation (cable directionality is my favorite example--makes no sense at all, logically speaking) that it might...just might be worth investigating. "

Possible rational explanation is that temperature is affecting noise characteristics of output devices, combined with negative loop feedback system causing change in final output noise.
Concept of wire direction is perfectly rational according to a nuclear physicist I knew.
His explanation was that the copper ingot takes up a magnetic alignment during manufacture and this stays in the copper.
I have done audio experiments that demonstrated this directional characteristic very clearly.
Remember, all materials have some magnetic property.
Search "paramagnetic and diamagnetic, ferromagnetic etc " http://hyperphysics.phy-astr.gsu.edu...ids/magpr.html

Maybe this helps,
Regards , Eric.