i have 4 SK 102 heatsinks
the height is 100mm
http://www.schuro.de/Daten/Fischer/SK47.pdf
is there any amp i could bould.. i havde the aleph p 1.7 preamp and i was thinking of yet an aleph
can it be done???
the height is 100mm
http://www.schuro.de/Daten/Fischer/SK47.pdf
is there any amp i could bould.. i havde the aleph p 1.7 preamp and i was thinking of yet an aleph
can it be done???
at 100mm height - the factor is 0,4K/W - so I would take 2 and 2 heatsinks per each channel - if you want to stay on the very "safe" side - A3 is the choice - you'll have 20°C rise above ambient temp - with A5 that goes to 30°C ... Still very very OK!
The only thing is that the chassis will be very wide and depth - but quite low ... But personnaly - I like that ...
The only thing is that the chassis will be very wide and depth - but quite low ... But personnaly - I like that ...
caution
if you mount 2 of these heatsinks in close proximity and thermally coupled (like making an amp side of 430mm length) the thermal resistance is not reduced in a linear manner. this means that you won´t achieve 0.2 K/W in this case, but rather 0.27 K/W - 0.3 K/W because the 2 heatsinks heat each other.
Uli
if you mount 2 of these heatsinks in close proximity and thermally coupled (like making an amp side of 430mm length) the thermal resistance is not reduced in a linear manner. this means that you won´t achieve 0.2 K/W in this case, but rather 0.27 K/W - 0.3 K/W because the 2 heatsinks heat each other.
Uli
Re: caution
If you look at the graph for this sinkprofile your story is correct if both heatsinks are placed on top of eachother, that's when 0.27 to 0.30 applies.
Heatsinks act like chimneys, place two next to eachother and each retains its function.
More important with Class A amplifiers is to put the heatsinks well above floorlevel so that cool air may flow to the bottom of the heatsink, warm and rise.
If placed parallel to eachother there is the effect of cross influence but with minimal effect.
Put a vertical plate to the open flange side of the heatsink and the thermal factor would be reduced to a significantly lower level than 0.20 at 100 mill (creating a true chimney).
A heatsink like that is extremely costly in production, therefore only done on special order.
Some manufacturers get around that problem by placing the heatsink in a duct, for example take a look at the Vincent SP991.
An idea for the true DIY'r, build a heatsink from 1/2 inch thick alloy plates, tapped and bolted together, with thermal grease at the joints. A heatsink shaped to favor, with excellent cooling ability, and a base plate of more favorable thickness for cooling high bias demands than regular alloy heatsinks.
The really cheeky use copper plates.
uli said:
If you look at the graph for this sinkprofile your story is correct if both heatsinks are placed on top of eachother, that's when 0.27 to 0.30 applies.
Heatsinks act like chimneys, place two next to eachother and each retains its function.
More important with Class A amplifiers is to put the heatsinks well above floorlevel so that cool air may flow to the bottom of the heatsink, warm and rise.
If placed parallel to eachother there is the effect of cross influence but with minimal effect.
Put a vertical plate to the open flange side of the heatsink and the thermal factor would be reduced to a significantly lower level than 0.20 at 100 mill (creating a true chimney).
A heatsink like that is extremely costly in production, therefore only done on special order.
Some manufacturers get around that problem by placing the heatsink in a duct, for example take a look at the Vincent SP991.
An idea for the true DIY'r, build a heatsink from 1/2 inch thick alloy plates, tapped and bolted together, with thermal grease at the joints. A heatsink shaped to favor, with excellent cooling ability, and a base plate of more favorable thickness for cooling high bias demands than regular alloy heatsinks.
The really cheeky use copper plates.
Re: Re: Re: caution
As you want to drill and tap holes in order to bolt the plates together think of 2/5 to 1/2 inch plates (~ 10-12 mill).
The lowest i use is M4 imbus,10 mm leaves some 3 mm on either side, alloy is so soft( unless you are using aircraft alloy or DurAl) that going lower than 3 is tricky.
As you bolt the flanges on the base and outer plate it is 10-12 all around.
I worked in an Alloy factory(way back when), and welded Argon Arc (or Tig,Mig).
If you have welding skills you could use thinner inner flanges and weld them to the thicker outer plates.
In my poor student days i obtained huge heatsinks for less than a buck of ethyleen or argon gas, the alloy came free of charge.
Downside is that the thick plates deform, the inner plate needs to be flathened, with no access to a machine shop that is difficult.
I flattened the heads of car engines with a mirror glass and diamant paste, but you need lots of patience for that.
For the surface mounting plate, like with extruded alloy heatsinks, the best have a thick base plate.
1/2 '' minimum, but if you can get hold of thicker go for it.
There is an article on the internet that describes a technique for welding output transistors on a sheet of copper and glueing the copper on heatsinks with thermal epoxy.
According to the writer this has the lowest case to heatsink resistance of all, the article is a couple of years old.
Enjoy, jacco.
wuffer said:
As you want to drill and tap holes in order to bolt the plates together think of 2/5 to 1/2 inch plates (~ 10-12 mill).
The lowest i use is M4 imbus,10 mm leaves some 3 mm on either side, alloy is so soft( unless you are using aircraft alloy or DurAl) that going lower than 3 is tricky.
As you bolt the flanges on the base and outer plate it is 10-12 all around.
I worked in an Alloy factory(way back when), and welded Argon Arc (or Tig,Mig).
If you have welding skills you could use thinner inner flanges and weld them to the thicker outer plates.
In my poor student days i obtained huge heatsinks for less than a buck of ethyleen or argon gas, the alloy came free of charge.
Downside is that the thick plates deform, the inner plate needs to be flathened, with no access to a machine shop that is difficult.
I flattened the heads of car engines with a mirror glass and diamant paste, but you need lots of patience for that.
For the surface mounting plate, like with extruded alloy heatsinks, the best have a thick base plate.
1/2 '' minimum, but if you can get hold of thicker go for it.
There is an article on the internet that describes a technique for welding output transistors on a sheet of copper and glueing the copper on heatsinks with thermal epoxy.
According to the writer this has the lowest case to heatsink resistance of all, the article is a couple of years old.
Enjoy, jacco.
btw, if you are thinking of an alloy heatsink conversion:
Place a sheet of rubber between the duct and the flange ends, that dampens microphonics of the heatsink flange.
(for those who are still into that stuff)
Some even attach a thick copper plate to the base plate of a heatsink for better thermal disctribution.
On some threads here you can see pictures of heatsinks with an extra alloy plate for mounting and/or heat distribution.
I remember infrared heat pictures from the Threshold SA/4, done at the end of the 80's by a German magazine.
Besides thermal output from the powersupply a distinct spread in heat was visible at the heatsinks.
Lovely picture, lovely amplifier, an amplifier designed by god.
Place a sheet of rubber between the duct and the flange ends, that dampens microphonics of the heatsink flange.
(for those who are still into that stuff)
Some even attach a thick copper plate to the base plate of a heatsink for better thermal disctribution.
On some threads here you can see pictures of heatsinks with an extra alloy plate for mounting and/or heat distribution.
I remember infrared heat pictures from the Threshold SA/4, done at the end of the 80's by a German magazine.
Besides thermal output from the powersupply a distinct spread in heat was visible at the heatsinks.
Lovely picture, lovely amplifier, an amplifier designed by god.
wuffer said:
Generally for class A amps: Yes(but thats my opinion)
Mr Pass designs for heatsink temperatures of 50-55 C, in his Aleph series as with the Threshold amps.
50 to 55 C is the thermal froniter of human skin, where feeling hot switches to feeling burnt.
That is why thermostatic watertaps limit at 50 C.
Jeff Rowland did the same with his class A amps, as did Accuphase ans so on. I do not know what krell did/does, i never liked those amps.
Other designers go beyond that mark, Le Classe A by Jean Hiraga was filthy hot, Musical Fidelity's were known as egg friers.
It depends on whether you would like the heatsinks to be the base of your amplifier chassis, and on your preference on the amps geometry.
And on your budget, there are several heatsinks that have the same coolingfactor(k/w) but one may be 3 times more expensive.
For example, a SK53 is still hard to beat moneywise, but its hard to build a decent looking amplifier with the heatsinks as part of the chassis.
A couple pro amp builders used them within the casing, take the first series Electrocompaniet AW250 by Per Abrahamsen.
I added a little sheet below for you to download, with it you can calculate heatsinks requirement fast, it gives you an option for different Fisher sinks.
The input k/w data for the transitors and mica isolators you can read from spec sheets.
If you desire to build one of Mr Pass's designs just fill in the heat an output transistor dissipates and it will give you the required heatsink factor for 1 transistor.
Divide that number for the number of transistors on 1 sink and you have the required K/W number for the sink you need.
btw:
Put 2 of your heatsinks on top of eachother and you can build two sides of an amplifier with a height of 20 centimeters, ample room for the powersupply parts.
Look for 4 more SK102's and you have enough for two channels each exhaling 200 Watts of heat.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.