Krell KSA 50 PCB

Jump to Latest
#3,985
still4given said:
I know this is probably been asked before but I wonder if we could do a GB parts order for everything but the transistors for those of us who already have our supply of those. Would that make things too difficult to sort out? Since the semis are all coming from Digikey we could just reduce that order and keep the others the same.
What do you think?

Thanks, Terry
Click to expand...


I'm on the wife’s brand new Dell PC( just pulled it out of the box and am.... Uh "testing" it) so I can't check the spread sheet.

I don't recall if output transistors were on it or not ( I didn't think so since some use TO-3, and some use TO-247).

I ASSUME you were referring to output transistors and not ALL the transistors right?
 
#3,986
rabstg said:



I'm on the wife’s brand new Dell PC( just pulled it out of the box and am.... Uh "testing" it) so I can't check the spread sheet.

I don't recall if output transistors were on it or not ( I didn't think so since some use TO-3, and some use TO-247).

I ASSUME you were referring to output transistors and not ALL the transistors right?
Click to expand...


Hi Troy,

Actually I WAS referring to ALL of the transistors, but you are right, the outputs aren't included and that means that the parts that are being bought from Digikey only amount to less than $6 per board. That's only $12 for a stereo amp and since you will have all of those transistors is should be a simple thing for you to make "Matched sets" for all of us.
:Popworm: :clown:

Sorry, I've been warned about thinking. 😀

Blessings, Terry
 
#3,987
Hi Still4given,
re passive heatsinks:-
for Iq=1.9A, I guess you would need three of your 8 by 12 sinks per channel.
If you sawed them in half then you may manage with four 8 by 6 per channel. two 8 by 6 will work 40% better than one 8 by 12.
If you increase the output stage bias then the number of sinks goes up in proportion.
The 5/16ths inch thick is quite light and means this particular sink will work better with a larger quantity of transistors spread across it's width and even better as a cut down version. e.g. 8 by 6 with three transistors at 2.5inch centres on four sinks gives a 6 pairs output stage but with long driver wires at low impedance. The sink will work slightly better with the lower edge lifted an inch or so above the supporting surface. The ribs must work vertically to give the lowest Tc temp.
Finally, no one has mentioned the biggest advantage of a multi transistor output stage. Lower junction temp. With plastic cased outputs this is fairly important for classA reliability.
 
#3,988
Hi Andrew,

That is an interesting idea. I would prefer the size of the case with shorter heatsinks as well. I'm not really into 12" tall amps. I would prefer to use the output boards for the devices and I'm not sure what layout they have. I would rather keep the wires to a minimum and I think using the boards will help with that. I didn't think about it, but if I cut the heatsinks in half they would make much smaller tunnels if I needed to use fans. Thanks for the idea, I think I'll try it. I would rather not have a fan if I can get away with it. I'm really beginning to appreciate the sound of this Krell I just finished, but that whine of the fan is rally starting to bug me. I know I am going to have to replace it and rearrange the way it works so the new one doesn't die as well.

I'm going to try and build the nest one with two PSUs and better parts and see if there is a noticeable improvement. It should be fun.

I just noticed that you said 6 pairs across 4 heatsinks. That won't work for me. I need to do 6 pairs across 2 heatsinks. I won't have 4 heatsinks if I cut the two I have in half, only 4. I suppose if I only use three pair per channel then that would work. I was still planning on trying 6 pair per channel since I have the extra output boards. Maybe I'll reconsider.

Blessings, Terry
 
#3,989
still4given said:
...since you will have all of those transistors is should be a simple thing for you to make "Matched sets" for all of us.
Blessings, Terry...
Click to expand...


Well it shouldn't be too hard to measure 250 transistors although I HAVE NEVER DONE IT BEFORE. Therefore, I will be needing advice when the parts ARRIVE.

As for a kit without transistors, well help out KMJ on setting up the order list and "design" a kit or two without transistors. I can just "forget" to add them to an order or two...

ALL-

So as it stands now:

We have sources for all the parts.

We have not priced all the parts and divided for 50 kits/channels.
We have not had a part price to add the 3% paypal, $6 domestic shipping, and $12 out of country shipping, and $1 for envelopes, markers and whatever else is required for shipment.

I am now planning on matching Xistors, color coding all resistors, and kitting / mailing the kits for the money LEFT OVER from the supplies fund which is a MAXIMUM of $50... As you can see, it is a non-profit GB being done in the spirit of DIY and that is why I am requiring assistance from other GB'ers.

Someone crunch the numbers, everyone else validate the numbers and I will start collecting money once the price is set.

I will have a 1 week turn around once I receive the parts but reserve the right for 2-3 days extra since I am also prepping the house for sale.

EDIT:

Doah! I thought for sure this would have been the first post of page 400.
 
#3,991
Terry,
here is a picture of McLean's homepage.
http://home.ca.inter.net/~lloyd.maclean/Krell/small1129.jpg

You can see the arrow on the side of the Papst fan pointing upwards, blowing air in from the bottom.
And blowing dust in the amplifier from the top of the heat duct, check out the amount of accumulated dust on the other pictures of the Krell on that site.

If you take ventilation systems in housing, the fan there is at the top, sucking air instead of blowing.
As posted before, fans do that better than blowing.
The fan is bolted to the heatsink, with such short heatsinks temperature of the fan is practically the same, whether the fan is bolted on top or on the bottom of the heatsink makes little difference.
And, with a fan on top, it does not blow dust in the chassis.

So a better choice is to have the air going out at the top, but with the fan on top, not under the heatsink.
 
#3,992
Jacco and others, the statement "hot air goes up by itself, pushing it down means lower efficiency" is correct but it is not a very big deal.

The calculated presure difference is about 0,22 Pa. (calculated with the next parameters):

heatsing 15 cm high
temperature in 21 degrees C
temperature out 60 degrees C.
rho air 60 degrees C 1,05 kg/m3
rho air 20 degrees C 1,20 kg/m3
g is 9,81 m/s*s

There is another misunderstanding that i saw many posts ago about the best colour of heatsinks (or a radiator, or .....):

The truth about it is that the color of a heatsink does NOT significant affect the heattransfer of it. Only paintings with metal in it and the thikness of the paintcoating do affect.

Do not forget that we talk about heat transmision and not about receiving of heat. The trick is the direction of the heattransfer.

So feel free to paint your heatsinks in any color you want, it does NOT affect the heattransfer significant.

Emissioncoefficient of all paints is about 0,9-0,96, what ever the color is.
 
#3,993
jacco vermeulen said:
Terry,
***
So a better choice is to have the air going out at the top, but with the fan on top, not under the heatsink.
Click to expand...

That's what I thought until I read a post that said that a fan on top would create a more laminar flow of air across the heat sink, whereas a fan underneath blowing onto the heat sink creates a more turbulent flow of air, it being known that turbulent airflow is much better at transferring the heat from the sink to the air.

So I put a small 80mm fan underneath the heat sink blowing up on the fins to create turbulence, and a big 120mm fan on the top blowing up as well. (pics at my www link below-- as usual) Then I saw a post that the uneven pressure caused by unmatched fans would cause one to work harder and wear out sooner than the other. But I have both running at less than full voltage via a fan controller so don't think it will be a problem. Moral of the story- theres a lot of info here!

I also saw those KSA - 50 pics years ago, and I never forgot about all that dust, which is one reason I have the output stage in a separate section of the amp, separated by a big shielding plate. Even if there were a massive amount of dust accumulated it would not make it into the front section of the amp where the PS and input stage are located.
 
#3,994
Quite right, Mr Green, it is the dust that should be the main issue with a forced cooling setup.
The rest is just nitpicking(i am guilty)

(i can not imagine that the flow will be other than laminar with the average windspeed of a fan.
Average windspeed of a 4.75" diameter fan that does 30 cbft/min is like 4 feet per second, 2.75 mph.
For turbulent flow an idea would be to roughen the inner surface of the heatsinks, like sanding it)

-->> I am still waiting for Terry's final result of the P101 versus KSA50 match
 
You must log in or register to reply here.
Top Bottom