swiss cheese my sink?

Cooling Processors quietly

Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee

Post Reply
Bean
Posts: 204
Joined: Fri Jan 17, 2003 3:14 pm
Location: USA

swiss cheese my sink?

Post by Bean » Sat May 31, 2003 9:27 am

Why is it that anything I get my hands on, I have to modify it?

Ok back to my post. I've been wondering about modifying my heatsink, sk7, by drilling a bunch of small holes through it (except no holes in the area that touches the CPU). My reasoning is it will increase the surface area of the sink. Maybe its not a good idea because it will reduce the mass.

So whats more important increasing surface area or mass? and is it a bad idea?

regularfries
Posts: 6
Joined: Thu Apr 24, 2003 2:15 pm
Location: UK

Post by regularfries » Sat May 31, 2003 10:48 am

Funny, I was thinking about that the other day. Not sure about mass vs surface area.

I worked out that to increase the surface area the diameter of the hole would have to be less than the thickness of the material times two:
d < 2 * t

IIRC the material of the SK7 is quite thin so that would mean quite small holes.

Hope this is helpful somehow!

Bean
Posts: 204
Joined: Fri Jan 17, 2003 3:14 pm
Location: USA

Post by Bean » Sat May 31, 2003 11:57 am

I'd be drilling 3/32" holes thru the 1/4" thick base . well under your size constraints per your formula and not very tiny (a 3/32" drill bit doesnt flex too much). drilling thru copper shouldn't be too bad.

I would have to drill about 20 holes to get a 25% increase of surface area. Wish I knew how much temp decrease would occur with increase in surface area then I'd know if it was worth the effort.

I've got a mile long to do list and I'm thinking about this - and on a saturday no less. I'm sick for sure :(

Bean
Posts: 204
Joined: Fri Jan 17, 2003 3:14 pm
Location: USA

Post by Bean » Sun Jun 01, 2003 9:06 am

[quote="Bean"]I would have to drill about 20 holes to get a 25% increase of surface area. quote]
---------------------------------------------------------------------
Ooops. thats not right - I was only considering the base and not the surface area of all the fins which makes the surface area of the holes negligible. The only merit to this idea is ........is.....??

I was just thinking the extra air flowing in the holes of the base was beneficial somehow and mistakingly tried to justify the idea by the increase in surface area.

I guess the base acts as a sink and just like holes in a sink dont hold water, holes in a heatsink don't hold heat.

DryFire
Posts: 1076
Joined: Sun May 25, 2003 8:29 am
Location: USA

Post by DryFire » Sun Jun 01, 2003 9:25 am

I think the holes wold only hamper airflow. doing more harm then help. i could be wrong

scalar
Posts: 90
Joined: Sat May 17, 2003 12:54 am

So you want MORE noise?

Post by scalar » Sun Jun 01, 2003 11:17 am

If silence is your goal, drilling holes is sure not going to help. The holes will create turbulence in the airflow that will both create noise and slow down airflow. This will make the fan work harder to move the air and so it too will make more noise.


Additionally, just using a drill bit only to do this will leave sharp edges on the exit side of the holes. These sharp edges will catch the air and disturb the airflow, making still more noise.

To do it right, you'd have to use a deburring tool used by machinists to remove hole burrs and chamfer the edges around EACH drilled hole:

A deburring tool is shown at the end of this webpage:
http://www.parktool.com/repair_help/how ... face.shtml


However, since you are likely drilling through many plates, it would be very hard to stick the deburring tool deep inside the heatsink. What you really need is a metalworking tool that is known as a reamer. It is shoved through the holes in the opposite direction you drilled them.

British site for hand reamers:
http://www.chronos.ltd.uk/acatalog/Chro ... c_115.html

American site for adjustable hand reamers:
http://www.houseoftools.com/product.htm?pid=170086


Here is an in-depth discussion of burrs, with pictures:

Elementary Knowledge of Metailworking
Chapter 1. Basic Machining and Tips - Removal of Burrs

-Scalar

SometimesWarrior
Patron of SPCR
Posts: 700
Joined: Thu Mar 13, 2003 2:38 pm
Location: California, US
Contact:

Post by SometimesWarrior » Sun Jun 01, 2003 11:34 am

I thought the base was supposed to be high-mass, to suck away all the heat from the CPU, and then to spread it to the fins, which do the real cooling.

Bean
Posts: 204
Joined: Fri Jan 17, 2003 3:14 pm
Location: USA

Re: So you want MORE noise?

Post by Bean » Sun Jun 01, 2003 11:54 am

scalar:

I agree, noise may be an issue but the fan would be low rpm.
disagree on slow down of airflow. If the holes relieve the back pressure caused by air flow that hits the base and has to change direction, I would think thats a good thing. It would do the opposite and increase airflow by decrease the load on the fan.

Burrs, good point. Thanks for the links. they were educational.
-------------------------------
SometimesWarrior:

You're prob right. it does seem to be the classic heatsink design although the designs for low noise heatsinks are not as mature. it may be counterproductive to induce the holes

scalar
Posts: 90
Joined: Sat May 17, 2003 12:54 am

Post by scalar » Sun Jun 01, 2003 11:56 am

High mass does you no good once it has become saturated with heat. You could use a brick to cool the CPU, but once it becomes saturated you'd have to turn the computer off several hours until the brick could slowly cool off.

What you really want is a material with high heat conductivity, where heat moves quickly through the material. Heat capacity is important, but not as important as conductivity. The hype about copper and all-copper is just that -- hype.

The actual engineering knowledge of such details is so vague that I doubt there are many people who really understand what makes a good, efficient heatsink design. The heatsink manufacturers know this, which is why there are so many ridiculous looking heatsinks out there. They're selling on looks and hype, not actual design efficiency. (ooh! blinky blue lights on the fan! A marketing name like "arctic"! it MUST be good!)

-Scalar

Bean
Posts: 204
Joined: Fri Jan 17, 2003 3:14 pm
Location: USA

Post by Bean » Sun Jun 01, 2003 12:37 pm

scalar said: "Heat capacity is important, but not as important as conductivity. The hype about copper and all-copper is just that -- hype."

LOL Now I know you're trying to pull my chain. but I dont want to go there.

DryFire
Posts: 1076
Joined: Sun May 25, 2003 8:29 am
Location: USA

Post by DryFire » Sun Jun 01, 2003 12:41 pm

I'll have to side with scalar on the copper hype. you won't see a massive increase in cooling ability from aluminuim to copper.

Bean
Posts: 204
Joined: Fri Jan 17, 2003 3:14 pm
Location: USA

Post by Bean » Sun Jun 01, 2003 12:49 pm

Copper vs aluminum was debated before:

http://forums.silentpcreview.com/viewto ... rface+area

Alistair
*Lifetime Patron*
Posts: 125
Joined: Sun Jan 05, 2003 12:59 am
Location: Vancouver BC Canada

Post by Alistair » Sun Jun 01, 2003 11:13 pm

its a thermalright heatsink - if holes would improve things, they would be there in the first place.

Bean
Posts: 204
Joined: Fri Jan 17, 2003 3:14 pm
Location: USA

Post by Bean » Mon Jun 02, 2003 10:14 am

IYNSHO
In your not so humble opinion :)

johnoh
Posts: 68
Joined: Wed Apr 30, 2003 1:21 pm
Location: texas

Post by johnoh » Wed Jun 04, 2003 6:32 pm

"The only merit to this idea is ........is.....?? "

The ability for slow moving air to flow through that thing. That's the beauty of the slk800 over the sk7 when each have weak airflow - a less restrictive path than the sk7 which just has a wall of metal at the end of the road for the air. Come on give it a try and post back with your temps.

Liquidated
Posts: 110
Joined: Fri Apr 11, 2003 12:53 am

wow

Post by Liquidated » Thu Jun 05, 2003 8:01 am

remind me not to annoy rusty, he destroyed poor bluefront in that thread lol

um *looks about nervously* nice desk rus!

Seems to me that if you want to increase surface area on an slk 800 type hs, driving channels into the fins would be a better way to go over drilling holes. I don't pretend to understand if this would help or hinder though.

Seems to me that thermalright and zalman could skive (or channel) indididual fins before assembling them instead of them having flat surfaces if the added area helps. Would be much easier to do that sort of work before assembling the heatsink.



Cheers!
-Liq

pingu666
Friend of SPCR
Posts: 739
Joined: Sun Aug 11, 2002 3:26 pm
Location: swindon- england :/
Contact:

Post by pingu666 » Sat Jun 07, 2003 12:48 pm

drilling might help, copper work hardens tho, its not a nice material to work with :(.

Post Reply