• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Hybrid Cooling Idea

Rudy Toody

Rudy Toody

Diamond Member
I have this cooling setup that has eight heat pipes poking their condensing ends out of the top.

If I took some water-cooling tubing and drilled 8 holes so I could make a U-shape (4 holes to each side) and press the tubing onto the top, would the subsequent extra cooling at the heat pipe ends make any noticeable difference?
 
Originally posted by: Rudy Toody
could make a U-shape (4 holes to each side) and press the tubing onto the top
Click to expand...

Could you repeat that ...in English.

Perhaps a pickey..............

 
Ok first of all here is what Heatpipes consist of.
http://forum.xcpus.com/air/269...-about-heat-pipes.html

If you open the heatpipes and run a water loop threw it it stops being a heatpipe and you turn the cooler into a very poor water block. If you wanted to combine liquid and air better then get a nice all copper waterblock on the cpu and mount a nice air cooler on top. Not that it would make much, if any, difference.
 
I'm not puncturing the heatpipes. I'm puncturing the plastic tubing and pressing it over the heatpipe ends.

If you visualize yourself flowing with the water through the tubing, first you will pass 4 heatpipe ends, then around the bend, and finally 4 more heatpipe ends.

The heatpipe ends will look like tiny mountain peaks poking through the side of the tubing.

Edit: the intent is to create a larger delta between hot and cold ends of the heatpipes.
 
OK i see. The surface area exsposed to the water would not be enough to make any impact on the level of cooling.

Here is an idea for you. Cut out a space in the center of the air cooler where you cn solder in a U bend of copper pipeto all the fins. Then run your water loop threw there. This will eliminate the need for a fan on the CPU cooler. But in the end a water block added to the loop is much more effective. But its a cool project just for the sake of doing it and showing it off.
 
Generally, a good engineering principle for non-engineers to follow: Avoid adding additional complexity to anything.

If you add a heatpipe cooler to the top of a water-block (assuming it is flat), the additional mass may provide an addition of cooling that is almost unmeasurable (I would think).

I've advocated the use of simple duct-panels for motherboard components to cool them -- where such items would be missed without their own water-loop. That's not much in the way of complexity, though . . .

The more productive path may be "chilled water" and the addition of TEC or other approaches to provide the chill. But chill -- still -- more complexity.
 
Originally posted by: BonzaiDuck
Generally, a good engineering principle for non-engineers to follow: Avoid adding additional complexity to anything.

If you add a heatpipe cooler to the top of a water-block (assuming it is flat), the additional mass may provide an addition of cooling that is almost unmeasurable (I would think).

I've advocated the use of simple duct-panels for motherboard components to cool them -- where such items would be missed without their own water-loop. That's not much in the way of complexity, though . . .

The more productive path may be "chilled water" and the addition of TEC or other approaches to provide the chill. But chill -- still -- more complexity.
Click to expand...

Here is another idea!

Most water blocks I've seen have a copper heatsink and plastic/Lucite attached to hold the fittings. Why not eliminate the copper? You could mount the Lucite directly to the CPU chip with a simple gasket to contain the water. It would be cheap and the only down-side is removal -- it will be sloppy.
 
Originally posted by: Rudy Toody

Here is another idea!

Most water blocks I've seen have a copper heatsink and plastic/Lucite attached to hold the fittings. Why not eliminate the copper? You could mount the Lucite directly to the CPU chip with a simple gasket to contain the water. It would be cheap and the only down-side is removal -- it will be sloppy.
Click to expand...

That's often been done....with dissapointing results. It's referred to as "Direct Die Cooling ((An example). Just having water flow over the CPU's case is less effective than the complex methods used inside modern water blocks.(ie channels/jets/inserts/etc.).
 
you do know how heat pipes work right?

Watercooling the tips of pipes wont help anything.

And watercooling the IHS directly wont help.

Also incase you think it, putting a mini reservoir ontop of your cpu, and having a mini sink with fins inside the res, that wont work great either.

Water is at its limit like air is. The last possible gains one will get will be at the radiator or coolant, or high pressure waterblock.

Can i ask if you ever done watercooling?
If you had a rig setup, you'll understand what watercooling exactly is, unless you bought a cheap kit.
 
I had two watercooled FX-60s for a while. Now I have three aircooled Phenom 9850BEs.

I was just trying to think outside the box. I thought that cooling the condensing end of a heatpipe would increase the temperature difference between the ends and possibly make it more efficient.

--Fred
 
For extra cooling, build a wind tunnel that surrounds the CPU cooler and has intake and exhaust from front to back on case as I have.....the tunnel isolates any heat or air from rest of case and also provides cool air from outside the case directly to the cooler....reckon its worth 5-10c depending on room temps
 
Originally posted by: SolMiester
For extra cooling, build a wind tunnel that surrounds the CPU cooler and has intake and exhaust from front to back on case as I have.....the tunnel isolates any heat or air from rest of case and also provides cool air from outside the case directly to the cooler....reckon its worth 5-10c depending on room temps
Click to expand...

:thumbsup: :thumbsup: :thumbsup: :thumbsup:

[Two up on Siskel and Ebert: my big toes.]

I just replaced a 680i board with a 780i. The chipset on the 680i seems to be the culprit, and I have myself to blame for trying to get "ballpark" OC settings before better attention to cooling -- in my case, with ducting in an extension of Solmeister's suggestion.

Now I find that the superb fit I had with a TR "blue-rubber-duct-ie," CM 830 and Noctua NH-U12P cooler is gone: the Noctua sits closer to -- and off-center of -- the rear exhaust fan. I'll need to cut a duct-box like a parallelogram that mates to the rear fan.

All of this had two purposes. Reduce the number of fans, noise and wattage while increasing cooling effectiveness on specific components.

Thinking again of AigoMorla's constant advice to me that I should jump on a water-cooled setup, I started looking again at high-end kits. I trust Aigo's reports about lower temperatures, but I follow the ratings of thermal-resistance -- C/W. With these 45nm CPUs, it hardly seems worth it. The Noctua seems close or equal to $300 WC-kits in thermal resistance around 0.09 C/W. But with the clunky 780i chipset-cooler and ding-bat flimsy "Magic" fan that goes with it, water seems slightly more worth it for chipset cooling. Go figure. And I don't want that fan on there: I want to pull air through the chipset HS fins with a duct-box for immediate exhaust from the case with a larger fan.

All in all, I have a lot of work to do that would be partly eliminated or displaced with a water-cooled setup. That means pretending like I'm the "CSI" TV-series' "miniature serial killer" -- designing little boxes that fit together and channel air. I could take another month to complete this nonsense the way I want it done.

To recapitulate -- with the most efficient heatpipe coolers available today, the gains from ducting the cooler to immediately exhaust the air do not so directly affect CPU temperature as much as it keeps case-interior air-temperature down, keeping all component temperatures lower.

On the matter of "direct-IHS-water-cooling" -- AigoMorla is the authority on this, and I defer to him. I'll add this: in the interface between IHS and heatsink or waterblock base, thermal resistance is going to increase even if you replace the copper with solid synthetic diamond. That is, an impossible-to-obtain solid-diamond heatsink base may lower overall thermal resistance, but the thermal paste -- even with micronized diamond, will raise it slightly. Ask me about the solid silver-dollar I destroyed (adding two interfaces to replace one) -- hoping to capture the better thermal properties and mass of silver. More complexity to no effect, a total waste of time, and about $40 of a collector's-edition coin.
 
Originally posted by: Rudy Toody
Just my luck! I re-invent the wheel and it has a flat!
Click to expand...

actually your trying to reinvent the spoke.

Your wheel aint turning yet. :X
 
Originally posted by: aigomorla
Originally posted by: Rudy Toody
Just my luck! I re-invent the wheel and it has a flat!
Click to expand...

actually your trying to reinvent the spoke.

Your wheel aint turning yet. :X
Click to expand...

Sure . . . . and these forums are like a World Exposition of Rube-Goldberg Extremists.

He needs to do some research and go through the hoops of frustration. Eventually, he may actually come up with something we all find useful -- if the maelstrom and vortex sucking money from his wallet doesn't take him along with it, like Maximillian Schell in "Black-Hole" . . . . . [heh-heh]
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top