Watercooling without pump?

The alternative to direct air cooling

Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee

Post Reply
*Lifetime Patron*
Posts: 1822
Joined: Thu Feb 13, 2003 6:53 am
Location: Sweden

Watercooling without pump?

Post by Vicotnik » Fri Mar 12, 2010 9:38 pm

Is it possible to build a water cooling setup with no pump? A closed system with only temperature differences to get some type of flow going. Not for very hot components of course, say a 45W TDP CPU, NB and perhaps a GPU.
Main: ASRock B85M-ITX | i3-4330 | 16GB DDR3 | Intel 730 240GB | HDPLEX H1-S | picoPSU | No moving parts | Idle 13.9W
HTPC: ASRock J4105-ITX | 4GB DDR4 | Intel 535 120GB | picoPSU | No moving parts
Gaming: Intel DH77EB | i5-3570K | GTX 1060 6GB | 16GB DDR3 | TJ08-E | RM750X
Server: ASRock N3150-ITX | ~30TB | G-360 | Idle ~25W

Posts: 3395
Joined: Thu Feb 04, 2010 6:06 pm
Location: US

Post by ces » Sat Mar 13, 2010 12:21 am

I doubt that. Water is too viscous for that to work.

Posts: 28
Joined: Fri Aug 14, 2009 9:14 am
Location: UK

Post by spartan117 » Sat Mar 13, 2010 1:51 am

not withing a closed system, using an open system you could use evaporation to pull the water through the system, but within a closed system the water simply wouldn't flow, as the water would not know which way around the system to flow.

Best bet is to get a quite pump, but with such a low TDP cpu, is there any need to water cool ? you might be better off with an air cooling solution such as a fanless Ninja :)
P182, Q6600 G0 2.4Ghz, XFX 8800GTS 320MB, Zalman reserator XT, 6GB DDR2, 1Tb WD green, 250Gb Seagate, 80Gb Intel xm25m,Corsair Hx520, Windows 7 Professional, Dell 2407, LG Flatron 17"

Posts: 279
Joined: Sun Jan 05, 2003 12:09 pm
Location: U.K.

Post by Bat » Sun Mar 14, 2010 4:37 am

A few people have done it.
I forget the details, but you'll find examples if you search.
It's essential to have the heat-generating components lower than the radiator.

*Lifetime Patron*
Posts: 2269
Joined: Sun May 21, 2006 9:09 am
Location: Northern California.

Post by xan_user » Sun Mar 14, 2010 7:43 am

Posts: 36
Joined: Fri Feb 25, 2005 10:36 pm

Post by ToasterIQ2000 » Tue Mar 16, 2010 11:18 am

I remembered reading about a 'thermosiphon' at overclockers, did a search, and came up with something different than I remember. ( Looks like that site has changed a great deal since I read there regularly. )

http://www.overclockers.com/cpu-cooling ... for-water/

The idea I remebered was to use the cpu to boil ether and use that boiling as a pump. ( I started watercooling for the silence and wonder: how 'quiet' is a boiling ether pump? )

Sterling engines can also be made that operate on a temperature differential well below boiling a medium, be it either water or ether.

Posts: 20
Joined: Mon Jun 22, 2009 12:28 am
Location: England

Post by jonj678 » Tue May 18, 2010 2:54 am

There's no absolute reason against this. When water heats up, it will decrease in density, and this can be used to move it around. It would be rather difficult to estimate how well it would work, but it's probably an interesting thing to attempt.

Radiator would need to be horizontal, and physically above the water block. I'm reasonably sure you want one short, vertical tube connecting the rad to block together, and a longer loop on the other side (preferably with a 90 degree barb just above the return to the cpu block).

The reasoning is that in a symmetrical loop, there's no motivation for the water to move through the loop so it wont. Limited to conduction through 10cm of water will lead to terrible temperatures. An asymmetric loop stands a better chance of establishing a flow, though I'm not convinced the forces involved would be enough to significantly overcome friction. I'd be more inclined to set it up and test than to calculate things in this case.

Have you given it a try yet, or just thinking out loud?

Posts: 39
Joined: Mon Jun 08, 2009 7:58 am
Location: Sweden

Post by Vetotat » Tue May 18, 2010 7:50 am

As far as I know, pump-less solutions has been used to cool old car engines. The old Saab 92 (1949-56) had a passive watercooling system for its 2-stroke engine. It doesn't seem not physically impossible, but I guess cooling a computer is quite different from a car engine.

Also found this on Wiki:

Posts: 279
Joined: Tue May 05, 2009 1:15 am
Location: Sydney, Australia

Post by ascl » Tue May 18, 2010 1:29 pm

Hmm the thermosyphon idea is very cool! I was thinking about something roughly similar, but I was thinking big heat pipes.

I found this:

Posts: 199
Joined: Wed Nov 10, 2004 1:44 pm
Location: Sweden

Post by datapappan » Mon Jul 05, 2010 9:51 pm

Yes it can be done, theoretically. As per above, the driving force would be the difference in density, caused by the difference in temperature. It's not needed to have an assymetrical loop though, more important is to make sure the friction loss in the tubes is kept at a minimum, i.e. wide tubes (1" maybe?).

This has been used in buildings for central heating, so it works. Problem with a PC would be relation between power (W) and resultant water flow - you might not get enough heat dissipation.

I guess you could argue against this idea much the same as against the "going fanless" idea: even the slightest mechanical help (pump/fan) will increase the effectiveness tremendously, so if you can bear (or drown out) the pump/fan noise, there's no real need for trying this (but it would be fun, of course).

[size=75][CPU: 2.4 GHz P4 Northwood w. AC Freezer 4 (fanless-Bluefront PSU mod inspired) / MB: ABIT VT7 (Via PT880) / GPU: ATI Radeon X800XL w Zalman ZM80D-HP / HDD: Seagate Barracuda IV SATA 80GB / CD: Samsung CD-DVD combo / Seahawk ALU Case / PSU: FSP 350-THN (fan mod-Everflow connected to CPU fan header)[/size]

Monkey Puzzle
Posts: 37
Joined: Tue Apr 13, 2010 5:14 am
Location: UK

Post by Monkey Puzzle » Sat Jul 31, 2010 5:41 am

ces wrote:I doubt that. Water is too viscous for that to work.
Another gem. Thanks.
Project: Hush! - A scratch-built, passive radiator water-cooling case...[url]http://www.silentpcreview.com/forums/viewtopic.php?t=58457[/url]

Posts: 273
Joined: Tue Aug 04, 2009 7:03 am
Location: Brooklyn, NY

Post by b_rubenstein » Thu Sep 02, 2010 10:09 am

Bultaco road racer with no water pump:

http://books.google.com/books?id=7bGuud ... &q&f=false

The difference between cooling a CPU and an internal combustion engine is there is a much lower temp differential with a CPU. The flow rate won't be high, but it should flow through tubes.

Posts: 166
Joined: Tue May 27, 2008 12:04 am
Location: Finland

Post by maalitehdas » Thu Sep 02, 2010 11:23 am

It should be easy if you have a change to setup water hoses high above your system. Even better if you have a cool place up there which water must pass during it's journey. I'm using same idea with air cooling and I have been able to remove all fans from main chamber of P182.
[size=75]E8400 @ 1,8GHz, passive cooling | CPU Queen 7163 points | CPU stress test two hours, temperature 43C | fan rpm's [b]ZERO[/b][/size]

Posts: 4
Joined: Mon Jul 12, 2010 1:16 am
Location: Czech Republic

Re: Watercooling without pump?

Post by Hgi » Mon Sep 13, 2010 1:14 am

Vicotnik wrote:Is it possible to build a water cooling setup with no pump? A closed system with only temperature differences to get some type of flow going. Not for very hot components of course, say a 45W TDP CPU, NB and perhaps a GPU.
no, is not possible. Always need to pump.

Posts: 7
Joined: Wed Apr 15, 2009 12:57 pm
Location: UK

Post by tobsco » Mon Sep 13, 2010 1:05 pm

Yes it is possible, this guy did it http://www.vonslatt.com/proj-cc.shtml

As other people have mentioned you need the radiator at the top and use nice big tubing to reduce impedance.

Post Reply