Common Tips for Watercooling - Page 3

bigben2k · 06-25-2002, 03:58 PM

Quote:

Originally posted by Cova
Aye - it is getting to be rather long. But it gives me something to do at work (though I gotta go to a meeting in 10 minutes)

I'm afraid I'm still going to have to disagree with you on the definition of a closed-loop system. I couldn't find a good definition on dictionary.com, but IMHO a system has to be closed and sealed to qualify as closed-loop.

And under my definition of closed-loop (and not counting evaporation right through silicon hose - technically that is an opening in the system) - you have no evaporation, and expansion/contraction of coolant with temp is very small - likely beyond our ability to measure as the pressure inside the system will increase then and all the tubing will stretch a bit to compensate (at least mine would, tygon is stretchy). And so water-volume does not change by any measureable amount (mass of coolant is constant, volume that it occupies may fluctuate incredibly slightly, but not enough that I'm gonna worry about it).

Well, you're almost there... the water line seals the inlets, that's why it's a closed loop. It's not plumber's goop or silicone, but it does seal the loop.

As for coolant volume, I thought it changed a bit, until I read this thread here about the guy with the silicone tubing.
Silicone bad... the volume shouldn't change, like you said, not noticeably.

Quickmcj · 06-25-2002, 03:58 PM

dcrainman>>>

You cant be true

That drawing shows what I have to do.....and that is alot of hose, and the flow will deffently go down because the water is pushed up...then down and then up again......this is alot more work for the pump, and now I also have to connect the outlets from the Grapich and Chipset.....in a Y or T and this will also slow down the flow.

Quickmcj · 06-25-2002, 04:03 PM

NEVER thought watercooling would bring this kind of trouble

But i see that i NEED an airtrap as the highest point. And I can have my res. at buttom without refill, cause i can use the Airtrap for that.

Cova · 06-25-2002, 04:05 PM

Quote:

Originally posted by bigben2k

Well, you're almost there... the water line seals the inlets, that's why it's a closed loop. It's not plumber's goop or silicone, but it does seal the loop.

As for coolant volume, I thought it changed a bit, until I read this thread here about the guy with the silicone tubing.
Silicone bad... the volume shouldn't change, like you said, not noticeably.

That just sounds wrong in my mind... The contents of the loop itself can't form a seal for the loop. Obviously that is whats occuring in a system with no roof on the res - but I still don't consider it closed-loop.

Perhaps this would be a good topic to make a new poll on (the what is closed-loop topic, not the how to trap the air in a system topic that this thread is supposed to be about).

Anyways, late for my meeting now - I'll argue more later

Quickmcj · 06-25-2002, 04:07 PM

And a little question...

How do you know what hose size for the different connections should be ?

Do you place the system in the case and cut the hoses, so they fit, then take the hole thing out and then test it for leak ?

bigben2k · 06-25-2002, 04:18 PM

Quote:

Originally posted by Quickmcj
dcrainman>>>

You cant be true

That drawing shows what I have to do.....and that is alot of hose, and the flow will deffently go down because the water is pushed up...then down and then up again......this is alot more work for the pump, and now I also have to connect the outlets from the Grapich and Chipset.....in a Y or T and this will also slow down the flow.

Actually, the height has nothing to do with it. The gravity has the same effect on both sides, which cancels itself out.

The pump needs to be powerful so that it can push a lot of water through the restrictions, like the waterblock. That's it, nothing else.

I agree about the airtrap, there's still the problem of the rad pouring water into the tube, when the pump is off. The only way around it, is for that tube to extend as far up as possible, and fill it with lots of water / coolant .

As for the hose sizes, it all depends on the fittings. Just try using 1/2 inch tubing, as much as possible.

Yes, you'd test it first.

1-Do a dry fit. Install everything, except the water.
2-Do a wet fit, without the electronics. Test for leaks. You can test it in the case, or outside the case, it's up to you. It depends if it bothers you to get your case wet.
3-Assemble everything.
4-Test for leaks (computer off)
and so on...

dcrainman · 06-25-2002, 04:35 PM

Bigben>>
the rad can poor all the water it wants, it'll clean all the air out each time the pump starts again

Quickmcj · 06-25-2002, 04:43 PM

I have a proposal...

Cant you all...if you have the time, make a drawing of what you think my system should look like ?

I know some og you have....but it is just so that I can see the "holl picture".

This is the MUST things for my system...

The pump MUST be in a reservoir.

The reservoir MUST be at the bottom of the case.

The radiator CANNOT be placed anywhere else.

The CPU block has ONE inlet and TWO outlet....one for Grapichs and one for Chipset...

Some sort of refill/airtrap anywhere in the system.

That would be it.....hope you will draw some art for me.....i need it, cause im kind of confused

bigben2k · 06-25-2002, 04:57 PM

There you go.

Fill it up all the way to the top.

Quickmcj · 06-25-2002, 05:03 PM

Bigben...

I cant do it that way. I have one inlet and two outlet on the CPUblock....therefor it would look like this...

Quickmcj · 06-25-2002, 05:05 PM

But is this one better ???

jtroutma · 06-25-2002, 05:11 PM

Why not this?

bigben2k · 06-25-2002, 05:13 PM

Well, it might be better, but I don't think you want to cut more holes. How tall is the fan? 1 inch?

If you can cut a hole, then you can very easily refill the system, with the case closed. Nice.

Cova · 06-25-2002, 05:22 PM

Ok - here's my version. Just make sure that the fill point is a good cap that can be taken off to fill/drain, and than when screwed on is air/water tight. The only thing you should have to worry about is the air bubble in the res getting into a tube if the system is laying on it's side/upside down when moving it to a LAN party or something.

Most of this is based on the theory that if you put an upside-down cup in water, the air is trapped in the cup - it doesn't magically find it's way to the top of the system (in this case bubble back up above the water). So - air trapped in the tall part of the res will stay trapped so long as the res stays with that side up. Since the water will be moving the slowest through the res that is where the bubbles in the water will separate from the flow and gather at the top of the res. Having an angled res roof makes it much nicer (all the bubbles will collect in one point) but is not required so long as you keep everything submerged.

Quickmcj · 06-25-2002, 05:29 PM

Jtroutma>>>

Isnt there WAY to many Y fittings in your sugestion ?

Anotherthing that i have forgotten......SPACE!

Now that i think of it.......i cant have all this in my MEDI chieftec. The "tube" thing cannot be realised because I would have to move it futher back in the case, and it would now block/interfere the rest of the watercooling system hoses.

It would deffently be the best to do like this then.....

Quickmcj · 06-25-2002, 05:34 PM

LOL

Nice drawings

OK then...

COVA>>>

But some says tat I need the top airtrap...it also makes sense, because its the highest point. But again.....how will u make the bubbles from the rad. go all the way down to the res. in your system ? Its against the forces of nature. But i really like the res...that could fit nice inside my case

but the only thing is what about the top airtrap that I have drawed ?

jtroutma · 06-25-2002, 05:36 PM

Last pic seems to be the best......

bigben2k · 06-25-2002, 05:46 PM

Agreed. Last pic is best. The res would be full (or not, it doesn't matter).

Cova · 06-25-2002, 05:55 PM

How to get the air bubbles out of the rad?

When you originally fill the system (no matter where your air trap is) you are going to have to move things around, lay parts on the side, etc. to try and get most of the air bubbles to come out of them. Once 90% of the air is out, you mount all the stuff in your system (or somewhere else for leak testing) and turn on the pump. The pressure of the water will force all of the tiny bubbles out of the rad/blocks - and your water may look kinda foamy - thousands of really tiny air bubbles being forced along with the flow. When the water slows down in the res those tiny bubbles will be released. During the initial fill you may find it useful to turn the pump on for a couple minutes to get the bubbles moving, off for a minute to let the bubbles float to the top of wherever they are, on for a few minutes to move the now much larger bubbles (the small ones will join together when the pump is off in the nearest high-point) to the res where they will leave the flow, etc.

Even with an air trap at the top, you still have to get the bubbles out of the rad. If the bubbles are small enough to be picked up by the flow then the flow may as well carry them down to the res. If the bubbles are in a dead-spot in the flow somewhere, then they will stay in the rad and not make it to the top air-trap either.

bigben2k · 06-25-2002, 06:02 PM

Cycling the pump on and off should take care of the dead spots.

It'd probably be better to fill the system, and once the air is out, add the additives.

Quickmcj · 06-25-2002, 06:49 PM

I will do this....

I will NOT have an top airtrap. This is because of Cova's very good explanation, on how to get the air out and into the res.

I WILL use 1/2 inch hoses....silicone....its easier for me

I will TRY to seal the res. as good as I can because, if the systems leaks at a higher point all the water will come down to the res. and overflow if I dont do it.

I now have a feeling of......"This will work".......

And a very good tip you shared with me there bigben. To mix the addictive later on, when Im done with taking the air out.

OK.....you have ALL been VERY friendly....I really have no reward or something

The only thing i could do, is to tell how to hole thing performs when it is installed.

I will use Distilled water with very little addictive/glow.... any objections ?

(Sorry for the awful spelling )

Quickmcj · 06-25-2002, 06:52 PM

When I think of the time you all have spend on me trying to learn me what "Watercooling" means..........thanks.

jtroutma · 06-25-2002, 07:06 PM

Hey... it all comes down to the final product. We all want to see what you come up with and most importantly, how well it works

Good luck.

bigben2k · 06-26-2002, 09:10 AM

Quote:

Originally posted by Quickmcj
I really have no reward or something The only thing i could do, is to tell how to hole thing performs when it is installed.

The reward is a job well done. Post some pictures, if you can!

Quickmcj · 06-26-2002, 10:38 AM

I'll see if I can get hold of a DIGICAM

06-25-2002, 05:13 PM	#63
bigben2k Responsible for 2% of all the posts here. Join Date: May 2002 Location: Texas, U.S.A. Posts: 8,302	Well, it might be better, but I don't think you want to cut more holes. How tall is the fan? 1 inch? If you can cut a hole, then you can very easily refill the system, with the case closed. Nice. __________________ WBTA (Water Block Testing Alliance) Nordic Hardware WC 101

06-25-2002, 05:34 PM	#66
Quickmcj Cooling Savant Join Date: Jun 2002 Location: Denmark Posts: 109	LOL Nice drawings OK then... COVA>>> But some says tat I need the top airtrap...it also makes sense, because its the highest point. But again.....how will u make the bubbles from the rad. go all the way down to the res. in your system ? Its against the forces of nature. But i really like the res...that could fit nice inside my case but the only thing is what about the top airtrap that I have drawed ? Last edited by Quickmcj; 06-25-2002 at 05:38 PM.

06-25-2002, 05:36 PM	#67
jtroutma Cooling Savant Join Date: Aug 2001 Location: SLO, CA Posts: 837	Last pic seems to be the best...... __________________ Athlon64 X2 4200+ @ 2.5Ghz (250FSB x 10) OCZ VX 1GB 4000 @ 250FSB (6-2-2-2 timmings) DFI LANParty nForce4 Ultra-D SCSI Raid 5 x (3) Cheetah 15K HDDs LSI Express 500 (128MB cache) OCZ PowerStream 520W PSU ATI X850XT PE (Stock) DTEK WhiteWater + DTEK Custom Radiator Eheim 1250

06-25-2002, 05:46 PM	#68
bigben2k Responsible for 2% of all the posts here. Join Date: May 2002 Location: Texas, U.S.A. Posts: 8,302	Agreed. Last pic is best. The res would be full (or not, it doesn't matter). __________________ WBTA (Water Block Testing Alliance) Nordic Hardware WC 101

06-25-2002, 06:02 PM	#70
bigben2k Responsible for 2% of all the posts here. Join Date: May 2002 Location: Texas, U.S.A. Posts: 8,302	Cycling the pump on and off should take care of the dead spots. It'd probably be better to fill the system, and once the air is out, add the additives. __________________ WBTA (Water Block Testing Alliance) Nordic Hardware WC 101

06-25-2002, 03:58 PM	#52
Quickmcj Cooling Savant Join Date: Jun 2002 Location: Denmark Posts: 109	dcrainman>>> You cant be true That drawing shows what I have to do.....and that is alot of hose, and the flow will deffently go down because the water is pushed up...then down and then up again......this is alot more work for the pump, and now I also have to connect the outlets from the Grapich and Chipset.....in a Y or T and this will also slow down the flow.

06-25-2002, 04:03 PM	#53
Quickmcj Cooling Savant Join Date: Jun 2002 Location: Denmark Posts: 109	NEVER thought watercooling would bring this kind of trouble But i see that i NEED an airtrap as the highest point. And I can have my res. at buttom without refill, cause i can use the Airtrap for that.

06-25-2002, 04:07 PM	#55
Quickmcj Cooling Savant Join Date: Jun 2002 Location: Denmark Posts: 109	And a little question... How do you know what hose size for the different connections should be ? Do you place the system in the case and cut the hoses, so they fit, then take the hole thing out and then test it for leak ?

06-25-2002, 04:35 PM	#57
dcrainman Cooling Neophyte Join Date: Jun 2002 Location: ma, usa Posts: 50	Bigben>> the rad can poor all the water it wants, it'll clean all the air out each time the pump starts again

06-25-2002, 04:43 PM	#58
Quickmcj Cooling Savant Join Date: Jun 2002 Location: Denmark Posts: 109	I have a proposal... Cant you all...if you have the time, make a drawing of what you think my system should look like ? I know some og you have....but it is just so that I can see the "holl picture". This is the MUST things for my system... The pump MUST be in a reservoir. The reservoir MUST be at the bottom of the case. The radiator CANNOT be placed anywhere else. The CPU block has ONE inlet and TWO outlet....one for Grapichs and one for Chipset... Some sort of refill/airtrap anywhere in the system. That would be it.....hope you will draw some art for me.....i need it, cause im kind of confused

06-25-2002, 05:55 PM	#69
Cova Cooling Savant Join Date: May 2002 Location: Canada Posts: 247	How to get the air bubbles out of the rad? When you originally fill the system (no matter where your air trap is) you are going to have to move things around, lay parts on the side, etc. to try and get most of the air bubbles to come out of them. Once 90% of the air is out, you mount all the stuff in your system (or somewhere else for leak testing) and turn on the pump. The pressure of the water will force all of the tiny bubbles out of the rad/blocks - and your water may look kinda foamy - thousands of really tiny air bubbles being forced along with the flow. When the water slows down in the res those tiny bubbles will be released. During the initial fill you may find it useful to turn the pump on for a couple minutes to get the bubbles moving, off for a minute to let the bubbles float to the top of wherever they are, on for a few minutes to move the now much larger bubbles (the small ones will join together when the pump is off in the nearest high-point) to the res where they will leave the flow, etc. Even with an air trap at the top, you still have to get the bubbles out of the rad. If the bubbles are small enough to be picked up by the flow then the flow may as well carry them down to the res. If the bubbles are in a dead-spot in the flow somewhere, then they will stay in the rad and not make it to the top air-trap either.

06-25-2002, 06:49 PM	#71
Quickmcj Cooling Savant Join Date: Jun 2002 Location: Denmark Posts: 109	I will do this.... I will NOT have an top airtrap. This is because of Cova's very good explanation, on how to get the air out and into the res. I WILL use 1/2 inch hoses....silicone....its easier for me I will TRY to seal the res. as good as I can because, if the systems leaks at a higher point all the water will come down to the res. and overflow if I dont do it. I now have a feeling of......"This will work"....... And a very good tip you shared with me there bigben. To mix the addictive later on, when Im done with taking the air out. OK.....you have ALL been VERY friendly....I really have no reward or something The only thing i could do, is to tell how to hole thing performs when it is installed. I will use Distilled water with very little addictive/glow.... any objections ? (Sorry for the awful spelling )

06-25-2002, 06:52 PM	#72
Quickmcj Cooling Savant Join Date: Jun 2002 Location: Denmark Posts: 109	When I think of the time you all have spend on me trying to learn me what "Watercooling" means..........thanks.

06-25-2002, 07:06 PM	#73
jtroutma Cooling Savant Join Date: Aug 2001 Location: SLO, CA Posts: 837	Hey... it all comes down to the final product. We all want to see what you come up with and most importantly, how well it works Good luck. __________________ Athlon64 X2 4200+ @ 2.5Ghz (250FSB x 10) OCZ VX 1GB 4000 @ 250FSB (6-2-2-2 timmings) DFI LANParty nForce4 Ultra-D SCSI Raid 5 x (3) Cheetah 15K HDDs LSI Express 500 (128MB cache) OCZ PowerStream 520W PSU ATI X850XT PE (Stock) DTEK WhiteWater + DTEK Custom Radiator Eheim 1250

06-26-2002, 10:38 AM	#75
Quickmcj Cooling Savant Join Date: Jun 2002 Location: Denmark Posts: 109	I'll see if I can get hold of a DIGICAM

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)