• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Setting up water cooling, now. Help.

J

Jakalwarrior

2[H]4U
Joined
Aug 6, 2005
Messages
3,189
I have mine set up with hose going from rad > cpu > res > ??

I have the MCP655. I remeber reading where you want gravity to take the water from the res down to the pump. ( my res is up by the bays and pump at bottom of case. But now on the pump, discharge is the up one and intake in the one on the side. So, should I turn the pump sideways?

Or does the whole gravity thing, doesnt matter?

Also, is it better to have the water going from the...

Rad > cpu > res > pump > back to rad

or

Rad > pump > res > cpu > back to rad?
 
It won't matter... Leave the pump sitting up up normally... It will pull the water from the res thru the side barb.

I personally do Rad > cpu > res > pump > back to rad... That way the water is cooler (doesnt have heat from the pump) when it hits the CPU block.
 
If I leave the pump sitting up normally. Then the pump would be pushing water up to the res, since discharge is the end sticking up.
 
No no no... look at some of the pictures people have of their machines on these forums...

Attach the reservoir to the SIDE barb of the pump... Attach the TOP barb of the pump to the Radiator... Then attach the other barb of the radiator to the CPU block... then the other barb of the CPU block back to the reservoir to complete your loop
 
Anything but a res or t-line before the pump will make filling the system a PITA and possibly reduce flow because centrifugal pumps do not handle inlet restriction well.
 
Thats how I want the flow. But to have mine like that. I would need to have the pump on the side.

Would it hurt/effect the pump to have it on it side??. It will me mounted down, good.
 
Ya the res. are ok but there is a much better way . I pressuize my system . just like they do in nascar.

its rather easy use a small canister 1/2 gallon(make sure your coolant leaves an air pocket in the top of the canister. I use 1 gallon hook an air supply line to the canister set airpressure to the canister at 25psi . use 2 DD fill ports . from the the canister install a line to the first fill port . the second fill port is used to bleed of all the air.
After all the air is bleed out close the bleed port . Reduce the pressure to 5 psi and charge system close the petcock and remove your fill line and close the fill port. Now you have a pressurized water cooling system.
 
$BangforThe$ said:
Ya the res. are ok but there is a much better way . I pressuize my system . just like they do in nascar.

its rather easy use a small canister 1/2 gallon(make sure your coolant leaves an air pocket in the top of the canister. I use 1 gallon hook an air supply line to the canister set airpressure to the canister at 25psi . use 2 DD fill ports . from the the canister install a line to the first fill port . the second fill port is used to bleed of all the air.
After all the air is bleed out close the bleed port . Reduce the pressure to 5 psi and charge system close the petcock and remove your fill line and close the fill port. Now you have a pressurized water cooling system.
Click to expand...
I hope you're kidding. Seriously. You're joking, right?

Pressurized systems do not increase performance until the temperature becomes so great that the extra pressure prevents the water from becoming steam.

But you really were joking, right?

Using a pressurized container of water would be a good way to instantly fill your loop without even powering up the pump, though.
 
if its the same as a d-5 the instructions say that you can mount it any way, as long as the inlet isn't pointing down
 
thewhiteguy said:
I hope you're kidding. Seriously. You're joking, right?

Pressurized systems do not increase performance until the temperature becomes so great that the extra pressure prevents the water from becoming steam.

But you really were joking, right?

Using a pressurized container of water would be a good way to instantly fill your loop without even powering up the pump, though.
Click to expand...

Ya what you said why nascars run pressurized system is fact so they can run 250f without steam. Yes filling under pressure will allow no pump starts it also is better at removing air in the circuit thats why I said 25 psi to fill with. Than reduce the pressure to 5 psi .
leave the system alone for 24 hours if you see liquid the next day you have a leak. If you use a pressure guage like I do. If the system doesn't have 5psi the next day you have a leak . If at any time you shut your pc off and the system isn't at 5psi on the guage check for leaks. Its just a safty I came up with because those components are $$$$$$.
 
OK I understand. Do you see any pressure increase when the system is at max load? I'm just curious whether or not water really expands very much at the low temperatures it will get to in a computer.
 
Yes pressure most definitly goes up when you have a closed loop system I like to see about 20-25 psi at the manifold. and it also serves as a flow check if you have pressure you have flow . I am some what of an expert in hydralics and its amazing how much its help me with setting up water cooling on pc's . I can't believe the number of false statements i have read in forums about paralle vs. serious circuits. And paths of least resistance . Their are mechanical laws that apply to flow and pressure and it would serve people well to learn these basic principals befor saying serial is better than parallel cooling . First thing is risistance to flow . this has many variables . id of tubing restictive water blocks length of flow path ect. ect. Resistance to flow is over come by head pressure. Thats one reason i use 2 pumps the other is its a safty net . If I lose a pump it will show up on my pressure gage. So in a PC a parallel set up well be taking cool water to each component . In a series sytstem every water block that the water flows threw carries a litle warmer water to the next cooling block. not real efficient is it. since I cool all the components in a system that could mean i might be heating the last component in a series
 
Bang how about some pics I am interested. I have a Mightyvac brake Bleeding pressure pump thing for testing automotive systems by pressurizing them. I also have an air compressor would either of these work?

I have a hard time visualizing this stuff and I am abotu to build my first system, thanks.
 
$BangforThe$ said:
Yes pressure most definitly goes up when you have a closed loop sysrem I like to see about 20-25 psi at the manifold. and it also serves as a flow check if you have pressure you have flow
Click to expand...
Of course there will be backpressure from the block, but I was wondering about static pressure, say if you turned off the pump right after running the computer at full load for a while, is the water pressure measurably greater than it is when the loop has sat overnight?
 
Justintoxicated said:
Bang how about some pics I am interested. I have a Mightyvac brake Bleeding pressure pump thing for testing automotive systems by pressurizing them. I also have an air compressor would either of these work?

I have a hard time visualizing this stuff and I am abotu to build my first system, thanks.
Click to expand...

The brake bleeding system you have is perfect . The problem I do see however is contaminasion from a dirty bleeder. This is in fact what I use but I bought it new just for this reason. And you need the compressor to charge the bleeder.

Sorry I have no digital camera . unless i can make my wifes work for a still shot.(I no zero about camera's) also the only system I have right now that well show the settup were talking about is in a case that I just had custom built that I would not want to show in public right now . It really going to be kinda special . Something no one has ever done with a PC . But I will do a nice diagram of a proper parrellel system and post it tomorrow or sunday
 
thewhiteguy said:
Of course there will be backpressure from the block, but I was wondering about static pressure, say if you turned off the pump right after running the computer at full load for a while, is the water pressure measurably greater than it is when the loop has sat overnight?
Click to expand...

when I shut down Hal the only pressure i should see is the 5 psi i charged it at no more no less . if i see less I know I have a small pressure leak. When hal shuts down it will never show more pressure than what I charged it at. Buy the way I don't know if that 5 psi is harnful or not it has caused me no problems . I do that just as a safty . If I don't have 5psi showing I know I have a leak.It has zero cooling effect .
 
thewhiteguy said:
Of course there will be backpressure from the block, but I was wondering about static pressure, say if you turned off the pump right after running the computer at full load for a while, is the water pressure measurably greater than it is when the loop has sat overnight?
Click to expand...
If it did than your radeators are not doing the job as intended . as your talking about heat pressure. there is no doubt a small amount there but i have not seen it show itself on the pressure guage.
 
$BangforThe$ said:
If it did than your radeators are not doing the job as intended . as your talking about heat pressure. there is no doubt a small amount there but i have not seen it show itself on the pressure guage.
Click to expand...
Pressure from heat expansion is what I've been talking about all along, what did you think I was talking about? I have no idea what the heat expansion properties of water are and that's why I asked whether or not there was any measurable expansion, which you just answered.

You call your computer "Hal"?
 
I truely would like to show off hal. its is a large PC complety water cooled including memory. it also uses a tec water chiller setup. For this system though it is giving me fits .

As my orginal design was for 200 watts cooling/ 400 watts heating . I need to increase the cooling to 300watts / 600 watts heating. . I have ran into 2 problems here the PS to run it . and the fact I will need to add 1 more PA 160 rad to get rid of the 600 watts of heat produced.

My orginal plan was to show this thing with the release of yonah. But because of the chiller issue I have to have new water blocks made and find a PS suitable for a pc case.I may have to scrap the chiller thing and go straight water but I haven't given up yet. The air intake on this is an air cleaner like whats used on submarines . I want this to be a low maintenance pc. The water cooling set up itself is in great shape and works wonderfully.(without chiller)

Yonah was less than I hoped for so I have to wait for Conroe. I tried the a P4 670 but it created so much heat along with the CF X1800 xt that Raditors couldn't handle the 400 watts of heat . and temps went way to high on the cooling side 45c If I ran everthing @ stock speeds It would stay at about 35C I was looking for 10c I think with 300 watts cooling with conroe I can get 10c I have to bet good O/C for CPU GPU

My target for this PC are the rich and powerful people that have to have the best without making a lot of noise. This thing is a struggle I been at it more than a year and have spent over $10,000 trying to get it developed. I am close but yet so far away . I well probable buy a yonah cpu and a desk top mother board just to test the water chillers with it . The chiller is very close to being canned.

I am seriously considering this Rad as I can make it work in my case without many mods and a low speed fan should be quite. Modify the water intake side it should work.
Its small if you look at the measure ments and it will fit my case design . The problem i run into here is the air cleaner intake . I can use the rad for other purposes so if i don't use it i won't be throughing my money away. I will just have to test it and see . the air cleaner does infact take in a lot of air

http://www.speedwaymotors.com/xq/aspx/dept_id.167/display_id.3989/qx/Product.htm
 
thewhiteguy said:
Pressure from heat expansion is what I've been talking about all along, what did you think I was talking about? I have no idea what the heat expansion properties of water are and that's why I asked whether or not there was any measurable expansion, which you just answered.

You call your computer "Hal"?
Click to expand...

Ya I am a 2001 freak/ Actually the model name will be Nautilus and the custom paint will support that theme.
Working out the quirks on this thing has me talking to the computer so I started calling it hal.
 
thewhiteguy said:
OK time for pics or bullshit.
Click to expand...
Ya I bet a lot would like the pics of this . its a totaly new PC case design and until I have conroe installed it stays under wraps. those that have seen say its the coolest thing thet ever seen as far as pc's go . I don't mind you call bullshit. I except that reaction .

You will see it when everyone else see's it if I can get A site to review it. I am pretty sure fud will review it . That wouldn't be so bad but I want good hardware site to do it and really run it threw the mill. It will be an intel ati only pc . the only thing I think going against it is its size. but its coolness could overcome its size. I don't want gamers to buy these I want Madana and that type to buy. It will be priced accordingly.
 
Thanks I am interested in the Diagram, but could not figure out where I would mount the pressure gauge. I have decided to get a stacker so plenty of room to play.

The Mightyvac uses a pump to create pressure.

it sounds like the system you are describing would be a great way to test a water cooling loop when doing maintanence and installation of new parts. I will probably pressurize my loop before installing it into my PC, rather than hook it up and check for leaks!

No new concepts here (with this idea), just your basic home made leak down testor. I have used one before on my ATV to see if the leak in the head was pressurizing my coolant tank. 5 PSI is plenty to check for leaks.

I like the concept of having a bleeder built into the pC.

My Mightyvac has only been used one time and only for brake fluid so it should be good to go. Not sure why the compressor would be needed though.
 
thewhiteguy said:
Pressure from heat expansion is what I've been talking about all along, what did you think I was talking about? I have no idea what the heat expansion properties of water are and that's why I asked whether or not there was any measurable expansion, which you just answered.

You call your computer "Hal"?
Click to expand...

Heat expands cold contracts . So your thinking that as the molecues seperate pressure is built. It doesn't quite work that way but close. Ever see plastc siding improperly installed on a house that damage is not done buy pressure its actually force and yes force does = pressure but its a litle differant than that. Many plastic topped water blocks have broken . Heat does cause that crack but its not really a pressure crack its an expansion crack.A pressure gauge really won't measure expansion and contraction it is a force but its not a pressure force . its more like a molecuture force. so well just call it expansion.. in a pc water cooling environment that would not show on a water pressure guage. But expansion of 2 unlike materials in a pc environment those force's do rear its ugly head thats why you get cracked plastic tops.
 
Justintoxicated said:
Thanks I am interested in the Diagram, but could not figure out where I would mount the pressure gauge. I have decided to get a stacker so plenty of room to play.

The Mightyvac uses a pump to create pressure.

it sounds like the system you are describing would be a great way to test a water cooling loop when doing maintanence and installation of new parts. I will probably pressurize my loop before installing it into my PC, rather than hook it up and check for leaks!

No new concepts here (with this idea), just your basic home made leak down testor. I have used one before on my ATV to see if the leak in the head was pressurizing my coolant tank. 5 PSI is plenty to check for leaks.

I like the concept of having a bleeder built into the pC.

My Mightyvac has only been used one time and only for brake fluid so it should be good to go. Not sure why the compressor would be needed though.
Click to expand...

Are all yor drive bays in use? Something like this would work great its a bay braket and its only intended to show you 1 of the many things you can do

http://www.frozencpu.com/scan/se=Water Cooling/se=Innovatek Parts & Accessories/mp=menu_search.html

I have a cheaper bleeder it is pressurized with air. that were the compressor comes in . well I am off to make the diagram I well post. it tomorrow. Can you post from Word onto the thread post. I really don't want to link from my wifes PC. I will cross that bridge when I get to it .
 
$BangforThe$ said:
A pressure gauge really won't measure expansion and contraction it is a force but its not a pressure force
Click to expand...
For being some "expert" you are seeming more and more like a BSer. Maybe English isn't your native language or you are drunk. When a liquid is warmed in a fixed volume container its pressure must increase. It's simple chemistry.

A computer for Madonna .... thanks for the laugh. Send your XBOX SI HUGE lol!!111 computer to Kyle and let's see a [H] review.
 
I hope you didn't hurt yourself laughing so hard . I didn't relize that we were building boiler system here . I stated clear there would be a small pressure build but not enough to show on a pressure guage. I tried to insert the diagram but I guess I can't unload it here. can I reformat it .. Madana was an example . I have sold gaming machines to some fairly famous people. Have you . Now if you lived in MN what famous people would game online???. That 1 sell lead to many more. Thats why I want to build the big bucks box. :eek:
 
thewhiteguy said:
For being some "expert" you are seeming more and more like a BSer. Maybe English isn't your native language or you are drunk. When a liquid is warmed in a fixed volume container its pressure must increase. It's simple chemistry.

A computer for Madonna .... thanks for the laugh. Send your XBOX SI HUGE lol!!111 computer to Kyle and let's see a [H] review.
Click to expand...

I would rather spell poorly than be rude. But acadimacs such as your self do live in the clouds. I don't drink so I can't use that as an excuse I guess I just bad speller or maybe I might be playing you . Now thats enough of that from myself. But you go ahead and blast away. Enjoy yourself rival is your superior intelect.
No seriously now were the mispelled words intelligible if they are no need to comment . I see people mispell all the time . never say anything . You kinda sound like your from AT. :eek: I do have a class A boilers licenes . But I bet your a chief right or you just book smart with no real experiance.
 
Justintoxicated said:
Thanks I am interested in the Diagram, but could not figure out where I would mount the pressure gauge. I have decided to get a stacker so plenty of room to play.

The Mightyvac uses a pump to create pressure.

it sounds like the system you are describing would be a great way to test a water cooling loop when doing maintanence and installation of new parts. I will probably pressurize my loop before installing it into my PC, rather than hook it up and check for leaks!

No new concepts here (with this idea), just your basic home made leak down testor. I have used one before on my ATV to see if the leak in the head was pressurizing my coolant tank. 5 PSI is plenty to check for leaks.

I like the concept of having a bleeder built into the pC.

My Mightyvac has only been used one time and only for brake fluid so it should be good to go. Not sure why the compressor would be needed though.
Click to expand...

PM me and we can exchange e-mails adderesses and i well attach that document.
 
First of all, can we try and keep this conversation civil folks?

Now onto addressing this...
$BangforThe$ said:
Their are mechanical laws that apply to flow and pressure and it would serve people well to learn these basic principals befor saying serial is better than parallel cooling . First thing is risistance to flow . this has many variables . id of tubing restictive water blocks length of flow path ect. ect. Resistance to flow is over come by head pressure. Thats one reason i use 2 pumps the other is its a safty net . If I lose a pump it will show up on my pressure gage. So in a PC a parallel set up well be taking cool water to each component . In a series sytstem every water block that the water flows threw carries a litle warmer water to the next cooling block. not real efficient is it. since I cool all the components in a system that could mean i might be heating the last component in a series
Click to expand...

You are correct, there are mechanical laws that apply to flow, and head is the way to overcome pressure in a loop. However you seem to be missing the point of the main argument against parallel flow. In a watercooling system, much the same as in an electrical system, when your flow splits off into two different flow paths each of the paths does not recieve an equal amount of flow. The majority of the flow will take the path of least resistance, this is proportional to the combined resistance, or pressure, of each separate flow path. The problem in parallel cooling is that correctly balancing these different flows can often times be difficult. Most of the time it might not make a very large difference, but if done poorly you could see some bad results.

Also, you are correct that the water heats up between individual components...however do you know how much the water heats up? After having shown the mathematical calculations on this forum many times in the past I tire of doing it again. Anyways, if you have a relatively low flowing loop, and a high power CPU, the water will increase in temperature by no more than around 0.5C. A typical watercooling loop will never vary in coolant temperature by more than 2C. This is really a miniscule difference in temperature, which is made possible by the very high specific heat of water. Therefore, the miniscule temperature difference between components, and the possible difficulties with parallel cooling is why running a loop in serial is most often recommended. This isn't saying that parallel cooling, when done properly, is bad...in fact it can be quite good. The problem is that doing it properly can be tricky.
 
Exactly and that is why I use 2 pumps so as to have enough flow and pressure to overcome a water block such as a Storm jet block when running in a parrallel set up . If you have enough flow the storms being more restrictive well not divert enough water in to the less resrictive blocks the line themselves equal out that restriction if you have enouch pressure and flow to overcome the restrictive flow path. . this is what I like todo . 3/8 lines to the Cpu and gpu 1/4 line to the N/B 1/4 line to the harddrives. Running that many parallel lines requires the use of 2 pumps or 1 much larger pump.

I would like to measure the pressure of a storm at inlet than compare it at the outlet i bet it would be interesting. The first manifolds i used had valves at each port on the return manifolds as i didn't know if i could balance the flow but I soon discovered with 2 good pumps and using the right sized line I didn't need the flow control valves . I also tried 1/2 line to the storm . than 3/8 to the gpu and it didn't lower the cpu temp at all. This made me happy as the smaller lines in my eye's looks better.

For me electromotive force is a little easier to figure out because its easier to test the impedance with a multimeter. But your right flow is flow and the path with the least resitance is were flow well go.
 
You can find differential pressure information here : http://www.systemcooling.com/swiftech_storm-10.html

Your ideas about parallel cooling are a bit flawed. Yes having two pumps in series gives you more head pressure, and hence you dont lose as much flow due to restriction. However, it does nothing to overcome the fact that only 20% of your total flow may be going through the CPU. This is why you run into problems with parallel cooling. The storm for example is around 3 times more restrictive than a block like the mcw6000. Which reasons that it is likely much more restrictive than a number of chipset blocks, or a typical straight through ramblocks.

It seems that you paid close attention in your setup to making sure you were getting adequate flow through the hottest components. However, many people dont have the equipment to test that themselves, which is why series is usually the best option.
 
You must log in or register to reply here.
Back
Top