Forgive me for not suppling pictures to help illsutrate my question. I don't have any imaging programs on this box, so I'll try my best to make sence. I want to know if there is any forseeable problem with mounting my radiator to the top of my case? Meaning, cutting a 120mm blowhole in the top and mounting the radiator fan-facing-upwards blowing out of the blowhole. I've seen quite a few people either mount it in the front of the case or on the back wall of the case. Is there a right and wrong way of doing this? Or is it just preference and availible room? Thanks in advance.
![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
-
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.
Radiator Placement?
- Thread starter [em]HEAT
- Start date
In page 4 of the watercooling gallery sticky you can see the variety of radiator placements for different computer cases. I think it is mostly based on how to work around your case dimensions and existing components(like if your 120mm rad will fit between the psu and drives). In the Lian Li v1000 case, in page 4, the motherboard is "higher" on the case and you can see the radiator is overlapping some pci slots (still an awsome setup). Also consider that "blowing out" air thru the top means your case's hot air will be blowing out thru your radiator also. Maybe "sucking" outside air and blowing it out somewhere else in the case would be better?
there is no problem with mounting a rad at the top except for the fill and bleed process... since the rad will be at the top part of the loop, the air will gather there. to correct this problem you're going to have to turn on your system and tilt your case so that the res/t line is the top of the system and let it run for a little like that.
and I agree with lunar, your fans should be sucking the cooler outside air to blow over your rad, not the inside hotter air.
and I agree with lunar, your fans should be sucking the cooler outside air to blow over your rad, not the inside hotter air.
You can always mount the rad/fans on top of the case on the outside so you have cool air going through the rad and you are not ehausting warm air into the case. Its the best of both worlds. And it only requires 2 small holes in the top of the case for the tubes.
killernoodle
2[H]4U
- Joined
- Aug 4, 2002
- Messages
- 2,634
If you are watercooling, you will not have the issue of having a hot case.
You are removing from the case the main heating elements, the CPU, GPU and NB, so they do not heat up the computer. The hard drives produce very little heat in comparison, and the power supply normally exhausts all of its own hot air outside anways. The radiator is then the largest hotspot in the case and it's placement doesnt matter.
People always make that argument but dont think about it first.
You are removing from the case the main heating elements, the CPU, GPU and NB, so they do not heat up the computer. The hard drives produce very little heat in comparison, and the power supply normally exhausts all of its own hot air outside anways. The radiator is then the largest hotspot in the case and it's placement doesnt matter.
People always make that argument but dont think about it first.
except the PSU dumps a lot of heat into the case right next to the top rad
the fill and bleed process basically removes all air from your loop. this allows better flow and temp exchanges from the hot parts, cpu/gpu, to the colder water. also it puts less strain on your pump. basically you fill you system and let it run trying to get the air out. the air will gather at the top part of the loop, which in this case will be your rad. are you using a t line or a res? you basically have to tilt your computer until the tline/res is the top most part so the air can gather in there. then you keep adding water until you displace most of the air out.
the fill and bleed process basically removes all air from your loop. this allows better flow and temp exchanges from the hot parts, cpu/gpu, to the colder water. also it puts less strain on your pump. basically you fill you system and let it run trying to get the air out. the air will gather at the top part of the loop, which in this case will be your rad. are you using a t line or a res? you basically have to tilt your computer until the tline/res is the top most part so the air can gather in there. then you keep adding water until you displace most of the air out.
Ok, thanks for all the help on the placement of the rad. Now I have an even bigger problem. In the illustrations from the DangerDen instructions PDF, they have a pump with 2 hose outlets, assuming one is for intake and the other out take. The pump that came with my system has only 1 outlet on the pump for the tubing to be attached. How am I suspose to hook this up?
Also, if I can use this pump, I'm assuming that the tubing is suspose to go through the collar of the pumps tubing attachment then you press the tubing on the nipple(for the lack of the correct word,sorry) then screw down the collar right? If so then why would I be given a pump that will not and cannot house the tubing? The tubing is entirely to big to fit through the collar first the be seated on the pump housing.
Any help with this will be greatly appreciated. Thanks. I've already contacted DangerDen asking the exact same questions, still waiting for a reply.
Also, if I can use this pump, I'm assuming that the tubing is suspose to go through the collar of the pumps tubing attachment then you press the tubing on the nipple(for the lack of the correct word,sorry) then screw down the collar right? If so then why would I be given a pump that will not and cannot house the tubing? The tubing is entirely to big to fit through the collar first the be seated on the pump housing.
Any help with this will be greatly appreciated. Thanks. I've already contacted DangerDen asking the exact same questions, still waiting for a reply.
i would suggest first going over to www.ocforums.com and going into the watercooling section and then the stickies... it is pretty apparent you have no idea what you're doing and you just bought a lot of stuff from DD. read the stickes which should help you out a lot.
see where the blue thing is on the side.. that's the other outlet.
see where the blue thing is on the side.. that's the other outlet.
killernoodle said:
Thinking about it will prove it is still the best possible solution. Maybe you will notice that the Mosfets, northbridge, ram all start putting out a lot of heat once they start getting serious Volts going through them and dont need the extra heat from a rad exhausting onto them. There is no reason to NOT put the rad outside on top of the case. Is it harder to mount than inside? NO. Is it noisier? NO, does it take up a lot of room out there? NO.
That is all.
killernoodle
2[H]4U
- Joined
- Aug 4, 2002
- Messages
- 2,634
MikeP said:
How is this any different than what I'm saying? I'm talking about having the radiator on the top exhausting air out of the case.
And no, the items you mentioned do not put out nearly the heat the CPU, GPU, and powersupply do. Also, whoever told you that your overclocking potential relies on how cool your ram and fets are must have been joking. If the case temps are 2C warmer because the radiator is exhausting into the case, it is not going to do anything to your OC. Just as the water temp is only slightly higher between the barbs on the waterblock, the temps between the intake and exhaust sides of the radiator vary only very slightly, especially as the velocity of the air going through the radiator gets higher. The heat inside the case, therefore, is removed from the case pretty quickly in any reasonable setup.
I happen to have a double heatercore mounted on the top inside my case exhausting air out the top of the computer. All I can say is that when running at 100%, the air coming out of the raidator is no warmer than the room air temp and I am getting a relatively low delta T between CPU temp and ambient temp.
bellevegasj
Limp Gawd
- Joined
- Aug 10, 2004
- Messages
- 237
I didn't read the whole thread but I would suggest setting it up to PULL air through the top of the case and radiator.
Case
Radiator
Fan
Case
Radiator
Fan
Giblet Plus!
Limp Gawd
- Joined
- Feb 18, 2003
- Messages
- 345
MikeP said:
is it uglier? yes. is it easier to damage? yes. are the fans on it going to be louder? yes.
with creative methods inside mounting works great.
killernoodle said:
You must have missed my posts on before and after Voltage fluctuation results when using passive Mosfet coolers. Above 1.9V especially it becomes very important. But then what can wide voltage fluctuations possibly have to do with stability and OCing ability...
Giblet Plus! said:
Uglier? Ive seen some beautifully made Rad enclosures. It all has to do with how you build it. No more ugly than 95% of the cases out there.
Easier to damage? You can stand on top of my rad enclosure which is more than I can say about the case itself. Or are you talking about not attaching it to the case and it falling off?
Louder fans? If they are on the inside of the enclosure they will be just as loud ( or quiet ) as a fan mounted inside the top of the case exhausting out.
Giblet Plus!
Limp Gawd
- Joined
- Feb 18, 2003
- Messages
- 345
most people who mount them on top just use extra long bolts to extend them out
an external box or enclosure is a different animal
an external box or enclosure is a different animal
killernoodle
2[H]4U
- Joined
- Aug 4, 2002
- Messages
- 2,634
MikeP said:
Has anyone here proven that putting heatsinks on mosfets improves overclockability? No.
Experience says that people do this for bling.
Thanks alot guys for all the very useful help. Now that I have everything measured and hooked up I'm waiting to goto work to get some real metal hoseclamps before I fill it and test it. Those plastic claps that came with the kit just don't seem like thy'd be very secure. After getting the clamps I'll run my leak test and keep you all posted. Thanks again.
Gongo said:
Thanks for the link. I bought this kit from DD, but the one advertised was a tad bit different, that's why I was confused. And yes, I have no idea what I'm doing , but is'nt this how you learn? By asking questions and making posts to kool smart people like you all here? Now that I've done quite a bit of reading and asking I believe I have the fundamental knowledge to set this kit up with out any damage to my box. I really apreciate all the help people.
killernoodle said:
The heat is dissipating into the air passing thru the radiator. The greater the difference between the temperature of the air and the hot water going thru it, the better cooling capacity you will have. If you add an air conditioner into the mix...
About you reading temps: Your instruments are probably not precise enough.
If we go to the molecular level, you would see that molecules trapped in a closed box (lower volume) collide more that other in a large container (like your complete room).
The temperature inside your case is greater than the one outside, unless its completely open or has a great intake and blow out hole.
killernoodle said:
Time you start thinking about it.
not to mention that WCing doesnt take 100% of the heat.. which the rest spills into the trapped case.. not to mention the PSU generates a lot of heat, and the pump/CD drive/HDD all generate heat that is trapped inside the case. not to mention the heat actually radiates fromt the tubing... but keep thinking that it still doesnt matter what direction your fan blows....
killernoodle
2[H]4U
- Joined
- Aug 4, 2002
- Messages
- 2,634
Lets think about this some more:
Consider an average watercooled computer: the GPU and the CPU are cooled. The radiator is not yet factored into the heat load. The pump resides at the bottom of the case with the hard drives. The motherboard layout is standard and has standard components on board.
When the CPU (~100w) and the GPU (~80w, these values are theoretical) are watercooled, almost no heat is added to the ambient case temp due to these items. What is left over? The NB (if it is not watercooled), the SB, the RAM, the FETS, the HDDs, and the PSU. The PSU vents its own heat ouside, so this is not a factor. The SB and ram may seem to be hot, but they dissipate very little of this heat to the air because they have very little surface area and are not actively cooled. They only seem to generate a ton of heat because they cannot dissipate this heat without a heatsink. If you sink these items and cool them actively with a fan, you will notice a dramatic difference in temperatures. Still, these items produce very little heat and definitely do not dissipate this heat into the case. Then we have the NB. This item is sinked because it produces more heat than most of the components on the board. Lets just say that the NB is now the leading heat source in the case at 30w (im being generous here). The FETS seem to be very hot because they also have no surface area or active cooling, so they seem to produce a ton of heat. This is not true, just as with the SB and RAM. Hard drives are similar, they produce little heat but this heat cannot be effectively removed so they seem to be hot. Notice, these items are hot to the touch, but the air around them is no warmer than any other area in the case.
So we have a radiator now that has to dissipate the pump's heat (most of the heat from the pump goes to the water because most mag drive pumps are essentially water cooling themselves) as well as the CPU and GPU, and because this area now has to dissipate say 200W, that is 200W that is now not being used to heat the case. By putting the rad on top exhausting air out, you are taking advantage of convection, you are not adding to the case temp (actually reducing it), it is simple and it is elegant.
I remember seeing thermal pictures of a computer that was being watercooled, and the CPU was colder than the rest of the motherboard by a large margin. Essentially, the case temp is much less than air cooling, and because the major heat producers are moved to a different location the air going to a rad at the top of the case to the radiator will be far cooler than you might think.
Consider an average watercooled computer: the GPU and the CPU are cooled. The radiator is not yet factored into the heat load. The pump resides at the bottom of the case with the hard drives. The motherboard layout is standard and has standard components on board.
When the CPU (~100w) and the GPU (~80w, these values are theoretical) are watercooled, almost no heat is added to the ambient case temp due to these items. What is left over? The NB (if it is not watercooled), the SB, the RAM, the FETS, the HDDs, and the PSU. The PSU vents its own heat ouside, so this is not a factor. The SB and ram may seem to be hot, but they dissipate very little of this heat to the air because they have very little surface area and are not actively cooled. They only seem to generate a ton of heat because they cannot dissipate this heat without a heatsink. If you sink these items and cool them actively with a fan, you will notice a dramatic difference in temperatures. Still, these items produce very little heat and definitely do not dissipate this heat into the case. Then we have the NB. This item is sinked because it produces more heat than most of the components on the board. Lets just say that the NB is now the leading heat source in the case at 30w (im being generous here). The FETS seem to be very hot because they also have no surface area or active cooling, so they seem to produce a ton of heat. This is not true, just as with the SB and RAM. Hard drives are similar, they produce little heat but this heat cannot be effectively removed so they seem to be hot. Notice, these items are hot to the touch, but the air around them is no warmer than any other area in the case.
So we have a radiator now that has to dissipate the pump's heat (most of the heat from the pump goes to the water because most mag drive pumps are essentially water cooling themselves) as well as the CPU and GPU, and because this area now has to dissipate say 200W, that is 200W that is now not being used to heat the case. By putting the rad on top exhausting air out, you are taking advantage of convection, you are not adding to the case temp (actually reducing it), it is simple and it is elegant.
I remember seeing thermal pictures of a computer that was being watercooled, and the CPU was colder than the rest of the motherboard by a large margin. Essentially, the case temp is much less than air cooling, and because the major heat producers are moved to a different location the air going to a rad at the top of the case to the radiator will be far cooler than you might think.
Like you have stated, killernoodle, when you watercool the overall inside case temperature drops. As you say, the vast majority of heat is going to the radiator. What you want to really cool down you are already cooling it with the water blocks; it doesn't matter throwing some hot air from your rad inside the case. The elements being watercooled, as you recall seeing in that thermal picture, are mostly "insulated" as in their temps are overall lower than the rest of the case. That is happening because their heat is going to the water passing thru the block. The great difference in temp of between water and block makes the water "suck" the heat out of the processor. This is an example of convection between a liquid and a metal.
The heat has to go somewhere--> rad. The cooling capacity of it will also be affected by the difference in the air going thru it and the copper of the radiator.
If the fans are "sucking" thru the top, there will be an area of lower pressure on top of your case. As an example if we imagine a wind of air passing by, it will go down and cool your radiator. If you are blowing out you are actually decreasing the effectivness of that wind, you are pushing it away from your rad. You are decreasing effective convection.
Many people report increase in performance with external rads. Wonder why?
The things inside your case are producing some heat, like you say. On the contrary, heat doesn't just go away, or go thru an invisible "wormhole" to another part of the universe. It radiates to the surroundings (like air inside your case).
Anyways the difference isn't so huge(not measurable with normal instruments) so just mount in a way that you are pleased with your setup. Im talking more about "optimum" and people here would use dog saliva for thermal paste if that lowered their cpu temp by 0.5 C. Even if it isn't as noticeable.
Just want to offer the other perspective.
The heat has to go somewhere--> rad. The cooling capacity of it will also be affected by the difference in the air going thru it and the copper of the radiator.
If the fans are "sucking" thru the top, there will be an area of lower pressure on top of your case. As an example if we imagine a wind of air passing by, it will go down and cool your radiator. If you are blowing out you are actually decreasing the effectivness of that wind, you are pushing it away from your rad. You are decreasing effective convection.
Many people report increase in performance with external rads. Wonder why?
The things inside your case are producing some heat, like you say. On the contrary, heat doesn't just go away, or go thru an invisible "wormhole" to another part of the universe. It radiates to the surroundings (like air inside your case).
Anyways the difference isn't so huge(not measurable with normal instruments) so just mount in a way that you are pleased with your setup. Im talking more about "optimum" and people here would use dog saliva for thermal paste if that lowered their cpu temp by 0.5 C. Even if it isn't as noticeable.
Just want to offer the other perspective.
zer0signal667
2[H]4U
- Joined
- Jan 11, 2002
- Messages
- 3,847
LunarJ said:
This hurts my brain... can you restate? What is the "wind" that you're talking about, you make it seem as if it's some mystical force that is unexplainable. Keep in mind there is no such thing as negative pressure- that is tension. If you use the terms "lower pressure" or "greater pressure", you will vastly clarify your statements.
You are correct, I should say lower pressure and not "negative pressure". I'll edit it.
The mystical wind is just that. I said "if you imagine a wind". I am inventing it just to make an example. What I want to say is that if you create an area of lower pressure air would get sucked in (Boyle's). If you have air blowing out it will be deflecting this mysterious wind. You can create the mystical wind yourself if you put an air conditioner blowing in the path by the top of your case. If you agree that the air outside the case has a lower temperature it will make sense to you that it is more effective.
The mystical wind is just that. I said "if you imagine a wind". I am inventing it just to make an example. What I want to say is that if you create an area of lower pressure air would get sucked in (Boyle's). If you have air blowing out it will be deflecting this mysterious wind. You can create the mystical wind yourself if you put an air conditioner blowing in the path by the top of your case. If you agree that the air outside the case has a lower temperature it will make sense to you that it is more effective.
zer0signal667
2[H]4U
- Joined
- Jan 11, 2002
- Messages
- 3,847
LunarJ said:
Ah, I think I understand what you mean... Similar to how an airplane creates lift, a pressure difference is created by the two fluid "streams" having different velocities. Is this what you're getting at?
Still, I don't see how anything less than a strong blast of air would affect someone's watercooling setup in this manner...
I didn't mean to discredit you in my prior post, I was just pretty confused about what principle you were trying to simplify with your example. I'm interested in fluid dynamics and hat flow systems, otherwise I probably wouldn't have given it a second though