Correct waterblock sequence

Maybe this will help, think of it this way:

Most of us overclock for one purpose, to get more for our money and crazy speeds.

Now because most of us do this for gaming....what is the best way for us to "maximize" our gaming fps, scores and etc....

Try running your CPU/GPU at stock speeds and see what kinda score you get (3dmark).

Then leave the GPU at stock speeds and turn your CPU to its max OC.

Then leave the CPU at stock speeds and turn your GPU to its max OC.

I always get more gain from teh overclocked GPU. So in my system and in my opinion in general about this....why not cool yoru GPU first??? SInce it helps in framerate the most.


And besides my loop distance (tubing and etc...) is so small and my flow is moving so fast (or fast enough) that the "cool" water coming off my radiator swirls into my GPU fist and then almost instantaneously to my CPU. I've seen air bubbles in my system go all the way through the entire loop so fast i can't keep up with them.

So what im saying is, in my opinion it doesnt really matter what block you go to fist, CPU OR GPU...but if i had to get that "little bit of nonsense" extra cooler water then it would go to the GPU since i get more fps out of a higher GPU than CPU, it just so happens that my system is better organized for looks, ease of arrangment and flow to have it going into my GPU first. I'm happy with my OC's. Especially my 9800pro @ 475mhz....

thats my $0.02

--Will
 
That is understood; your GPU temp is nowhere near the temp of your CPU, so the water will still absorb a good bit of your CPU's heat.

But why deal with it if you could just use a 'Y' to hit them both at the same time? Again, I really don't know how much dividing the flow in 1/2 will effect the system.
 
I've never done a "Y" splitter either. I say just stick with one direction, thats whats worked best for me. However, why don't you guys build it normal then add the "Y" and show us the results.
 
Well, if you think about it in terms of water temperature, and this whole post explained that the water temperature becomes constant throughout the loop. I would say your doing nothing but halfing your flow rate.
 
I took it to be a bunch of different opinions on whether the temperature remains constant, with no real conclusion or numerical data (temperature readings!) to back up these claims. Without such data, I will continue to think that this equilibrium mumbo-jumbo is just that: mumbo jumbo. As much as I thank each respective contributer for adding his/her insight, I'd as soon not spend any more time thinking about it until I get into my own thermodynamics classes.

OMP did say that he had seen data supporting this, but I'm no fan of second-hand facts -- no offense intended. I checked procooling, but couldn't find anything; I can't seem to search their forums, and I got a bit lost in their search tool.
 
Ok Mr. Technical.

hrm...go research and save that extra 0.3 C, then let us know you've won.
 
The only ones who 'win' from either result of an argument such as this is the community as whole . . .

I'm not trying to prove that anybody's right or wrong; I'm asking if anybody can show us data supporting their claims, so that we all know without a doubt that the coolant in a system such as ours either is or isn't in thermal equilibrium.
 
im working on it, so far I have is a heat transfer equation through convection...

q=h*(t1-t2)

The problem is the h variable is composed of many variables (one of which is flow rate) which of I cannot find a formula for, im going to look in my old physics book.

edit: My old physic's book is very brief when it comes to convection, it explained more about density changes and prosed no formula's so im back to the net. (no wonder I didnt remember that formula)

edit: im gunna keep looking =)
 
i'll dig out my physics book too.

I'll go back to my Organic Professor when the term starts, he will have a good method to do this.
 
wow....

This went from being a moderatly advanced discussion including bits of physics and thermodynamics, but somehow the spirit of the origional discussion was lost and here we are throwing sh*t on observational facts and speculating again.

Vellicator: Actually READ what we've been saying. Once you've read it, then try and understand what has been said. The physics is undeniable and just because you don't understand it doesn't make your theories correct and give you the right to question what has been said.

Here are some good discussions over at procooling that deal with the new questions on this thread (and help Vellicator see about equilibrium as well)

http://forums.procooling.com/vbb/showpost.php?p=115664&postcount=17

http://forums.procooling.com/vbb/showthread.php?t=6385&page=1&pp=25

-OMP
 
gclg2000 could you give me a realistic relationship between flow rate and efficiency? I tried with the calculator, but later realized that a square mm isnt the same thing as a square meter. =)
 
Well, since you're being bull-headed, I guess I may as well follow suit.

If you'd shut up and actually READ my post, you'd see that I neither "threw sh*t on observational facts" nor speculated. Actually, I did the exact opposite by asking for access to those observational facts. Also, I did READ and UNDERSTAND and, in fact, THANK YOU for your explanation. I never said that you were wrong, and I never "questioned what has been said" -- I simply said I wouldn't believe it until I saw the data that you spoke of.

If you want to know why I ask, look here.
 
I think I did more of the speculation and threw my pile on the convection formula. Im interested in physics, even though I dont know what im talking about half the time.
 
Lord, i havent started researching htis yet, because i have a pathophys and genetics final next week in which im studying for now, however i think its going to depend on finding out the effeciency of the radiator to remove "heat" from the water. I'm going to put a thermometer in my res to make sure it is at room temperature. Becasue in theory, especially with my system my radiator is so huge that it shoudl get the water to room temp. Once that is confirmed, the volume is going to have to be the next question which will vary from system to system,then the pump flowrate on top of that. Im interested enough to do the research and come up with a ~ forumale table we can all enter our specs into to find out the "hair line" specifics that some of us are so damned curious about. This will be a good little research project. I'll let you know when i get started on it.
 
Vellicator said:
Well, since you're being bull-headed, I guess I may as well follow suit.

If you'd shut up and actually READ my post, you'd see that I neither "threw sh*t on observational facts" nor speculated. Actually, I did the exact opposite by asking for access to those observational facts. Also, I did READ and UNDERSTAND and, in fact, THANK YOU for your explanation. I never said that you were wrong, and I never "questioned what has been said" -- I simply said I wouldn't believe it until I saw the data that you spoke of.

If you want to know why I ask, look here.
Click to expand...

Haha....I like how you link to the U of R laser lab page, which is where I got my undergraduate degree. Are you a student there now? Great place to learn.

Sorry for the direct approach, I was mainly reacting to your other post which basically said I'm not sure whats going on and I don't see any scientific data to support the claims so it's not true ("mumbo jumbo" I believe you said). Even before this you post about watercooling loop orders without a hint of understanding of the physics involved (i.e. running "hot" water over the GPU....).

Another thing: How can you claim you aren't questioning the validity of something when you plainly say you won't believe it until someone shows it to you. That makes no sense at all. Get your story straight.
 
As I'm sure you know, there have been countless scientific hypotheses, derived equations, and-so on that were proven inaccurate by thorough experimentation.

Your explanation seems perfectly sound, but just as sound, in my mind, is the feeling that it "just ain't right". The only way I'll believe otherwise is if that feeling is proven wrong, which, truthfully, wouldn't be much of a suprise for me. An example would be "Gabriel's Horn", which you've likely studied. As many times as I do the math, it still doesn't make sense, thus it "just ain't right".

Anyway, I do not question the validity of your stance. What I'm interested in is the grey-area where theory meets fact.
 
How true...

Everyone is entitled to have their own opinions and theories...That fact is not in question here. It's just that some people have to be correct, and some people aren't....And most of us know whats correct and whats not. There is no use convincing people who are dead set one way or another.

My last comment on this subject is this: Read what Cathar has to say on the subject (links in previous posts) and try and dispute his testing methods. BTW, if you don't know who Cathar is, he designed the Whitewater block (now sold through Dtek) and the best waterblock on the market today in the Cascade. His work in watercooling testing and theory is unsurpassed and easily available at the procooling forums.

P.S. I have studied Gabriel's Horn, and I really don't see the relevance to this discussion. The result of the y = 1/x paradox (infinate surface area and finite volume) is both mathematiclly sound and quite possible when you think about it. The math is flawless, it's just that the result is paradoxical and not easily understood. Your statement that just because it doesn't make sense in a conventional matter then it "just ain't right" is a complete load. Take some quantum physics and quantum optics and try and understand that. Even now, the majority of quantum isn't understood theoretically, the fact that major scientists get the same experimental results over and over again proves that just because a theory remains questioned, the experimental results speak for themselvs.
 
My goodness, I guess I shall have to repeat it.

I do not question anything that you have said!

I simply would like to know if anybody has taken ACTUAL temperatures after the waterblock and after the radiator, and whether they were the SAME. Physical data is all I'm asking for, not further explanation of something that I already understand.
 
Vellicator said:
My goodness, I guess I shall have to repeat it.

I do not question anything that you have said!

I simply would like to know if anybody has taken ACTUAL temperatures after the waterblock and after the radiator, and whether they were the SAME. Physical data is all I'm asking for, not further explanation of something that I already understand.
Click to expand...

I know you aren't questioning what I have said. All I'm objecting to is your constant posting that contradicts previous postings. You say you aren't questioning what has been said, yet you seem to throw into every post how a lack of understanding or rational explination means a theory is incorrect. Prove me wrong, or explain what you meant by your previous 3 posts.

Here is a good test simulation that found a maximum water difference in a watercooling loop at 1 degree EVEN WITHOUT WAITING FOR THE WATER TO REACH EQUILIBRIUM. : http://www.hydrocool.com/Testapparatus.html

That only took me a 2 minute search at procooling. Imagine what you can find when you search for a longer period of time.

I'm tired of wasting my time here. Feel free to continue debating. My work and instigating here are done.

-OMP
 
Vellicator I think the problem here is simply your lack of understanding of the underlying physical phenomena in thermodynamic systems (I don't mean that as an offense, you said yourself you're not experienced with the math). The fact of the matter is that our conclusions stem from a number of physical properties that have been studied, proven, and utilized for in some cases hundreds of years (such as the basic gas laws, etc., from which the laws of liquid systems have been elicited). There is a large wealth of experimental evidence supporting these laws, and the general progression of scientific study has led to their culmination and combination into very rugged and reliable methods for studying thermodynamic systems.

The fact that no one has provided data for a specific experiment testing these effects in a water cooling loop does not mean that data seen from similar experiments or experiments from which our model has stemmed doesn't prove this. In our day in age we are being forced to study certain phenomena via not-so direct methods. Current quantum mechanics is a good example of this, we can elicit specific properties ONLY through non direct means, eg. it is IMPOSSIBLE to study sub atomic 'materials' at planks distance. It is believed that it is at this length, or close to it that the most fundamental component of matter (namely strings) exist, since we can never actually visualize or test the existence of these strings we have turned to the study of phenomena in the MACROWORLD, namely effects seen in black holes, gravitational distortion (and its connection to general relativity) etc. There has been mounting evidence to support this theory, and most of this evidence comes from mathematical models, and properties seen in systems quadrillions of times bigger then the strings themselves. While on the quantum topic, you'd be amazed at the FUCKED UP properties elicited by quantum mechanics, such as quantum entanglement (one particle reacting to a change in another particle that could be trillions of miles away), higgs field mechanics, quantum duality etc. Most of these properties defy belief and dont seem to carry any logical reasoning, yet both direct and indirect evidence has been produced to explain these phenomena.

One thing you havnt noticed is the resemblance watercooling setups have to industrial or commercial equipment. Eg., your fridge relies on thermodynamic equilibria for it to work (for the effects of sublimation and condensation to operate on the cooling of the refridgerant).

So, no we havn't provided you with direct proof, but to do so would mean we would need to teach you step by step starting with basic gas laws basically every thermodynamic law that can be applied to our 'relatively' complex thermodynamic system (aka our watercooling loop). Just because you don't see something, doesn't mean its not true, and I can personally (and im sure OMP would agree) attest to the validity of thr thermodynamic laws dictating the physical effect we see. So until you've learned these yourself in a thermodynamics course, and until you also have a firm understanding of calculus, not only how to differentiate and integrate, but what doing these functions actually MEANS I don't think you have any authority in disputing OMP (or myself) and in fact all I see if a troll who has come to stir the pot.
 
Vellicator said:
I simply would like to know if anybody has taken ACTUAL temperatures after the waterblock and after the radiator, and whether they were the SAME.
Click to expand...
Over at ProCooling in pHaestus's waterblock reviews, he has based his estimates for the CPU's heat on measuring the water temperature before and after the water block. You'll have to do a bit of digging in their forums to find it, but the information is there. While you are there, read also Incoherent's heat measurement stuff, and pretty much everything else in the 'testing and benchmarking' subforum.
 
Interesting discussion, and without any violent outbursts or anything of the sort thus far :D
Let's keep it that way, and that means no more pseudo-insults, gclg2000. We're here to exchange useful information, not argue or instigate argument.
 
Don't single me about guy. When people say immature things that make no sense, thus not helping the community, they set themselves up for it. ALL of us got fed up with it.

Obviously you got lazy and only read page 4 of this thread which of course shows my NASA and dunce remark. Read the whole thread and not my short easy posts alone before you isolate me into a "problem."

Regardless i'll stop.
 
I don't know if this will help, but here is a log from about a year ago of my research / quest for optimal watercooling heat transfer within my setup.with interesting results.

06-16-03
Water cooling component placement test for optimal heat transfer.

Test #1
Cooling Components used:

SwiftTech MCW50 VGA water block ½ inch ID fittings

SwiftTech MCW5000 CPU water block ½ inch ID fittings

SwiftTech FBK525 fill and bleeds kit

BlackICE Extreme (BIX) Radiator with 1- 120mm Panaflo M1A Fan

Danner Mag3 water pump ½ inch ID fittings

Shin Etsu G751 thermal compound

System setup

SwiftTech FBK525 fill and bleeds kit >> Danner Mag3 >> SwiftTech MCW5000 CPU water block >> SwiftTech MCW50 VGA water block >> BlackICE Extreme >> SwiftTech FBK525 fill and bleeds kit.

System being used for test:
Abit NF7-S v2.0
ATI 9700Pro
AMD Athlon 2500+ over clocked to 3200+
Room temp 74f
Idle temp: 37c
Full load 45c (prime95 and 3DMark 2003 for 1 hour)
06-18-03 No change in temps
06-20-03
Test #2
Cooling Components used:

SwiftTech MCW50 VGA water block ½ inch ID fittings

SwiftTech MCW5000 CPU water block ½ inch ID fittings

SwiftTech FBK525 fill and bleeds kit

BlackICE Extreme (BIX) Radiator with 1- 120mm Panaflo M1A Fan

Danner Mag3 water pump ½ inch ID fittings

Shin Etsu G751 thermal compound

SwiftTech FBK525 fill and bleeds kit >> Danner Mag3 >>
BlackICE Extreme >> SwiftTech MCW5000 CPU water block >> SwiftTech MCW50 VGA water block >> SwiftTech FBK525 fill and bleeds kit. *Note* Water blocks were not removed from system.

Room temp 74f
Idle temp: 35c
Full load 43c (prime95 and 3DMark 2003 for 1 hour)
*Note* 2c lower temps in both idle and full load
06-22-03 no change in temps







06-23-03
Test #3
Received Danger Den Typhoon-Quad DUAL Reservoir
My theory is, by adding the quite large bay reservoir and removing the fill bleed kit and putting the radiator before the water blocks I can drop the temps a little more, thus optimizing the heat transfer.
Cooling Components used:

SwiftTech MCW50 VGA water block ½ inch ID fittings

SwiftTech MCW5000 CPU water block ½ inch ID fittings

Danger Den Typhoon-Quad DUAL Reservoir

BlackICE Extreme (BIX) Radiator with 1- 120mm Panaflo M1A Fan

Danner Mag3 water pump ½ inch ID fittings

Shin Etsu G751 thermal compound

**Both water blocks left in place**

Typhoon-Quad DUAL Reservoir >> Danner Mag3 >>
BlackICE Extreme >> SwiftTech MCW5000 CPU water block >> SwiftTech MCW50 VGA water block >> Typhoon-Quad DUAL Reservoir
Room temp 74f
Idle temp: 33c
Full load 41c (prime95 and 3DMark 2003 for 1 hour)
*note* clear difference in temps, "4c lower across the board" than the original setup.
Just for sake of argument, I then moved the Radiator.

Test #4

Typhoon-Quad DUAL Reservoir >> Danner Mag3 >> SwiftTech MCW5000 CPU water block >> SwiftTech MCW50 VGA water block >> BlackICE Extreme >> Typhoon-Quad DUAL Reservoir

Room temp 74f
Idle temp: 34c
Full load 43c (prime95 and 3DMark 2003 for 1 hour)

Conclusion:
From my test it appears that for optimal heat transfer for my water cooling setup, I get much better transfer with the bay reservoir and the radiator in the system loop before the water blocks.
Click to expand...

Of course your mileage may very :cool:
 
gclg2000 said:
Don't single me about guy. When people say immature things that make no sense, thus not helping the community, they set themselves up for it. ALL of us got fed up with it.

Obviously you got lazy and only read page 4 of this thread which of course shows my NASA and dunce remark. Read the whole thread and not my short easy posts alone before you isolate me into a "problem."

Regardless i'll stop.
Click to expand...


Actually, I've been reading this thread since the discussion began. I didn't have a problem with your NASA remark, other than the fact that you are suggesting that there is no science behind watercooling - but that's your own opinion. I do have a problem with intentional insults, however. As long as you express your opinions in a civil manner, then you're free to keep doing so. That goes for EVERYBODY here.
 
ok whatever you say. A better choice for this would have been PM, as to not start a small fire (flame).

As far as the science, i've been on board with it ever since....see my post with physics?
Also have posted many in the extreme cooling section under the physical properties of alcohols for liquid cooling loops.

I won't make "direct" quotes i guess anymore, but i don't see how the NASA is ok and that one pushed the bar....lol

Anyways im sorry, i will not do it again bud.
 
I tried to stay out of this (especially since NASA pays a large part of my funding), but I guess our "uber" scientists are still at it. Noone disputes that the laws of thermodynamics or quantum physics hold true for this. They are just wholy inappropriate to describe the system. The laws of thermodynamics deal primarily with conversion of energy into work and back (a water cooling system performs basically no work other then overcoming the flow resistance) or phase changes (which I hope you don't have going on in your system). You could also solve the full transport equations or the Schroedinger equation for the system without gaining any of the useful parameters of the system. If you want to discuss mathermatical models for this, look at heat transfer coefficients and Reynolds numbers and you might get something out of it. And no, I'm not saying this because I can't distinguish my Lagrange polynom from a Bessel function.
 
Coming from a nuclear power background and using the very same type of system that's being discussed I can tell some of you are probably misguided. (water goes into reactor and heat is added, water goes out through pumps to steam generators where heat is removed, then water goes back to reactor)

As a general statement water coming out of a heat removal device will be cooler than the water coming in to that device. (it's called the cold leg of the system)

For small systems, such as your typical water cooled computer, those temps most likely will be small or negligible because the line runs are short.

For large systems, such as a power plant, those temps will be very different and significant.

I just wanted to clear that up for those that are actually trying to learn something here. :)

F.Y.I. - Just because a system is in equilibrium does NOT mean that it's all the same temperature. It just means that at any one point in the system the properties of the system are not changing over time.
 
Correct me if I'm wrong, but I thought that in a power plant that uses steam to drive turbines (such as in a nuclear plant), that when water is added to the reactor core, the water is converted to steam which is channeled to turn turbines. If this is correct, then we can not compare this to a watercooling loop because in a nuclear plant, the steam is a product of the loop, which removes mass from the loop, and by the laws of thermodynamics, is treated as an open system where temperature differences between heat sources is expected. We made the point earlier that watercooling loops are basically closed because they don't lose and gain mass.

Maybe I'm wrong about the steam part, but if I'm right, then I would expect a cold and hot leg, where in a watercooling loop I would not (and proven not to).

EDIT:

In all the above posts, the word Equilibrium is used to express thermal equilibrium, meaning a null temperature difference across the loop. It is possible to be in a type of equilibrium without being in thermal equilibrium, but here we are only talking about thermal.
 
No. Water does not convert to steam in the reactor. If it did, you would have a fuel element failure and possibly some magnitude of a meltdown. The water is pressurized so that it does not boil inside the reactor nor anywhere in the loop. The steam generator would be the equivilant of the radiator in a computer cooled system. It is a seperate system. Just as the air in a radiator does not come in physical contact to the water in the cooling system, the steam in a steam generator does not come into physical contact with the water of the reactor cooling system. The steam generator takes heat away from the reactor coolant just like a radiator takes heat away from the computer water coolant.

AND

You can be in thermal equilibrium and have different temperatures throughout a system as long as the net thermal value of the system does not change.
 
divbyzero said:
No. Water does not convert to steam in the reactor. If it did, you would have a fuel element failure and possibly some magnitude of a meltdown. The water is pressurized so that it does not boil inside the reactor nor anywhere in the loop. The steam generator would be the equivilant of the radiator in a computer cooled system. It is a seperate system. Just as the air in a radiator does not come in physical contact to the water in the cooling system, the steam in a steam generator does not come into physical contact with the water of the reactor cooling system. The steam generator takes heat away from the reactor coolant just like a radiator takes heat away from the computer water coolant.

AND

You can be in thermal equilibrium and have different temperatures throughout a system as long as the net thermal value of the system does not change.
Click to expand...

I see. My mistake....Thanks for pointing that out about the reactor.

It is not true, however, that different parts of a thermally closed loop can be in thermal equilibrium and have different temperatures. This is forbidden by the zeroth law of thermodynamics. A system will ALWAYS attempt to reach a stable equilibrium point in regards to temperature. The key word here is ATTEMPT. Because of the extraordinary temperatures in reactor cores, and (I would assume) a relativly high flow rate through the legs in the loop, the system is overwhelmed and cannot fully reach a stable temperature because of the lack of time for the water to absorb and transfer energy before being recycled back into the loop. The same is true in watercooling loops, its just the effect is much smaller because of the lower heat energies involved and as you said before, the relative smallness of the system. Once again we see the differences inherent to relative micro and macroscopic systems.
 
That's the problem with the statement. The loop isn't thermally closed (heat enters the loop from the CPU waterblock and heat leaves the loop through the radiator), so the zeroth law does not apply. What's more, work is done within the loop to the system(pump).

I can see that you're an intelligent person. I think there is just a little confusion as to how the laws apply. The only way that the loop could have the same temperature everywhere was if there was no heat transfer occuring to or from the loop and no work was being done by or to the loop.

EDIT

It has been some time since I've been in school, but I believe the ideal thermal system to apply the zeroth law to is when you have blocks A, B, and C at different temperatures, T1, T2, and T3, and you place them in contact with eachother in a vacuum (thermally closed). At some time t, all three blocks will be the same temperature T123' based on their masses, initial temperatures(kinetic energy), and thermal conductivities(affects only time), assuming no heat transfer to or from the vacuum.
 
You must log in or register to reply here.
Back
Top