i need your waterblock desings, to test

im starting a project where i test a bunch of different waterblock desings, to get some hard data, as apposed to people arguing over better desings, or people running simulations whihc dont reflect real world data.

i need some imput form yall on waterblock desings, first ill start with basic desings, but i need any ideas yall have, nomatter how strange.

im building a die simulator, planning on ahving a .5" square die. heat load can go up to 100W. im using 3/8" hose from home depot.

flow rate will be adjustable, without affecting the radiator performance, ive got the radiator on a seperate pump, with a water mixer thingy joining the rad water with block water, so the flow rates of each pump dont coflict.

i can mill blocks, im using aluminum. some desings might be too complicated for me to make though.

gimme your desings!!!!!!!!!

haha I can only imagine what BillA is thinking as he reads that

Check out the WaterBlock Picture gallery

Oh jeez!! Duck!! Here come the flames...........................

why?

oh, billa is doing similar, didnt realize. no intent to offend anyone. however, i get the impression that billa is using his setup more as a means of judging a competition of whose block is best, this aint my intention. what im doing is testing to see if one general desing type works better than another, for instance, maze 1 style versus gemini style. this is why im donig all aluminum. its more an engineering approach, to gain information of general wb desing performance trends, which can be used to make a block which blows all others away in billas setup.

I'm having major dejavu here.

Was it you or someone else who asked to do exactly the same thing about 3 months ago?

no Cathar, new crop every month
(but I too am thinking this is a repeat)

maybe he needs the WW dimensions also ?
(remember, this is "more an engineering approach")

be cool

http://forums.procooling.com/vbb/showthread.php?s=&threadid=5184

check this design... I't sperfoming ok, and I miht have JFettig test it soemtime

athlonnerd said:

oh, billa is doing similar, didnt realize. no intent to offend anyone. however, i get the impression that billa is using his setup more as a means of judging a competition of whose block is best, this aint my intention. what im doing is testing to see if one general desing type works better than another, for instance, maze 1 style versus gemini style. this is why im donig all aluminum. its more an engineering approach, to gain information of general wb desing performance trends, which can be used to make a block which blows all others away in billas setup.

Click to expand...

But how are you going to quantify the performance of a design vs. another? From what you say, I get the impression that you will build one block with each design and compare them to one another.

But even if you stick with one single design (let's say, maze) there are so many parameters that can vary: size of the channel, shape of the channel (square? Rounded?) Thickness of the bottom, thickness of the channel walls, number of turns before the water gets out, smoothness/roughness of the channel, varying smoothness along the way (rougher at the inlet for more turbulence, smoother near the outlet for better flow). According to BillA, even a small variation in the position of the block on the chip makes a difference. Obviously, the cpu shape will make a difference too: a WB might beat another on a specific chip, and the other way around on another chip. And this is only the WB! Performance will also vary with the pump, radiator, system layout... How do you plan to make general deductions?

Save yourself some major pain and cash, athlonnerd.
Find something else to do. Trust us. You can't do this.

why do you engineers have to be so hard minded? he is doing this partially to be fun, and partially to see what one works best, If this is what he wants to do, dont discourage him, whats rong with doing this anyway? He isnt checking to what hundreth of a degree something outperforms something else....

reeeellllllaaaaaaxxxx... if someone wants to do a rough project then let them, it should not concern you whether they wish to invest their time or not, or to what degree they wish to be accurate. in the meantime, your posts only hijack someone else's thread.

JFettig said:

why do you engineers have to be so hard minded? he is doing this partially to be fun, and partially to see what one works best, If this is what he wants to do, dont discourage him, whats rong with doing this anyway? He isnt checking to what hundreth of a degree something outperforms something else....

Click to expand...

Training in science and engineering is like Zen Samurai training,
or in modern terms, Jedi training my young Padawan.

In simple terms none of this stuff can be valid if a whole long
list of issues are not addressed. We may seem hard minded
but our Masters drove us to see things in a clearer light.
Nobody wants to take the fun out of stuff, but statements made
about performance and winners is just not valid for casual
testing. And it's not just about the hundreth of a degree.
As Bill has shown may times the mounting variation can
easily skew the results. People might not like it but only
a master can get this stuff right. And sometimes even the
masters screw up. And how about the cost of the test
setup. And, yes, if you don't have the expensive stuff you
wont have valid results.

This is the way it is. Go with the flow.

VALID is a relative term, don't forget

i hate to push the subject any further, but seeing as how the subject of the thread is already dead...

i understand your explanation tecumseh, it is very clear and makes sense. but what ever happened that made it wrong to perform a test you want to do, so long as you make it clear what you are doing and do not mislead anyone?

If BillA wants to perform tests to a masochistic level of accuracy let him (and many on the forums, including myself, are thankful he does), he states his methods and tolerances so anyone who can get use out of the results understands how much faith to put into them.

In the same way, if someone else wants to perform tests that can show that one block design is relatively better than another design in a given case let them, so long as that person also makes the methods and tolerances of his tests clear.

This perspective hinges on one requirement... that the tester plainly and clearly conveys the level to which his measurements are accurate and the method by which they are obtained. Anyone capable of finding utility in an individual's test results must also be capable of determining for himself the quality of the work done to obtain the results.

I always attempt to convey what i assert as my own opinion, and let others decide what to accept as fact. I will break from this stratum now to make a point that seems to be willfully ignored by many in this thread, and generally around the forum as of late:

***no one can define VALIDITY for someone else, this is a decision that each individual must make on his own, in consideration of his own set of requirements, in his own application.***

There are two sets of responsibilities that we all learn in K-12 that are applicable here... the responsibilities of a professor and the responsibilities of a listener.

1) It is the responsibility of the professor to be clear in his methods and tolerances, and make them known, in order to validate his work. So long as he does that, then his work is valid.

2) It is the responsibility of the listener to discern the utility he can take from a professors work given the set of methods and tolerances the professor employed. In this way, if the methods and tolerances can not be determined or they do not fit the listeners application, then the listener must determine for himself that the work is invalid.

There is nothing fundamentally wrong with performing a test the way you want to and maybe enjoying it, so long as you do not mislead about the methods in which you obtain your results. In all fairness, i think this is most appropriate. just my own philosophical waxing... i do appreciate the desire for acute scientific accuracy around the forums, im not trying to fight anyone, just sharing perspective. see you all around.

Well, my thought is "Why?"

If even one thing is altered, and that includes the material used to make the block with, then the results will be skewed.

What has been learnt then? Why conduct the experiment if there is a fundamental flaw like that?

What is the goal of this experiment? To learn which is the better design, or to learn which is the better design for aluminium?

What will be learnt that a review that tests the pertinent blocks/designs that exist in the marketplace won't tell you?

That's what I don't understand...

are we equating children playing patty cakes in the mud with civil engineers building a suspension bridge ?
(more accurately analyzing which of several designs will be more effective)

come on kiddos, get a grip
you guys speak 'computerese' right ?
GIGO ring any bells ? - as in 'garbage in, garbage out'

who said "its more an engineering approach, to gain information of general wb desing performance trends, which can be used to make a block which blows all others away" ?

not me, try athlonnerd

he is going to 'get some dimensions', 'make some pieces', test the pieces',
AND THEN analyze the results to derive optimum solutions

now I do indeed know a bit about wbs, and I would not even dream of making such an assessment, or being able to integrate the results into a unified waterblock theory

I.M.O.G.
I do concern myself with accuracy
be advised: I call bullsh*t just that
I do not accept the premise that 'its all good', junk work is still junk and is appropriately identified as such

perhaps a rephrasing of your admonition is appropriate
BTW, threads are public and anyone may comment
and telling athlonnerd that he is way over his head is quite on-topic

be cool

athlonnerd said:

oh, billa is doing similar, didnt realize. no intent to offend anyone. however, i get the impression that billa is using his setup more as a means of judging a competition of whose block is best, this aint my intention. what im doing is testing to see if one general desing type works better than another, for instance, maze 1 style versus gemini style. this is why im donig all aluminum. its more an engineering approach, to gain information of general wb desing performance trends, which can be used to make a block which blows all others away in billas setup.

Click to expand...

So this test would be more along the lines of which block has the best flow path and such, not overall cooling performance?

no skin off my back billa, i know you call 'em like you see 'em.

my main point was just that i think it is within every individuals own realm to determine for himself what is sh*t work, much like you determined for yourself that this would be sh*t work.

my own opinion is in agreement with yours, this work wouldn't be of much value to me, and i suspect it wouldn't be of much value to anyone - but i will leave that up to everyone else to decide.

i am well aware of the complicity of this project, i realize the exorbitant amount of factors which can throw such testing off.

Billa, I do not care much for your air of arrogance and antagonism.

I had indeed posted a similar thread a while ago, I got little response, so I though I would ask again. Why is this so looked down upon by you?

My intent is not for this to be a one time project, I mean to expand upon it later.

Redpepper, I could not agree more, I do mean to test all those types of factors, I was hoping to get a discussion going so as to compile a list of these factors to test. I’m hoping to do enough testing such that we can get an idea of what factors make a wb better than others, and with that knowledge design a wb that performs well, this is what I alluded to when i said it was a more engineering approach; gain enough knowledge/understanding to build a good block, rather than trial and error, guessing as to what would make a block good and testing your theory.

I think I have a good plan of how to make testing as objective as possible, although it should be understood that there is always error involved, but you have to start somewhere.

“Save yourself some major pain and cash, athlonnerd.
Find something else to do. Trust us. You can't do this.”

I appreciate your concern for my time and wallet size, but if you knew me better you’d know that negative advice like trust us you cant do this only ever deepens my resolve. Why can’t I do this? You’ve never even met me; please don’t presume to pass judgment on my abilities.

What if the Wright brothers adopted an attitude of “screw it its too complicated?” where would we be today. I’m not presuming my self or my project to be as important as that of the Wright brothers, but we need to be more open minded.

Who do you then consider to be the master? No one should be revered as master, and we must especially not limit projects to those supposed masters.

Why is accuracy limited to the expensive stuff? Compaq computers are expensive, yet I doubt if any one here would attest to their superiority. I can take silicon thermal grease and sell it for 50bux a tube, is this going to make it magically better than as3? Professionalism does not always constitute superiority or accuracy. I work at TI, in multiprobe, and am astonished at some of the low tech solutions and equipment that are used, it being one of the worlds leading electronics manufacturers, yet it still works. I understand that my setup will not be perfect, but that is not necessarily going to degrade its validity.

Are we all to live in billas shadow?

No this is not a flame, just a rebuttal, or clarification.

Now, rather than allowing the scale of such a project to discourage us, lets discuss ways to make it work better.

“So this test would be more along the lines of which block has the best flow path and such, not overall cooling performance?”

That is indeed part of my intension, experiment to determine what design factors allow for what kind of operational differences, with the final result hopefully being knowledge to design superior water blocks.

“he is going to 'get some dimensions', 'make some pieces', test the pieces',
AND THEN analyze the results to derive optimum solutions”

Who are you to summarize my plans, I have not even committed to word yet half of what I intend to do, if your referring to my request of dimensions of commercial wb’s, I have decided to not go to that extent, I see now that it really would not have served any purpose.

Who are you to decide what junk work is? As IMOG explained, it is up to the reader to decide whether or not my or your work is garbage. I couldn’t agree more.

To be honest I don’t expect to be terribly successful, but can I not try?

“are we equating children playing patty cakes in the mud with civil engineers building a suspension bridge ?
(more accurately analyzing which of several designs will be more effective)”

are we to conclude you consider your work to be building bridges, and my work, and others for that instance, to be Childs play? How arrogant can you get?