Need suggestions Copper or Aluminum

[BillA]

Originally posted by maskedgeek
. . . . and billa you saw that fixittt made a aluminum spir@l and got same performance as the copper spir@l?

so ?
you are asking for my assessment of the comparability of the wbs ? dunno
his test setup ? dunno
his calibration ? dunno
his test procedure ? dunno

and even if provided with this info, in great detail, I doubt that I could be convinced of a "same performance" claim

I've been doing this for 2 years, every day, and understand just how difficult it is to generate accurate and reproducible data

for example:
right now I'm set to take data after a 4 1/2hr. warmup; this is the 18th install of the Cooltech WB75 wb, and I have many to go
(then, I get to re-do 5 weeks of work because I just identified, and controlled, another variable !)

I KNOW my numbers are 'good', what do you know about your's ?
let alone Fixitt's

[bigben2k]

I have to side with BillA on this one: how the temps were taken, and how the whole testing was done needs to be explained, otherwise the results are "rough", with a large margin of error that could go either way.

That being said, I still believe that Alu is a valid material to use for watercooling.

[maskedgeek]

i honestly dont get why you go thru all this crazy madness, a as long as you have all the variables the same except the block or the part your testing isnt that all you need? and acurate readings... maby flow rate and given flow rate but you have all this other jibberish that half of us even look at... i know when i see your 100000 numbers i just overlook it
and what is your other variable you found?

[maskedgeek]

oh yeah, and why are you soo stuck on copper? you give me no information that leads to how much better copper is, that gives me NOTHING the whole thread has turned pointless...

[maskedgeek]

oh and bill, go back and re-read all my posts and why dont you actually answer the questions i have instead of just being all prowd and looking over them

[bigben2k]

I can't speak for BillA, but let me refer you to this article by myv65. Specifically, check out pages 3 and 4 (it's a very long read). I can vouge for the accuracy/truth in this article, I have verified the information.

That should clear up the Al vs Cu question.

(PS: I still believe that Aluminium is good for a waterblock: it may not be the best in performance, but it will provide anyone with a cheap ability to watercool their rigs, albeit at the cost of a large overclock)

[maskedgeek]

Originally posted by bigben2k

(PS: I still believe that Aluminium is good for a waterblock: it may not be the best in performance, but it will provide anyone with a cheap ability to watercool their rigs, albeit at the cost of a large overclock)

thanx, i saw that above but i had to argue bill and forgot

[tokamac]

did anyone even see my post?
Tokamac

[maskedgeek]

Originally posted by tokamac
did anyone even see my post?
Tokamac

yes i did see your post, and i kinda responded to a few at once... yes its pretty much 2x more conductive...in theory pretty much, but when it comes to real life situations of watercooling computers its not exactly like that

[myv65]

Originally posted by maskedgeek
yes i did see your post, and i kinda responded to a few at once... yes its pretty much 2x more conductive...in theory pretty much, but when it comes to real life situations of watercooling computers its not exactly like that

It's statements like your last one that bring exasperation to engineers. Metals have no brain. They do not know if they are cooling a CPU, being used to conduct electricity, or even sitting in the ground in varied states. Pure copper will always behave the same in conduction. Pure aluminum will always behave the same in conduction.

Pure copper will always have nearly twice the heat conductivity coefficient of aluminum. This means that in equal shapes, copper will always have slightly more than 1/2 the temperature differential as aluminum when conducting heat through the same pathway.

Why you will not see a precise difference for a given power is that the flow pathway of the heat will not be precisely the same for the two materials. Copper will cause heat to spread over a larger area than aluminum, all other things being constant. For this reason, you will frequently see differentials less than 1/2 for copper versus aluminum.

In air cooled heat sinks, you'll actually find that aluminum does a better job as fins/pins. Not because of some absurd idea that "aluminum gets rid of heat better than copper". It's simply a question of surface area. For a given weight, aluminum has far more surface area and gains the advantage in convection. I'll guarantee you that less than 5% of the people that consider themselves cooling gurus really understand this basic priciple.

In water cooled blocks, this advantage is largely muted because of the inherently higher convection coefficients of water. This puts an even higher premium on a material's conductivity.

Now, as BigBen2k and BillA noted, you can do just fine with aluminum, but you'll never do as well as copper. I'll guarantee that you'll tend to see a variation of 2-5°C for most blocks and power levels if you made the blocks from both materials. Is the extra cost of the raw material and machining difficulty worth 2-5°C? Not for most people, but that doesn't change the fact that it exists.

What you see as an irate attitude towards your questions folks like BillA see as ignorance in search of an easy answer. All the answers you seek are available, but it doesn't do a whole lotta good to simply tell you the answers. Unless you understand why things are the way they are, you doom yourself to blindly proceeding randomly. You may make a few different blocks with some performing better than others and have no idea why. Rather than pop in and ask for all the answers, invest some time to search for those answers on your own, both through hands on research and by reading. And I got news for ya, if you think you're going to find good information cruising forums you are sadly mistaken. There are plenty of very smart people that hang here and elsewhere, but they're outnumbered by morons that spout conjecture without proof. Again, unless you understand the real issues, you have no way to separate the good stuff from the garbage.

OK, I'll get off my soapbox now. . .

[maskedgeek]

actually that want the origional point of this thread, the origional point of the thread was: would you buy aluminum blocks? would you refrain from buying the block just because it is aluminum and would yield 2-5c dif from copper?

quote from my origional message

So what I want to know is if you peeps would still buy them if they were aluminum vs copper.... Like If you would refrain from buying them or anything, Im gonna order the metal prob monday or so... they are gonna be sealed by solder(aluminum solder) or a gasket and screwed together, and I can make them the rite sizes for your mounting....

[myv65]

OK, getting back to your original point. The obvious answer is that some people will and some people will not. It is also only one of the factors involved with deciding on a block. IMHO, there are far more factors to consider such as fittings, channel configuration, hold-down mechanism, company (or individual) reputation, cost, appropriateness to remaining system components, etc.

I'll tell ya who I wouldn't buy from. Someone that has no experience. Someone that doesn't exhibit a grasp on what separates a good block from a bad one. Someone that is fly-by-night and may not be around to help with problems. Someone that hasn't done their homework to develop a good block. The list goes on and on.

I guess that leaves only people that I know very well from forums AND can make blocks (this is a very, very short list) and the "big names". Even among the big names, I wouldn't buy from quite a few simply because they don't seem to know much about what they're doing either. Some seem to think it's enough to make a pretty block. OK, nothing inherently wrong with that as lots of people are more concerned with how their system looks through an acrylic window than how well it actually performs. That does not describe me, however.

There's all kinds of people out here that buy blocks for all kinds of different reasons. Many (most) couldn't tell you the difference between a "good" block and a "bad" block. With a portion of the buying public clueless about what matters, selling on looks alone has merit.

Anyway, aluminum and copper is but a small part of the equation. Develop a block that performs well, mounts well, and looks good. Show it off and support your customers. Do all this and you'd be on your way. Easily said, but tougher to execute.

[maskedgeek]

Originally posted by myv65
OK, getting back to your original point. The obvious answer is that some people will and some people will not. It is also only one of the factors involved with deciding on a block. IMHO, there are far more factors to consider such as fittings, channel configuration, hold-down mechanism, company (or individual) reputation, cost, appropriateness to remaining system components, etc.

I'll tell ya who I wouldn't buy from. Someone that has no experience. Someone that doesn't exhibit a grasp on what separates a good block from a bad one. Someone that is fly-by-night and may not be around to help with problems. Someone that hasn't done their homework to develop a good block. The list goes on and on.

i know exactly what your saying, i dont have much of a reputation here but over at overclockers and wc101.com i have a pretty good one, iv made somewere areound 8 blocks and that is a small number compared to the amount i would have made if i had the monies for mill and material, im here all the time to help out, iv done tons of homework, thats were i have decided on aluminum, for many reasons in above comments, have you seen any of my blocks???

I guess that leaves only people that I know very well from forums AND can make blocks (this is a very, very short list) and the "big names". Even among the big names, I wouldn't buy from quite a few simply because they don't seem to know much about what they're doing either. Some seem to think it's enough to make a pretty block. OK, nothing inherently wrong with that as lots of people are more concerned with how their system looks through an acrylic window than how well it actually performs. That does not describe me, however.

pretty blocks suck, iv made a few but its pointless im a function over form for watercooling, all my previous blocks(previus to the 2 solid acrylic with copper based blocks iv made were 100% soldered, the fittings were even soldered in, check out the gallery and waterblocks section in wc101.com forums, youll see 2 of my blocks, and in my avatar a few more... i also dont think that how it looks shoudl be paid attention over how it performs, although i do like to color my water etc

[quote
There's all kinds of people out here that buy blocks for all kinds of different reasons. Many (most) couldn't tell you the difference between a "good" block and a "bad" block. With a portion of the buying public clueless about what matters, selling on looks alone has merit.

Anyway, aluminum and copper is but a small part of the equation. Develop a block that performs well, mounts well, and looks good. Show it off and support your customers. Do all this and you'd be on your way. Easily said, but tougher to execute. [/quote]

this is my plan to make a good block and have great support, and make the good block aluminum is one of the things i wanna acomplish, i have another design cadded up that i mite take to the cnc at school if i can get access again(most likely)
my first kit that im building for someone is almost complete, waiting for the pump, and just to mod the intake and make it all fail safe,
theres a pic of a few parts of it leak testing, the h/c shroud (fan not included) and the waterblock 100% soldered together 1/2 barbs and clearflex 60 tubing(only 1 small segment of scrap shown)

-edit- the main reason i made those acrylic ones is becuause i was running out of material and didnt have cash for more and had scrap acrylic laying around i made use of....

[bigben2k]

I don't know about you myv65, but I'm the type of customer that wants to experiment with block designs, yet I don't have the facility to produce one, so if MaskedGeek is offering, I might very well take him up on it.

I think you might have misconstrued MaskedGeek's purpose here: beyond the Al vs Cu theory, there is a business aspect, and what I see is that Jon (MaskedGeek) is doing some market research, to get an idea of the number and type of customers who would go for Aluminium blocks. He's already outlined his business plan, which is to produce Al blocks, and he's going to redirect the profits to acquire the necessary tools to produce Copper blocks.

It all sounds good to me!

Jon, if it's of any interest, I might go for an acrylic prototype, so that I can test flow paths, and look into milling challenges. The block wouldn't be of any use otherwise.

[maskedgeek]

yeah i have done 2 designs in acrylic, it kinda sucks to machine tho, it melts and gives off fumes, im gonna try to get ahold of some poly sometime...

im glad someone is understanding me and is able to explain it better than i can

[Les]

Originally posted by unregistered


as part of a test series (still unfinished) I have looked at aluminum and copper bps of different thicknesses mounted on 'The King of Testing Waterblocks', an old style MCW462

Some performance predictions based on "my best fit model" for your Swiftech 462u data presented here http://forums.procooling.com/vbb/sho...5&pagenumber=8. Used a T/W value obtained by allowing 0.05 c/w of the TIM(C/W) for the offset of the die tc.
Best fit calculated assumming the turbulence created in a 6x6mm square entry nozzle is maintained over all the wb base. From the Reynolds number calculating the Convection Coeff (Sieder-Tate) and then adjusting the effective Convection area. :


The "best fit 80x80mm" .is larger than the internal base area but.............

Showing the predicted effect at diiferent flow rates of Cu and Al bp thickness and the change to the new smaller AMD (80sq mm ) die .






Edit: Added C/W v Thickness curves for 8x10mm Die

[BillA]

interesting numbers Les, the smaller die is going to make things hotter for sure
my take on the geek's lament ?

aluminum = the cheap 'solution' that doesn't work as well

save a buck and add some heat

[maskedgeek]

Originally posted by unregistered
interesting numbers Les, the smaller die is going to make things hotter for sure
my take on the geek's lament ?

aluminum = the cheap 'solution' that doesn't work as well

save a buck and add some heat

"beware he's grumpy" really? i couldnt tell?
im not even gonna say any more...

[BillA]

maskedgeek

what is the topic of this thread ?

you seem displeased that you cannot sustain a chorus of "aluminum is 'just as good' as copper"

you as a wannabe fabricator may prefer aluminum for various reasons,
but ultimately they all relate to lower cost

and such paths, predictably, also lead to lower performance

no biggie
no individual is 'winning' anything, we're discussing material properties
- if YOU want to ‘talk technical’ you had better start becoming comfortable with numbers, lots of them

"Need suggestions Copper or Aluminum"

[maskedgeek]

FINALLY you start making sence!!!

partially the reason i want aluminum is the tools i have to work with, the cost isnt a biggy, but the labor of working aluminum vs copper, have you ever done any copper and aluminum milling and whatnots? it takes experience to know these things,

I plan on using hi quality designs and blocks in aluminum, itll only be around 2-5c lower than it would be with copper, does 2-5c really matter? toward the 5c it may but 2c is no big deal...

[Heavy_Equipment]

Originally posted by maskedgeek
actually that want the origional point of this thread, the origional point of the thread was: would you buy aluminum blocks? would you refrain from buying the block just because it is aluminum and would yield 2-5c dif from copper?

quote from my origional message

would you refrain from buying the block just because it is aluminum and would yield 2-5c dif from copper?

I had to quote that line twice.

Yes, I would pass over a product that runs 2C (never mind 5C) hotter than the next one. I'd pay top dollar to better what I have by 2C, and I have proven that time, and time again. The line of heatsinks, and waterblocks on my shelf is my proof.

Aside from that, look up #Rotor... he makes two versions of his block, one copper, and one aluminum...same design.

(hmmm, his site appears to be down...)

[maskedgeek]

ok thanx for the comment, im glad i see some truth here, iv seen rotors stuff, but never seen his aluminum blocks, how do they compare to his copper ones?

[Heavy_Equipment]

I can't comment on the difference...I never inquired about the aluminum block when I bought mine (Cu)...we did talk about battery effect, but that was it.

Come to think of it though, here's a guy that has a few systems running on one water loop...one being a dually, and he's using his copper blocks on them.

IMO, he wouldn't use 2nd best (and more expensive) when he has both at his disposal. Would he?

[bigben2k]

Originally posted by unregistered
interesting numbers Les, the smaller die is going to make things hotter for sure
my take on the geek's lament ?

aluminum = the cheap 'solution' that doesn't work as well

save a buck and add some heat

Interesting indeed...

because it shows that copper is best at at least 10mm bp thickness, for most flow rates, where Aluminium is best at 10mm OR LESS, for flow rates above 3 gpm (180 gph).

I don't believe that the performance difference is worth all the bashing that ya'll are giving MaskedGeek here. Aluminium has its place, just not for serious overclocking. A Mercedes is better than a Lada, no one can deny that.

[utabintarbo]

Originally posted by bigben2k
Interesting indeed...

because it shows that copper is best at at least 10mm bp thickness, for most flow rates, where Aluminium is best at 10mm OR LESS, for flow rates above 3 gpm (180 gph).

I don't believe that the performance difference is worth all the bashing that ya'll are giving MaskedGeek here. Aluminium has its place, just not for serious overclocking. A Mercedes is better than a Lada, no one can deny that.

Perhaps a better analogy would be to a Honda - not flashy, probably won't win any street races, but will get you where you are going economically and reliably.

Bob