• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

PROFESSIONALLY done waterblock testing. (satisfied?) Cheap!

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.
gone_fishin said:
It's that crowd, the good enough crowd. They want to see shiny waterblocks in their perdyness displayed across a web page attributing great feats of wonder and awe to their likeness, with no attachment to the real world.:D

Could you try to be a little more helpful instead of posting something just a tad demeaning every few posts?

Since87 is actually helping me. I admit i am NOT well experienced at this and my reaserch has been pretty good up til now. i probably wont be buying this for another month.. i do have experience with WBs just nothing ever as specific as calculating C/w

Please if you want help me but if youare just going to post things saying how i dont want to do any work blah blah and such i dont need your suggestions.
 
Let me pop my head in here for a sec... you seem to be stuck on the amount of heat that's removed from the CPU for a given block, when really it's a trivial matter: since we're dealing with an equilibrium situation, the heat removed is exactly equal to the heat produced by the CPU (let's ignore secondary losses so we don't lose anyone ;) ). If it weren't, the temperature of the chip would be changing... i.e. a transient condition. What makes one block better than another is the temperature gradient between the CPU and coolant that's required to drive this transfer of heat.
 
Bingo.

Atleast EXCELSIOR didn't go to the frontpage and look how to calculate C/W up.

I guess you could take that either way though LoL!

Sorry, don't take anything i say seriously. 80% of the time i'm 80% joking!
 
Toysrme said:
Bingo.

Atleast EXCELSIOR didn't go to the frontpage and look how to calculate C/W up.

I guess you could take that either way though LoL!

Sorry, don't take anything i say seriously. 80% of the time i'm 80% joking!

:eh?:

....... Lost me there Chief...
 
xX~EXCELSIOR~Xx said:


Could you try to be a little more helpful instead of posting something just a tad demeaning every few posts?

Since87 is actually helping me. I admit i am NOT well experienced at this and my reaserch has been pretty good up til now. i probably wont be buying this for another month.. i do have experience with WBs just nothing ever as specific as calculating C/w

Please if you want help me but if youare just going to post things saying how i dont want to do any work blah blah and such i dont need your suggestions.

I'm sorry but I tried giving very specific help, only to be greeted with, it makes no sense. The other helpful advice is don't waste your money till you know and understand the concepts and what it will take to measure the differences in two waterblocks. Spending a couple hundred on a sensor and calibration then finding out the resolution is worthless would be a bummer for example.
 
Bill should have wrote a book on this before he left so others could take up the challenge. I know Joe is getting stuff together and I know he knows a lot, but I don't know that he knows a lot about the stuff he needs for this. In other words, why trust him?

Anyway, I am learning something from this thread. Although the information might have been posted a hundred times, it's the one you read that matters.
 
xX~EXCELSIOR~Xx said:
..................
If you have a desig nthat is not not that good at conducting heat.. lets say just a flat surface. Put it on top of a 200 degree F heat load.

you do have a chunk to think about. There could be a 200 degree heat load but it could cool to room temp by blowing it. or there could be a 200 degree heat load that takes massive insane waterblocks jetting thousands of gallons per second to cool it... witch one is it?
The temperature of the die doesnt determine the heat load. The wattage is what the heatload is measured by. Not the wattage in power consumption but heat dissipation. For a test to remain valid, That would have to remain constant.

If your wondering about c/w Its not totally acurate because of the TIM joint, like someone pointed out.

To calculate C/W, Its a simple calculation. all it is is how it looks. Delta C*/Wattage of the heatload.

btw, I havent told ya nothing yet:D
You got a LOT of homework to do, a few years of figuring out how your gonna work this thing 100% and maybe a little help.

Jon
 
I gotta agree with these guys, you've got the cart way ahead of the horse. If you are wanting to do this with the hopes of recovering your money, you are definitely wasting your time. On the other hand, if your are interesting in doing this as a hobby and spending/learning as you go, then go for it.
These guys are really taking it easy on you. Your thread wasn't made asking for advise, it was made offering a service. A service which they are pointing out, you can't really offer.
If you want to learn some leasons from BillA, pay close attention to the order he did things. He laid out his plans and methodolgy in great detail. Then he started building his testbed, and walked us through it in great detail again. Then he started testing blocks that he had at his disposal and did some fine tuning of his methods and equipment. Only after all this did he offer to start testing other blocks. All the steps and stages were well documented, this satisfied him and those of us who would be following along watching for his reports.
BillA invested blood, sweat and tears into his testbed, not to mention a butt-load of money. It was a labor of love. If you do some searching of his posts, you can find the aprox $ amount he spent, you can also find the amounts he charged for DIY blocks and Commercial blocks, and the number of each block he tested. Not even factoring in value of his time, I'd say he had less than a 1% return on his investment. If you want to make some money, you'd probably be better off buying Lotto tickets.
Well thats my $.02

peace.
unloaded

PS. Your price of $30 vs $90 he charged for DIY blocks won't make a difference because there was a standing offer to have DIY blocks sponsored so the maker would only have to pay $30. Sadly nobody ever took advantage of this offer.
 
Gotta chime in... just because Excelsior has good intentions, which is admirable.

First off, the heat source would be in Watts, and if you use a 70W source, that's what the rad will dissipate.

Given the above, it's not hard to figure out that the water temp would actually remain the same.

So in the total temperature difference between the CPU core and the ambient air (in case of a rad), it's easier to see that a good design will allow a lower CPU temp, because the temperature gradient between the CPU and the water will mostly sit in the block, and not the CPU. A bad design will allow for a higher temp of the CPU, while the block itself would have a larger temp difference, across its baseplate.


Excelsior: you really need to understand a few things, such as repeatability. Can you swear that your result will be the same the next day? Next month? BillA uses one block as a standard, and calibrated his test rig with it. He ran the calibration test before testing, to make sure everything is still the same.

What other calibrations do you see as necessary, in your proposed rig?


Otherwise, for $30, I'd send you a couple of block, for sure!
 
Excelsior, I'm probably gonna sound a little harsh here, but these comments are meant to help you. I apologize in advance.

1. In order to be taken seriously, you must present yourself well. Sen tenc es ty ped likethis arev ery h ard to re adand donotlookprofessionall.

2. Learn the differences between "heat" and "temperature" and be clear in their uses. For example, if a CPU is generating 50W of heat, all waterblocks will dissipate 50W of heat (ignoring secondary heat paths.) The equilibrium temperatures will be different, but they will all dissipate the same amount of heat in steady state.

3. Understand the necessity of repeatable results. Unless you spend the money to have all of your equipment calibrated regularly, and do a shunt cal before each test, your results will not be accurate. However, accuracy isn't as important as repeatability in this particular field of testing. (You're not going to publish these results in any journals, are you?:))

4. Understand that many people here will try to help you, but most people here have a very weak understanding of heat transfer and thermodynamics. Many will inadvertantly mislead you. You must be able to tell when someone posts false information.

5. Finally, realize that testing is some of the most boring work known to man. It's even more monotonous than assembly line work. If you are independently wealthy and have nothing better to do with your time, go ahead with your plan. Let's estimate your cost of testing a waterblock to be about $5, and it will take you about 10 hours per block once you are comfortable with your test setup. (Both are very conservative numbers...) Are you prepared to be paid $2.50/hr for your troubles? Trust me, after spending 1 month testing, you'll realize that a job at Wal-Mart is much more fun and rewarding.

Sorry to chastise you, but reality is usually harsh. Just realize what you're getting into before you spend too much money.
 
bigben2k said:

So in the total temperature difference between the CPU core and the ambient air (in case of a rad), it's easier to see that a good design will allow a lower CPU temp, because the temperature gradient between the CPU and the water will mostly sit in the block, and not the CPU. A bad design will allow for a higher temp of the CPU, while the block itself would have a larger temp difference, across its baseplate.


????? "it's easier to see that a good design will allow a lower CPU temp, because the temperature gradient between the CPU and the water will mostly sit in the block, and not the CPU"

What you just said,
Translation: temperature gradient=difference between two temps
The two temps?: CPU and water
???the gradiant sits in the block not the CPU??? excuse me but WTF are you talking about???

And Ben you gotta make one before you send one;)
 
bigben2k said:
Gotta chime in... just because Excelsior has good intentions, which is admirable.

The road to hell is paved with...

I believe there are a lot of people in this world who believe Osama Bin Laden has good intentions. (Including Osama himself.)

Admirable?

Well intentioned dissemination of misinformation is not particularly admirable IMO.

Also - what gf said.
 
Um, that was a joke.

>I guess you could take that either way though LoL!
As in was he honest enough to not run and look it up and pretend, or was he not quick enough to think about it?

Anywho's...

Don't get your feelings hurt, it's just that nobody here will trust you for a while. Sit back in the community for a few months and then reconsider LoL! If you haven't changed your mind, you'll at least have the possibility to have a better chance of knowing what's going on.

BTW This is an increadable amount of money devoted to this project LoL! I'm sure you have better things to spend it on too. Don't worry about being our savior because we don't have someone testing our blocks. We'll all be just fine!
 
I have been trying like hell to stay out of this. Is it better
to crush youthful exuberance with hardcore knowledge or
let people forge ahead in bliss, only to be crushed later?

xX~EXCELSIOR~Xx, there are some good people here
trying to give you the benefit of their hard-won experience.
Some of the people in this thread giving you the best advice
were once beat on by BillA and others. They got it and now
are helping to move water cooling forward with insightful
comments and tests. It's not a requirement that you be
a physicist or thermal engineer. Some of us are. I understand
your desire to "get going" on things, but you have put the
cart before the horse.

Dude, might I suggest that you chill somewhat and try to
get what people are telling you? Just a thought. :cool:
 
Amen, amen Tecumseh! --- the voice from the back of the room floats forward.

We had a fairly good discussion last fall called "Scientific Debate" and several of the folks here were part of the discussion/debate. The mods in their infinite wisdom delete all debate threads older than 3 months or I would point you at it. I would have been tempted to make a condensed version of it "required reading" for those that wish to cross the boundary from opinions to providing results.

Got to say that if you're studying the sciences or engineering and you think you've been put upon here then you might want to reconsider your choice of professions cause you really haven't seen anything yet compared to the world waiting for you. I think all have been been trying to be constructive. I know I was in the original thread to which you pointed as being such a negative experience. Science really is a contact sport xX~EXCELSIOR~Xx.

Be :cool:
 
Last edited:
Calvin-N-Hobbes said:
Excelsior, I'm probably gonna sound a little harsh here, but these comments are meant to help you. I apologize in advance.

1. In order to be taken seriously, you must present yourself well. Sen tenc es ty ped likethis arev ery h ard to re adand donotlookprofessionall.

2. Learn the differences between "heat" and "temperature" and be clear in their uses. For example, if a CPU is generating 50W of heat, all waterblocks will dissipate 50W of heat (ignoring secondary heat paths.) The equilibrium temperatures will be different, but they will all dissipate the same amount of heat in steady state.

3. Understand the necessity of repeatable results. Unless you spend the money to have all of your equipment calibrated regularly, and do a shunt cal before each test, your results will not be accurate. However, accuracy isn't as important as repeatability in this particular field of testing. (You're not going to publish these results in any journals, are you?:))

4. Understand that many people here will try to help you, but most people here have a very weak understanding of heat transfer and thermodynamics. Many will inadvertantly mislead you. You must be able to tell when someone posts false information.

5. Finally, realize that testing is some of the most boring work known to man. It's even more monotonous than assembly line work. If you are independently wealthy and have nothing better to do with your time, go ahead with your plan. Let's estimate your cost of testing a waterblock to be about $5, and it will take you about 10 hours per block once you are comfortable with your test setup. (Both are very conservative numbers...) Are you prepared to be paid $2.50/hr for your troubles? Trust me, after spending 1 month testing, you'll realize that a job at Wal-Mart is much more fun and rewarding.

Sorry to chastise you, but reality is usually harsh. Just realize what you're getting into before you spend too much money.

Dont scrutinize me for my chopped up scentences. IM on codine for my arm most of the day :D

The reason im doing this is for a fwe things. #1 i could learn a hell of a lot of things from this experience #2 to add content to my site

I could care less if i was getting 2.50 an hour to test each block because i wouldnt reall be making anything since it will all be going to reapying the money i spent.

I am not trying to do this because no one else is doing it as a savior as someone said but rather it would be a fantastic learning experience.

I am NOT going to spend a few years learning and come back and do it :rolleyes:

Basically this thread was strictly wondering if there was an interst in such a service. I am discussing a thread in other forums on how exactly to go about such and the most accurate methods. I didnt really ask for the ways to construct it and such but rather just wondering if ther eisan interest if i did it well. If i did this i would obviously do a ton more research first. If i did it i would definitely post my procedures for approval and suggestions first. All that i was really awnting was to know if people were interested in such.
 
Now, I've been a member of these forums for about 2 years now, I just don't post on everything. During my time here, I've learned a few things:
1.) The men/women/boys/girls in the O/C forums are, without a doubt, the most helpful, least likely-to-flame of anyone out on the 'net.
2.) When someone here says something, and he/she has 7 humpty-jillion stars, it doesn't always mean that they're supplying correct information. The people that DO, however, back up their claims with good, viable info.
3.) The best way to learn anything is to learn it yourself.

That being said, I would say to you, Excelsior, that you should probably go to the front page, read a decent amount under ALL of the archived articles, and soak up as much as you can about this. Personally, I doubt that you will get much (if any)interest from people on these forums, because we expect THE VERY BEST. Now, I'm not saying that you couldn't be the best one day, I'm saying that right now, you are starting out. After some time and experience, that interest-o-meter might just rise a bit. I don't mean to sound rude or overtly annoying in this post, but that is the way that this world works.

Another $.02 in the money jar...

Final piece of advice: Read over articles that you've already read through because you might have missed something. I do it all the time with the (ED)itorials and the archived pieces.
 
Back