Sigh... Water Cooling Testing......

[pHaestus]

jaydee:

Have you considered contributing to the Procooling guest writer program or to overclockers? Both provide some small compensation for one's work, and in all honesty it is hard for me to really get excited about "another cooling site" when the quality writers are already spread pretty thin throughout many sites.

Just a thought.

[jaydee]

Originally posted by pHaestus
jaydee:

Have you considered contributing to the Procooling guest writer program or to overclockers? Both provide some small compensation for one's work, and in all honesty it is hard for me to really get excited about "another cooling site" when the quality writers are already spread pretty thin throughout many sites.

Just a thought.

Yeah I have actually, havn't ruled it out yet either. Well I have ruled it out for overclockers, but not here or a few other sites. I am not expecting my site to take off, in fact almost rather it didn't. I have no plans for putting adds on and it will not have any type of income base on it. All paid out of my pocket. More of a place to direct people that ask "what should I do" type questions. Only advertizing will be the link in my sigs on various sites.

If I come up with an acticle good enough, I may attemp to put it in the guest writers program here. Certainly not against it. Overclockers have me beyond pissed off at them, so I am not at all interested in them. If it wasn't for BillA's articles and a couple others I would never go to that site again!

[BillA]

some months ago a fellow from Calif (?, whose name I don't recall) spoke with me about a java 'calculator'
apparently there is one for hot rod components which allows one to mix-and-match and predict the resultant hp (and torque curve ?)

while such is certainly doable, how worthwhile will be the results ?

lets consider one of the more egregious WCing 'sins'; the inappropriate selection of pumps

we (those in this thread anyway) understand that the pump output at zero head and deadhead pressure is inadequate to size/spec the correct pump
- but equally the P-Q curve itself is insufficient without also knowledge of the system backpressure, no problem - such can be calculated via the components
- but how to address where on the pump's efficiency curve one is, and how to define the limits of 'acceptability' ?
(this last is a substantially 'larger' problem than generally appreciated)

AI systems have long fascinated me, but their utility is hugely dependant on the size and quality of the supporting database
- and generating good data does cost $

here is an idea:
why don't some of the resellers subsidize this activity ?
for they are the ones who will benefit from such

incidentally, wrt the fans:
all that data is known (output vs. backpressure), CFM values all come from the P-Q curves which exist for ALL fans
what has to be generated are the rad flow resistance curves (for both liquid and air)

I second what pHaestus just said
impact (unfortunatly) is more related to the number of viewers than perhaps the intrinsic value of the content
you, JD, will have much more effect by writing for existing sites
- establish your credibility FIRST, eh ?

[jaydee]

Originally posted by unregistered

you, JD, will have much more effect by writing for existing sites
- establish your credibility FIRST, eh ?

Yes, good point and well taken. Of all the sites I frquent there are only 2-3 I would consider participating in. ProCooling, SystemCooling, and amdmb. Maybe OCAU to but I just signed up there and havn't had much time to see what it is about yet.

[BillA]

out of curosity, what is your difficulty with JoeC ?
(Ed is the non-technical editorial end)

I am not familiar with System Cooling - but I'll go look
and I would not really call amdmb a 'technical' site (despite Dave's massive contributions there)
OCAU is not technical in terms of a resource (which I'm presuming is your interest)

but I'm more than sure that JoeK would like some articles along the lines we're discussing

[jaydee]

Originally posted by unregistered
out of curosity, what is your difficulty with JoeC ?
(Ed is the non-technical editorial end)

I am not familiar with System Cooling - but I'll go look
and I would not really call amdmb a 'technical' site (despite Dave's massive contributions there)
OCAU is not technical in terms of a resource (which I'm presuming is your interest)

but I'm more than sure that JoeK would like some articles along the lines we're discussing

Whom ever is responsible for running the forums is who I have problems with. If they are going to ban my PAID e-mail address then that is enough. I don't sit well with stuff like that. I even tried to change my e-mail to one of my website POP accounts but it will not let me even do that. The forums are unorganized and not well kept up IMO. Banning my e-mail was the last straw.

amdmb isn't really that technical compared to some, but unless people step up it will never be. amdmb was the second forum I ever visited and first computer related one. I see a lot of clueless people wanting to water cool and they are not getting a lot of direction. Luckily there are a few like myv65, Robotech, ScomRacer, and lately Ben had been poping in. It is getting better. I feel a certain loyalty to them since I been there for a long time and apart of 3 of 4 DC teams there. Pretty good people and the forums are run real well and taken care of. Non of the BS overclockers forums seem to sufffer from.

SystemCooling is fairly new, but the people there are not. Pretty experienced people running it. Has potential.

ProCooling still tops my list though. Seems more directed at water cooling than anything else and most of the people that come here are looking for water cooling info unlike the other sites where people are looking for all types of info from mobo help to linux support. Cooling is just a subtopic.

[pHaestus]

You should ask around some of their former sponsors before affiliating with system cooling. Give KD Computers a ring and ask Dave about his opinions on those fine fellows. Also DO go back and read some of Bob's heatsink reviews. He makes hard|OCP look like NIST in my opinion. My opinion on that site is that they are all talk but more than willing to market and promote and write glowing reviews so he does well "in the industry". They have had "reviews" where they simply got a sponsor to take pictures of coolers and didn't even test them.


AMDMB has never done anything to make me consider them to be anything other than a reasonably upstanding enthusiast site. Dave's articles are all great and on occasion I find the brand specific mobo forums a good resource. I seem to recall their getting in some flack with Asus for a bad review a year or so ago right?

[gmat]

Getting flak from a mfger for a review is a sign of excellence, in my book...
As for a piece of software simulating a WCing loop, why not, but filling the (needed) database would take ages. A team could do it - think open source ...

[pHaestus]

a team? But the database ONLY works if one can trust the numbers that go into it. How many trustworthy number generators do we have around?

[gmat]

That's the problem. If you count on only one person to fill the data, the DB won't be nearly useful before 1 or 2 years, and i'm optimistic.
One needs the curves (actually, the data behind them) for every pump, piece of tubing, rad, fan, and waterblock around. The software can interpolate between curve points, using linear or more elaborate techniques, but too many missing points and you'll get inconsistent results.
That said, many curves show a nice inverse exponential look, which can be neatly approximated. But one needs to consider each case (pump, rad, etc) and each subcase to match the math functions as exactly as possible. Not a short work either.
Also, some parameters are very hard to code, like turbulence behaviour for example. This is the domain of finite elements calculus, and out of reach for a small java applet...
We need to define exactly what to expect from that piece of software.

[jaydee]

Thats news to me about SystemCooling. I will dig around some on that....

Yeah the Asus Control panel got a very bad review from them. Asus got pretty pissed about it. I havn't seen anything to think they are biased which is another reason a I stick around there.

[BillA]

JD
I too have SEVERE problems with the OCers forums,
a snapshot of which is described in this thread on Scientific Debate
it is my considered opinion that the forum managers mean well but don't have a clue, the mods are inconsistent and biased, and that despite being adults can act like the kiddies they supervise
- due to this the OCers forums are totally worthless as a technical resource, which is my interest (and as a sounding board for snippets of articles)

JoeC seems to have control of the 'technical' side and really is a prince of a fellow (integrity, goodwill, and knowledgeable)
- though I'm getting ready to BLAST a seriously stupid article he accepted on P4 lapping
OCers is a much larger operation than this site and the specific responsibilities are delegated to different individuals

you should reconsider (just shitcan the forums)

BTW, was not impressed with SystemCooling; the WCing stuff looked like mfgr fluff to me

[pHaestus]

Yes Joe Citrella is one of the few people who runs a website that I actually trust on reviews. And while Ed is certainly opinionated I also find him quite entertaining and about 2/3 of the time on target.

Even so, I find the Procooling forums and irc chat to be a very rare animal indeed. We have arguments and tempers flare, but VERY rarely is anything deleted or anyone banned. I know Bill and JoeK had some issues prior to my being here, but my experience over the last year or so is that this forum is about the perfect size for good discussion and debate.

[Blackeagle]

BillA

I understood your use of hit. Thank you for that and the link. I just checked out the graf you posted along with since87's info on the math.

Now let's see if I'm getting the use of C/W right.

If I O/C a CPU up high and it is putting out a estimated 140 watts of radiated heat, and I'm useing a strong pump with a flow of 3gpm and my C/W is as in the last point on BillA's graf aroung 2.19 then my heat load on die would be 30.66C over ambient. Or a real die temp of 55.66C at a room temp of 25C.

Would this be right? (Allowing for a smoke and mirrors starting heat output?) Thanks.

[BillA]

better to multiply your s&m 140W by 0.7 to address secondary path losses
and the "C/W" would be 0.217 or so
this all relates to the CPU/coolant temp differential

then you need to do so for the rad (Since87 added the wb "C/W" and the rad C/W)
and it is not the room temp that is the baseline, but the air temp through the rad - generally a bit higher

but yes, you have the idea

[Blackeagle]

Thanks BillA !

If I can get the math for more of this figured I'll be able to keep my learning curve going.

BTW, when I first found this site my idea then of a high performing system was: Eheim 1060 pump with a BIE rad & maze 3 block with 1/2" lines and standard nylon fittings of all types.

Now I know I still need more work on the "best cooling system" for what I'd like to set up as a system, but none of the above will be in it except some 1/2" line.

[BillA]

lol
I bought a 1060 and sold it the next day for a $60 loss

you owe me

[Since87]

Originally posted by gmat
That's the problem. If you count on only one person to fill the data, the DB won't be nearly useful before 1 or 2 years, and i'm optimistic.
One needs the curves (actually, the data behind them) for every pump, piece of tubing, rad, fan, and waterblock around. The software can interpolate between curve points, using linear or more elaborate techniques, but too many missing points and you'll get inconsistent results.
That said, many curves show a nice inverse exponential look, which can be neatly approximated. But one needs to consider each case (pump, rad, etc) and each subcase to match the math functions as exactly as possible. Not a short work either.
Also, some parameters are very hard to code, like turbulence behaviour for example. This is the domain of finite elements calculus, and out of reach for a small java applet...
We need to define exactly what to expect from that piece of software.

I'll start a new thread to discuss a possible simulator tomorrow.

If you don't mind gmat, I'll paste your post into that thread.