Newbie question: Water cooling vs. Air cooling

[Rusty075]

Bob, you're grasping now. Let it go. Expecting "pH" () level analysis from a normal user is unrealistic. Snutten's temps went down. Good for him. Everyone understands the concepts involved now, there's no point in beating the horse anymore.

[peterson]

Well, this discussion is over my knowledge for the moment, even though i studied flows....over 20 years ago...oh man.

Don't blame me if you get mad at eachother.

I'm not sure i'll try this parallell combination. Not to begin with anyway.
I think i'll stick to a 240mm rad and CPU/GPU block to begin with.

But i've recently spent so much on my PC, that i probably won't be able to afford it until after the summer. But atleast i'll have time to change my mind several times

And so...what's you knowledge of Alphacool's stuff? (check the english flag for english lang.) It looks really good to me. Could their pump AP 700 be ok? 19dB is probably not what it delivers. Their radiator HC-240 LE looks good too. (the links seems so wierd so i decided not to post them) I can't find any specifications on flow or anything. Perhaps manufacturers usually don't reveal that.

Does one usually regulate the flow in W/C? I.e. the voltage to the pump, of just the speed of the fans to keep temp/noise level??

Thanks!

[snutten]

Never measured the flow. It was such an obvious improvement, easily seen in my reservoir.

BOOM! This is a real problem, and doesn't look good after your earlier "I empirically know the flow from a 1048 (peterson's leaning towards a 1048 right?) will in fact increase in both loops compared to the single original loop if it passes 2 blocks and 2 rads." No doubt there was a useful increase in total flow, but double or better now seems a pretty dubious claim. The response at PC should be fun.

Hopefully the guy at procooling who said he wanted to try it out can confirm my findings. The improvement really was obvious enough for me but I certainly understand you need something more substantial than yet another nerds wishful thinking.
The 1048 can manage 600 lpm with only 1,50 m head. Heck, splitting into three lines probably wouldn't hurt either with such numbers. It's all about managing the head pressure!

[Rusty075]

Does one usually regulate the flow in W/C? I.e. the voltage to the pump, of just the speed of the fans to keep temp/noise level??

Since more flow is a good thing, you generally don't do anything to regulate it. Pump and fan voltage can be adjusted for noise as needed.

And nobody's mad at each other. This is how watercooling discussions usually go. Its the beauty and the curse of WC'ing: there's enough complexity to make sure that there's always something to argue about. Air cooling discussions are never as interesting.

[bobkoure]

Bob, you're grasping now. Let it go. ...

I wouldn't call it grasping, just pointing out that non-Ph temp readings may or may not mean anything. However, it's pretty straightforward for a semi-tech-literate person like me to stick hoses in a bucket (or buckets) and use a stopwatch to measure flow.
I read "I've tried both ways. I empirically know the flow from a 1048..." not to mean "my temps as reported by MBM were lower when set up this way", but instead to mean, well, measured flow - something that non-Ph types can do, if not with fantastic accuracy, at least good enough that doubled-flow is significant (meaning outside range of error), and, in this case, surprising enough that I started a thread on ProCooling about it. (As a BTW, I usually find surprising data to be the most interesting as it can point out some way or other that my mental model of the world is wrong.)

Guess it's one of those cross-forum cross-cultural things... the same way folks here read my guess at 17dB (based on how loud I know 20dB to be) to be a statement implying measurement in an anechoic chamber with decent spectrographic equipment - which it was not - just a guess based one one thing being louder than another.

[snutten]

Who reported DDC being louder than 1048? All the things I've read so far indicate them to be comparable, with neither pump winning out in acoustics.

Well, I own one of each and I can tell you the DDC is way noisier than a 1048

[Edward Ng]

Ah, there it is; another thread lost to me since I started my new job...

This is interesting to note. Thank you.

[BillA]

an outsider reading this thread would have to be amazed at the credence given to flow rates and temperatures, and the immediate skepticism with which db(A) 'values' are met

the same approach is valid for all measurements; they are either actual measurements, or not
-> why discuss someone's guessing accuracy ?
if they are unable or too lazy to measure it . . . . .

if a nonsensical claim is made, the burden of proof is on the claimant

[pdf27]

an outsider reading this thread would have to be amazed at the credence given to flow rates and temperatures, and the immediate skepticism with which db(A) 'values' are met

Yeah, it's a holdover from the history of SPCR - the watercooling forum is relatively new, but we're all used to people telling fibs with the dBA measurements quoted on fans. Then again, a lot of us go over to ProCooling for things like that and only really go over the silencing side on here.

BTW, Welcome to SPCR (in case nobody else got there first!). I've been hoping for ages you would turn up over here

[BillA]

whoa
did not mean that hard numbers are bad, and real #s on sound are as scarce as hen's teeth
rather that all 'fibbing' is (a strange euphemism for BS), whether temp, watts, noise, etc

[bobkoure]

I confess to guessing about dBs - I've been trying to make them educated guesses - but that's all they are.
BTW, there's a "rule of thumb" that for there to be a perceptible difference, there has to be at least a 1 dB(a) difference.
Anyone know if this "rule of thumb" actually applies in the lower ranges?

Oh, and Bill - welcome here! Even if you just want to visit and don't feel like being resident cynic. Fine either way!

[pdf27]

whoa
did not mean that hard numbers are bad, and real #s on sound are as scarce as hen's teeth
rather that all 'fibbing' is (a strange euphemism for BS), whether temp, watts, noise, etc

Yeah, I know - was just trying to explain why we're rather more... strict... about dBA numbers than anything else
As for "fibbing", I'm from the UK not the US so certain linguistic pecuilarities are to be expected

[snutten]

an outsider reading this thread would have to be amazed at the credence given to flow rates and temperatures, and the immediate skepticism with which db(A) 'values' are met

There are numerous sites out there discussing flow and general o/c. Turning SPCR into yet another one would be horrific. Let's stick to what we're good at here. Silent PCs!

the same approach is valid for all measurements; they are either actual measurements, or not
-> why discuss someone's guessing accuracy ?
if they are unable or too lazy to measure it . . . . .

if a nonsensical claim is made, the burden of proof is on the claimant

I dunno if you're referring to my split lines? You're right about everybody speculating, and yep, I'm too lazy to measure it only to present the numbers on the internet!
I guess you have higher standards. I'm too sloppy and simply tried this out until solved. Just don't agree on the word "nonsensical" in my case. Again, 600 lpm, 1.50 m head. Which is why I tried it in the first place. Made sense to me...
Never even guessed on my total flow in exact numbers, never found that interesting. Presenting numbers, like noise 22 dBa or flow 13 lpm, begs good measuring standards. Saying flow substantially increased, however, is like saying this Adda fan is noisier than this Nexus. Not how much noisier. Still don't trust it's noisier without precise numbers because "loud as hell" vs "barely audible" isn't good enough? (SPCR seldom bothers to measure up bad fans. Just confirms it's bad, and measure the good ones instead!)

The reason I pointed to Cathar's numbers is because it shows us how un-important flow is, and how small pumps are weak against restriction. I leave it to you to scientificly test and (dis-)prove it

[peterson]

Woohoo, time to put some fuel on this before it goes out.

I'm really close to a decision on going w/c or upgrade my aircooling.

Since last time i've bought a Stacker case (i know it's difficult to silence but it's really versatile and suits my needs. And it's a challenge to dampen )and a Nexus PSU (Not so impressed by this one. )

I feel the airflow for the PSU gets obstructed by my CPU HS (XP120) so a waterblock is one solution. One other is ofcourse to buy a passive PSU and/or and Scythe Ninja HS (or similar)

IF, i go for w/c, i'll definitely go for a CPU and GPU (not cooling memory) waterblock. And pulling air out of the case with rads at the back seems like a better idea than having them at the front. My case has some good space for 1 120mm rad and 1 2*80mm rad. That'll be enough right?
How much airflow does one lose w fans pushing vs pulling? I just imagine that it'd be quieter to have fans on the inside of the rads rather than on the outside of them. I'll make a shroud of some sort.



I also considering on a passive radiator. But i haven't recieved an answer on how much heat they can dissipate. I have space for 2*CORA Midi radiators (8 pipes total). They're available at Alphacool. Anyone her knows how much a passive radiator like that can handle?


Has anyone here ever considered mounting a radiator on the PSU fan??? I mean, the PSU fan rotates allways anyway. Perhaps it could do twice the job.

The pump is still under consideration. And silencing the HDDs are under construction.

First on my pump list

(on wishlist)

I really don't 'need' w/c-ing, but it'd be fun to play with i guess.

[bobkoure]

Those radiators look to be both water-flow-restrictive (which you might not care a lot about) and air-flow-restrictive (which, if you care about noise you do care about)
Have a look at the thermochill PA160 - optimized for use with a single low-noise fan.
No idea how much heat those passive rads might shed. If you find numbers, be sure that the temperature difference between coolant and air is something you might actually use - like 5C or less. Radiator capacity goes up as the temp difference goes up, so the numbers become less useful to you...

If you want to just run one fan in your system, put the radiator over the intake and seal the case - so the one fan pulls through the rad, through the case, through the power supply.

[pdf27]

Has anyone here ever considered mounting a radiator on the PSU fan??? I mean, the PSU fan rotates allways anyway. Perhaps it could do twice the job.

In case you were in any way serious, how exactly would you make the seals work?

[peterson]

I had a shroud of some kind in mind.

[pdf27]

I had a shroud of some kind in mind.

Ah, sudden "ping" noise in head. For some reason I was under the impression you were thinking of building the radiator into the fan blades
I suspect you're asking the poor PSU fan to do too much there - they're usually weak 25mm jobbies. If you were to use a standard rad/shroud and duct the air through the PSU to cool it (removing the fan) you might get away with it though. Not sure there'll be any accoustic benefit however...