This tube routing ok?

Hi,

am water cooling a bitfenix prodigy. At first just a cpu only loop with an xspc ex240 up top, later on gpu also with the addition of a 200mm phobya extreme rad on the front.

For the cpu loop, could this be better? The blue circles are 45 degree fittings, the green thing is a t-joint for my t-line (not going to use a res). I already tried one way, but I'm pretty sure I did the t-line wrong as bubbles keep shooting back into the pump as I routed it. Not sure how to upload a photo here so here's a link.

http://imagebin.org/245911

Suggestions welcome! Thanks.

you have a HUGE pump for such a small system. is it the variable model where you can turn it down all the way? and here is the problem of no reservoir, bleeding the system. you have to do it over and over and over again... fill t line>bump pump>fill line>bump pump... until the sound of flowing water disappears. if you can't turn that pump down to almost nothing (fan controller?) then i'm not sure you can do much but be patient and work slow.

i'm not sure changing the routing of the tubing will help much here.

+1 to that it took me atleast a few hours to bleed out my tline, its been a couple years so i dont really remember exactly how long.

get a cylindrical reservoir, like EK 150 or 250, is my suggestion.

I used t-lines for first few years of water cooling and to bleed, always had to turn pump on/off and manually at just right time, tilting case slowly with pump off, to get bubble going up the t line. Was a pita. I hated bleeding.

Reservoirs bleed themselves. ~7 years ago after trying first cylindrical reservoir, I would never use a t-line again. Just hook things up, fill, and your done...top off once bled. You might have to tilt case couple times while pump is running, which is easy enough.

even micro res is nice, small, and offers many mounting options.... bleeds easily. http://www.frozencpu.com/products/8...ctor_High-Flow_Reservoir_12_ID_and_38_ID.html

Bitspower 150 seems to be a nice little res. It's the one I'm going to be trying.

The pump is the vario one, I just put it on level 1 and it's totally silent. With just the pump in there it actually is alot 'smaller' than most people's pump/res combo. I'm not too much of a fan of how hot the ddc pump gets.

I've bled the system a couple of times now for practice and it takes about 30 minutes of shaking the case around etc... I think that isn't so bad. I make it a bit easier by fitting a res to the end of the t-line only during filling, still takes longer because the bubbles have to escape while flowing past the t joint but I don't need to use a syringe or fill the t-line half a dozen times bit by bit.

But does having a res actually increase the 'cooling' of the loop by increasing water volume? ATM because of the short tubing (and narrow tubing) and no res there's not much water needed to fill it. Maybe that's just a myth but I did read it somewhere.

No, the plastic sides of the res more insulate the water. Think about it. Physics. How many square FEET of cooling does a rad have with fins vs the surface area of the res?

Physics.

Everything is true on the Internets. I can say in my $1.2 million temp measuring chamber with a $1.3 million temps controlled lab I saw a .000000000002 C diff in having a larger res with more surface area.

So, I know that a larger res makes a diff in temps. I can't be wrong. It's the Internets.

Seriously, no measurable diff. A res has a small surface area and poor heat transfer properties. Physics and common knowledge if you think about it.

I know that a res itself doesn't passively cool the water, i meant in terms of total system water volume. I saw a res for sale once which was coated in aluminium heat sinks all over it and I just laughed. There is a link between volume of water and thermal capacity. If greater thermal capacity then greater cooling. I just don't know if it would be negligible on this scale. Q = mc * delta T, where Q is heat transfer, m is mass, c is specific heat constant of water, delta t change in temp.

Was going to post another thread about 'styling' (i have no artistic bones) but I guess I can ask here. Atm all that is set in stone in this rig is silverstone x560, assrock z77e-itx, clear 3/8 5/8 tygon tubing, ex240 and phobya 200. No idea how to make it look 'cool' lol. Do want some colour in there. Fittings etc I can change. Anyone has any ideas (or digital rendering skills) I can pay for that if you pm me

how about you t-line out to a quick disconnect... then you could have a micro rez/quick disconnect to attach just for filling purposes...???? wait, how often do you plan to fill/bleed it? standard once a year...? waste of time. choice of no reservoir = more work for you... haha... suck it up.

if a reservoir was metal, and had any sort of heat sink style/shape, then yes it could provide "SOME" sort of cooling capacity... comparable to a radiator, NO.

the "idea" you're thinking of is relates the capacity of the system and how it correlates to amount of time needed to "heat up" the water; this does not help cool the system, but rather spreading the heat load over more water increases the time to heat up. the system will still cool the same, based on radiator/fans combo. sure, if your reservoir is a 200gal fish tank, it will take some time to heat up... but given enough time... it would reach a comparable temp to a 500ml reservoir. ((or something like that))

baobeiiii said:

I know that a res itself doesn't passively cool the water, i meant in terms of total system water volume. I saw a res for sale once which was coated in aluminium heat sinks all over it and I just laughed. There is a link between volume of water and thermal capacity. If greater thermal capacity then greater cooling. I just don't know if it would be negligible on this scale. Q = mc * delta T, where Q is heat transfer, m is mass, c is specific heat constant of water, delta t change in temp.

Was going to post another thread about 'styling' (i have no artistic bones) but I guess I can ask here. Atm all that is set in stone in this rig is silverstone x560, assrock z77e-itx, clear 3/8 5/8 tygon tubing, ex240 and phobya 200. No idea how to make it look 'cool' lol. Do want some colour in there. Fittings etc I can change. Anyone has any ideas (or digital rendering skills) I can pay for that if you pm me

Click to expand...

You have a great grasp of the math involved. I don't. I been WC a lonng time.

I do know that if you increase the mass of the water your temps will be lower for a while. Been tests, rigs built that show even a 55 gallon drum in time will heat up to the same temps as a rig with a little res. Minus the surface area of a 55 gallon plastic drum set in the room with no active cooling. Like the early mineral oil coolers from Puget Sound PC.

It's about the surface area. A WC rad has thin/flat tubes that expose a LOT of the water to the fins on the rad. For example, a 55 gallon drum has a very very small surface are in relation to the amount of water. Like any res really, it's outside area is very small in comparison to the mass of the water.

A rad on the other hand has a very large surface area vs the mass of the water.

Thus resses are very ineffective in removing any heat, to the point of not measuring or worrying about in the WC world. I know it's an exercize about ideas, but not worth the trouble..

Styling is up to you and a learned skill plus looking at a bazillion posts and using others ideas.

The styling is a personal thing, a wonderful part of the hobby that is up to you.

baobeiiii said:

There is a link between volume of water and thermal capacity. If greater thermal capacity then greater cooling. I just don't know if it would be negligible on this scale. Q = mc * delta T, where Q is heat transfer, m is mass, c is specific heat constant of water, delta t change in temp.

Click to expand...

Despite conumdrum claiming he doesn't have a grasp of the math involved (haha...), his point is right. You aren't applying that equation to the system regardless.

The equation you mentioned holds true for heating a constant volume of water. To apply your equation to a loop, you have to look at it on a time basis, so you no longer have m, but dm/dt, which on a macro scale is simply your flow rate, which we generally convert to gpm because it's much easier to measure the volume pumped by the pump than the mass (and with only a degree or two change within the loop, density really isn't changing). With your pump pumping at a constant flow, dm/dt is now constant, regardless of a reservoir or not. So, res has no real impact form the point of having a greater volume of water (minus the initial startup temps), and like Conumdrum said, the heat transfer from the res to the air is going to be nil...

atclaeys, nicely explained!

The reason I want some help with styling, is i'm buying everything from frozencpu, shipping is like 100usd to China (I live in Shanghai) so thinking about that in advance = money saved.

I can't just buy stuff, hook it up and say 'oh it looks not too great' then order again. Got to do it right the first time. Why not just buy things in China you might wonder, well they don't have alot of these parts such as the phobya 200 rad.

Ton of effort getting the parts but it should allow me to overclock cpu and gpu well, the 240 up top plus what is almost equivalent in surface area to a 360 on the front

do you have a window on your case? if not, what is your plan to "display" all this style you speak of?

did you have any colors in mind? UV reactive tubing? thoughts on clean and simple?

i personally like when all the fittings are rightly angled(but not right angles, unless needed), tube runs are short and exact. i didn't have the $$$ for all the fittings so i went with black tub and red coil, see my sig for pictures.

PS: look here for ideas http://www.overclockers.com/forums/showthread.php?t=653729