Reservoir, How big to do whatever it does best

[croak5]

Hi,

I'm putting together a WC system. I've had a couple of threads going here and there about reservoirs, specifically what's the point of having one. Lot's of people just use a "T" .

The feed back was for the most part inconclusive so for convenience I figured I'd do both, easy enough.

My dilemma is I've got a pretty big reservoir, takes up 2 drive bays. Whereas some systems have a tiny reservoir built onto the pump (and of course some just have a "T").

My questions is to gain whatever pluses a reservoir gives you, is there some calculation to determine what it's size should be?

Ive got a 1 dangerous den CPU block, a 450 gph pump, basic 2 inch thick (single)120mm fan radiator, and this inconvenient reservoir

Also, any body know of a nice filler/ cap thingy for a "T" system (like to fill from the

top of the case)


Thanks, Croak

P.S. Space is a big factor because I'm trying to stuff everything (including my gig ds3) into a modded baby atx case

[jabski]

Also, any body know of a nice filler/ cap thingy for a "T" system (like to fill from the

top of the case)




here (http://www.dangerden.com/store/product.php?productid=56&cat=7&page=1)

[JeffnWV]

reservoir can be any size...just depends on how much more warm water you want to have sitting in your loop that contributes to warmer water temps....

just use a T line at the top of the loop and keep 2" of water in it.

[Sideroxylon]

reservoir can be any size...just depends on how much more warm water you want to have sitting in your loop that contributes to warmer water temps....

just use a T line at the top of the loop and keep 2" of water in it.

A res doesn't contribute to warmer or cooler temps. A res makes filling and bleeding much easier than a t-line. But makes no difference in performance. It really just comes down to preference. Personally, I wouldn't build a wc'ing loop without one.

[DepTi]

I got my reservoir because I wanted it to be easier to fill and bleed, but I also wanted it for the asthetic look.

[JeffnWV]

A res doesn't contribute to warmer or cooler temps. A res makes filling and bleeding much easier than a t-line. But makes no difference in performance. It really just comes down to preference. Personally, I wouldn't build a wc'ing loop without one.

A res does too contribute.. it contains warm water. That additional water helps retain the heat in the water which means the cooling components must deal with the additional load.
Reduce the amount of water in the loop and it cools better/more efficiently.

[JeffnWV]

I got my reservoir because I wanted it to be easier to fill and bleed, but I also wanted it for the asthetic look.

yea the looks are about the only benefit that I can reason.

Filling either a res or T is still the same process.. all a res does is hold a little more fluid per fillup (still have to repeat filling it up a couple times). I just connect a 2 foot piece of tubing to my T line... holds as much water as a small res...and for me, the T line is better because I can use my shut off and close the loop for better initial air bleeding (moving the case around and not get air back in the system...can't do that with a res)

[Polizei]

But at the same time, it takes longer to heat up that larger amount of water.

Waterchillers are usually built with having a large amount of water in mind... takes a long time to pull the temp down to 0C or -10C or whatever operating temp is, but once its that low, the water temp doesnt fluctuate because there is so much water to heat up.

[lowfat]

A res does too contribute.. it contains warm water. That additional water helps retain the heat in the water which means the cooling components must deal with the additional load.
Reduce the amount of water in the loop and it cools better/more efficiently.

I'd like to know where you get your information. A res should have no influence temperatures at all.

A person could say the exact opposite as you and state that having more water in the loop allows the temperatures more time to cool down, which is also incorrect.

[JeffnWV]

I'd like to know where you get your information. A res should have no influence temperatures at all.

A person could say the exact opposite as you and state that having more water in the loop allows the temperatures more time to cool down, which is also incorrect.

I reallyWON'T take the time to explain this...sheesh

[ranker]

A res does too contribute.. it contains warm water. That additional water helps retain the heat in the water which means the cooling components must deal with the additional load.
Reduce the amount of water in the loop and it cools better/more efficiently.

I've never heard of this either and would love to hear the math/physics breakdown for this theory.

[lowfat]

I've never heard of this either and would love to hear the math/physics breakdown for this theory.

exactly.

[Gabriel Noraa]

Based on my own experience with both, res and 't', there is no noticeable performance hit. As I like having 2 loops sharing the res, I had to increase the size a couple of times to avoid the 2 pumps sucking air from the return line. If you only use one pump I prefer 't', takes more time to bleed but also there is less space required. I didn't find a significant performance difference from one option to the other, the best way it's trying and find what you like. In the forums you always find a lot of help but also many contraditory opinions. I would say, try both and decide.

[JeffnWV]

ok...the guys got a single 120mm rad...it's needs as little water as possible to cool.
The physics are obvious. the water is in the radiator for only so long a period of time...the more that must be cooled in a small radiator the less efficient the system is. Granted the res won't make THAT much of a difference, but the principle is obvious...

[lasse.j5]

Dude you can't put it that way.

Think of it. Then the cpu also has more water to warm up?

The things you are saying doesn't make sense.

[nibble]

A res does too contribute.. it contains warm water. That additional water helps retain the heat in the water which means the cooling components must deal with the additional load.
Reduce the amount of water in the loop and it cools better/more efficiently.

So warm water helps to retain heat in water? Reservoir size has no impact on cooling performance PERIOD.

[JeffnWV]

OK then lets just agree that a single fan rad can cool a bathtub amount of water to the same temp as it can a cup of water.

I give !

[serialk11r]

Huh? WTF...
Bigger res=more water=higher heat capacity. However, after a while the water will still become warm. So it does affect performance, just not very much. I wouldn't worry about it.

[lasse.j5]

OK then lets just agree that a single fan rad can cool a bathtub amount of water to the same temp as it can a cup of water.

I give !

Yeah lets agree on that because it can. It just takes longer time which is irrelevant in this matter.

[JeffnWV]

Huh? WTF...
Bigger res=more water=higher heat capacity. However, after a while the water will still become warm. So it does affect performance, just not very much. I wouldn't worry about it.

nope, I am NOT right.. so don't try to convince me

[LMG03]

anyone want to Comment on the performance of XSPC 250mm Passive Aluminium Reservoir, acording to Tweakup.dk a danish site this passive reservoir is effektiv if used correctly.
if it is true i think that it could be helpfull in this case, if Croak5 have the space for it.

http://www.watercoolingshop.com/catalog/product_info.php?cPath=28&products_id=430

the test:
http://www.tweakup.dk/article/1128/dk/

[Kurz]

Having more water in a Res does not matter...
Since water takes the shortest route between two points.

From nozzle to nozzle.
The extra water is there just in case.

[serialk11r]

anyone want to Comment on the performance of XSPC 250mm Passive Aluminium Reservoir, acording to Tweakup.dk a danish site this passive reservoir is effektiv if used correctly.
if it is true i think that it could be helpfull in this case, if Croak5 have the space for it.

http://www.watercoolingshop.com/catalog/product_info.php?cPath=28&products_id=430

the test:
http://www.tweakup.dk/article/1128/dk/

Aluminum=[explosion]

[Colossous]

anyone want to Comment on the performance of XSPC 250mm Passive Aluminium Reservoir, acording to Tweakup.dk a danish site this passive reservoir is effektiv if used correctly.
if it is true i think that it could be helpfull in this case, if Croak5 have the space for it.

http://www.watercoolingshop.com/catalog/product_info.php?cPath=28&products_id=430

the test:
http://www.tweakup.dk/article/1128/dk/

Just taken delivery of one of these babies today, got to say i bought it more for looks than performance, its going into my mm UFO build and i think will look pretty sweet !! As for performance, i did'nt really expext it to help a lot but if it does then cool !! :ROTF:

http://i83.photobucket.com/albums/j303/richmawr/PICT0029res2.jpg

[SparkyJJO]

anyone want to Comment on the performance of XSPC 250mm Passive Aluminium Reservoir, acording to Tweakup.dk a danish site this passive reservoir is effektiv if used correctly.
if it is true i think that it could be helpfull in this case, if Croak5 have the space for it.

http://www.watercoolingshop.com/catalog/product_info.php?cPath=28&products_id=430

the test:
http://www.tweakup.dk/article/1128/dk/

The theory behind it would work, but the problem is the aluminum. Al + Cu in a water loop = corrosion. If it was brass or copper it would be better.

Oh, and welcome to XS :welcome:

[LMG03]

Well thanks sparky, but tjek my Join Data.

[zdog]

run a rez and put a fan on it. ive had my eye out for what i can use. like a really high flow ineffient radiater. it acts like a rez with a little cooling properties.

[ranker]

run a rez and put a fan on it. ive had my eye out for what i can use. like a really high flow ineffient radiater. it acts like a rez with a little cooling properties.

Can acrylic even be cooled by just a 120mm fan pointed at it?

[serialk11r]

lol that won't work...
If the res was a piece of copper tube with fins brazed on...maybe.

[SparkyJJO]

Well thanks sparky, but tjek my Join Data.

lol I see that now :p: I just saw that it was your first post heh. Welcome to posting then :D

[tw33ter]

lol that won't work...
If the res was a piece of copper tube with fins brazed on...maybe.


hmmmmm... passively cooled rez, you may be on to something... :fact:

[serialk11r]

hmmmmm... passively cooled rez, you may be on to something... :fact:

Aye, but it'll be expensive. Copper aint cheap :( Might be possible to get a ton of copper tubes, cut them up, and SOMEHOW braze tanks on. It'll look like a rad without fins and terribly large tubes...of course it won't be effective, it'll be ugly as crap, and it'll probably be hard as hell to bleed but...well I was thinking about making a bong cooled res in the future for asthetics AND cooling but well...that's in the future.

[Friedchicken]

For anyone that still thinks a res has an effect on temps, consider this...

a cup of 30C water or a gallon of 30C water... guess what- it's still 30C. What does matter, is the density of the water and the thermal dynamics of the fluid.

A radiator cooling a 4L loop to 30C is not going to lose effectiveness if it's attached to a 10L loop. It can only draw so much heat per volume within the radiator (not the loop volume).

That is why people never comment on shortening the loop to decrease temps- only in terms of water pressure.

[ranker]

For anyone that still thinks a res has an effect on temps, consider this...

a cup of 30C water or a gallon of 30C water... guess what- it's still 30C. What does matter, is the density of the water and the thermal dynamics of the fluid.

A radiator cooling a 4L loop to 30C is not going to lose effectiveness if it's attached to a 10L loop. It can only draw so much heat per volume within the radiator (not the loop volume).

That is why people never comment on shortening the loop to decrease temps- only in terms of water pressure.

This is the "money" post folks. Bookmarked for future usage.

[serialk11r]

This is the "money" post folks. Bookmarked for future usage.

Well ranker, lets say you have a rad that can keep the water temps at 2C above ambient with 100W of load (one hell of a rad :D). When your PC is heavily loaded lets say it puts 400W into the water. To heat up more water, its going to take longer. Say you were gaming for a sec, then left your PC idle. Your temps will initially pull down faster, because the water had less time to heat up, but the smaller res will "catch up" later because it had less water. But the difference should be TINY. And besides, a T causes less restriction than a res, because of obvious reasons (res pretty much wastes pumping power at the outlet, T only has some turbulence at the place where the fill line comes in, and maxxx said he measured less restriction with Ts than res) so you can probably expect a T to perform a TINY bit better because of the SLIGHTLY higher flow. The differences should be so small you can't see them, unless you get really Xtreme and throw in a 10 gallon res :D. So lets put it this way. Under NORMAL circumstances, given enough time, the difference between a bigger res and a smaller res will be very small.

[Agent11]

If you had an insanely huge res.. lets say 500gallon, the res would act as a heat dump.
As the warm water that just came through the loop hits the idle water it disipates the heat it carries into the cooler water in the res.

If you kept your pc at max load for a very long time and the rad was inefficient though, I could see you hitting a point when the temp of the water in the res could hurt temps.

It isn't the fault of the res though. The problem is that you have need a better rad, and not because you have the bigger res.

Because the rad couldn't cut it the res was compensating. As the water in the res warms up to the same temperature as the rest of the water in the system you realize your rad is :banana::banana::banana::banana:.

It would be an interesting experiment. I'm sure the delta's would be interesting.