pump not up to the job always buy bigger and better.. hmmmmn

[Tattysnuc]

Phew - that's a lot to read through!

Surely the amount heat that can be removed by a watercooling rig depends on the amount surface area where heat is removed and the time that the fluid spends in contact with those cooling surfaces?

Let's consider a hgigh flow setup - As coolant fluid is put in contact with the heatsource - it spends very little time in contact with that heatsource (CPU block etc) then the fluid is only heated a little (due to the brief contact and the thermal conductivity of the fluid)

However the reverse also stands true - when the fluid then passes to the radiator - high flow passes through the radiator spending little time being cooled.

If the pump is a low speed one then the temperature of the coolant fluid would rise more when in contact with the heat source, and drop more over the radiator. Either way, if there is insufficient cooling we would see this manifesting as high component temperatures, with a system that quickly reaches equilibrium.

I see water cooling rigs as a balancing act - balancing flow rate with fan speed and radiators - aka cooling capability vs the heating element of the setup. ie high flow "rated" blocks work better with a higher flow rate pump/fan/rad setup.

Now, back to your original point, why are we being pointed toward higher power/capacity pumps? I believe that is purely and simply marketing. After all, how many people at Koolance, EK, XSPC, or even our beloved OCUKhave degrees which include an element of thermodynamics?

I doubt that most of them have a working knowledge of the subject matter. I've forgotten more than I care to remember.

 

[Tattysnuc]

I hope that doesn't sound like a brain fart. I know what I mean, but am full of flu at the moment.

 

[craarc]

I hope that doesn't sound like a brain fart. I know what I mean, but am full of flu at the moment.

arrgh me to in work after not sleeping hardly cause of a damn sore throat and a cough .

yup quite a bit there theorising hard to explain what you mean sometimes and then people read it in a way that you think nope i didnt mean it like that ..
as i see it tho
basically its not about time in contact with the heat source to remove more heat its more important to have a higher flow over the surface i beleive.

below taken from an article

"The reason higher flow rates work better in computer water cooling is this:
There is more water with a larger temperature differential moving through the water block- this removes more heat.
There is more water with a larger temperature differential moving through the radiator- again removing more heat.

It is the Heat Transfer that we want to maintain as efficiently as possible, and that is best done with a higher flow rate. Rather than thinking that there won't be enough time for heat to move towards the cool water, and therefore compromising heat loss, it is better to think that there is more fresh water moving onto the CPU and therefore, there is increased cooling"

so its not about time in contact as in longer time makes it better, but less time more flow is better as i see it.

high flow rate will give better but a little pump like an ek2.2 give plenty im thinking.

i suspect it more to be marketing also to a certain extent.

i do think we could get adequate performance from equipment that we all ready have if we maximise its potential rather than jump on the upgrade badwagon to quickly sometimes

however as pc freeks we do always strive for the latest fastest highest performance we can squeeze out of something

 

[craarc]

just came across this info

The EHEIM 1250 is rated for 317 gph, or 5.28 gpm, with a delivery head of 6’7″. It features a ¾” intake and ½” on the output side. As soon as you hook up hoses etc., flow rates² drop quickly:


Flow Test Measurements – EHEIM 1250

Setup GPM
% of Spec

Spec Rating 5.28
100%

1/2″ Tubing 2.62
49.6%

3/8″ Tubing 2.04
38.6%

Waterblock Alone 1.54
29.1%

Radiator Alone 1.35
25.6%

GPU Waterblock Alone 0.81
15.3%

WB + Rad 1.13
21.4%

WB + Rad + Gpu 0.67
12.7%



Note: 3/8″ tubing was used with radiator and waterblocks, which feature 3/8″ nipples.

Obviously this is only for this setup and does not address pressure, but I think it’s representative of typical watercooling systems many of our readers are using. Whatever waterpump you have, figure at a minimum actual waterflow through your system is something like 15 – 30% of the pump’s rating with just a radiator and waterblock.

The more that’s included in the cooling loop, including GPU waterblocks and 90 degree bends, the more it reduces waterflow. In addition, the smaller the tubing diameter, the lower the flow. Waterblocks, GPUs and radiators with anything below 3/8″ fittings are not conducive to high flow rates.


CONCLUSIONS

Overall, the larger the diameter, the better. However, I feel that anything over 1/2″ tubing is difficult to work and fit, but anything below 3/8″ ID will severely compromise flow rates. Placing a small diameter GPU in a large diameter cooling loop will slow things down appreciably.

¹King Instrument Company – Stainless Steel, Liquid Flowmeter. It has 1″ Female NPT connections, both in and out, and has viewing windows on two sides. The overall length is 19-1/2″ long, with the flowtube box, 14-1/2″ long x 4-1/2″ square. The viewing windows are 2-1/2″ x 11″, and the scale for the flowmeter, is 0-3.0 GPM, marked in 0.05 GPM divisions. The float is # 3/4-GPV-S2 for a liquid with a specific gravity of 1.0 (water).

²I should also note that the instrument itself will impact flow rates (5″ H2O).

interesting also

Even though you use 1/2″ ID tubing throughout your system, the short piece that connects the pump suction to the reservoir can be 5/8″ ID or 3/4″ ID. The larger the tubing ID, the lower the water velocity and therefore the less flow resistance or pressure drop. This in turn minimizes the pumps NPSH requirements.


from that above i would say bigger the better for pipework and fittings (especially just on suction side)to keep your pump near its maximum flowrate as they seem to drop of a hell of a lot.

of for more research

look at the flow in this baby http://youtu.be/vTnFbHaUAnk

 

[rjk]

Aren't you demanding quite a lot from a puny 2.2 pump?

nope not at all

Cleee is right.

having loads of spare tubing etc just hanging around is not the best thing especially if there is any component in the loop that impinges flow a lot, like a dense radiator or a low flow waterblock. it looks like your tubing layout is long too rather than going for the shortest route as would be the suggested method.

the 2.2 is a good pump and does the job for simple loops, but from the looks of your pictures, what i can make out of them anyway, is that your loop would maybe be better suited with a DDC or similar

 

[craarc]

pump is doing fine not a murmor and it hardly gets warm but
yer im probably going to get another pump soon anyway to try a few things out but atm the 2.2 is throwing it around great stuff got lots of fluid turbulance in the top resevoir so im not really worried about the pump and my temps are fine sometime hit 31c when gaming

im thinking of making a multi resevoir loop with maybe 5 or 6 resevoirs at diferent heights with various fluid levels a bit like this so the base of the res are in the desk along with all pipework and they flow into each other filling at diferent levels

as long as the tops have a good seal the air trapped will buffer the height of fluid in each and create a nice flowing effect and would look nice in diferent colors also so 2 loops combined maybe, and a couple leds in there somewhere

 

[craarc]

Cleee is right.

having loads of spare tubing etc just hanging around is not the best thing especially if there is any component in the loop that impinges flow a lot, like a dense radiator or a low flow waterblock. it looks like your tubing layout is long too rather than going for the shortest route as would be the suggested method.

the 2.2 is a good pump and does the job for simple loops, but from the looks of your pictures, what i can make out of them anyway, is that your loop would maybe be better suited with a DDC or similar

hence the reason to go big tubing not small, even with all the tubing all my fittings etc the pump is doing it easiliy there is plenty of flow and my temps are fine water is flying around and the tubing acts as extra resevoir volume.

if it were small bore it would offer more resistence not what we want so with larger bore you can get higher flow rates at lower pressure

back to the hose pips analagy in the first post

1 open a hose pipe up and u get high flow but low pressure..

awesome cooling for a pc

2 close the hose pipe as in make it a nozzle put your finger over the end making it high pressure = low flow and bad for pc cooling.

but so many people talk of high flow systems with restrictive flow and small bore tubing..

so can we have a 25mm id cooling block and a 25mm id pipe system please


thats the point

im getting high flow and good cooling just with a cheapo pump £30 odd quid

going to get a few new bits next month hopefully to mess with and will see what i can make

really want a block that will take 3/4 inch + tubing and fittings it would be easy to modify a pump out let to take larger bore stuff or modify the impellor and make larger holes in the resevoirs, im thinking of making a res out of acrylic that will hold a few litres tho so will see.. but making a block would be more difficult i guess

 

[Biffa]

You need low flow blocks, thats a given.

Yep, biggest problem you have is the largest bore in your loop is going to be the holes where your fittings attach. So G1/4 is the thread size, the internal diameter of nearly any attachment screwing into that hole is 10/11mm

So no matter if you put a 2ft wide pipe on the end of your attachment, the fluid still has to go through a 10mm hole to get into and out of the block/rad/res/pump.

Good luck finding 25mm bore fittings, but again its a good project if you have the time, energy and commitment

 

[craarc]

yup not only thinking of going up in pipe size but modifying fittings also so wont be limited by 1/4 bsp thread sizes looking at some size options now then need a block design with huge base area that can be adapted rads and resevoirs and pumps can be sorted easily enough its fitting it onto a block will be hardest may have to make a block not sure yet

along these lines so instead of a shallow block with 1/4 bsp fittings a deeper one that will take larger bore fittings as if its not deeper than a normal block it wont take the flow that the larger pipes can provide the trick will be making sure that as much of the volume as poss passes over the vanes while not restricting flow but i think this will be easier with a higher flow in the first place also the vanes will be a lot longer than in a traditional block etc