Temperatures in relation to flow-rates. Data on several CPU-Blocks

[Hondacity]

great testing... i also have the supreme beating that gtz....

[TouGe]

Great post HESmelaugh. I enjoy these types of threads.

[skinnee]

180l/h = 1 GPM

the scaling is so good because it starts with such a low flowrate, testing needs to be extended to at least 360 l/h (2 gpm)

1 liter = 0.264172052 US gallons

180 l/h = 3 l/m

3 l/m = .79 GPM

So, 360l/h is 1.58GPM.

Your tests top out at 210l/h or 0.924 GPM...good tests for the market you're in. Thanks for sharing!

[majestik]

How do I get 2 GPM anyway? Will a dual-Laing do it? I'm hoping that I wont have to replace all the tubing since that's a lot of bother... -.-

You use an Iwaki, duh!

[NaeKuh]

You use an Iwaki, duh!

its big tho

I thought about posting that also, then realized he lives in Germany and would more than likely be revoked of his citizenship and forceably evicted from his country...

the two countries cant even agree on cars.

[Martinm210]

Exemplary testing, I really like the flow rate relation work...that's a huge undertaking and extremely valuable information. AWESOME!!!!!!!!!

[HESmelaugh]

You guys are crazy. :-)

The strongest pump I can find in local watercooling shops is the Laing D5 Vario. Those other monstrosities are from gardening-sections in DIY-stores, I assume?
A package containing a second MCP355 is on it's way and I'll see how far I can take flow-rates with that. If I don't quite reach a gazillion litres per second, or whatever 3 GPM converts to, I'll consider further options.

Thanks for the positive feedback everyone!

[skinnee]

If you can hit 475-485 l/h that is a right around the 2GPM mark. There is really no reason to go above 2 GPM, the only thing you get over 2 GPM is diminishing returns and more heat dump from the pump. I'm not certain at which point the heat dump starts netting losses, something I have been wanting to test for some time now...might just have to do that sooner than later.

[HESmelaugh]

Ok, so approximately 500 l/h would be the goal. With 253 l/h being the highest flow I got in my current loop (with the Apogee GTX), that means I would need to more than double the flow to still have enough when the EK Supreme is in the loop.
I'll have to redo the loop with fatter (and shorter, if possible) tubing. But it might just be achievable. So far, I'm planning to test the Supreme vs. the Apogee GTZ at very high flowrates. Make that very high flow-rates in German terms/barely adequate flow-rates for XS.
Should I also test different orientations with the EK Supreme? NaeKuh pointed out, that it apparently makes a difference. The Apogee GTZ has instructions on what the optimal orientation is and just out of curiosity, I tried a mount with it 90° rotated and really did see higher temps. But for the EK Supreme, I didn't find any specifications on how to mount it. I had it mounted like this:


If orientation-testing has already been done extensively with the EK Supreme, I don't see the need to redo this. But if optimal orientation is something we're still not sure about, I'll gladly do some testing for that too.

By the way: Are those Iwaki pumps not annoyingly loud?

[evilsponge]

MMM supreme... Yes it's true the supreme give different result per 90 deg rotation> I don't know why...

-sponge

[Bojamijams]

From what I hear around here HES, is that no two EK supremes are the same. Which is a strange issue in itself, but just because someone rotated it 90deg and got better temps doesnt' you will. You might have the better temps if you rotate 270.. or 180.. or 'normal'.. really you're gonna have to 4 tests and see which is best if you want to be REALLY thorough.

[HESmelaugh]

Hmmm... are we talking about significant differences here? Like, 0.5° or more? If we are, I think it's definetely worth testing.

[Martinm210]

Hmmm... are we talking about significant differences here? Like, 0.5° or more? If we are, I think it's definetely worth testing.

Yes it is...try it.

I tested one EK that liked the inlet on the left side, and another that liked it on the right side, but both preferred the nozzle alignment to match the core orientation as you have tested already.

I would just try one or two mounts 180 degrees from your current configuration and see which is better. If they are the same I wouldn't worry about it.

[G-UNIT91]

Yes it is...try it.

I tested one EK that liked the inlet on the left side, and another that liked it on the right side, but both preferred the nozzle alignment to match the core orientation as you have tested already.

I would just try one or two mounts 180 degrees from your current configuration and see which is better. If they are the same I wouldn't worry about it.

you have also tested ACS, it's true?

my friends they don't believe it

[HESmelaugh]

[x] Check different orientations with EK Supreme. :-)

Thanks for the input, Martin.

[G-UNIT91]

you have also tested ACS, it's true?

my friends they don't believe it

up

[NaeKuh]

[x] Check different orientations with EK Supreme. :-)

Thanks for the input, Martin.

im the one that told martin and, the entire forum for that matter, on the EK orientation, and made him and everyone else go OMGWTFBBQ

[Hondacity]

HESmelaugh

excellent job...what about this setup?

[HESmelaugh]

you have also tested ACS, it's true?

my friends they don't believe it

The Ybris A.C.S.? Yes, it's in the roundup. Black trendline on the graph. Here are the links to two pages with lot's of pictures showing that block:

Page 1

Page 2

HESmelaugh

excellent job...what about this setup?

Interesting. A bit confusing, though. Am I seeing this correctly: The first pump with the XSPC Res top send water directly to the second pump with the "normal" XSPC top, and from there, the water goes through the entire loop? Do you know what your flow-rate is in that setup?
I really like the look of it, and would love to see some more pics! :-)

EDIT: now I just saw a third pump in the background! Now I'm really confused. So is it Pump #1 with Res -> Pump #2 with Res -> Pump #3 with XSPC top -> Radiator #1 -> Radiator #2 -> System?

[G-UNIT91]

The Ybris A.C.S.? Yes, it's in the roundup. Black trendline on the graph. Here are the links to two pages with lot's of pictures showing that block:

Page 1

Page 2

I said to Martin ....

[HESmelaugh]

im the one that told martin and, the entire forum for that matter, on the EK orientation, and made him and everyone else go OMGWTFBBQ

Thanks for the tip, NaeKuh, in that case!
Didn't mean to discredit the source, just didn't know the source...

Cheers,
Shane

[Hondacity]

The Ybris A.C.S.? Yes, it's in the roundup. Black trendline on the graph. Here are the links to two pages with lot's of pictures showing that block:

Page 1

Page 2




Interesting. A bit confusing, though. Am I seeing this correctly: The first pump with the XSPC Res top send water directly to the second pump with the "normal" XSPC top, and from there, the water goes through the entire loop? Do you know what your flow-rate is in that setup?
I really like the look of it, and would love to see some more pics! :-)

EDIT: now I just saw a third pump in the background! Now I'm really confused. So is it Pump #1 with Res -> Pump #2 with Res -> Pump #3 with XSPC top -> Radiator #1 -> Radiator #2 -> System?

you are correct...

[Hondacity]

what was your room temp at that time? and are the temps measured on cpu idle?

[HESmelaugh]

The temperatures shown are the difference calculated from CPU-Temperature minus Water-Temperautre.
CPU-Temperatures are under full load (Everest stress test), measured and averaged out during a 20 minute time frame, after a 45 minute warmup.

The ambient temperature (more precisely the temperature of the air moving through the radiator) was between 21° Celcius and 22.55° Celcius during the test runs.

Hope this answers your questions. :-)

[Spankyfart]

Laing DDC 1T+ (Swiftech MCP355)

I am confused, isn't the 1T the MCP350.
The Laing DDC-1 plus is the MCP355 innit?

Or is this a regional thing.