Flowrates of Koolance 350 / EK Supreme / DTek FusionV2

Test Configuration

Pump1: Swiftech 355 with XSPC reservoir top
Pump2: Swiftech 355 with XSPC Acryllic top
Radiator1: HWLabs 480GTX
Radiator2: Thermochill PA120.3

http://i24.photobucket.com/albums/c5...4flowrates.jpg


other stuff

http://i24.photobucket.com/albums/c5...rblocccccc.jpg
Some notes:
- test was done purely for flow
- when I ran single pump, pump heat was greater compared to having two pumps in series. ( less stress on pumps in series?)
- water temp got warm when i had the rad fans off...18w is alot, goods fans and rad is a must. my overclocked q9550 eats around 95w(stock)+90w(additional watts consumed when oc'd)
- i still(or might) have to replace the pump with the swiftech 655 + EK X-top
- i still(or might) have to interchange rads mcr320 and/or pa120.3
- waterblock testing on I7 will come too(when newegg ships my stuff)

Thanks! Interesting results. I didn't know the Koolance block was more restrictive than the Supreme.

Test results are always appreciated :up: Might I ask how flow was measured?


For those not familiar w/ the Koolance 350
http://www.hardocp.com/images/news/1...cZzxIa_1_1.jpghttp://www.hardocp.com/images/news/1...cZzxIa_1_5.jpg

Wow... that is uber restrictive.

i used a flow meter

http://i24.photobucket.com/albums/c5...5/P1010908.jpg

Any chance to redo the test with a different rad like the Thermochill or Feser to see if it is the rad causing some of the restriction?

Quote:

---

Originally Posted by Utnorris

Any chance to redo the test with a different rad like the Thermochill or Feser to see if it is the rad causing some of the restriction?

---

Actually, it'd be better to remove the rad altogether to see actual flow rates of each block.

Quote:

---

Originally Posted by Waterlogged

Actually, it'd be better to remove the rad altogether to see actual flow rates of each block.

---

True, but how much does that matter when no one runs a loop without a radiator?

whew, man that block really slams on the brakes. It's less than half the flow with the same rad and pump compared to the DTek.

.8gpm is kinda low. Dual pumps with that setup looks like a smart move...or the uber Iwaki.

Quote:

---

Originally Posted by Polizei

True, but how much does that matter when no one runs a loop without a radiator?

---

Also very true....but, these results are for only 6 out of a myriad of possible combos. For better results testing should be stripped down to basic elements so the variables that could possibly skew results are limited.

I'm not saying this testing is worthless (it's actually pretty decent), I'm saying it only gives results for some specific layouts, which doesn't help everybody.

RD30 is coming too, I'll switch the radiator to a Thermochill tonight (Utnoriss' request)

Quote:

---

Originally Posted by Hondacity

- when I ran single pump, pump heat was greater compared to having two pumps in series. ( less stress on pumps in series?)

---

I'm probably stupid, but what do you mean here? And how was this measured?

This is probably a really dumb question, but isn't the midplate installed 45 degrees off here?

Wow, that is some restrictive block !! :shocked:
I think we got a new block entitled as a flow lover :rofl:

Thanks for the test results Hondacity :up:

Quote:

---

Originally Posted by warriorpoet

This is probably a really dumb question, but isn't the midplate installed 45 degrees off here?

---

Looks like it to me but IDK.

Quote:

---

Originally Posted by warriorpoet

This is probably a really dumb question, but isn't the midplate installed 45 degrees off here?

---

I think it can only go on one way. You see the little hole in the upper right corner and the little pin that aligns it.

Quote:

---

Originally Posted by Hondacity

Test Configuration

Pump1: Swiftech 355 with XSPC reservoir top
Pump2: Swiftech 355 with XSPC Acryllic top
Radiator: HWLabs 480GTX

http://i24.photobucket.com/albums/c5...ump355xspc.jpg

Some notes:
- test was done purely for flow
- when I ran single pump, pump heat was greater compared to having two pumps in series. ( less stress on pumps in series?) - water temp got warm when i had the rad fans off...18w is alot, goods fans and rad is a must. my overclocked q9550 eats around 95w(stock)+90w(additional watts consumed when oc'd)
- i still(or might) have to replace the pump with the swiftech 655 + EK X-top
- i still(or might) have to interchange rads mcr320 and/or pa120.3
- waterblock testing on I7 will come too(when newegg ships my stuff)

---

Came to the same conclusion and told dean this a while ago.
However it should be the other way around im guessing. (more restriction does not mean more work on the pump impeller.)

Didnt honestly think the KL-350 would be more restrictive then the supreme.

i just added the pa120.3

enjoy:)

Yeah very nice honda.. thanks for your flow testing..

Really suprised me how restrictive it is.

NOTE:
Blah sorry to derail this thread, but seeing how a lot of noobs will read this and still not know what cpu block to get.

How many people would want me to write out a nice short note on how to pick out the proper cpu block?
You guys are ignoring classes in blocks when picking, and were back in the 2006 when everyone got a Storm without knowing what it was and got worse temps then a apogeeGT.

Quote:

---

Originally Posted by NaeKuh

How many people would want me to write out a nice short note on how to pick out the proper cpu block?
You guys are ignoring classes in blocks when picking, and were back in the 2006 when everyone got a Storm without knowing what it was and got worse temps then a apogeeGT.

---

I would really like to see that. I think I have a decent handle on the current blocks but, I still consider myself a noob and I think this would benefit many people here including myself.

Quote:

---

Originally Posted by faster3200

I would really like to see that. I think I have a decent handle on the current blocks but, I still consider myself a noob and I think this would benefit many people here including myself.

---

You need to match the block to the loop. CPU only loop it's ok for the block to be restrictive. Multi-block loop such as chipset, GPU and CPU you should get a low restriction block. IMO of course.

these only show flowrates...

i will test them on the i7 when it arrives, hopefully dtek will have the mounting bracket for the fusionv2...

at the moment i have this q9550...and been waiting for a good time thermal testing these three

and i'm contemplating on cpu+gpu results, i haven't seen real data on how flow rates affect the dtek gfx waterblock..

Quote:

---

Originally Posted by warriorpoet

This is probably a really dumb question, but isn't the midplate installed 45 degrees off here?

---

actually no.

theres a dot whick you allign the base plate with the mid plate.

I think a lot of ppl are missing the recessed channel at the end of the grooves that the water dumps into to be whisked to the block outlet.

Quote:

---

Originally Posted by Waterlogged

I think a lot of ppl are missing the recessed channel at the end of the grooves that the water dumps into to be whisked to the block outlet.

---

I know I missed it until you pointed that out. :(

Wow, I never thought the 350 was going to be more restrictive than a Supreme.

Quote:

---

Originally Posted by NaeKuh

actually no.

theres a dot whick you allign the base plate with the mid plate.

---

Looks like you'd get a heck of a lot better flow 45 degrees away.

Quote:

---

Originally Posted by warriorpoet

Looks like you'd get a heck of a lot better flow 45 degrees away.

---

hmmm i see where your getting at with this statement.

ummmm OP, how is the supreme less restrictive then the D-tek V2 on a dual pump?

:rofl:

I think WL is correct.. You need to pull everything off the loop but the block and pump solo. Your last 2 graphs make no sense..

Quote:

---

Originally Posted by NaeKuh

hmmm i see where your getting at with this statement.

---

TEST! :D

Quote:

---

Originally Posted by NaeKuh

hmmm i see where your getting at with this statement.

ummmm OP, how is the supreme less restrictive then the D-tek V2 on a dual pump?

:rofl:

---

Actually, now that I look at it, water wouldn't flow all the way down all the channels is it were simply rotated, but I bet a more open outlet would help things quite a bit. Like this:

Quote:

---

Originally Posted by NaeKuh

hmmm i see where your getting at with this statement.

ummmm OP, how is the supreme less restrictive then the D-tek V2 on a dual pump?

:rofl:

I think WL is correct.. You need to pull everything off the loop but the block and pump solo. Your last 2 graphs make no sense..

---

1. ...2 graphs make no sense? elaborate on your statement.

2. supreme is less restrictive than the fusionv2? i don't see that...where is that?

I have the 655+ EK X top on the loop...will post some numbers

I just added the 655+xtop

Quote:

---

Originally Posted by Hondacity

1. ...2 graphs make no sense? elaborate on your statement.

2. supreme is less restrictive than the fusionv2? i don't see that...where is that?

---

oh woops.. i could of sworn you had the supreme outflowing the fussion.

wow 2 pumps and it still can't break 1.5GPM w/ 480gtx!!!

that koolance block is hella restrictive, but I wonder at what GPM it reaches it's peak efficiency before diminishing returns.

Needs temperature results

If flowrate is a concern, we also have the CPU-345 available. On our product pages we also supply a flow-rate to pressure drop graphs as well.

http://www.koolance.com/water-coolin...u-350ac_o0.gif

http://www.koolance.com/water-coolin...u-345at_o0.gif

Quote:

---

Originally Posted by Hondacity

i used a flow meter

http://i24.photobucket.com/albums/c5...5/P1010908.jpg

---

can you link me where to get one?

Quote:

---

Originally Posted by Hybridchemistry

Needs temperature results

---

+1. ANY tests such as flow only matter as far as they impact temps. I can make a loop that will blow away anyone's loop here in terms of flow . . .the downside is that it would be a loop of the pump inlet to outlet with one piece of tubing. It would not cool the CPU block ;)

Hats off to the OP. Please add to your great testing bud!

i'll search my emails..then pm you the link, i'll include the fittings that you need also....

oh yeah temps on i7 will come when my i7 comes..newegg says thursday...

Quote:

---

Originally Posted by [Koolance] Dean

If flowrate is a concern, we also have the CPU-345 available. On our product pages we also supply a flow-rate to pressure drop graphs as well.

---

Dean, I just want to publicly say that I'm impressed by the way you guys are changing your product line, posting flowcharts that aren't total BS, eliminating aluminum, etc. You guys are obviously busting your butts and it shows. Thanks!

Quote:

---

Originally Posted by headala

Dean, I just want to publicly say that I'm impressed by the way you guys are changing your product line, posting flowcharts that aren't total BS, eliminating aluminum, etc. You guys are obviously busting your butts and it shows. Thanks!

---

I agree. It's been a long time coming, but I'm glad to see things changing. I can only hope the guy at Koolance that created the craptastic article about Koolance radiators 2 years ago is fired.

Quote:

---

Originally Posted by ranker

I agree. It's been a long time coming, but I'm glad to see things changing. I can only hope the guy at Koolance that created the craptastic article about Koolance radiators 2 years ago is fired.

---

everyone needs a second chance..

:rofl:

Ranker honestly, when have you seen me use second par eq on my setups?

Great post, thanks HondaCity...very informative....

I wonder where Swiftech GTZ would fall on this graph (b/n Fuzion and Supreme maybe?)

@anzial yes very close to that..if you interpolate martins data it would be in between the ek supreme and the fusion....

Great post. I would have like to have seen the Aqua Di.

DTek Gfx2 + 355 + xspc top + 480gtx will be added today

DTek Gfx2 + 355 + xspc top + 480gtx added today

+ Dtek fusionv2 + Gfx2 + 480gtx + 355 + xspc top

Quote:

---

Originally Posted by Hondacity

DTek Gfx2 + 355 + xspc top + 480gtx added today

+ Dtek fusionv2 + Gfx2 + 480gtx + 355 + xspc top

---

:up:

keep up the good work!

i currently have the mcr320 on the setup....seems to me martins testing and my data are abit off... anybody with confirmation?

i have the 2nd and 3rd pump...all are reproducing the same values on the mcr320....

Quote:

---

Originally Posted by Hondacity

i currently have the mcr320 on the setup....seems to me martins testing and my data are abit off... anybody with confirmation?

---

:X

no... then id say something is throwing you off. :X

trust me i know martin and his methodology. Hes dayam accurate.

He's an engineer for the government.

mcr320 data added

*tried with 3 different pumps...different waterblocks...

my mcr320 has no dents, its a single pass radiator...even if the one pipe is dented it couldn't cause this much restriction. martin might have a different mcr320 than mine...

Can you test the 350 flowrate without the plastic mid plate? I would bet that with a proper nozzle that the 350 would perform better without the plastic piece on single slug with a nozzle.

Quote:

---

Originally Posted by nikhsub1

Can you test the 350 flowrate without the plastic mid plate? I would bet that with a proper nozzle that the 350 would perform better without the plastic piece on single slug with a nozzle.

---

I'm willing to bet you're right.

Quote:

---

Originally Posted by warriorpoet

I'm willing to bet you're right.

---

I've been known to be wrong once or twice :D My brain just doesnt see the function of the plastic bit... it only causes unnecessary restriction. On dual slug 'maybe' but i7, I don't see it. Flow rates would be better too.

Quote:

---

Originally Posted by nikhsub1

I've been known to be wrong once or twice :D My brain just doesnt see the function of the plastic bit... it only causes unnecessary restriction. On dual slug 'maybe' but i7, I don't see it. Flow rates would be better too.

---

For rev.2 I think it'd probably work better if the plate (if necessary) were threaded into the top. They'd get a much freer outlet path and still have some degree of control over directional flow. Bonus: no inter-plate leakage.

Before launch we did try many different shaped and sized holes on the accelerator plate, what is being produced is what yielded the best overall temperatures. But due to feedback from many enthusiasts we are making available blank impingement plates here so anyone can try their hand at improving the block!

Quote:

---

Originally Posted by [Koolance] Dean

Before launch we did try many different shaped and sized holes on the accelerator plate, what is being produced is what yielded the best overall temperatures. But due to feedback from many enthusiasts we are making available blank impingement plates here so anyone can try their hand at improving the block!

---

im actually thinkn of making a series of tiny holes instead of 1 big hole.

Seeing if having focused jets instead of 1 big jet will be any better.

Quote:

---

Originally Posted by ranker

I agree. It's been a long time coming, but I'm glad to see things changing. I can only hope the guy at Koolance that created the craptastic article about Koolance radiators 2 years ago is fired.

---

You know what was the most interesting thing about that whole debacle? The fact that the competitors who screamed bloody murder and said how much bs the koolance tests were... and said how they were going to prove it... Well guess what the comparison tests they promised never ever materialized. That spoke volumes :)

Anyway, as Dean noted earlier the 350 block is designed as a more restrictive block and is stated on the site. The 345 is a higher flow block, also as stated. It would be interesting to see that tested as well.

Cheers

Quote:

---

Originally Posted by NaeKuh

im actually thinkn of making a series of tiny holes instead of 1 big hole.

Seeing if having focused jets instead of 1 big jet will be any better.

---

As I have said before, the more the industry tried to run from Cathar's designs the more the engineering brought them back. :)

I finally giving in and buy a Core i7 setup, I just can't fight the sickness any longer. I will do one last shabang on LGA775 CPU blocks before moving to LGA1366. Plus, I was getting close to violating my contract with the XS WCG team. :D

I am very excited to start testing a couple new LGA775 blocks, and just geeked to finally see what the LGA1366 blocks have to offer.

Quote:

---

Originally Posted by [Koolance] Dean

Before launch we did try many different shaped and sized holes on the accelerator plate, what is being produced is what yielded the best overall temperatures. But due to feedback from many enthusiasts we are making available blank impingement plates here so anyone can try their hand at improving the block!

---

:up:

Quote:

---

Originally Posted by skinnee

I finally giving in and buy a Core i7 setup, I just can't fight the sickness any longer. I will do one last shabang on LGA775 CPU blocks before moving to LGA1366. Plus, I was getting close to violating my contract with the XS WCG team. :D

I am very excited to start testing a couple new LGA775 blocks, and just geeked to finally see what the LGA1366 blocks have to offer.

---

I am glad to see that the counseling sessions have finally paid off. :up: :ROTF:

Take two quads and call me in the morning. :D

Quote:

---

Originally Posted by SiGfever

As I have said before, the more the industry tried to run from Cathar's designs the more the engineering brought them back. :)

---

thats why i keep saying the guy is a legend.

:up:

Quote:

---

Originally Posted by skinnee

I finally giving in and buy a Core i7 setup, I just can't fight the sickness any longer. I will do one last shabang on LGA775 CPU blocks before moving to LGA1366. Plus, I was getting close to violating my contract with the XS WCG team. :D

I am very excited to start testing a couple new LGA775 blocks, and just geeked to finally see what the LGA1366 blocks have to offer.

---

good to hear...:yepp:

Quote:

---

Originally Posted by santiagodraco

You know what was the most interesting thing about that whole debacle? The fact that the competitors who screamed bloody murder and said how much bs the koolance tests were... and said how they were going to prove it... Well guess what the comparison tests they promised never ever materialized. That spoke volumes :)

Anyway, as Dean noted earlier the 350 block is designed as a more restrictive block and is stated on the site. The 345 is a higher flow block, also as stated. It would be interesting to see that tested as well.

Cheers

---

Ahh the old KL fanboy. The tests were BS. Cathar even pointed it out itself. KL did a good job by just sweeping the whole ordeal under the rug and starting over at square one. The new 350 block is a fantastic step in the right direction. Not only have they gotten rid of their ALU crap, but they've also pushed the performance envelope. That gets two thumbs up from me.

However, you'd be a fool to think that KL's test conditions on their radiator debacle weren't comical.

Quote:

---

Originally Posted by SiGfever

As I have said before, the more the industry tried to run from Cathar's designs the more the engineering brought them back. :)

---

Makes me smirk every time I see this play out. =)

Is the LT a I7 block only? Because it would seem thats a better choise than the Fuzion V2

its for 775 and 1366

read this thread...thanks

http://www.xtremesystems.org/Forums/...d.php?t=217412