• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Koolance EXOS 2?

M

MapleSyrupMods

Bring on Titan X Volta
Joined
Oct 3, 2002
Messages
1,224
I am currently looking to invest in the Koolance EXOS 2 cooling system. Is this a good wc setup? I have looked at some reviews and they say its good, but how would this system fair against others?
 
It is a very good watercooling system. But you can do better if you build a custom setup. The only drawback to the EXOS2 is the price. I bought a EXOS1 system because I did not have the confidence to build a custom kit. Although I am very pleased with the results from my Koolance system, I am looking forward to building a custom 1/2" setup in the future.

P4C 2.4@3.4, 43C Load.
 
That is the only reason why I am considering getting this, I dont know about a custom setup. Does the Koolance not leave room to change parts etc?
 
one of the advantages of a system like the EXOS2 is that its an external
thus your not preheating air with internal components before running it through the rad or preheating your case by running the air through the rad

as far as swapping components, you could change the block or lines but the rest of the unit is basically not easy to mod
and since its a integrated solution swapping out the block might not accually increase performance
since its all about a balance in flowrate\pressure to a rad interface with a given airflow volume (at variable ambient room temperatures)

the reason youd get the EXOS2 is that its an integrated solution
(meaning that you have a reasonable idea of its performance)
and that its very simple to migrate it from box to box IMO
 
If you are looking an easy to install and maintain setup with good performance, then the Exos II is a great unit... it's rated at 720 watts of heat dissipation, and because it's external, it does not pull or push hot air in or out of your case.. which is nice. :) If you are interested in MAX performance, this might not be the setup for you, but I was pleasantly surprized with my koolance case (which has an Exos II built into the top of it) in that it performs exceedingly well with the danger den blocks I got for it... even with SLi! I think the idea for getting the best performance from a koolance setup is getting good flow blocks which do not drop the loop pressure, and it will perform very well.. You want to stay as least restrictive as possible so the loop keeps a decent flow rate. The danger den TDX (some folks think it's fugly - but I kind of like mine) cpu block and maze4 gpu block seem to compliment the koolance quite well... Granted there are many more alternatives out there, but I cannot speak from experience about them...
Anyways... my $0.02 :)
 
Ice Czar said:
one of the advantages of a system like the EXOS2 is that its an external
thus your not preheating air with internal components before running it through the rad or preheating your case by running the air through the rad
Click to expand...
well.....i've heard this argument more than a few times from various people, and i don't think that it has much merit, even coming from you. yes, you have shown how i was wrong in previous examples, however i think that you are going to need to make one hell of a strong case this time.

once you have eliminated the heat from both CPU and GPU from the inside the case, and you are most likely moving something like 100 CFM at the bare minimum through a 120.2 rad and your PSU, both with quiet fans, your case air is not going to be much above ambient.

i'm not going to deny that there is going to be SOME difference, all other thing being equal, however you'd be talking something less than a 0.2 C CPU temp difference, probably much less than that. keeping the room's door open vs keeping it shut would probably make more of a difference.

the exos systems make things easy, and there is no denying that. as for the performace gain of using outside of case air......i don't think that is anything to preach.

as for great pre-fab kit that will out-perform the exos 2, in the same price range, you might want to look at the swiftech apex kit, and add a MCW 55 waterblock. it comes with instructions, and the only tool that you need other than standard PC building gear is a hand drill and the proper drill bit. this particular kit even has the added benifit of mounting the rad such that it uses external air, meaning that it get the same marginal benifit as the exos 2 from that factor.
 
revenant said:
do I have pictures???? lol... yes, I sure do... this is a work log for el rolio and my first watercooling builds... he's using a swiftech apex kit, and have the koolance.. so you'll see the apex also. :)

http://www.hardforum.com/showthread.php?t=947868
Click to expand...
i expect that you already know it damn well, however that build has both looks and performance. congrats on having one WICKED machine.
 
Hey thanks a lot! :)
I am really happy with this build... even though it was a pre-mod kit, I have put a LOT of additional work into it to enhance the looks and performance. I love that lian-li case so much... that is the ultimate watercooling case imo... well any vXXXX series... all the perferated surfaces let it breath so well... really nice case! :)
 
I plan on getting the exos-2 as well and I've been wondering if its possible to replace the radiator fans for it and if it will affect performance at all. Any input on that rev? Or anyone else with the exos-2 for that matter.
 
I am sure you can, but do not reverse the direction of flow, because I hear that does not work well at all... it's always better to pull air through the fins. Anyways, I am sure it will affect the top end performance, but I do most of my top end runs with the fans on a setting of 6 out of 10 and it seems to get the job done nicely while remaining very quiet. 10 really moves some air, quite a bit, and does make some noise, but it's like a rushing wind sound.. not a typical fan whine noise... and I watched my temps fall immediately (1c in about 2 mins, then 1c more over 10 mins or so) when I put it on 10 while on a hot day when my room was like 85... so it does perform pretty well. Settings 1-5 are silent... 6 is barely audible over the other 120mm fans I have, which are very quiet, but I can hear them a little... 7 is where it gets to be where you can really notice that sound is coming from the top. heh...

Anyways - are you looking to improve performance with higher CFM fans? you might be happy with the 10 setting.. it moves a decent amount of air. I think the best thing you can do with the Exos2 is get some good performing non-restrictive blocks for it. :)
 
Noise isnt really a factor for me so if it will affect performance I might get 2 Delta 151 cfm fans for the rad.:) If I use different fans I'll still be able to hook them up to the controller on the unit and control the speed right?
 
ok - I am sure you will be able to... as long as they are 12v DC... just be sure to check for small animals and loose articles of clothing stuck to your air intakes ;)
 
They are, I have one of them in my current system and its on a fan controller. Its a monster. Thanks for the info rev.:)

EDIT: 1 more thing, can you tell if the stock fans are 25mm or 38mm thick? Because the Deltas are 38mm
 
np... let me know how yours goes... with the koolance controller and those fans, I had bet you'll have one helluva wide range of performance options as your disposal. I might do the same. :) FYI - I am trying to hunt down info on what the current/default fans are rated at... just for grins...
 
The fans are rated at 94 cfm each at full rpm, as stated on the review here at [H].:) The only thing I can't seem to find is the thickness of the standard fan. If its only 25mm I'll have to scrap the Delta idea.:(
 
*Sigh* I found a review where they took the enclosure apart and they even listed all the specs on the fans and they are only 25mm thick.:( rev, do you think theres any way at all you could get a 38mm fan to fit in there? lol
 
DFI Daishi said:
well.....i've heard this argument more than a few times from various people, and i don't think that it has much merit, even coming from you. yes, you have shown how i was wrong in previous examples, however i think that you are going to need to make one hell of a strong case this time.

once you have eliminated the heat from both CPU and GPU from the inside the case, and you are most likely moving something like 100 CFM at the bare minimum through a 120.2 rad and your PSU, both with quiet fans, your case air is not going to be much above ambient.

i'm not going to deny that there is going to be SOME difference, all other thing being equal, however you'd be talking something less than a 0.2 C CPU temp difference, probably much less than that. keeping the room's door open vs keeping it shut would probably make more of a difference.
Click to expand...


we shall see as Im putting this to the test directly
a few variable factors to consider

as the deltaT between the case\room air and the heatsource decreases the importance of a seperate airflow will increase (thus in a cool room your likely correct but in a hot room the impact will I predict be more pronounced)

obviously the airflow into and out of cases of various sizes isnt the same, at some critical threshold of resistance for a given ambient the difference between pre\post and seperate will be increasingly noticable

VRMs (mosfets\voltage regulation moduals) are getting berry berry hot these days in excess of northbridges, to the point where we now have OATES and other cooling strategies for them

additional heatsources that maybe present in other peoples cases, lights, drives ect add another shade of variable

a seperate or external airpath that maintains the greatest possible deltaT may not be critical under "normal" well thoughtout circumstances, with a decent room temp, good airflow and reasonable number of heatsources in the case, but its also in theory the ideal solution and will at some threshold manifest itself IMO ;)

and Im going to see if I can sketch out a few such realworld seneroios
as ambient temperature has a direct and proportinal effect past most other design considerations I suspect

if youd like a peek at what is cooking in my little brain
http://www.bae.uky.edu/~snokes/BAE549thermo/psychrometrics.htm
http://www.humiditycontrol.com/howitworks.pdf
http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=7544977062&rd=1&sspagename=STRK:MEWA:IT&rd=1
 
As an eBay Associate, HardForum may earn from qualifying purchases.
Water-cooling is indeed an interesting sport. I have been playing with this and that for quite a while and the fact is there are a lot of good kits and various bits and pieces to assemble a custom unit.

I like to keep things conservative as far a voltages and temperatures and at the same time I like good performance, so finding that balance of what will work well and yet not over tax components (I’m not into life shortening) sometimes isn’t easy, but when you find it it’s worth while.

That brings me to the Exos 2. After building many units I got lazy a few months ago and decided the heck with it, these guys are pros, how bad can it be.

Installation from start to finish took about an hour, part of that time removing extra fans from inside the case. The results are a super clean install, total simplicity of operation and great temps.

First CPU was a 3800 Venice at 2.82 gig. Ambient room temps are 23 to 24c for reference. All my machines run Folding At Home 24/7 to they never see less then 100% load 24/7. The 3800 hit a max temp of 36c which I feel is definitely in the safe zone and that was at 1.55 volts, a tad higher then I like but livable.

Then I decided to swap to a 3800 X-2 in that machine. Swap time was 5 full minutes because of the simplicity of the Koolance CPU mounting system.

3800 X-2 at 2.5 gigs per side at 1.48 volts with temps at 33c, same ambient room temps and 100% CPU load 24/7.

For a simple bolt on unit that comes with everything except an AMD bracket (6.95 extra) and an absolutely no hassle install I think it’s a good value.

As for capacity, just for S’s and G’s the unit comes with some tubing adaptors for any other things (GPU, Chipset) that you might want to cool. I tapped into that part of the system and ran a set of tubes over to a second machine and added the old style water block (1/4” fittings) so the system could cool two CPU’s. What I got was a 1c rise in temps, again, two systems being cooled and a 1c temp rise.

Keeping in mind everything IceCzar said is totally correct in my opinion it’s still a great value, and really doesn't need modded.

Yup, I like Koolance a lot.
 
naturally, the proof is going to be in the numbers. i look forward to seeing your results.

normally, the resistance of adding a bit of tubing length and convolution is going to have minimal effect on real world temps, and can almost be left out of consideration with loop design, as long as you are in the right ballpark. i think that it is worth taking added tubing lengths into consideration as well, given how marginal i predict that any temp improvement will be, having the rad outside of the case.

perhaps take that into account by running a few tubing lengths for each scenario. eg, internal rad with normal lengths of tubing, internal rad with extra tubing, external rad with minimal tubing (preferrably same as config with rad internally mounted, with typical tube lengths) external rad with typical tubing length to get up to a top-mounted external box, etc.

no denying that mosfets, memory, chipset and all the other errata in there put off a decent bit of heat, however they don't get THAT hot with decent, targeted, airflow. add to that the fact that a rad taking air out of the case will likely give a case somewhat better airflow than a typical air cooled config, and things run pretty cool in there.

have you discussed your project with Cathar or BillR? they might have some input, if you haven't already talked to them. if i'm remembering correctly, Cathar has an uber-quiet water cooled rig that uses a large external rad, so he might very well have run the numbers and testing already.
 
I also have a Exos 2 cooling my system. This is the first time I tried watercooling. I am running a Dual Opteron 252 (2.6Ghz each) and a 7800 GTX @525/1.4GHz and it idels at 28c load at 40c. I am very happy with the system, it looks great, performs great and also is very quiet. At fan setting 8-10 the system can be loud but at 5-7 it is very quiet and cool's my whole system just fine.
 
well the "project" is Im the new cooling editor :p

and while I havent talked to BillR or Cathar, I have been attempting to read everything ever posted at procooling and been discussing things will BillA

"real world" articles will start very soon
the uberl33t {H]igh Altitude Test Lab is in the works but that level of precision and automation will take awhile

so far
Keithley 2700
E&H Massflow Meter

been playing cat and mouse with a Rosemount Differential Pressure Meter, Vaccum Pumps,Watlow Cartridge Heaters, ect.
 
As an eBay Associate, HardForum may earn from qualifying purchases.
LOL BillR is no slouch either ;)

but BillA has a bit more PLC\PID\HART experience
(I gather)
which up till a few weeks ago was all Greek to me :p

now I know exactly how ignorant I am about feildbus architecture
 
I have further delusions as well, but lack the funds currently
things like thermal imaging and measuring airflow resistance
first things first

BillA saved my butt already once pointing me to an integrated high res DMM\data logger
best I get it online and repeatable results first before the next phase
which is an ambient control test chamber I can pressurize
 
p0intblank said:
Noise isnt really a factor for me so if it will affect performance I might get 2 Delta 151 cfm fans for the rad.:) If I use different fans I'll still be able to hook them up to the controller on the unit and control the speed right?
Click to expand...

I’ve tried bigger fans, not the deltas but 125 cfm fans non-the less. First, it made about a 2c temp change VS running the Exos at half speed. Second, I didn’t try to use the Exos’s fan controllers because they can’t handle that much amperage. Third, why on earth put up with the noise for such a small gain? The unit is all but dead quiet and works well stock.

RE inside case temps; I was going to add the chipset cooler and GPU cooler to the loop, but since the Exos is external my case temps dropped so far I found I didn’t need do to that. My chipset runs at 28c on air, my GPU even over clocked stays well below 45c.

I am using a super lan boy case with dual 120mm fans (one front, one rear) capable of over 100 CFM each. Both are now turned down all the way rendering them dead quiet. Let me add my hard drives are (Raptor 74’s run no higher then 30c)

Don’t know if this helps the tubing length question but Koolance was adamant that you can use the unit up to 15’ away from the computer, I did try it with two 12’ pieces of tubing with no real detectable temp changes. Considering the unit is basically a low flow unit I was quite surprised that those lengths could be used.

Why you might ask would you want to use 15’ of tubing? How about putting the cooling unit in the direct path of an air conditioning unit? How about sitting the cooling unit on a windowsill in the winter? While I had the 12’ pieces added I did the AC thing and watched while my “under load” CPU temp dropped to 20c…. I’m seriously thinking of computer relocation because the AC and the window are in the same location, but I don’t want to deal with all the extra tubing.

I’m in no way promoting Koolance as the ultimate answer, but for a simple out of the box experience it simply works and works well, more then can be said of many other things we buy for our computers. ;) :D
 
p0intblank said:
The fans are rated at 94 cfm each at full rpm, as stated on the review here at [H].:) The only thing I can't seem to find is the thickness of the standard fan. If its only 25mm I'll have to scrap the Delta idea.:(
Click to expand...

They appear to be right around 1.25 inches thick (little less than)... seems like the most common thickness for a 120mm fan to me. :)
 
BillR said:
I’ve tried bigger fans, not the deltas but 125 cfm fans non-the less. First, it made about a 2c temp change VS running the Exos at half speed. Second, I didn’t try to use the Exos’s fan controllers because they can’t handle that much amperage. Third, why on earth put up with the noise for such a small gain? The unit is all but dead quiet and works well stock.
<snip>

:D
Click to expand...

ok - well good to hear it from someone who's tried it. I said earlier that I figured the best thing for that kit was to find some good non-restrictive blocks for it. The 10 setting is quite windy and the 1-6 settings even work amazingly well and are so quiet, :)
 
revenant said:
ok - well good to hear it from someone who's tried it. I said earlier that I figured the best thing for that kit was to find some good non-restrictive blocks for it. The 10 setting is quite windy and the 1-6 settings even work amazingly well and are so quiet, :)
Click to expand...

Keeping in mind the system is “low flow” I’m not sure you can find a more non-restrictive block then what they supply.

Let me know if you try any that work better.
 
BillR said:
RE inside case temps; I was going to add the chipset cooler and GPU cooler to the loop, but since the Exos is external my case temps dropped so far I found I didn’t need do to that. My chipset runs at 28c on air, my GPU even over clocked stays well below 45c.
Click to expand...
i just want to mention that my reported chipset temps are around 32 C load, with my more-or-less passive chipset cooling solution.

i have a fan nearby, to cool my vid card memory, and my chipset, however it's not audible above the rest of my system, which i consider to be pretty quiet with my present config.
 
Top Nurse said:
Now that looks like some seriously fun [H]ardware. :D
Click to expand...
i suppose that depends on personal preferance.

i don't consider anal-retentive attention to accuracy/precision to be all that pleasant.

then again, i know plenty of people, with plenty of things that float their boat. (sometimes, i wish that i DIDN'T know) :)
 
BillR said:
Keeping in mind the system is “low flow” I’m not sure you can find a more non-restrictive block then what they supply.

Let me know if you try any that work better.
Click to expand...

Well, they don't supply blocks with the kit... so went with all danger den blocks. they were described to me as being "good non-restrictive bocks" (TDX and two maze4's). I am very happy with my temps... with my room at 20c my cpu temps are 32c idle and 36c load... gpus are 40c & 41c idle and 44c load. I don't have the koolance blocks to compare these numbers with, but I have a feeling these are good readings. (I doubt they will do better) I know the Koolance blocks use 1/4 fittings and my DD bocks are all 3/8 which does promote better flow... (more coolant, not faster speed) so I guess that is what I mean... go with 3/8" fitting bocks rather than the 1/4" ones...
 
Ice Czar said:
LOL BillR is no slouch either ;)

but BillA has a bit more PLC\PID\HART experience
(I gather)
which up till a few weeks ago was all Greek to me :p

now I know exactly how ignorant I am about feildbus architecture
Click to expand...

Thanks for the compliment

You and I both share a love for the good stuff and a lack of funds to fulfill all our desires, sucks huh?

One thing I find very interesting though is the difference between the engineer and the technician and I’ll use Top Nurse’s “Top Fuel” metaphor to try to explain.

I’m a big Drag Racing fan, and I used to drive a racer and work on the engines. I still hang in the pits as often as possible.

I love to watch the “College boys” (no disrespect intended) with millions of dollars of equipment take every measurement possible making every adjustment possible right up to within 2 minuets of the actual race, then watch as the “good ole boy” who’s been doing this forever basically shakes his head, sticks his finger in the air, feels the track and makes a major adjustment at the very last second, usually resulting in a race win.

What we do is very much the same, the theory VS the practical and at the same time we share and learn, win and lose.

I’d still love to have some of the test gear and really learn how to use it properly though if for no other reason then to actually know the right answer, or at least have a solid basis for one ;) :D
 
Ice Czar said:
LOL BillR is no slouch either ;)
Click to expand...
and, just to make it clear that i am not trying to knock anyone, i am not trying to say that BillR is a slouch. he is simply not the one that i was thinking of, when i was posting people who might have the resources to run the simulations and have checked the sims vs the real world.
 
BillR said:
I’ve tried bigger fans, not the deltas but 125 cfm fans non-the less. First, it made about a 2c temp change VS running the Exos at half speed. Second, I didn’t try to use the Exos’s fan controllers because they can’t handle that much amperage. Third, why on earth put up with the noise for such a small gain? The unit is all but dead quiet and works well stock.

RE inside case temps; I was going to add the chipset cooler and GPU cooler to the loop, but since the Exos is external my case temps dropped so far I found I didn’t need do to that. My chipset runs at 28c on air, my GPU even over clocked stays well below 45c.

I am using a super lan boy case with dual 120mm fans (one front, one rear) capable of over 100 CFM each. Both are now turned down all the way rendering them dead quiet. Let me add my hard drives are (Raptor 74’s run no higher then 30c)

Don’t know if this helps the tubing length question but Koolance was adamant that you can use the unit up to 15’ away from the computer, I did try it with two 12’ pieces of tubing with no real detectable temp changes. Considering the unit is basically a low flow unit I was quite surprised that those lengths could be used.

Why you might ask would you want to use 15’ of tubing? How about putting the cooling unit in the direct path of an air conditioning unit? How about sitting the cooling unit on a windowsill in the winter? While I had the 12’ pieces added I did the AC thing and watched while my “under load” CPU temp dropped to 20c…. I’m seriously thinking of computer relocation because the AC and the window are in the same location, but I don’t want to deal with all the extra tubing.

I’m in no way promoting Koolance as the ultimate answer, but for a simple out of the box experience it simply works and works well, more then can be said of many other things we buy for our computers. ;) :D
Click to expand...
Wow thanks for all the info. 12 feet you say? I might consider using a good deal of length since I wont be able to keep the unit on top of my case since it has a blowhole at the top. And when you put it in front of the AC there wasnt an issue of condensation was there? Also using that much tubing would allow me to tuck the unit away and always have the fans at high and still not hear it.:)
 
BillR said:
Thanks for the compliment

One thing I find very interesting though is the difference between the engineer and the technician and I’ll use Top Nurse’s “Top Fuel” metaphor to try to explain.
<SNIP>
Click to expand...

I like that metaphor. :)
 
revenant said:
Well, they don't supply blocks with the kit... so went with all danger den blocks. they were described to me as being "good non-restrictive bocks" (TDX and two maze4's). I am very happy with my temps... with my room at 20c my cpu temps are 32c idle and 36c load... gpus are 40c & 41c idle and 44c load. I don't have the koolance blocks to compare these numbers with, but I have a feeling these are good readings. (I doubt they will do better) I know the Koolance blocks use 1/4 fittings and my DD bocks are all 3/8 which does promote better flow... (more coolant, not faster speed) so I guess that is what I mean... go with 3/8" fitting bocks rather than the 1/4" ones...
Click to expand...

The CPU block for the Exos-2 is indeed 3/8, it was the original that was ¼. In the post where I mentioned hooking up a second computer I am using the ¼ “ tubing with supplied adaptors (making it a parallel system) and still achieving good cooling, as I pointed out only a 1c rise in total temps. ;)
 
You must log in or register to reply here.
Back
Top