CoolIT Freezone Chilled Liquid CPU Cooler

HeavyH20 said:
I had similar results with my QX6700. It tips over about 3.5 GHz, however. Can't keep up after that.
Click to expand...

I thought you were getting idle 46C load 85C @ 3466 (your post #68 above)?

What date code do you guys have on your Freezones?
 
theelectic said:
I thought you were getting idle 46C load 85C @ 3466 (your post #68 above)?

What date code do you guys have on your Freezones?
Click to expand...

Mine is still running at 39 c at windows and full load it goes to 60 c with orthos running for 20 minutes. e6600 at 1.375 v

Mine also came preset to the coldest setting. Mine reads a date code of 1806
 
No Offense but I think these guys who are getting these awesome cold temps must either be living in the North Pole or have their windows open in the winter.
 
I was actually about to send the freezone to Steve as a really late X-mas present. (I have been meaning it to send it for a while).
Instead I am going to pull it out and do a little more investigation. There are just too many people getting too many different results to ignore here. I will keep you guys up to date with what happens here. While we do not normally include overclocking in our thermal testing due to the numerous variables and difficult reproducability among other things, I am going to take this 4200+ out for a run with some extra voltage and see what happens.
 
Justin_Diduch said:
I was actually about to send the freezone to Steve as a really late X-mas present. (I have been meaning it to send it for a while).
Instead I am going to pull it out and do a little more investigation. There are just too many people getting too many different results to ignore here. I will keep you guys up to date with what happens here. While we do not normally include overclocking in our thermal testing due to the numerous variables and difficult reproducability among other things, I am going to take this 4200+ out for a run with some extra voltage and see what happens.
Click to expand...
cool
I think a lot of people are interested to see your results
 
Justin_Diduch said:
I was actually about to send the freezone to Steve as a really late X-mas present. (I have been meaning it to send it for a while).
Instead I am going to pull it out and do a little more investigation. There are just too many people getting too many different results to ignore here. I will keep you guys up to date with what happens here. While we do not normally include overclocking in our thermal testing due to the numerous variables and difficult reproducability among other things, I am going to take this 4200+ out for a run with some extra voltage and see what happens.
Click to expand...

Thanks for going the extra mile :)

Normally I wouldn't have been concerned with the results when overclocked either - however CoolIT specifically states on their website the Eliminator is targeted for overclocking, and the Freezone is capable of cooling quad core:

Cooling Capabilities:
Freezone is recommended for CPUs with total heat output of up to 175W. Higher loads can be accommodated but the result will be comparable to water cooling. The Eliminator is recommended for CPUs with a total heat output of up to 125W. The Eliminator is targeted for overclocking nearly any dual core CPU while the Freezone is more than capable of cooling the new quad core kentsfield from Intel.
Click to expand...

According to the Extreme PSU Calculator, an AM2 X2 3800+ overclocked from is regular 2.0GHz, 1.35V (89W) setting to 2.7GHZ, 1.4V puts out 143W. That is well under 175W and I am not getting anything close to regular radiator based water cooling.
 
Wait a minute here boys...i really have some issues with the comments made about 22c delats of the 1.55v E6600....I am running a QX6700 that starts out130wats at stock but is less actually because the board actually undervolts under load so much. However you will see I get close to 10-11c dltas up to the points where I go over 175watts where the freezone limit is said to be.

If you do the math that places the heat dissapation of an E6600 @3.6ghz w/ 1.55v (which mine can do) as being 133watts. Now this is at the capacity of the Elminator which is designed for 125watts. the Freezone is designed for 175watts max. Also all of the watercooling companies and Coolit test there systems at 100watts of load to up with their CW coefficients.

Look down below. I use reliable Coretemp (which match my TAT app) and I dont have 22c deltas from idle to load.

8-9c at 2.66ghz(stock)ghz 1.23v load and 108watts
10-12c at 3.2ghz 1.29v load and 145watts
11-12c at 3.4ghz and 174watts 1.38v load (***this reaches the limits of the freezone***)
15-16c at 3.5ghz and 195 watts 1.44v load

Look at intels cpu spec pages and you will see default voltage for an E6600 is 1.325v but the QX6700 is 1.35v.

Actually 1.55v if actual is pushing 20% vcore OC and the limit of ost water cooling period. Likely had to set 1.6v+ to hold 1.55v under load.

My potentiometer is at high but that is where it should be expected on a quad core...I am still using the 92mm fan. Will be switching out for my 100cfm 120mm fans. My guess is that these 92mm fans based on rpm are about 50cfm max fans...

I will give him the fact I was not running 4 instances of prime95 on all of them, but I was runnig 4 instances of 149x work units from F@H which load up significant amount of program up into cache and actually loads the cpu 3-4c higher then normal F@H work units from my testing. I dont run orthos or prime95 as an app. So I test apps I will run



http://forums.anandtech.com/messageview.aspx?catid=28&threadid=1998273&enterthread=y

Ambient room temp was 22c
All idle temps are taken 10minutes after bootup.
All load temps are taken after 30-60min of 4core loading with F@H (4 instances)
All temps are measured with Asus Probe, coretemp (TAT has confirmed to be identical)

QX6700ES @ 2.66ghz w/ 800ddr2

Asus P5B-D
1.3v set, 1.25v idle, 1.23v load (108watts)
Idle:
Asus PCprobe CPU 12c / MB 27c
Coretemp = 30-34c (core 2 is always the coolest)
Load:
Asus PCprobe CPU 21c / MB 28c
Coretemp = 38-43c

QX6700ES @ 3.2ghz (10x320) w/ 800ddr2

Asus P5B-D
1.375v set, 1.31v idle, 1.29v load (145watts)
Idle:
Asus PCprobe CPU 20c / MB 28c
Coretemp = 36-42c(core 2 is always the coolest)
Load:
Asus PCprobe CPU 29c / MB 29c
Coretemp = 48-52c

Gigabyte DS3
1.4125v set, 1.36v idle, 1.33v load
Idle:
Coretemp = 55-59c (w/ Scythe Ninja and Antec 3 120mm (High))
Load:
Coretemp = 66-71c (w/ Scythe Ninja and Antec 3 120mm (High))

QX6700ES @ 3.4ghz (10x340) w/ 850ddr2

Asus P5B-D
1.475v set, 1.4v idle, 1.38v load (174watts)
Idle:
Asus PCprobe CPU 28c / MB 29c
Coretemp = 46-50c(core 2 is always the coolest)
Load:
Asus PCprobe CPU 41c / MB 30c
Coretemp = 61-65c

Gigabyte DS3 ***never got fully stable***
1.4625v set, 1.40v idle, 1.38v load
Idle:
Coretemp = 64-68c (w/ Scythe Ninja and Antec 3 120mm (High))
Load:
Coretemp = 74-78c (w/ Scythe Ninja and Antec 3 120mm (High))***chip was throttling according to RMClock, and not likely was hottest temp, coretemp load at 3.33ghz was 74-77c and did not throttle at 3.33ghz with 1.45v set load 1.37v though still not 100% stable***

QX6700ES @ 3.5ghz (10x350) w/ 875ddr2

Asus P5B-D
1.55v set, 1.46v idle, 1.44v load (195watts)
Idle:
Asus PCprobe CPU 36c / MB 29c
Coretemp = 54-59c(core 2 is always the coolest)
Load:
Asus PCprobe CPU 49c / MB 30c
Coretemp = 70-74c


i will be testing same 120mm theory here within a few days.....I would take the 1-4c drop and hoping a 100cfm fan versus the 50cfm 92mm fan may be mostly on the high side of that.
 
By the way if you notice it is running 15c+ lower then mY Scythe NInja air cooler did with a 100cfm 120mm (not stock to the cooler) fan at high.

Based n some of the review I saw it beat out a swiftech with apogee block and double to triple radiator which I am sure by the pics would have been 2x as loud minimum....

I would like to see them come out with a 120mm fan version and 8-9 tecs with the silver shield removed or perforated to let my side case fan blow insome air. Hopefully it would have the capacity of 200+watts of heat dissapation. Or at least allow people to add a gpu block for oveclocked dual cores. I have an aeroengine case with a 160cfm side 25cfm fan and it basically is blocked.
 
Duvie,

I hope you don't mind that I've taken the data and put it into 2 charts... I am more of a visual learner/person ;-)

Idle Temps
Hardocp-idletemps.gif


Load Temps
Hardocp-loadtemps.gif



EDIT: Here is the excel spreadsheet file with charts if someone wants to download and make better looking charts: http://www.megaupload.com/?d=PQTVLKS0
 
Asus P5B-D
1.375v set, 1.31v idle, 1.29v load (145watts)
Idle:
Asus PCprobe CPU 20c / MB 28c
Coretemp = 36-42c(core 2 is always the coolest)
Load:
Asus PCprobe CPU 29c / MB 29c
Coretemp = 48-52c
Click to expand...

This result roughly matches what I got with my overclocked X2 3800+, which is around 143W dissipation.

I know Duvie you said in that other forum post that your Freezone came with a 92x25mm fan - mine definitely had a 92x38mm fan, Yeung Lin model DFB923812H.
 
MIne is definitely 25mm...It measures 1 inch and 1 inch is 25mm...It also is the exact same depth as my 120mm Tricool antec which is 25mm...

NIce charts...
 
Hey guys.

This is my first post here. I saw this thread while googling for freezone. Been watching it for the past few days. I got my freezone this past friday and been running test on it. I'm just really surprised that some of you guys have such bad scores.

Here's my system

Asus PW5 DH Deluxe
Intel Core 2 Duo 6600
Geil 8500 Ram
XFX 8800 GTX
Thermaltake Armor Case

Bios Settings

RAM = 2.2v
CPU = 1.5v
North Bridge = 1.65
RAM settings = auto 5-6-6-18
FSB = 400

400 is the max (stable) that I've been able to do. I manage to do 450 but it fails the Orthos test. This is the only time I've seen the system go over 50C.

Before I installed the block I removed the paste that came with it and replaced it with Arctic MX-1. I also have a small fan blowing on the northbridge. The bridge seems to be the weakest link in the whole setup. It gets way too hot for my liking. I must say that I haven't put on the little fan that came with the board. Since I already installed the freezone I didn't feel like removing it to install it on the bridge. Either way I don't think it would be enough to cool it down.

When I was benchmarking last night, the temp got as high as 50. It was switching around 48-50. I ran Orthos for 2 hours.

Before I went to sleep I decided to leave Orthos running over night. Because I was somewhat paranoid, I decided to turn off my floor heater just in case. From the screenshot you can see that the score was around 43-44 with Orthos running for 10+ hours. I'm sure this is because I turned off the heater but regardless, I would still say I haven't seen it over 50C yet since I did stressed it last night before I went to sleep.

Also remember that I do have an 8800 GTX installed and this does generate some heat.

Here's a screenshot of the result.



I want to get 4ghz (stable) but I don't know if I want to push it anymore.

How did you guys even get such high FSB?(to get 4ghz) What were your voltage settings?

Also, what would be a good way to cool the north bridge?
 
I just reseated my block again with the freezone and still no luck. Idle at windows 39 to 41 c at 1.375V E6600 360fsb x 9. On a striker board NB V to 1.525. Full load orthos 58c to 60c

Im begining to think its the damn heat pipes on this striker board doing the damage.

Im also thinking of remoding the fan to a 120mm silverstone fan http://www.performance-pcs.com/catalog/index.php?main_page=product_info&products_id=2888
Now before I do something stupid do you guys think this fan will hold this weight of the freezone?

You guys are saying the freezone is capable of cooling the quad QX6700. I have this chip on the side ready to build my second rig on a commando board, but if it cant cool the dual core how the hell is gonna cool my quad core?:eek:

Thanks Any help is Appreciated
 
infospace said:
I just reseated my block again with the freezone and still no luck. Idle at windows 39 to 41 c at 1.375V E6600 360fsb x 9. On a striker board NB V to 1.525. Full load orthos 58c to 60c

Im begining to think its the damn heat pipes on this striker board doing the damage.
Click to expand...

I don't think it's the motherboard heatpipes, your general Freezone experiences sounds like mine. The Freezone can't cope with your CPU load. If you can run your computer with the side panel off, if temps don't drop by a significant degree, you know it's not an inside case airflow problem.
 
You may need to lap the IHS. It may be rather uneven and hench the block may not be making good contact. 360x9=3240 and at 1.375v (sounds like a bios settting so under load it is less) you are dealing with far less heat then I was dealing with at 3.2ghz...

That is only 94watts of heat to dissipate....At stock 2.66ghz even at an undervolted 1.23v I hae 108watts...Look at my numbers...NOWHERE NEAR AS BAD AS YOURS...It appears the cooler is good as in mine an others cases..

It clearly has to be other factors. I dont want to say user error, becuase it could be other things.

BOTTOMLINE....Your numbers should be like mine for my QX6700 at stock

Try this:

1) reapply small amount of AS per Artic Sivers instructions for a dual core cpu
2)Turn potentiometer to high as per instructions in Coolit manual ( I notice virtually no audible difference in mine. I actually have called coolit (they haven't called me back) as to why I noticed no change in fan speeds when going from low-med-high
3)Lap the IHS
4) position a fan inside the case to blow some air under theat ridiculous larghe and obstructive shield to allow the 92mm exhaust fan to pull some cooler air through the fins of the heatsink.
 
So which of you is going to be the first to void your warranty cut into this thing? I wanna see results of a modded 2-stage water/freezone cooler like we mentioned earlier in this thread! :D
 
Torsten said:
So which of you is going to be the first to void your warranty cut into this thing? I wanna see results of a modded 2-stage water/freezone cooler like we mentioned earlier in this thread! :D
Click to expand...

I already returned mine for a refund.

I'm waiting to see Justin_Diduch's results w/overclocked CPU, his setup is very very similar to mine which is why it made no sense to me why I wasn't getting temps at least on par with air cooling. Pretty much nothing in the review matched up with my experience at all.

OTOH I still can't see how 56W of air cooled TECs can possibly cope with any kind of CPU load. Look at the Swiftech MCW-CHILL:

http://swiftnets.com/products/MCWCHILL-452.asp

It contains TWO 226W water cooled TECs and with an 80W heat load, cools to 4C below ambient. That's 452W of well cooled (hot side water cooled) TECs. So 56W of air cooled TECs is supposed to be able to handle a 175W heat load with comparable results to water cooling? Doesn't make sense to me unless the air cooling on the TECs can handle the heat from the TECs, the pump, AND the CPU.
 
theelectic said:
I already returned mine for a refund.

I'm waiting to see Justin_Diduch's results w/overclocked CPU, his setup is very very similar to mine which is why it made no sense to me why I wasn't getting temps at least on par with air cooling. Pretty much nothing in the review matched up with my experience at all.

OTOH I still can't see how 56W of air cooled TECs can possibly cope with any kind of CPU load. Look at the Swiftech MCW-CHILL:

http://swiftnets.com/products/MCWCHILL-452.asp

It contains TWO 226W water cooled TECs and with an 80W heat load, cools to 4C below ambient. That's 452W of well cooled (hot side water cooled) TECs. So 56W of air cooled TECs is supposed to be able to handle a 175W heat load with comparable results to water cooling? Doesn't make sense to me unless the air cooling on the TECs can handle the heat from the TECs, the pump, AND the CPU.
Click to expand...

It's going to be a couple of days, I will keep you posted.
 
theelectic said:
I don't think it's the motherboard heatpipes, your general Freezone experiences sounds like mine. The Freezone can't cope with your CPU load. If you can run your computer with the side panel off, if temps don't drop by a significant degree, you know it's not an inside case airflow problem.
Click to expand...

I opened the side door on my case and the temps dropped from 39c to 34c does this signify a case cooling problem?
 
Duvie said:
You may need to lap the IHS. It may be rather uneven and hench the block may not be making good contact. 360x9=3240 and at 1.375v (sounds like a bios settting so under load it is less) you are dealing with far less heat then I was dealing with at 3.2ghz...

That is only 94watts of heat to dissipate....At stock 2.66ghz even at an undervolted 1.23v I hae 108watts...Look at my numbers...NOWHERE NEAR AS BAD AS YOURS...It appears the cooler is good as in mine an others cases..

It clearly has to be other factors. I dont want to say user error, becuase it could be other things.

BOTTOMLINE....Your numbers should be like mine for my QX6700 at stock

Try this:

1) reapply small amount of AS per Artic Sivers instructions for a dual core cpu
2)Turn potentiometer to high as per instructions in Coolit manual ( I notice virtually no audible difference in mine. I actually have called coolit (they haven't called me back) as to why I noticed no change in fan speeds when going from low-med-high
3)Lap the IHS
4) position a fan inside the case to blow some air under theat ridiculous larghe and obstructive shield to allow the 92mm exhaust fan to pull some cooler air through the fins of the heatsink.
Click to expand...

I did apply the artic silver just like it says on the instructions it didn't make a difference.
I have the potentionmeter adjusted to the coldest setting. My fan speed dont change either when I adjust the potentionmeter.
I think Im just gonna remove that shield completely. How about removing the 92mm fan and replacing it with that 120mm silverstone, you think thats a good idea?
 
infospace said:
I did apply the artic silver just like it says on the instructions it didn't make a difference.
I have the potentionmeter adjusted to the coldest setting. My fan speed dont change either when I adjust the potentionmeter.
I think Im just gonna remove that shield completely. How about removing the 92mm fan and replacing it with that 120mm silverstone, you think thats a good idea?
Click to expand...

Temporarily put a high performance Delta fan on there just for gathering potential data, I'd love to see the results ;)

What is your system temp reading when running orthos runs for awhile? Sounds like your case is warming up quite a bit, easy to do with todays hardware.
 
infospace said:
I opened the side door on my case and the temps dropped from 39c to 34c does this signify a case cooling problem?
Click to expand...

With temps dropping 5C I'd say you have a slight case cooling problem. A stronger/larger intake fan, and/or re-routing internal cabling out of the way would be a start.
 
Im moving my system into a new case the coolermaster 830 evolution. It will let me blast off 8 fans. And I'm also gonna rip that metal side panel off the freezone and remove the 90mm fan and replace it with a delta 120mm and then I will come back with results. I just hope my comp don't fly away with all this fan power:D
 
People have been posting some nice results when upping the fan cpm ability of the freezone, either by 120mm or 92x38. Good luck :)
 
I would probably like to do this too, is there a guide on how to accomplish this? fan to buy, etc. thank you in advance.
 
As far as wich fan, It all depends how far you want to take this..... I chose a 49db 114cfm fan, if you really dont care about noise Delta has 220cfm 59db fans.

I just made my own adapters from old case slot covers,(one on top of the rad, one on the bottom) drilled 2 sets of holes,(in the adapters) one as if it was for the stock 92mm fan, another for the 120mm fan. I offset the 120mm just enough to clear the tube from lower to upper waterblock. then just screwed it down and made a fan duct from duct tape (how ironic) so that the few cm of the fan that stuck out past the rad still had a sucking effect on air through the rad. The silver aluminim plates on either side of the rad just pull off with a razorblade (on there with double sided tape)
 
I don't own one, but I see the potential. Sounds like CoolIT has the right idea, wrong pelters on it.

I am just thinking the whole thing is getting too much heat saturation.

I would be willing to hack one up :D
 
That adapter seems like a BAD idea, the entire unit hangs off your case by the fan, I doubt that plastic adapter can hold the weight of the unit. I was afraid of using a 120x25mm fan for the same strength reason, the stock 92mm fan is 38mm deep as well, not only for added cfm but strength.

Plus all 92-120 fan adapters are centered, I dont think that will work. You're going to be pinching the hose that runs from the lower water block to the upper one, its located right beside the stock fan. Thats the main reason I made my own adapter, so I could move it off center and not kink the hoses.
 
4DoorGTZ said:
That adapter seems like a BAD idea, the entire unit hangs off your case by the fan, I doubt that plastic adapter can hold the weight of the unit. I was afraid of using a 120x25mm fan for the same strength reason, the stock 92mm fan is 38mm deep as well, not only for added cfm but strength.

Plus all 92-120 fan adapters are centered, I dont think that will work. You're going to be pinching the hose that runs from the lower water block to the upper one, its located right beside the stock fan. Thats the main reason I made my own adapter, so I could move it off center and not kink the hoses.
Click to expand...

Good points. Why not just leave the stock 92mm fan in there, use the 120mm bracket, and attach another 120mm fan to the bracket? That way you get TWO fans pulling air with very little need for mods, except for another 25-38mm of fan.
 
It may still work like that, however the "fan adapter" is really a joke. Its only meant to make the 92mm bolt to a 120mm exhaust hole, it blocks everything that isnt 92mm fan.

Plus, if you just add another fan you're moving the whole rig further to the front of the case, depends on your setup but as is in my Stacker 830 the tubes to the cpu block are "gently" bent quite sharply as my CPU is located right behind the front edge of the rad, In my case if I doubled up on fans I believe the rad would then be right over the cpu and cause even tighter bends. Its not like you can plubm a 90deg bend ontop of the cpu block. (well I'm not touching the innards.

Again there's also the strength issue there, adding another fan just pushes all the weight out further from the mounting point (your fan) and you gotta make sure your fan-to-fan is strong for the same reason.

ps. I took pics of my setup this afternoon as I was installing a nothbridge and ram cooler, I'll try and get them up so people who dont have this know what I'm talking about.
Edit pics are here http://www3.telus.net/public/janmraz/Computer/ I'll apologize now for the shaky hand, but you get the jist of it. 2/6/9 show the tightness of the tube bends. After going to the 120mm I can no longer swing my fan door open, I can pop it open a couple inches then I have to remove the hinge and take the whole thing out, not a biggie.
 
Can anybody tell me if the Freezone or the Eliminator for that matter will fit in a Qpack. I thought I remember reading somewhere where someone tried this but can't find it now. If it won't fit naturally can it be modded easily? I was thinking of using this for my SFF rig in my sig.
 
Has anyone installed the freezone and had 80mm rear case fans? I have 80mm rear fans and I'm not sure how I would install it.
 
All what I did to add a 120mm fan is take some double sided poster tape, put them on the fan, then stick it to the back of the case. Running prime95 at 33C on amd fx-62
 
I modded with a 120mm fan a delta still no help. Same temps as usual. I wouldnt recomend this cooler to my worst enemy. Its garbage.
 
infospace said:
I modded with a 120mm fan a delta still no help. Same temps as usual. I wouldnt recomend this cooler to my worst enemy. Its garbage.
Click to expand...

Same here, but WTF, here's a recent review of the Eliminator (LESS TEC cooling power than Freezone) that shows 27C overclocked load temps?!?!

http://techgage.com/article/coolit_eliminator_cpu_cooler/4

Someone needs to explain heat conduction principles to me, since I still don't get how the Freezone is supposed to keep up with a high processor load, and it's been decades since I've taken first year physics!
 
I've found a way to make it barable.... Its modded with a 114cfm 120mm fan on the back for exhaust then the stock fan is resting ontop of my video card blowing at the rad. I've also got 3x120mm and 1x140mm sucking in outside air from the front of my case blowing towards the stock fan wich speeds up the airflow, then there's 3x140mm on the side door blowing inwards. Yeah its a lot of fans in my stacker 830 but:

Ambient 19c
QX6700 @3425 1.42v Idle 50c load 68c before I added the stock fan in a blowing position load was 73-78c
 
Recent Bjorn3D review shows results in line with what I got, 56C load temps w/E6600 @ 3.2GHz (approx 100W load). The description of the load temp jumping then slowly rising is also familiar:

http://www.bjorn3d.com/read.php?cID=1055&pageID=3122

It did noticeably better compared to a Thermalright SI-128 with SFF21F fan (64CFM), but I'm thinking a Tuniq Tower would beat a SI-128 easily, so what's the point?

Even with all those problems they do end up "strongly" recommending it as a "almost a perferct product" - WTF?
 
You must log in or register to reply here.
Back
Top