![]() |
![]() |
|||
|
||||
|
|||||||
| Register | Forum Rules | FAQ | Search | Today's Posts | Mark Forums Read |
| Welcome Guest Visitor! Please Register, It's Free and Fun To Participate! | |
|
The EXTREME Overclocking Forums are a place for people to learn how to overclock and tweak their PC's components like the CPU, memory (RAM), or video card in order to gain the maximum performance out of their system. There are lots of discussions about new processors, graphics cards, cooling products, power supplies, cases, and so much more!
You are currently viewing our boards as a "guest" which gives you limited access to view most discussions. You need to register before you can post: click the register link to proceed. Before you register, please read the forum rules. By joining our free community you will have access to post topics, communicate privately with other members (PM), respond to polls, upload your own pictures, and access many other special features. Registration is fast, simple, and absolutely free! To start viewing messages, select the forum that you want to visit from the selection below. After you have registered and read the forum rules, you can check out the FAQ for more information on using the forum. We hope you enjoy your stay here! Note To Spammers: We do not allow unsolicited advertising! Spam is usually reported & deleted within minutes of it being posted, so don't waste your time (or ours)! |
|
| Please Register to Post a Reply |
|
|
Thread Tools |
|
|
#1 | ||||
|
DR Sheldon Cooper FTW
Senior Member
|
Swiftech H220 AIO Water Cooler Review
![]() What sets this kit apart from the rest of them, is that it can be expanded upon. This is possible due to Swiftech designing their own pump, that is five times more powerful then the ones used in other closed loop coolers. The pump is PWM controlled, and has a 60,000 hour MTBF. Swiftech includes a three year warranty for the H220, even if you take it apart to expand from it. The radiator has a copper core, is 29mm thick (54mm with fans installed), and has a built in reservoir for draining and refilling. The barbs used are completely swivel-able, and fit the included 3/8 ID, 5/8 OD tubing with compression clamps. This kit comes with an 8-way PWM splitter, allowing you to run the pump and up to 7 PWM fans off of one PWM header on your motherboard. They can also be daisy chained if you need more (they are sold separately for $9.95). Swiftech includes two Helix PWM fans, which can run from 800-1800RPM, with a static pressure rating of 2.29mmH2O at full speed. All screws needed for installation are included, as is a 1g tube of TIM-Mate 2 thermal paste. The fans come pre-installed for easy plug and play operation. I removed the IHS from the CPU and repasted with MX-2. Doing so dropped temps by 11C. Ive also experimented with different methods (stress test options, fan options, TIM application, etc...) and have decided on the following. So all comparisons between coolers will use these methods. Testing Methods Cooler mounted to the roof in an intake/push orientation (meaning the fans are intaking air from out side the case, and pushing through the radiator in to the case). Push/pull is used for the CNPS20LQ when using the 2x fans. Fans are 3ft away from my head in normal seating position. Prime95 AVX small FFTs 15 minute runs are used for load testing. A thermometer is placed right next to the top fans to measure ambient temperature. Core Temp is used to record temperatures of all cores, idle and load. Idle will be taken at 800RPM fan speed (as low as the Helix fans can go) 30 minutes after the Prime95 run was done. Sound readings will also be done at 800RPM. Thermal Paste Small line of MX-2 between 3570K die and IHS. The CPU will not be removed from the motherboard, so this will not change at all. Small dot of MX-2 between IHS and cooler block. Fans First testing will be done with included factory fans, limited to 1800RPM. 1800RPM was chosen, as that is the max speed of the Helix fans from the H220. Also, anything over that is just unreasonably loud with the SP120Ls. I will also test the SP120Ls, Helix, and Zalman fans on a quiet setting. The RPM will be different from each, but will set them so they produce the same DB rating (29Db +/- 1Db) as NF-F12s 7V at 3ft. Idle testing will be 800RPM from all sets of fans. Then I will test with Noctua NF-F12s. These are highly regarded fans, that are very quiet and smooth sounding, while offering good air flow and high static pressure. These will be tested at 1100RPM, which is the speed they run with the LNA (Low Noise Adapter). Only other case fan installed is a stock Fractal rear exhaust at 7V. There is no perceivable noise from this fan at sitting distance. The Twin Frozr fans on the MSI GTX-670 is manually set to 30% fan speed (lowest setting). PSU fan will not be spinning. For sound testing numbers, I will be using the sound meter app on my Nexus 4. This isn't the most accurate way of getting true decibel readings, but since it can be used to tell whether one is louder then the other. It will be placed directly next to my keyboard on my desk, and sitting directly above the fans on top of the case, with no other ambient noise. Components Intel I5 3570K IHS repasted with MX-2 running at 1.225V in BIOS, 1.224V load, 4.5Ghz Gigabyte Z77X-UP4-TH motherboard 4x4GB Crucial Tactical Tracer 1866Mhz 9-9-9-27-2T MSI GTX-670 PE idle Seasonic X-650 Gold running in Hybrid mode so fan does not spin 2x Samsung SSDs behind motherboard tray Fractal Arc Midi R2 case, drive cages removed, stock fans removed aside from rear exhaust position at 7V, all panels installed on case Cooling And Sound Results Zalman Stock Fan - 23C Ambient Load 1800RPM - 3ft 38Db - 1" 52Db 68 / 75 / 77 / 76 Idle 800RPM - 3ft 28Db - 1" 39Db 35 / 41 / 38 / 40 Zalman Stock Fan Quiet Setting - 23C Ambient Load 1000RPM - 3ft 30Db - 1" 41Db 79 / 85 / 87 / 87 (DNF, apps were crashing once 85C was hit) Zalman NF-F12s 7v - 23C Ambient Load 1100RPM - 3ft 28Db - 1" 42Db (fans are spaced with push/pull rather then side by side) 65 / 72 / 74 / 73 Idle 800RPM - 21Db - 1" 36Db 33 / 36 / 34 / 35 Corsair H100i SP120L Fans - 23C Ambient Load 1800RPM - 3ft 40Db - 1" 61Db 57 / 64 / 65 / 65 Idle 800RPM - 3ft 26Db - 1" 42Db 31 / 33 / 32 / 33 Corsair H100i SP120Ls Quiet Setting - 23C Ambient Load 1100RPM - 3ft 30Db - 1" 50Db 64 / 70 / 70 / 70 Corsair H100i NF-F12s 7v - 23C Ambient Load 1100RPM - 3ft 29Db - 1" 50Db 63 / 69 / 69 / 69 Idle 800RPM - 3ft 22Db - 1" 41Db 29 / 33 / 30 / 33 Swiftech H220 Helix Fans - 24C Ambient Load 1800RPM - 3ft 36Db - 1" 65Db 59 / 67 / 66 / 66 Idle 800RPM - 3ft 25Db - 1" 39Db 31 / 33 / 30 / 33 Swiftech H220 Helix Fans Quiet Setting - 23C Ambient Load 1250RPM - 3ft 30Db - 1" 50Db 62 / 69 / 69 / 68 Swiftech H220 NF-F12s 7v - 24C Ambient Load 1100RPM - 3ft 25Db - 1" 49Db 61 / 68 / 68 / 67 Idle 800RPM - 3ft 21Db - 1" 39Db 30 / 32 / 28 / 32 Conclusion So much to talk about with the H220, don't know where to start. Guess we'll start with the cooling performance. With the same fans on all three coolers (the Noctua NF-F12s), the H220 leads the H100i by 2C over ambient, and 7C over the Zalman. When comparing the three at the same noise level, the H220 is still 1C cooler then the H100i, and vastly cooler then the Zalman (more to discus on this later). Now, when we look at the performance of them with the included fans, at the same speed (1800RPM), the picture changes. Here the H100i beats the H220 by 1C over ambient, and 12C over the Zalman. I can see a couple of reasons for this. First, the biggest reason that I can think of, is the fans performance in a restrictive environment. Each fan is designed a different way, and some fans are better then others when they are "blocked" from where they are trying to get the air to pull through. The Arc Midi R2 case has a pretty thick fan filter in the top panel, and not much room between the fan and filter. The Noctua NF-F12s and SP120Ls don't seem to be effected by this as much as the Zalman and Helix fans. Before I got this case, I had a NZXT Phantom 630, which had practically no restriction behind the fans. In that case, the Zalman with its stock fan turned down had no trouble at all keeping temps under 75C. The difference in performance was also much smaller compared to the H100i. The second reason, is even though they are spinning at the same speed, the Zalman and SP120L fans are much louder then the Helix fans. They are able to do more at the same speed, but at the expense of more noise. With an AIO cooler there are three main components to be compared. The block/pump unit, the rad, and the fans. Clearly, the H220 has the best block/pump and rad of all the AIO coolers. The fans that come with it just don't cool as well as some other options out there. They are pretty quiet (though not as quiet or nearly as smooth sounding as the NF-F12s), look nice, and are PWM controlled though. If you put the H220 up against the H100i with the same set of fans, or where both are making the same amount of noise, the H220 is the better performer. But if you just want straight cooling performance, don't care so much about noise, and don't want to spend more money on other fans, then the H100i is very slightly better. However, the major point to the H220, is the expandability of it. Ive already ordered an other radiator, new tubing, and have a block ready to cool my GTX 670. This will continue to cut down on noise, and also improve cooling performance. On to build quality and installation. There is a very heft weight to this unit that other AIOs don't have. Especially the block/pump. The rad of my unit was spotless with no bent fins, and all the air bubbles escaped from the pump in a matter of seconds. Negative points. The tubing was very dirty/stained. Its "gunked" in to the outside of the tubing, and effects the look of it. My unit also came with the pin for the blue PWM wire of the pump being bent, and not making good contact. This took me a while to figure out why it was running full speed all the time. After fiddling with the wire a bit, and installing it in a way that puts tension on it, it now works. The mounting of the block/pump could be much easier. The included back plate has four little foam squares that don't do anything but make it harder to instal. They are sticky, but so small when you try to thread the screws through, it pushes it right off. So you have to either hold the back plate with one hand, and try to hold the block/pump and screw it in with the other, or prop something against the back plate while you mount it. Also, because the radiator has swivel barbs, and the res built in, its longer then most 240mm radiators. It was a very tight fit in the top of my Arc Midi R2 case. The radiator is 29mm thick instead of 25mm thick like the H100i and most other 240mm AIO coolers. This means that cases with 50mm of space from the top to the motherboard may not fit (depends on your motherboard). To close this up, overall I'm very happy with this kit. Its a great way to for first timers to get their feet wet with water cooling, and is a great bargain for enthusiasts who don't want to blow a ton of money on a full custom kit all at once. After having many custom kits before, I can tell it would take at least double the money to get a custom kit that would perform noticeably better then this. Id also trust it to last a very long time, and Swiftech's customer support is top notch. They have been in the water cooling game for a long time, and really know what they are doing. Expect an H320 to be released soon as well, with a 3x120mm radiator. I believe a white version may be coming as well. You can also check their site for extra fittings, tubing, PWM splitters, rads, and Hydrx 2 coolant for expanding it. H220 gets a full recommendation from me. H220 Continued Testing Push/Pull NF-F12s 7v - 21C Ambient Load 1100RPM 4x fans 56 / 64 / 63 / 63 Improvement over 2x fans - 1C cooler over ambient Exhaust Pull NF-F12s 7v - 21C Ambient Load 1100RPM 60 / 66 / 66 / 66 Difference from intake push method, 1C warmer over ambient Last edited by Cecil : 03-16-2013 at 09:07 PM. |
||||
|
|
|
|
#2 | ||||
|
Partial to LUNAR
Senior Member
|
Looks good. Looking forward to your results.
|
||||
|
|
|
|
#3 | ||||
|
DR Sheldon Cooper FTW
Senior Member
|
Got the testing method established, and have started on the final results.
Additional Comment: All testing is done! Last edited by Cecil : 03-13-2013 at 03:13 AM. Reason: Automerged Doublepost |
||||
|
|
|
|
#4 | ||||
|
"Think"
Senior Member
|
This looks like the Cadillac of the closed loop systems out there,
Nice review C |
||||
|
|
|
|
#5 | ||||
|
Mad Warranty Voider
Senior Member
|
Very nice. Hope Amazon carries it soon, I have some gift cards to use.
|
||||
|
|
|
|
#6 | ||||
|
ambient water-cooling
Senior Member
|
Nice review Cecil. Thanks.
|
||||
|
|
|
|
#7 | ||||
|
EA Ports
Senior Member
|
|||||
|
|
|
|
#8 | ||||
|
"Think"
Senior Member
|
It is an awsome product as it has great preformance and with adding more rad surface easily expandible for other duties
Warrenty will be dicy if tampered with though I am sure |
||||
|
|
|
|
#9 | ||||
|
DR Sheldon Cooper FTW
Senior Member
|
Quote:
3 year warranty on all parts, regardless if you expand on it or not. |
||||
|
|
|
|
#10 | ||||
|
"Think"
Senior Member
|
this would be the cats meow for a mITX build IMO
|
||||
|
|
|
|
#11 | ||||
|
DR Sheldon Cooper FTW
Senior Member
|
Just got to be careful about the board used. Someone tried it on the Asrock ITX Z77, and it blocked a ram slot unless using ultra low profile ram.
I got an XSPC EX240 crossflow, and MCW-80 to add in to the loop. Using Primochill blue tubing. All the stuff should be here on Tuesday. More testing next week. |
||||
|
|
|
|
#12 | ||||
|
"Think"
Senior Member
|
I'm in for the results
|
||||
|
|
|
|
#13 | ||||
|
DR Sheldon Cooper FTW
Senior Member
|
Quote:
|
||||
|
|
|
|
#14 | ||||
|
"Think"
Senior Member
|
Gotta think when they drop the tri in the market they will not be able to make them fast enough for demand
|
||||
|
|
|
|
#15 | ||||
|
ambient water-cooling
Senior Member
|
Thanks Cecil for the link. I usually don't groove on reading Annual Reports but that was interesting.
Maybe true Witchdoctor but I wonder about the pump. They are going to have to price it so it won't compete with their own 35x, right? Might be a real bargain for a quiet PWM pump. Speculation is fun.
|
||||
|
|
|
|
#16 | ||||
|
"Think"
Senior Member
|
I kicked around the idea of using a Drive2 I already had for a current build I am doing, it is a great pump block combo, ran 2011 @ 4.5 for a year rock solid. I think this product is a good comprimise. The tri will give entusiast level water cooling to the masses. With all the closed loops and now this expandable system that can remain closed if you so choose air to air solution are running out ideas to remain viable
|
||||
|
|
|
|
#17 | ||||
|
DR Sheldon Cooper FTW
Senior Member
|
I believe the stand alone pump/block would be priced about $100, but dont know for sure. Its $40 short of the AD2, which is enough to make it worth it to not go for the AD2, but also not so much to steal sales from the AD2. Also, $40 saved against the H220, which for those who would use other fans and rad, it would be worth it, but again, not steal sales from those who want the AIO.
The H320 is only supposed to be $20 more then the H220, at $160. |
||||
|
|
|
|
#18 | ||||
|
ΜΟΛΩΝ ΛΑΒΕ
Senior Member
|
Great review Cecil, looking forward to the expanded results. Contemplating waiting for the 320 to CF some 7970s so i can postpone modifying my 800D. That being said if i can achieve 60s like Gabe says for this pair of 680s and a 3770k through just the h220 alone i'll be stoked.
http://youtu.be/N5mJTGnrNWc?t=26m4s |
||||
|
|
|
|
#19 | ||||
|
DR Sheldon Cooper FTW
Senior Member
|
Quote:
While gaming, you would be able to get away with two cards and the CPU, but adding another rad would help a lot. Stuff I ordered for mine. ![]() Instead of compression fittings, I went with clamps and painted them black to match the look of the pump and rad fittings. ![]() ![]() Layout,
|
||||
|
|
|
|
#20 | ||||
|
ΜΟΛΩΝ ΛΑΒΕ
Senior Member
|
Quote:
http://www.youtube.com/watch?feature...&v=tZgctchIQ7M H220 + 3770K OCed and stock SLI 680s running at 81C forgot about this in the video. Guy states in the comments that the temps aren't representative of two 7970s on the last custom rig, because Uniengine is being run in window mode, in which CFX doesn't work? If that is so does 1 7970 get 120+ frames per second in heaven? My single 7970 even at 1600x900 doesn't get close to 120+ Last edited by Blackops_2 : 03-20-2013 at 06:39 PM. |
||||
|
|
|
|
| Please Register to Post a Reply |
| Thread Tools | |
|
|