Swiftech H220 AIO Review

[NRG!]

Well, I just purchased one of the H220s. May configure it with Cougar PWM fans in a push/pull configuration as they worked so well in the Corsair H100.

Planning to purchase a pair of Komodo 7970 waterblocks and another radiator from his product line.

M. Grouchon appears to be someone that is dedicated that the term mediocrity shall never be associated with his Swiftech brand.

The komodo/hydrocopper waterblocks aren't exactly the pinnacle of waterblock design tbh. Middle of the pack performers with high restriction and a high price. Aesthetics to each their own, but I think they are very plain compared to offerings from some other brands.

I'd like to see swiftech improve those in the next iteration because I like their complete full cover style, and the only other offering like that is the dreaded EK CSQ. I just can't stand the current plastic guff with a sticker on it.

 

[BramSLI1]

Can I just ask a really important question?

What is your protocol in the case of a leak? do you take care of your customers?

I ask because I had a coolit leak on me last year causing £300 worth of damage and they simply told me to dunk all of my stuff in distilled water and when I told them I wasn't comfortable doing that (as it would void my warranties) they told me to f**k off.

Do you cover your customers? I know leaks are very rare but it's always good to know that a company will do the right thing.

Our warranty doesn't cover damage to components if our product leaks. This is pretty much the way we've always done things and most other water cooling companies won't cover hardware damage either. In fact, I don't know of any off of the top of my head that do. I hope this answers your question and I'm sorry that it isn't the one that you were probably looking for.

 

[Fetzie]

http://www.youtube.com/watch?feature=player_embedded&v=HJQBzMALZ7s

Linus just posted a video, saying that the engineering sample he was sent could have been faulty (different filling procedure for engineering samples leading to air-bubbles in the loop), and that the fan speed adapters used throttled the stock fans right down to 600 RPM. It still beat the competition, so any lead it has would be even bigger if it were indeed faulty. He is going to wait for a retail unit to retest.

Could TTL's sample also have had this issue (which would certainly reduce its performance)?

 

[Greenback]

http://www.youtube.com/watch?feature=player_embedded&v=HJQBzMALZ7s

Linus just posted a video, saying that the engineering sample he was sent could have been faulty (different filling procedure for engineering samples leading to air-bubbles in the loop), and that the fan speed adapters used throttled the stock fans right down to 600 RPM. It still beat the competition, so any lead it has would be even bigger if it were indeed faulty. He is going to wait for a retail unit to retest.

Could TTL's sample also have had this issue (which would certainly reduce its performance)?

I could of been faulty! not it was.

With the stock fans it had no lead, it was with the noctua fans it had a lead. And also he mentions he uses the fan reducers in every test which means that they also wouldn't of been running at full speed, any comparison is flawed unless he retests everything the same

 

[BradKing]

http://www.youtube.com/watch?feature=player_embedded&v=HJQBzMALZ7s

Linus just posted a video, saying that the engineering sample he was sent could have been faulty (different filling procedure for engineering samples leading to air-bubbles in the loop), and that the fan speed adapters used throttled the stock fans right down to 600 RPM. It still beat the competition, so any lead it has would be even bigger if it were indeed faulty. He is going to wait for a retail unit to retest.

Could TTL's sample also have had this issue (which would certainly reduce its performance)?

Not sure if OC3D got an engineering sample or not, if there would be air in the loop that obviously could impact the results; that said, I think the rad was in the top and the reservoir bit was the highest in the loop, so it should have been pretty OK after a while. Even then, I'd cut back the hose length myself so it would be hard to get it vacuum filled at home anyway

The other part is a lot more interesting, the fact the fans didn't seem to react correctly to lower/direct voltages and that Swiftech asked Linus to do a pwm run too, since the OC3D tests were at 12V for the fans and the pump. If that screws up the performance a bit it could well explain how Swiftech feels it should perform better.

It's looking better already Now we need to get some resellers in NL (H220 is 140 euro before delivery, h100i under 100 euro delivered, that's not a gold award )

 

[jarryy]

so much hassle over a AIO, all these test results has just taught me to stick with my NH-d14 :3

 

[Phelan]

I could of been faulty! not it was.

With the stock fans it had no lead, it was with the noctua fans it had a lead. And also he mentions he uses the fan reducers in every test which means that they also wouldn't of been running at full speed, any comparison is flawed unless he retests everything the same

He is doing a complete retest. The difference is that while the other fans worked fined at 7-9v (whatever the reducers were) and ran at about 1100-1200 rpm, the Helix fans do not. The reducers dropped their RPMs all the way down to 600 rpm. He also mentioned that the review units weren't filled using Swiftech's new vacuum fill system, and it had an air bubbles in it, and that running the pump at max speed was pushing the air around causing excess noise and having an adverse effect on temps, which is why he got better temps running the pump at 50%. As I mentioned earlier, with a pump this strong, there's no reason to run it full speed when using it as a CLC. 60% speed (about 1800 rpm, what other AIO pumps run at) will provide the same or better temps with much less noise.

 

[Greenback]

He is doing a complete retest. The difference is that while the other fans worked fined at 7-9v (whatever the reducers were) and ran at about 1100-1200 rpm, the Helix fans do not. The reducers dropped their RPMs all the way down to 600 rpm. He also mentioned that the review units weren't filled using Swiftech's new vacuum fill system, and it had an air bubbles in it, and that running the pump at max speed was pushing the air around causing excess noise and having an adverse effect on temps, which is why he got better temps running the pump at 50%. As I mentioned earlier, with a pump this strong, there's no reason to run it full speed when using it as a CLC. 60% speed (about 1800 rpm, what other AIO pumps run at) will provide the same or better temps with much less noise.

If the stock fans on the kraken's and corsair had a reducer and ran at about 1100-1200 rpm when he tested them then it isn't running at it's best, he didn't notice on this who is to say he missed it on the others

He says that it may have air bubles not it has

 

[d6bmg]

I would say it is a descent cooler. We all have hoped for improvement, but alas, it's not there. Still good as second choice cooler.

 

[Phelan]

If the stock fans on the kraken's and corsair had a reducer and ran at about 1100-1200 rpm when he tested them then it isn't running at it's best, he didn't notice on this who is to say he missed it on the others

He says that it may have air bubles not it has

He said in the followup video that after investigating the other fans ran fine with the reducers. The Helix PWM fans were the only ones he found to have issue with them in the investigation. But he's retesting all of them afaik.

 

[Greenback]

He said in the followup video that after investigating the other fans ran fine with the reducers. The Helix PWM fans were the only ones he found to have issue with them in the investigation. But he's retesting all of them afaik.

He says it's been "fine in the past" Not after investigating, but with out retesting who knows?.
But still comes back to if the stock fans had reducers then they weren't running at best thus a comparison is flawed, The H100i uses high speed fans H220 uses lower speed fans thus if you ran fans on both of them at 1100-1200 rpm thats nearly full speed for H220 and less then 50% for H100i

 

[Resident]

... it full speed when using it as a CLC. 60% speed (about 1800 rpm, what other AIO pumps run at)

1800rpm for all other AIO pumps? You're sure about that are you?

Pretty sure that states just under 2300rpm

 

[Phelan]

1800rpm for all other AIO pumps? You're sure about that are you?

Pretty sure that states just under 2300rpm

Whoops my bad; I was misinformed on pump's speed. Regardless, I'm pretty sure the H220 pushes more water @1800 rpm than the cool-it/asetek units @2300, or at least similar.

 

[JettonX]

So how is it that these results were so different to other reviews?

 

[tinytomlogan]

So how is it that these results were so different to other reviews?

Because we keep strict testing procedures and wouldnt change them to make this look different.

All info you need is in the conclusion. Its also worth noting Gabe understood and explained our results validating them.

 

[JettonX]

Guess that's the H220 off the list then!