Swiftech H220 Status Update - We Made a Mistake

Linus Tech Tips ·Linus Tech Tips ·2013-05-07 · 838 words · ~4 min read
Floatplane YouTube

Transcript

JSON SRT VTT 61
0:02 in our video review of the swifttech h220 we had a 20 Dee difference between
0:08 our Noctua n12 fans and the swifttech
0:12 Helix fans now we thought those results were pretty suspect at the time but we
0:17 reren them a few times um the problem
0:20 though is that we made a critical error we were using no to as low speeed
0:24 adapters on all of the fans and in the past it's never been an issue because
0:28 we've always just run KN to a fans but what we discovered is that while the
0:32 lowp speed adapters which we thought were just your standard sort of
0:35 undervolt from 12vt to 9vt sort of adapters we thought everything was hunky
0:40 dory um what we discovered is that while it'll tone these guys down from around
0:45 1400 RPM to around 1100 RPM it turns
0:48 swiftex Helix fans all the way down from 1,00 RPM to 600 RPM so the Helix fans to
0:54 get that 20° difference we're actually running at half of the RPM of the Noctua
1:00 fans Not only that but the swifttech fans were only run with the pump at full
1:04 speed and what we found out is that our units here may actually be not quite
1:10 optimal in terms of their performance because these are early samples that are
1:14 not using Swift Tech new vacuum filling
1:18 procedure that basically amounts to zero
1:21 air bubbles inside the system so we might have had air bubbles that were
1:24 getting caught in the pump causing the performance to degrade um with the pump
1:28 running at full speed which might be part of why we got our best results when
1:32 running the pump at half speed so we're waiting for a retail unit of the h220
1:38 and so that we can actually take it and retest it in a number of different
1:43 configurations so we're going to run our noctu a fans at 600 RPM to see how
1:49 comparable that is just you know let's see what happens if we both fans
1:53 we're going to run our Noctua fans at 1100 RPM which is our standard testing
1:57 procedure then we're going to run swiftex fans at 600 RPM again with the
2:02 new fixed unit all with the pump at half because that doesn't seem to affect
2:05 performance much and then we're going to run Swift text Unit at 1100 RPM then
2:10 we're going to take some additional fans we're going to grab some gentle typhoons
2:14 some GED Wings some air penetrator 121s we're going to run all of those at 1100
2:18 RPM this is all in a pull configuration because that's the configuration that I
2:22 prefer to use we've added something to our test bench here that we didn't have
2:26 before so check it out we have a variable fan speed control ER in the
2:30 front so we can use software to exactly
2:33 monitor the RPMs of the fans and fine-tune them to make sure that noise
2:37 levels and RPM levels are the same and
2:40 then last but not least swifttech has also asked us to test theirs in a push
2:45 configuration which is the way it ships stock and to also try using pwm control
2:51 for the whole unit itself because part of the problem may be that swiftex fans
2:56 are not designed for voltage control but are designed for PW m control using a
3:01 four pin connector and a compatible pwm chipset so we're going to plug
3:05 everything into the CPU fan header and then we're going to plug both fans and
3:10 the pump into swiftex pwm splitter and
3:13 then report that result as well and on
3:17 that one we're going to Target 1100 RPM for the fans so this should give us the
3:21 most complete and clear picture of how the Swift Tech cooler performs now I
3:26 it's going to take us a little while to get this done so that's why I'm releasing this video so you guys know
3:31 that it's coming but in the meantime remember guys our result with it
3:35 performing better than everything else is still valid because this engineering
3:41 sample unit had everything going against it and with the Noctua cooler compared
3:45 to every other liquid cooler that we've tested due to its copper radiator
3:49 compared to the aluminum finned radiators that are used in other lowcost
3:52 Solutions due to just the overall design of the unit the better pump the better
3:57 uh see swifttech designed base PL
4:00 dramatically better design dramatically better performance than other lowcost
4:04 coolers it is the best performing cooler on the market if you add aftermarket
4:08 fans we just need to do more investigation into the stock ones that
4:12 are included thanks for watching this video sort of a Blog Style video and as
4:17 always don't forget to subscribe to lus Tech tips for more unboxings reviews and
4:20 other computer videos