1
00:00:07,919 --> 00:00:16,240
so i'm in the middle of setting up my my heatsink test bench this is going to be

2
00:00:12,320 --> 00:00:18,400
an x79 platform with the gtx 580 as well

3
00:00:16,240 --> 00:00:23,279
as a vengeance c70 case because it'll allow me to easily mount tall air

4
00:00:20,320 --> 00:00:27,680
cooling heat sinks dual radiator sort of all-in-one units like the h-100 as well

5
00:00:25,840 --> 00:00:32,160
as single radiator all in one unit such as the h80 so the reason i'm setting

6
00:00:30,400 --> 00:00:37,360
this up is so that i can do some testing but while i was working on getting it

7
00:00:34,559 --> 00:00:42,160
going i pulled my h100 off of my test bench and i noticed something that i

8
00:00:39,680 --> 00:00:46,480
wanted to sort of talk about so i get a lot of questions about should i

9
00:00:44,079 --> 00:00:49,280
use my fans in push you can see this is configured and pushed the blades are

10
00:00:47,920 --> 00:00:53,280
going to be rotating this way pushing air through

11
00:00:51,520 --> 00:00:58,000
the fins of the rail through the fins of the radiator out the

12
00:00:55,120 --> 00:01:01,520
other side or do i want my fans in pull which is where you'd flip them around

13
00:00:59,760 --> 00:01:04,879
and have them pulling air through the radiator so i don't know if i have an

14
00:01:03,440 --> 00:01:09,200
example of that yeah see these ones are configured in push as well right here so

15
00:01:07,439 --> 00:01:13,280
they're pushing air that way or do i want to have my fans configured

16
00:01:11,360 --> 00:01:16,960
in push pull so that's where you actually have fans pushing on one side

17
00:01:16,000 --> 00:01:23,520
and pulling through the other side of the rad in order to

18
00:01:20,799 --> 00:01:26,720
get well greater static pressure so the answer to that is actually not that

19
00:01:25,360 --> 00:01:31,680
simple so there's a few different things to consider so number one

20
00:01:28,560 --> 00:01:33,439
is performance for performance push or

21
00:01:31,680 --> 00:01:37,600
pull sort of if you have two fans on a dual 120mm grad either on this side or

22
00:01:35,920 --> 00:01:41,520
on the other side actually it doesn't matter this has been proven time and

23
00:01:39,360 --> 00:01:44,960
time and time and time again it's been debated until

24
00:01:43,200 --> 00:01:47,360
you know you can debate it till the cows come home it's not going to change the

25
00:01:46,240 --> 00:01:51,360
fact that within margin of error it performs

26
00:01:49,600 --> 00:01:55,520
pretty much the same now when it comes to push pull a lot of people are taking

27
00:01:54,159 --> 00:02:01,680
sort of i mean especially it was fashionable back in the day and i bought some myself to grab something cheap like

28
00:01:59,280 --> 00:02:06,399
yate loon fans and then you know you do a layer of fans a layer of radiator a

29
00:02:04,000 --> 00:02:10,879
layer of fans and you can turn the voltages of those fans down

30
00:02:08,640 --> 00:02:15,760
in order to make them quieter even though you have more of them and you

31
00:02:12,400 --> 00:02:18,239
still get decent performance because

32
00:02:15,760 --> 00:02:22,400
the push-pull configuration allows more pressure to be built up by the fans now

33
00:02:20,480 --> 00:02:26,319
think about it this way guys if i take two fans that okay here these two fans

34
00:02:25,040 --> 00:02:30,879
are a great example because of the way they're positioned if i take these two fans they're spinning at the same RPM

35
00:02:29,440 --> 00:02:34,959
they're capable of moving the same amount of air i take this heatsink out of the middle and i put them next to

36
00:02:33,680 --> 00:02:40,800
each other they're not going to make the air go any faster like the actual

37
00:02:38,319 --> 00:02:44,400
cfm is not going to change what's going to happen though is if there's

38
00:02:42,160 --> 00:02:48,000
resistance in the middle then they're going to get the benefit of helping each

39
00:02:46,239 --> 00:02:51,440
other to move air through the resistance they're going to gain more pressure so

40
00:02:49,760 --> 00:02:56,000
people often ask me should i run in push pull i say no the reason i say that is

41
00:02:54,160 --> 00:03:00,080
because there are so many fans on the market that are optimized for pressure

42
00:02:57,920 --> 00:03:03,440
such as the sp series from Corsair although those aren't necessarily my

43
00:03:01,360 --> 00:03:06,480
favorite Noctua's f series are excellent scythe gentle typhoons are excellent

44
00:03:05,120 --> 00:03:12,239
silverstone air penetrators are excellent lots of good pressure optimized fans

45
00:03:09,760 --> 00:03:15,200
so you actually have all this pressure available to you by putting another

46
00:03:13,920 --> 00:03:20,080
layer of fans on the other side all you're doing is increasing the noise because remember you can't make the air

47
00:03:18,239 --> 00:03:23,599
go any faster in an open environment and if you already have a pressure optimized

48
00:03:21,680 --> 00:03:26,959
fan then you're good i mean if you had a

49
00:03:24,879 --> 00:03:30,959
non-pressure optimized fan so this older Noctua s series is a great example see

50
00:03:29,040 --> 00:03:34,239
that you could spin that at 10 000 RPM it would never generate any pressure

51
00:03:32,640 --> 00:03:38,560
because there's so much room for air to leak around the outside this is not a

52
00:03:36,480 --> 00:03:45,280
good radiator fan whereas you see these nice broad blades on the sp-120 that is

53
00:03:41,920 --> 00:03:47,760
an optimized radiator fan so anyway

54
00:03:45,280 --> 00:03:52,799
long story short is push-pull necessary i would make the argument no get a

55
00:03:49,440 --> 00:03:54,480
proper fan run in pull and there's a

56
00:03:52,799 --> 00:03:59,120
reason for that now when i was taking this radiator off of my test bench i was

57
00:03:56,879 --> 00:04:02,319
reminded why i normally configure radiators in pole

58
00:04:02,799 --> 00:04:07,760
see you can see the dead spot in the middle where the fan hub is and then you

59
00:04:06,159 --> 00:04:11,599
can also see where all that dust is built up this system was actually

60
00:04:09,280 --> 00:04:16,720
running for about a year on the floor so that's why there is so much disgusting

61
00:04:14,159 --> 00:04:20,880
dust buildup on it so why do i always run radiators in pole configuration

62
00:04:19,280 --> 00:04:26,000
because once you've got a fan girl on there and a fan on here without actually

63
00:04:23,520 --> 00:04:29,040
taking the fan off it is impossible to clean

64
00:04:26,880 --> 00:04:32,080
on the other hand whoops

65
00:04:29,919 --> 00:04:36,240
if you run a configuration like what i did with here let me just go over here

66
00:04:34,639 --> 00:04:40,880
i'm just gonna have to adjust my exposure a little bit there we go if you

67
00:04:38,560 --> 00:04:44,720
run a configuration like what i did with my thermo-chill here you can see this

68
00:04:42,479 --> 00:04:49,360
radiator was also run for a couple years and hasn't been extensively cleaned

69
00:04:46,479 --> 00:04:53,040
because the fans on the other side there those are my silverstone air

70
00:04:50,800 --> 00:04:57,440
penetrators are pulling air through it so it's very very simple to just go

71
00:04:55,280 --> 00:05:01,600
through once in a while and you know blow a can of dust at it or use a vacuum

72
00:04:59,520 --> 00:05:06,240
to clean it off whereas if the fan's in the way it is pretty much

73
00:05:04,000 --> 00:05:09,120
impossible to do it and you end up with something that looks like this and

74
00:05:07,520 --> 00:05:12,479
without taking the whole thing apart is very difficult to clean now this is easy

75
00:05:10,720 --> 00:05:17,199
when it's on a test bench like this but what if it was in a case what if i had

76
00:05:14,639 --> 00:05:20,000
it configured in push with the fans on the bottom here blowing out how am i

77
00:05:18,400 --> 00:05:23,919
going to clean that it's like in the middle of a case whereas if the fans are

78
00:05:22,240 --> 00:05:26,400
right up against the edge of the case and all i have to do is get in there and

79
00:05:25,280 --> 00:05:30,639
kind of go take the dust out of it and you are good

80
00:05:28,880 --> 00:05:35,280
to go so thank you for checking out my little sort of spiel about push-pull pressure

81
00:05:34,080 --> 00:05:40,560
fans non-pressure fans don't forget to subscribe to Linus tech tips for more

82
00:05:38,000 --> 00:05:44,080
unboxings reviews and other disgusting videos i mean look at that

83
00:05:42,320 --> 00:05:47,120
it's disgusting it's like cat hair i think it's mostly

84
00:05:45,680 --> 00:05:50,600
cat hair stupid cats