re: PC case fan config

Alright. I will try a couple of those things and post any results.

Anyway, to put a more concise thought process on it, being I've been all over the place:

Your case sounds like it's choking for air, which is causing negative pressure. You could shut down exhaust fans, like you did, but that means you're getting less air flow, because you're not increasing intake, but you're no longer going to get air sucking in through holes in the case, which is unfiltered airflow, but airflow nonetheless.

If your cpu temps can handle it, I think moving your rad to the top, and keeping all of your fans oriented the same (front is still intake, top and back are still exhaust) will significantly increase your airflow for your entire system, reduce dust intake, and probably allow you to slow your fans down a bit if noise is an issue. But ultimately, that's all in theory. Every case ecosystem is its own animal. Run stress test benchmarks, take notes on all your temps, and enjoy yourself. Hell, this kind of nitpicky bullshite is half the fun of diy pc building

If you need any tips on benching software, monitoring software, etc, just let me know. But if you can handle an oc, I assume you've got that already.

My knowledge on fans only goes so far. If you had to choose between these fans for the radiator which would you choose? Would any of them be much different in terms of cooling? As long as I have a top mounted rad, do the fans even matter as much?

Fan 1

CFM: 55

SP: 2.60


Fan 2

CFM: 87.6

SP: 2.30


Fan 3

CFM: 75

SP: 4.20

quote:

---

My knowledge on fans only goes so far. If you had to choose between these fans for the radiator which would you choose? Would any of them be much different in terms of cooling? As long as I have a top mounted rad, do the fans even matter as much?

---

Assuming you're keeping the case front panel on, I would be inclined to go SP both for the top and front. The top because of the rad and the front because of that super choky front panel. But that's, as I've mentioned a few times, in theory.

Reality can be a different animal.

The important thing to remember about standard cooling (so either air or aio, not aio using ice water, etc) is that you can never cool anything below the temperature of the air you're moving over it. That's just how physics works. So when you're considering fans and whether they will help, think of this thought experiment:

Let's say you have your rad, but instead of having air pass freely through it, you gate the air, and only open the gates to replace the trapped air every X Seconds. If you go from five minutes to 30 seconds, you'll see a huge improvement. But if you go from two seconds to one second, you probably aren't going to see much change, because the air probably didn't reach equilibrium with the temp of your rad in that time period, anyway.

Now, that's a silly thought experiment, but it illustrates the point that, once you reach a certain point, moving more air, or faster air, across a heatsink won't give you much, if any, better performance. The trick is finding that point.

GN ran thermal results on aios a while back, focusing on fan normalized performance, and something jumped out at me: for every aio, 25% fan speed had super high temps. When you bumped to 50%, the temperature dropped significantly, generally 5-15 degrees. But when you bumped up to 75%? Maybe a degree. To 100%? Maybe one more. To give you an example, one aio was 81 - 74 - 72 - 71. Once you reached a certain threshold, moving the air faster just didn't matter a ton.

Now, none of that mattered when your rad was your intake, because you needed as much air as possible, regardless, Now that it's exhaust, you only need what you need. And here's where it starts to matter whether you're price conscious. If you aren't particularly price sensitive, buy six static pressure fans, slap them on there, and see where they take you. If you are, I would buy three. Put them on the front intake; put them on the top exhaust. See what your temps do. Put your hand behind your intake and above your rad and see if you can feel any difference in the amount of air moving. Just like with moving the rad, it's all about figuring out the best scenario for your specific usage case.

As for which fans, those numbers are going to be meaningless, assuming they came from different manufacturers, as they all have different methodologies. If they are from the same company, I would probably lean towards Fan 3 in a vacuum, but that's ignoring aesthetics, noise, vibration, etc. Fans are something that I spend very little time looking at stats for and focus far more on trusted reviewers whose methodology I trust.

quote:

---

GN ran thermal results on aios a while back, focusing on fan normalized performance, and something jumped out at me: for every aio, 25% fan speed had super high temps. When you bumped to 50%, the temperature dropped significantly, generally 5-15 degrees. But when you bumped up to 75%? Maybe a degree. To 100%? Maybe one more. To give you an example, one aio was 81 - 74 - 72 - 71. Once you reached a certain threshold, moving the air faster just didn't matter a ton.

---

This caught my eye because I just looked at a chart that seems to describe the same concept.

quote:

---

you're usually better off finding the good for your usage case instead of chasing the perfect.

---

I have this problem every time I want to upgrade something in my PC. I get consumed with all the #s and stats and sometimes try too hard to be perfect.

I think I've settled on fans, all Noctuas.

140mm NF A14 in the rear

120mm Nf A12x25s on the rad

moving the current fans on the rad to the front b/c they're actually pretty good.

Might get some Nf f12s for the front later on or for a push/pull if I wanna switch it up.

*4 years later*

Going to be testing the NF-A14x25s and either Phantek T30s or the Lian Li P28s on the front intake soon. Yes I'm still using this exact setup aside from a few hardware upgrades.

Still have A12-25s on the rad and stock fans on the front intake.