PSU fan up or down?

[martin-w 3,235]

So what do you say lads, better up or down?

 

This is one of those arguments that's raged for years, a bit like the positive or negative case pressure debate.

 

I have to say, I'm totally against PSU fans facing down. There was a time, before the bottom mounted PSU trend, when PSU's were mounted at the top of the enclosure with fan orientated so it ingested air and vented it outside. Our PSU's got along just fine, they didn't overheat, the fans spun at the RPM you would expect and PSU lifespan wasn't compromised at all.

 

Then along came the bottom mounted PSU trend, enclosure manufactures thought it would be a great idea if the PSU faced down and sucked in cool air from outside. Sounds like a great idea, but wait, what if the PSU malfunctions and spays a shower of sparks onto a flammable material beneath the enclosure? So manufactures like Lian Li designed their intakes with this in mind. Small holes drilled into the bottom of the enclosure [very restricting] or worse still canted vents like an almost closed venetian blind, resulting in a very restricted intake and the PSU fan providing far less airflow than it's CFM rating would suggest. To combat this the manufactures raised the PSU off the bottom of the enclosure. Great idea, now the PSU is sucking in air out of the enclosure, negating the entire purpose of a down facing fan.

 

Then of course we have dust to consider. Not only do we have a situation where the vents at the bottom of the enclosure restrict the fan, but we add a dust filter to the mix too.

 

I've just bought a new Lain Li X510 for my new Skylake build. Awesome enclosure, absolutely love it. However, I fully expected the vents for the PSU to be the typical Lian Li almost closed venetian blind type, no problem, I'll mount it fan up anyway. What I wasn't expecting was for the vents to be badly aligned with my PSU to the extent that the 1/3 of the PSU fan misses the vents. I could mod it of course, I would have no issue doing this if I were a fan of down facing PSU's.

 

So has anyone else discovered that their PSU's run hotter with fan down, restricted by the dodgy vents and filter? Does anyone think as I do, or am I on my own. Perhaps I'm missing something?

 

As the saying goes, if it ain't broke don't fix it. Mount the PSU with fan up.

[TurboKen 221]

Hi Martin,

 

I would agree, mount the fan up (pulling air from under the case out the back). As the heat will want to rise up and out the back naturally why fight it. And because I'm picky I would also modify the case to have better fan air flow. But I bet its not really needed. Good Luck with the new build!

[martin-w 3,235]

I would agree, mount the fan up (pulling air from under the case out the back). As the heat will want to rise up and out the back naturally why fight it. And because I'm picky I would also modify the case to have better fan air flow. But I bet its not really needed. Good Luck with the new build!

 

Hi Ken.

 

If the PSU fan is up, orientated so it's on top, it will draw air out of the PC enclosure and vent it out of the enclosure. This is may favoured orientation. Not pulling air from under the PC case.

 

I think you may have misunderstood me. :smile: Pulling air from under the PC case, through restrictive grills and filters compromises PSU cooling in my view.

 

Yes the air entering the PSU will be somewhat warmer with the PSU fan up, because it's being cooled by PC enclosure air, but not enough to be an issue at all.

[stans 703]

The way I have always set up my PSU's:

If mounted at the top of the case, fan faces down.

If mounted at the bottom of the case, fan faces up.

[martin-w 3,235]

The way I have always set up my PSU's:

If mounted at the top of the case, fan faces down.

If mounted at the bottom of the case, fan faces up.

Yep would agree with your choices.

 

These days though cases are designed with vents at the bottom, encouraging a downward facing fan. Great idea in principle, but not in practice.

 

Looked again at my X510 yesterday, and also my old Lian Li case. Utterly restricting are the vents, plus filter of course.

 

It's possible to grab a Dremel and cut them out, thus improving the intake no end, but then there's little defence against a shower of sparks from a failed PSU falling onto anything potentially flammable beneath the case.

 

I shall most definitely stick to my preferred method, namely fan up sucking air into the PSU from inside the case. Will have to mod the X510 a bit though. The bottom compartment/shelf is attached with rivets, will need to drill them out and remove. It doesn't do much anyway to be honest. It's supposed to be a separate PSU compartment, but it's not that at all, the front is open and the end is open, no thermal isolation whatsoever. Awesome case though!

 

I'd be interested if anyone on the forum has experienced issues with PSU down. I recall someone had high PSU fan RPM and had to switch to fan up.

 

It seems case manufactures aren't experts in fluid dynamics, we can't trust them to get it 100% spot on.

[onebob 125]

Yep would agree with your choices.

 

These days though cases are designed with vents at the bottom, encouraging a downward facing fan. Great idea in principle, but not in practice.

 

Looked again at my X510 yesterday, and also my old Lian Li case. Utterly restricting are the vents, plus filter of course.

 

It's possible to grab a Dremel and cut them out, thus improving the intake no end, but then there's little defence against a shower of sparks from a failed PSU falling onto anything potentially flammable beneath the case.

 

I shall most definitely stick to my preferred method, namely fan up sucking air into the PSU from inside the case. Will have to mod the X510 a bit though. The bottom compartment/shelf is attached with rivets, will need to drill them out and remove. It doesn't do much anyway to be honest. It's supposed to be a separate PSU compartment, but it's not that at all, the front is open and the end is open, no thermal isolation whatsoever. Awesome case though!

 

I'd be interested if anyone on the forum has experienced issues with PSU down. I recall someone had high PSU fan RPM and had to switch to fan up.

 

It seems case manufactures aren't experts in fluid dynamics, we can't trust them to get it 100% spot on.

 

My system was built for me at a well known firm here in the North of England, and the fans are as follows:-

 

2 fans on top of case ( water cooled stuff) air coming out and upward.

1 fan on rear of case at top,  (adjacent to above ) air going in

psu fan placed at rear and at bot, warm air coming out.

bot of case has mesh grill under psu and also a mesh grill in center of bot of case ( where a addition fan can be fitted) .

 

bob

[martin-w 3,235]

Sounds like PSU fan down Bob. In  from the bottom through restrictive vents and filter and out the rear. Common these days. Not my preference.

 

 

 

1 fan on rear of case at top, (adjacent to above ) air going in

 

Air will be exhausting out of the rear fan Bob. It's an exhaust. Unless they've done something weird.

[onebob 125]

Sounds like PSU fan down Bob. In  from the bottom through restrictive vents and filter and out the rear. Common these days. Not my preference.

 

 

 

 

 

Air will be exhausting out of the rear fan Bob. It's an exhaust. Unless they've done something weird.

 just check again with a piece of paper, air input ( paper clings outside of rear fan ) ,

 

https://www.dropbox.com/s/76ilrqfay0rai7s/2015-12-17%2018.35.28.jpg?dl=0

 

bob

[martin-w 3,235]

Well that's not the normal direction of airflow. Should be from front to back. Do you have any front fans?

 

Front fans should blow in and rear exhaust.

 

There's usually an arrow on the fan body indicating direction of airflow.

 

It was either installed the wrong way round by mistake, or someone decided to defy conventional wisdom.

[gboz 157]

Well that's not the normal direction of airflow. Should be from front to back. Do you have any front fans?

 

Front fans should blow in and rear exhaust.

 

There's usually an arrow on the fan body indicating direction of airflow.

 

It was either installed the wrong way round by mistake, or someone decided to defy conventional wisdom.

 

Not necessarily so once a top mounted radiator is used to water cool the CPU.

In this case you might give cooling priority to the CPU and have the air flow coming in over the CPU radiator.

After this it's logical to have every other fan acting as exhaust, lowering the overall case pressure and assisting CPU radiator intake.

 

gb.

[fppilot 4,371]

Whew! I thought this might be another prolonged rant against Pennsylvania State University!

[garrett_frank 150]

I'm always in search of a problem but I can't say I've ever entertained one that involved the PSU. And it's because of an answer that I now pose to the rest of you: is anyone having any sort of heat-related issue with their PSU?

[martin-w 3,235]

And it's because of an answer that I now pose to the rest of you: is anyone having any sort of heat-related issue with their PSU?

 

 

 

Yes... the individual resides on this forum. I recall he mounted his PSU fan down and the temperature increased markedly along with the fan RPM. He mounted the PSU fan up, ingesting air from inside the case and all was well. I was hoping he'd respond to this thread. I can't recall his name.

 

Having said that, it's not so much about "having an issue" probably more about averting a less than optimal installation. It's also about hardware enthusiasts debating an aspect of custom PC design.

 

I'm about to build a new PC. My opinion of the fan down installation is in the first post. However, if my logic is flawed I'd gladly change my opinion. I don't think it is.

[martin-w 3,235]

Not necessarily so once a top mounted radiator is used to water cool the CPU.

In this case you might give cooling priority to the CPU and have the air flow coming in over the CPU radiator.

After this it's logical to have every other fan acting as exhaust, lowering the overall case pressure and assisting CPU radiator intake.

 

gb.

 

No, he didn't say that, he didn't say "every other fan acting as an exhaust". His rear fan is configured as an intake.

 

It may well be that his radiator fans are intakes [blowing warm radiator air into the enclosure] and that his rear fan is an intake and perhaps front fans exhaust. Effectively reversing the conventional airflow direction. This has been done before, what I said was that it's not the "normal direction of airflow" and that it's "contrary to conventional wisdom". I didn't say it's never done.

 

 

 

 

[gboz 157]

No, he didn't say that, he didn't say "every other fan acting as an exhaust". His rear fan is configured as an intake.

 

It may well be that his radiator fans are intakes [blowing warm radiator air into the enclosure] and that his rear fan is an intake and perhaps front fans exhaust. Effectively reversing the conventional airflow direction. This has been done before, what I said was that it's not the "normal direction of airflow" and that it's "contrary to conventional wisdom". I didn't say it's never done.

 

Well no, I did not say that he had "every fan acting as reverse" either.

I was just citing an example of where front to back airflow was not warranted.

 

onebob says he has his CPU water cooler fans exhausting.

Presumably the case builder has given priority to CPU cooling and set up the rear fan as an intake to assist the outflow of the CPU radiator fans and provide it directly with cooler outside air.

Looking at his photo of the case internals you can see that he has a very large rear fan sitting very close to his CPU fans. Running the rear fan as exhaust could slow CPU air flow, at least through the closest CPU fan.

 

gb.