VOGONS


Top or bottom mounted PSU?

Topic actions

Reply 20 of 33, by gdjacobs

User metadata
Rank l33t++
Rank
l33t++
dr_st wrote:
cyclone3d wrote:

For every 7C cooler, you about double the life expectancy of the capacitors inside a PSU.

I think it said 10C in one of the articles you linked to.

10 deg C drops is the accepted figure for the lifespan of capacitors, chokes, and transformers to double.

All hail the Great Capacitor Brand Finder

Reply 21 of 33, by cyclone3d

User metadata
Rank l33t++
Rank
l33t++
gdjacobs wrote:
dr_st wrote:
cyclone3d wrote:

For every 7C cooler, you about double the life expectancy of the capacitors inside a PSU.

I think it said 10C in one of the articles you linked to.

10 deg C drops is the accepted figure for the lifespan of capacitors, chokes, and transformers to double.

Ok, I am pretty sure I saw a 7C figure somewhere but not finding it now so I'll just use the standard 10C from now on.

Yamaha YMF modified setupds and drivers
Yamaha XG resource repository - updated November 27, 2018
Yamaha YMF7x4 Guide
AW744L II - YMF744 - AOpen Cobra Sound Card - Install SB-Link Header

Reply 22 of 33, by Scubs

User metadata
Rank Newbie
Rank
Newbie
cyclone3d wrote:
Yes it does. On bottom mounted PSUs, the PSU draws cool air through the PSU from the outside of the case. […]
Show full quote
Scubs wrote:

We all should know the hotter a part runs the shorter the life span. But that has nothing to do with top vs. bottom mounted PSU's

Yes it does. On bottom mounted PSUs, the PSU draws cool air through the PSU from the outside of the case.

On top mounted PSUs, the PSU draws in warm/hot air though the PSU from inside the case.

It has EVERYTHING to do with top vs. bottom mounted.

Your assuming a top mounted psu cant pull air from outside the case as well. Your also assuming the bottom mounted psu has a side mounted fan and not a rear mounted fan.

Reply 23 of 33, by dr_st

User metadata
Rank l33t
Rank
l33t
Scubs wrote:

Your assuming a top mounted psu cant pull air from outside the case as well. Your also assuming the bottom mounted psu has a side mounted fan and not a rear mounted fan.

I find his assumptions quite reasonable.

In most bottom-mount PSU cases the airflow is set up as follows, which is consistent with his comments.:
$

A rear-mounted fan makes little sense, as that's where the cables / connectors go. Even if you found such a PSU, it would still be sucking in cooler air at the bottom of the case than at the top.

And how exactly is the top mounted PSU going to pull air from the outside of the case? Through the exhaust? Or do you suggest an intake fan at the top of the case (which is generally not done, because it would interfere with normal airflow).

https://cloakedthargoid.wordpress.com/ - Random content on hardware, software, games and toys

Reply 24 of 33, by squiggly

User metadata
Rank Member
Rank
Member
gdjacobs wrote:
dr_st wrote:
cyclone3d wrote:

For every 7C cooler, you about double the life expectancy of the capacitors inside a PSU.

I think it said 10C in one of the articles you linked to.

10 deg C drops is the accepted figure for the lifespan of capacitors, chokes, and transformers to double.

Think I would rather extend the lifespan of everything else in the case by using the PSU as another heat exhaust fan for the entire case.

Reply 25 of 33, by Radical Vision

User metadata
Rank Member
Rank
Member

My machines are very cool, with proper coolers, and the PSU is mounted on the top on all of them, the 462, slot 1 Pentium II, Slot 1 Pentium III, 939 Opteron on 2.7GHz , nothing is overheating or so, i have zero problems with that, but seems other people have.......

zj3y5r-6.png
ptdfb9-3.png
s3yhar-2.png
9s6qlr-3.png
rd0y5f.png

Reply 26 of 33, by dr_st

User metadata
Rank l33t
Rank
l33t

^^^
That's exactly what my initial claim is based on. The assumption that a bottom-mounted PSU will run cooler in most circumstances is very plausible. Whether it has practical (not merely theoretical) effects on the longevity of anything is not so firmly established, in my view (and in light of some of the data cyclone3d linked to).

If the life expectancy of a capacitor running at 55C ambient is 13 years (*), and every 10 degree drop doubles it, then for 45C ambient (which is probably closer to what you would get, even in a hot case with a top-mounted PSU), then it should be ~26 years? Way beyond what one would deem necessary. Almost every other component in a system has significantly shorter life spans.

(*) It may be that I misunderstood the bottom line of the article where this number was mentioned, as I only skimmed it; If I did, please enlighten me.

https://cloakedthargoid.wordpress.com/ - Random content on hardware, software, games and toys

Reply 27 of 33, by buckeye

User metadata
Rank Oldbie
Rank
Oldbie
dr_st wrote:
I find his assumptions quite reasonable. […]
Show full quote
Scubs wrote:

Your assuming a top mounted psu cant pull air from outside the case as well. Your also assuming the bottom mounted psu has a side mounted fan and not a rear mounted fan.

I find his assumptions quite reasonable.

In most bottom-mount PSU cases the airflow is set up as follows, which is consistent with his comments.:
$

A rear-mounted fan makes little sense, as that's where the cables / connectors go. Even if you found such a PSU, it would still be sucking in cooler air at the bottom of the case than at the top.

And how exactly is the top mounted PSU going to pull air from the outside of the case? Through the exhaust? Or do you suggest an intake fan at the top of the case (which is generally not done, because it would interfere with normal airflow).

Ok, got a confession to make. On my XP rig my bottom mounted PSU has the fan TOPSIDE, so it's sucking hot air from the parts above it? Should I flip it?

Foxconn Q45M E8400 Core2 Duo 3GB DDR2 800 BFG Geforce 7950GT 512MB X-Fi Xtreme Music 500W XP SP3
Intel SE440BX P3 450 256MB 40GB Voodoo 3000 16MB SB 32pnp 350W 98SE
MSI x570 Gaming Pro Carbon Ryzen 3700x 16GB DDR4 Zotac RTX 3070 8GB WD Black 1TB 650W

Reply 28 of 33, by cyclone3d

User metadata
Rank l33t++
Rank
l33t++

It will make the PSU run cooler.

Yamaha YMF modified setupds and drivers
Yamaha XG resource repository - updated November 27, 2018
Yamaha YMF7x4 Guide
AW744L II - YMF744 - AOpen Cobra Sound Card - Install SB-Link Header

Reply 29 of 33, by vvbee

User metadata
Rank Oldbie
Rank
Oldbie

I never buy bad quality power supplies that need their airflow micromanaged for heat and I also don't aim to keep them around for decades like I do some other components. That said I mount my psus whichever way I deem beneficial for the particular system. Certainly in designing airflow that doesn't generate harmful pockets of warm air you need to consider more than the placement of the psu, ideally having first obtained 3d heat maps of the case in various configurations.

Reply 30 of 33, by appiah4

User metadata
Rank l33t++
Rank
l33t++

Just wanted to chime in to share this PC I have.. Apparently bottom mounted PSUs were a thing all the way back in 1994:

AIDATAP133_01.jpg AIDATAP133_02.jpg

Retronautics: A digital gallery of my retro computers, hardware and projects.

Reply 32 of 33, by shamino

User metadata
Rank Oldbie
Rank
Oldbie

I have an old HP P2 Xeon machine with a bottom mounted PSU. I like it aesthetically and for the physical stability it adds, but that system was designed for it. The PSU connector is on the bottom of the board and each set of wires from the PSU is the appopriate length for where it needs to go. Unlike modern implementations, it pulls air from inside the case, which I think is more practical. It's 20 years old and still works perfectly, but it wasn't cheaply made and wasn't marketed for quietness.

If a PSU fails from heat within a reasonable lifetime, I blame it on any of these things:
1) The use of junk capacitors that will never be reliable at any temperature. I'm not going to try very hard to extend the life of a bunch of Fuhjyyus. The only real solution is to replace the caps or get a PSU that has good ones to start with (admittedly that can be hard to find).
2) The use of a cheap fan that fails. Basically the same complaint as #1 - cheap parts.
3) The use of a variable speed fan. In principle that's great, but in practice these setups seem to be skewed heavily in favor of quietness. They want NewEgg reviews to rave about how quiet the PSU is. Nobody's going to post back about long term issues from a chronically high temperature.

I don't like the idea of pulling air from under the case. This is asking for tons of dust and pet hair to get sucked in. If you keep your computer on your desk then this issue is greatly lessened. Even so, pulling air from the surface directly underneath the case is always going to be the most dust prone way to do it.
From what I've seen of modern cases, I think they tend to have filters down there nowadays. That's good, but it doesn't totally satisfy me with respect to the issue, and it's not a convenient place to need to be messing with filters.
On a conventional motherboard with top mounted power connectors, I prefer a top mounted PSU. If it's going to be in the bottom, I would prefer to pull the air from inside the case. I would also have a dust filtered intake fan in the front of the case.
But I also like external drive bays, a sealed top, and properly colored wires. Based on some recent window shopping, I think my preferences for cases are obsolete on all but the cheapest models nowadays. 😀

Reply 33 of 33, by Errius

User metadata
Rank l33t
Rank
l33t

The only time I've run a bottom-mounted PSU was as an auxiliary power supply in a system with a weak primary PSU. This was only temporary until I got a more powerful primary PSU. (Power consumption was unacceptably high with two PSUs.) The case was the original Cooler Master Stacker which has both top and bottom PSU slots.

“I like to dissect PCs. Don't you know I'm utterly insane?"