Removing PSU fan - very bad idea, hugely bad idea or just a bad idea?

Associate
Joined
3 Feb 2009
Posts
2,236
I just bought a Coolermaster 330 to replace my parent's Hewlett Packard "purple place" box - the innards are getting a bit toasty so I figured best to try to prolong the life of them.

Unfortunately, the power supply in the old box gets jammed on the io shield area of the new one, the culprit being the fan on the PSU.

Considering it's only a 300W PSU which was entrusted previously with drawing all of the heat out of the system, I can't imagine it actually produces THAT much heat itself. If I removed the fan - aside from any other safety aspects such as being able to touch the inside of the PSU - would there be any real danger of serious harm from the heat produced?

At the moment they simply can't afford £35 for a new PSU, but I'm getting more and more concerned about the rest of the components. There's nowhere to install a system fan and the CPU and PSU fans both don't run as fast as they used to.

If there's any risk there's no way I'll do it - but I'd like to get the kit into the new case.
 
Associate
Joined
26 Jan 2009
Posts
1,462
Location
Salisbury, Wilts
really bad idea. PSU's are still charged with current long after they have been switched off. Either pay an expert to do it or buy a new PSU. Would highly not reccommend replacing the fan unless you know what you are doing.
 
Soldato
Joined
30 Sep 2003
Posts
10,916
Location
London
Not good news. A standard PSU isn't designed to run without a fan. PSUs that are designed to be fanless tend to be quite expensive, because they're very efficient, have massive heatsinks and tend to use the PSU case for extra cooling.

If the PSU is a crappy generic one, that makes the situation worse. It's probably not very efficient, which means that it generates more heat, and the internal heatsinks will probably be much smaller than on a quality PSU. A bit of airflow goes a long way, and I wouldn't be surprised if the supply dies before long with the fan removed. That's a risk in itself, since cheap PSUs have a nasty habit of taking your system with them when they expire.
 
Last edited:
Associate
Joined
12 Mar 2009
Posts
490
As has been said, seriously do not open a PSU to do maintenance on it of any kind, the capacitors hold a serious amount of power in them even when having been left sat on a desk doing nothing for weeks; one slip and a nasty electric shock and possibly even a heart attack aren't worth the cost of getting a new PSU.
 
Associate
OP
Joined
3 Feb 2009
Posts
2,236
I'd be very careful to neutralise the capacitor first :) finally a use for my A-level physics.

But all things considered, I'll just wait until they can afford a new PSU. I'm determined to upgrade and replace the PC - a new case for cooling and a new PSU now could save them a lot of money in the near future - but I can't afford to fund it myself.

The heatinks are relatively small, my hope was basically that because it's so low wattage it wouldn't be a big problem with heat, necessarily. Ah well - they can stick with the old case for a few months. It's just a shame there's nowhere to rig a couple of fans on it :)

Thanks for the advice.

Oh, and 5abr3 - did you do anythign unusual to get your 720be stable on stock volts, or was it just a multiplier bump?
 
Soldato
Joined
3 Nov 2004
Posts
9,855
Location
UK
Even a good psu 15% of so of the power turns into heat. With a reasonable load your psu is going to be radiating 20-30W of heat or more. Without some airflow over heatsinks, that heat will just build up. Unless its designed to run passive then it won't, at least not for long.
 
Soldato
Joined
18 Apr 2007
Posts
3,736
Location
Probably in a river somewhere
I'm not sure I understand how the fan is getting in the way, do you mean it doesn't fit against the hole for the PSU in the back plate of the case? Or is it sticking out and hitting the motherboard below it?

It is a standard ATX sized power supply isnt it?

PK!
 
Soldato
Joined
31 May 2006
Posts
7,564
Location
West London
A good clean with a can of compressed air can work wonders - or the cheaper option is just a good lung full of air.
Might be an matx psu from the HP - picture would help or the HP model number.

As other have said do not run a cheap psu without a fan (actually don't run an expensive psu without either)
I've a passive psu - it uses the case (3mm aluminum not .3mm steel) aswell as internal fins as heatsinks.
it's more than 10x heavier that the half decent 380w eathwatts at work.
 
Last edited:
Associate
Joined
12 Mar 2009
Posts
490
I'd be very careful to neutralise the capacitor first :) finally a use for my A-level physics.

But all things considered, I'll just wait until they can afford a new PSU. I'm determined to upgrade and replace the PC - a new case for cooling and a new PSU now could save them a lot of money in the near future - but I can't afford to fund it myself.

The heatinks are relatively small, my hope was basically that because it's so low wattage it wouldn't be a big problem with heat, necessarily. Ah well - they can stick with the old case for a few months. It's just a shame there's nowhere to rig a couple of fans on it :)

Thanks for the advice.

Oh, and 5abr3 - did you do anythign unusual to get your 720be stable on stock volts, or was it just a multiplier bump?

There are a lot of issues with discharging power supples in a safe manner (I have degree level electronics modules concerned with one and 3 phase generation and consumption designing. I wouldn't open it at all without dedicated equipment for isolating and discharging (certainly don't ever do it on your own!).

Small heatsinks are being used as the airflow increases their ability to transfer heat for a given amount metal and is much cheaper than pure metal heatinks to just bang a fan on them ;)

Didn't do anything special and I only had the original fans from my Antec 300 at the time and I have more airflow and a cooler card in there now. I upped the multiplier and the FSB; the Gigabyte auto-clocking tool can push it further with voltage tweaks, I just dont fancy over-volting it. Think I got the FSB to about 223MHz and was on the edge of stability at that without more volts.
 

Aod

Aod

Soldato
Joined
7 Oct 2004
Posts
8,663
Location
London
I'd be very careful to neutralise the capacitor first :) finally a use for my A-level physics.

how do you think you'd go about this? shorting something without a high-enough resistance will kill the capacitor (and the PSU along with it).

you'd need a very chunky resistor in the order of a mega-ohm or so.
 
Associate
Joined
12 Mar 2004
Posts
294
Location
UK
I pick b) Hugely bad idea

I'm not sure how the PSU fan is causing the issue you mention, but because you may a) electrocute yourself and/or b) overheat your PSU I'd just go and get a new one.
 
Associate
Joined
18 Mar 2009
Posts
514
last year i did something so stupid im suprised im still alive. i dropped a screw into the psu so i opened the whole thing up, got it out and screwed it back together, seemed like an ok idea at the time but could have ended nastily. esspecialy as i carried on useing it
 
Associate
Joined
2 Jan 2009
Posts
416
I was thinking about opening up a psu to replace its fan, I dont think i will do it now.......... not worth it. I did it once before and got away with it.
 
Top Bottom