Is it safe to replace the fans in PSU?

PSUs: The source of DC power for all components in the PC & often a big noise source.

Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee, Devonavar

Post Reply
mountainpenguin
Posts: 7
Joined: Thu Apr 24, 2003 5:07 am

Is it safe to replace the fans in PSU?

Post by mountainpenguin » Thu May 15, 2003 4:10 am

I have two PSU that I like to replace the fans with Panaflo. They are Enermax EG465P-VE FC and PC Power Cooling Turbo-Cool 425 ATX. This article has detailed mod instructions on 365W model of the power supply but I was wondering if the same procedure can be done safely on 465W model. How do I determine if fan replacement can be done safely in general?

halcyon
Patron of SPCR
Posts: 1115
Joined: Wed Mar 26, 2003 3:52 am
Location: EU

Post by halcyon » Thu May 15, 2003 4:24 am

I have removed the 80mm fan completely on my 430W Enermax 465AX-VE(W) and it's still running, although it does generate more heat.

I removed the 80mm fan, put the fan grill back in place and blocked the vents in the PSU case that were directed back inside my computer case.

It hasn't broken yet, but I can give you no guarantees.

The supplies may shut down, the voltages may start to ripple and the maximum output power capability may go down, but I haven't heard of other problems myself.

regards,
Halcyon

mountainpenguin
Posts: 7
Joined: Thu Apr 24, 2003 5:07 am

Post by mountainpenguin » Thu May 15, 2003 4:39 am

That sounds pretty extreme to me. I certainly don't want to risk a meltdown or fire hazard. I'm thinking of just replacing the rear 80mm fan with Panaflo. I may just leave the internal 92mm fan alone since it's inside the case and it doesn't make that much noise. Oh well, I guess I'll just have to keep an eye on the thermometer after the mod.

halcyon
Patron of SPCR
Posts: 1115
Joined: Wed Mar 26, 2003 3:52 am
Location: EU

Post by halcyon » Thu May 15, 2003 4:53 am

mountainpenguin,

well even my model still has the 92mm fan active and it pushes out air at a faster rate than some of my old Ultra Quiet 300W power supplies.

I am a bit worried about the temps myself too, if not for else then for the increase in case temp and the problems that brings about.

I'm now trying to find even more efficient power supplies to be added to my evergrowing collection of PSUs :)

regards,
Halcyon

MikeC
Site Admin
Posts: 12285
Joined: Sun Aug 11, 2002 3:26 pm
Location: Vancouver, BC, Canada
Contact:

Re: Is it safe to replace the fans in PSU?

Post by MikeC » Thu May 15, 2003 8:45 am

mountainpenguin wrote:I have two PSU that I like to replace the fans with Panaflo. They are Enermax EG465P-VE FC and PC Power Cooling Turbo-Cool 425 ATX. This article has detailed mod instructions on 365W model of the power supply but I was wondering if the same procedure can be done safely on 465W model. How do I determine if fan replacement can be done safely in general?
In general you can't. The key word is safely. No way you can say it's safe or not, what's safe for one is not necessarily safe for another just because of willingness to risk, and because what you can do safely maybe I can't. So in general, assume it is not safe until proven so. In other words, any time you mod a PSU, unless you really know what you are doing, be aware that you are playing with something that is tied to your AC electrical system & all the dangers associated with that.

Here's some rules of thumb:

1) any mod that results in significantly lower airflow will likely shorten PSU life. If the PSU uses any marginal quality parts, this could mean very accelerated failure. It only takes one part to take out several, which usually makes a PSU worthy only of the dump. Plus there is the risk of the PSU taking out other components as it dies. "significantly lower": let's define this arbitrarily as 10~20% to be conservative.

2) it's best to combine reduced airflow with reduced impedance to that airflow. In other words, at least open up the in/outflow vents so the lower alirflow gets through & do the most cooling it can do.

3) any method to introduce cooler outside-of-case air into the PSU intake will help keep the PSU cooler, and avoid fans from speeding up due to thermistors getting hotter. But this has to be balanced against the question of how much noise is allowed to escape via this fresh air intake.

Usually experimentation is the only sure way of confirming all the above, and what works in one system may not work the same way in another similar system because of thermal / airflow differences, and differences in the working environment -- ambient temp, location, even user patterns. (If you never play games or do anything to push the video card, you may consistently have 25W less heat than a gamer with the same system.)

marc999
*Lifetime Patron*
Posts: 230
Joined: Sat Oct 26, 2002 4:05 pm
Location: Calgary, Alberta, Canada

Post by marc999 » Thu May 15, 2003 9:12 am

Mike,
I was wondering if I could get your input on my planned Q-tech PSU mod.
(I'm still not sure if I'm going to do this, I have to determine how loud the PSU is, so far it seems okay). In case I do go ahead with the mod, I herein release you of all liability. I can't afford to pay a lawyer anyways !! :D

1. Cover all side vents with electrical tape, so that a "tunnel" effect of airflow occurs from front to back of the PSU.
2. Mount a 80mm M1A @ 5V on the back of the PSU.

Whatcha think ?!

mountainpenguin
Posts: 7
Joined: Thu Apr 24, 2003 5:07 am

Re: Is it safe to replace the fans in PSU?

Post by mountainpenguin » Thu May 15, 2003 10:16 am

MikeC wrote: In general you can't. The key word is safely. No way you can say it's safe or not, what's safe for one is not necessarily safe for another just because of willingness to risk, and because what you can do safely maybe I can't. So in general, assume it is not safe until proven so. In other words, any time you mod a PSU, unless you really know what you are doing, be aware that you are playing with something that is tied to your AC electrical system & all the dangers associated with that.
Sounds like a heap of common sense is what's needed in doing a PSU mod. I've been told I don't have much of that "common sense" but I guess I'll scrounge up as much as I can. :)
MikeC wrote: Here's some rules of thumb:

1) any mod that results in significantly lower airflow will likely shorten PSU life. If the PSU uses any marginal quality parts, this could mean very accelerated failure. It only takes one part to take out several, which usually makes a PSU worthy only of the dump. Plus there is the risk of the PSU taking out other components as it dies. "significantly lower": let's define this arbitrarily as 10~20% to be conservative.
Hmm.. Dead PSU due to mod is one thing but dead computer due to mod is a whole different scale of problem. I think I'll be start with more conservative mods especially when it comes to PSU.
MikeC wrote: 2) it's best to combine reduced airflow with reduced impedance to that airflow. In other words, at least open up the in/outflow vents so the lower alirflow gets through & do the most cooling it can do.
So if I were to remove the outflow 80mm fan at the rear of the PSU and add an inflow 80mm fan on the other side of PSU, it would actually be better?
MikeC wrote: 3) any method to introduce cooler outside-of-case air into the PSU intake will help keep the PSU cooler, and avoid fans from speeding up due to thermistors getting hotter. But this has to be balanced against the question of how much noise is allowed to escape via this fresh air intake.
I've tried something slightly unconventional with air flow and it seems to have some mixed successes. I've made the two outflow fans on the back of the case to blow air inward to the case. This made the CPU run cooler since fresh air was blowing directly onto the CPU heatsink area. However I think it has a overall negative impact on the air flow for rest of the case. I also thought about doing something else. Perhaps creating a inflow and outflow air path to PSU completely separate from the rest of the system might be better. Currently warmed air already filtered through the hard drive and CPU is being once again filtered through the PSU thus reducing the cooling effectiveness on the PSU. By having a some sort of duct to bring in outside air just for PSU can provide better cooling for PSU. Essentially it's the parallel air flow system rather than serial one. Of course I should really try this out and see if makes any difference.
MikeC wrote: Usually experimentation is the only sure way of confirming all the above, and what works in one system may not work the same way in another similar system because of thermal / airflow differences, and differences in the working environment -- ambient temp, location, even user patterns. (If you never play games or do anything to push the video card, you may consistently have 25W less heat than a gamer with the same system.)
Speaking of experimentation, I should really get a decent multimeter with a thermal sensor to systemically measure the results. Any particular one you recommend? Is there also a device to measure the noise decibel levels that doesn't cost an arm and a leg?

energy
Posts: 192
Joined: Mon Apr 21, 2003 4:44 pm
Location: London, UK
Contact:

Post by energy » Sat May 17, 2003 5:32 pm

By having a some sort of duct to bring in outside air just for PSU can provide better cooling for PSU
I have my Zalman 300W PSU mounted on top of my case (actually not even mounted, just placed on a bit of foam) with a 15dB(a) NMB fan mounted externally... on the noise killer it gets around 7v which means it remains very very very quiet under constant load (XP1800, two 80gb 7200 drives, 512mb RAM, GeForce 4 Ti 4200).

Just waiting for someone to design a case with the PSU mounted in a way where it sticks out of the top and suck in cold air. Thats the key to silent PSUs... cool air, not the preheated CPU air.

Post Reply