Dampening vibration on Mugen 2 fan?

[mot524]

Overall I'm extremely happy with the quietness of my computer.

But I have noticed that at certain ambient temperatures, and near-idle loads, my computer will make a humming noise for ~20 seconds, then be quiet for ~20 seconds, then hum again, etc. (Not unlike water torture. You can see how this might drive a person insane.)

I have worked out two causes for this:

1) I have a Gigabyte motherboard with s****y fan control. It seems to spin the fan up too much to cool off the CPU, then spin it down too much and allow the CPU to warm back up, ad nauseum, causing the cyclical humming noise. There are no settings for the fan control--it's either on or off. My old, cheap MSI mobo was a million times better--it let you specify target temperature ranges and corresponding fan speeds. Unfortunately I'm forced to use the Gigabyte for reasons I won't go into.

2) I have a Scythe Mugen 2 CPU cooler bolted to the motherboard, which is bolted to my case, and then fan is held on to the front of the heatsink by the pair of crappy thin wires that came with the Mugen. This setup causes a tremendous amount of low frequency case vibration at certain fan speeds.

So, I was wondering if someone has come up with a good way of mounting a fan to a Mugen 2 that dampens vibration a bit? Thanks in advance!

For the time being I have plugged the CPU fan into a low-voltage case fan header so it runs at a constant 800 RPM (instead of 1400). Unfortunately this speed causes a lot of vibration, but it's constant, and placing a large pile of heavy books on the top of my case seems to keep things under control. (Classy, right?)

[KadazanPL]

Slipstreams (the fan that comes with Mugen2) are very well behaved. It shouldn't produce any significant vibration even at higher speeds. The fan that came with my Mugen works very smoothly; even at max 1400RPM all I can hear is the air rushing past fins, no vibrations.

There are two probable reasons for the phenomenon you described:
1. your fan is damaged or a bad sample - off balance and therefore starts vibrating,
2. the side panels on your case aren't secured or are very thin/light.

I would first try to hold the case's sides with my hands and see if that makes the hum go away. If it does - think about mass loading the panels (rubber tiles, bitumen, polymer sheets, etc.)

If it doesn't help - change the fan

[mot524]

Thanks for the reply and I understand what you're saying. I screwed down my side panel but unfortunately it didn't seem to help. Surely fan vibration must be an issue even with nice fans though, otherwise why would Nexus ship its fans with those silicone mounts?

I wonder if an Arctic Cooling fan would be a good solution to the problem. Not the best acoustically but they have those rubber mounts built into their frames...

[jhhoffma]

I've used very small pieces of double-sided sticky foam to insert between a fan and it's heatsink where vibrations are a problems. Usually, they are thin enough where they don't cause a problem with the wire clips.

Bigger issue is what you are running underneath where you need a fan on the Mugen to begin with? I've got a 125w Phenom II x4 under mine in a Antec 300 and I run it semi-passively, CPU never gets above 50C.

[mot524]

Thanks for the foam tape idea. Unfortunately I haven't gotten the hang of mounting fans on the Mugen and it has been kind of a pain both times I've done it, so I'm loathe to switch out the fan again. (I replaced the Scythe fan for a 3-pin Nexus fan so now I can control the speed with a Fan Mate instead of using my motherboard's crappy PWM control.) I have resigned myself to keeping some heavy books on the back of my case to reduce the vibration, which works well enough.

Interesting re: passively cooling the Phenom. I have a Q6600 (105W TDP) so in theory I should be able to do the same thing I guess. Right now the only fans I have are a 12cm Nexus exhaust fan and a 12cm Nexus CPU fan, both running at a constant 750 RPM, in an Antec Sonata III. My core temps are 36C/60C. I am surprised your CPU stays below 50C.

The early Core 2 Quads (like the Q6600) had a problem where the core temps reported 20C cooler than their actual temperatures. I think some BIOSs and temp monitoring software take this into account and some don't. I'm not sure if 60C is my actual load temp or if it's 80C. I am hesitant to let my load temp go above 60 (80?) though, which I imagine it would do if I was running passively. Although I guess having a fan on the heatsink doesn't necessarily mean it's doing a lot of cooling.

[NeilBlanchard]

Is the hard drive possibly the source of the vibration? Do you have just one, or more than one hard drive?

[mot524]

No. My hard drive did cause a lot of vibration but I put a square of foam under it and now it's basically impossible to tell if it's on or off when the case is closed. The packing materials from an old motherboard included a sheet of foam that worked perfectly for this. HD temps are only 5-10C hotter.

[NeilBlanchard]

Well, noise is higher frequency vibrations, but hard drives also vibrate at lower frequencies.

Can you see or feel the Mugen fan vibrating? Try unplugging the fan (briefly) and see if the case vibrates? If it is the fan, it would have to shake the heatsink, and shake the motherboard, in order to get through to the case -- and these are on different axis, so it would have to wobble the heatsink side to side (radially to the fan rotation) and this would have to flex the motherboard like a diaphragm, and then the motherboard tray would have to also flex.

It could be cavity resonance, which involves the exhaust and intake fans, but not the heatsink fan.

Try turning the hard drive up on it's long edge? Is there only one hard drive, or two, or more? The cycling you mentioned could be the beat frequency of two hard drive rotating at *almost* the same RPM.

[mot524]

Well, now I have to apologize for the message board spam because I unplugged the CPU fan (easy to do with the Fan Mate) and the vibration is still there... it must be from the hard drive... I thought the vibration went away when I unplugged the hard drive but I must be mistaken. Look like it's time for a 2.5" hard drive.

Switching the CPU fan to constant RPM definitely did eliminate the intermittent "water torture" resonance which might be the cavity resonance you're talking about.

[mot524]

Oh, also, my CPU temps with no CPU fan (only exhaust fan) were 45/65+. The idle temperature was stable but under load the temperature was increasing pretty steadily so I stopped it after a few minutes at 65.