I posted the following message on 25 Feb 92:
> My newly-installed 670MP produces quite a lot of fan noise. At 3 feet
> from the front of it, I get about 63 dB(A) in a normally 45 dB(A) room,
> and what's worse, some of this noise is a high-pitched whine which
> sounds like the grid under the fans resonating. Sort of like the rear
> seats in a DC-9, but not quite as loud. The Sun serviceman said it
> sounded loud to him, too, but we couldn't make it go away. Neither one
> of us had heard a 670MP before, so we didn't know if it was normal or
> not.
> Anybody out there know if this is normal or fixable?
> E-mail, don't call, I might be deaf by then. 8-)
The answer for me turned out to be replacement of the fan tray by my
authorized Sun field engineer. As a result, I get about 58 dB(A) SPL
compared to 63 dB(A) before. It's still a noisy machine, but the sound
of the grid vibrating is gone.
If you have a 670MP with a metallic whine coming from the fans, and you
have a service contract or warranty service, I recommend that you get
your service people to replace the fan tray. After the fix, all I hear
is good old fan noise. Also, I can hear the hard disk seeking; I
couldn't before without listening very closely.
BTW, that fan tray is awfully hard to remove. On my machine, it seemed
like there must have been some big bolts somewhere we hadn't undone.
The manual says that it takes a good pull, but we were about ready to
tie it to my truck bumper. Worst thing about it is that the tray is on
the bottom of the enclosure, right next to the floor, and you have to
pull it out horizontally. Of course, the enclosure is on casters. The
serviceman was able to get the tray out by rocking the it left and
right very slightly (it won't go far that way) using a big screwdriver
inserted behind the tray's handle. Sun service people should be used
to heavy labor by now, of course, with those big VME cards with
springfingers.
I had a number of suggestions, including:
1. Get a new tray from Sun (what we did)
2. Use some sort of acoustic baffle (this was said not to work well)
3. Tightening the screws in the fan tray (they're tight enough, okay?)
4. Modifying the fan tray by either disconnecting some fans or adding
series resistance to lower the rpm's (not recommended unless you know
what you're getting into)
I can't help thinking that some RTV silicone in the right places might
have helped. We didn't try it, and I wouldn't unless I was desperate.
Thanks go to these wonderful people for their suggestions:
(in no particular order)
sblair@upurbmw.dell.com (Steve Blair "UNIX Network Services")
ksp@maxwell.nde.swri.edu (Keith S. Pickens)
vss@mathcs.emory.edu (V.S.Sunderam) | [ Gee! Two guys
edgar@mathcs.emory.edu (Edgar Leon) | from Emory.--jpc ]
beric@sunvcr.canada.sun.com (Beric Maass [Sun Vancouver FSE])
Full responses available upon request.
-- J. Porter Clark jpc@avdms8.msfc.nasa.gov or jpc@eb23.msfc.nasa.gov NASA/MSFC Communications Systems Branch
This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:06:37 CDT