What are the consequences of using a power supply with dying fan?

[kony]

I just wanted to make sure it is OK to continue using the computer
while I decide on a replacement PSU and wait for it to arrive. Since
my CPU has a fan on it, and I also have 4 case fans, I wasn't sure if
the PSU fan was that important or not.

Generally speaking, it is not safe to do it. A very big
gamble that is unnecessary since the typical PSU fan failure
only needs a drop of oil in the fan bearing to revive it for
long enough to get you by till the replacement is available.
However, if the fan is what I expect from the brand of PSU,
it is a dual ball bearing Nidec fan which, since it's not
sleeve bearing, won't respond to the drop of oil... but it
makes me wonder what it's doing as ball bearing fans tend to
get whiney then sort of a grating sound right before they
lockup, not contining to run very long once they're really
loud.

Also, the motherboard
monitoring software reports the CPU temperature as well as voltages,
so that's why I asked those questions.

The fan seems to be spinning OK though, even though it is making loud
noises (sounds like it is struggling to spin).

So it's definitely that fan? If in doubt you might stick a
plastic straw in while the system is off, turn it on and see
if the noise persists while straw blocks the fan.

I'm just wondering because if it's spinning ok, not
drastically low RPM, those dual ball bearing fans tend to
get really whiney before they fail, a quite different sound
than most PSU fans make from running dry and chewing up
their bearing. I wouldn't call the whiney noise loud, just
annoying due to the pitch of it.


Disappointing, as the

PSU is a Fortron Source (FSP400-60PFN), and I had read a lot of
positive comments on that brand. And it is only one year old (and
actually had started to make noises several months ago, but they were
off and on, and now the noises are pretty much constant).

That's surprising, those PSU usually use dual ball bearing
Nidec fans which are about as good as it gets for a PC PSU.
Maybe you're just unlucky enough to have the one in a
million that's bad, I routinely replace crap fans in PSU
WITH the Nidecs and haven't had any fail.

Is it possible the system is just running so cool that the
PSU is trying to throttle the fan down more than (the
designers) could have anticipated? Some fans just don't
throttle down as well as others, unfortunately it's usually
the better ones that do worse at this as they have higher
torque motors and better magnets.

Anyway, it's just a standard fan. So long as the present
fan doesn't die and let the PSU bake, you could save quite
bit of money by just putting a replacement fan in it.
Should be 80x25mm, IIRC the current was probably around
0.20A on those Fortron used, but with a thermal control.
You might try something like this but I don't really if they
used a 0.1" spaced bi-pin connector on that fan or soldered
it to the circuit board, in which case you'd need to splice
the fan to the existing leads or pull whole circuit board
out to fix it. Anyway, this is about the lowest cost fix if
you're in the US as SVC has really cheap US postal rates for
a single small item.
http://www.svc.com/fba08a12m.html

 

[kony]

On Tue, 13 Mar 2007 10:56:50 -0700, "JAD" <john

you have blown it out with compressed air?
the fan blades are clean?
fortron is a good brand I'm a little surprised also. 1 year ...goes to show you...still a
crap shot when it comes to PSU's

Fortron is pretty good but the Nidec fans they often used
were even better, what you'd typically find in server PSU
besides Sanyos. Some Fortrons used NMB fans, also quite
good, and Yate Loons, quite bad fans. It's wierd they chose
opposite extremes like that, though the Yate Loons were
usually in their budget OEM PSUs like those bundled with
cases, though also in the 120mm sizes on recent PSU but I
thought they started using something else because of the
Yate Loons' lifespan problems.

 

[Guest]

Generally speaking, it is not safe to do it. A very big
gamble that is unnecessary since the typical PSU fan failure
only needs a drop of oil in the fan bearing to revive it for
long enough to get you by till the replacement is available.
However, if the fan is what I expect from the brand of PSU,
it is a dual ball bearing Nidec fan which, since it's not
sleeve bearing, won't respond to the drop of oil... but it
makes me wonder what it's doing as ball bearing fans tend to
get whiney then sort of a grating sound right before they
lockup, not contining to run very long once they're really
loud.

You are right, it is a Nidec fan.

So it's definitely that fan?

I suspect it is, because my computer didn't start making these noises
until after I installed that PSU.

But I can turn off some of the case fans to make sure one of them
isn't doing it.

My motherboard monitoring software also reports the CPU fan speed, and
it has been spinning at its normal RPM whenever the noise is being
made.

If in doubt you might stick a
plastic straw in while the system is off, turn it on and see
if the noise persists while straw blocks the fan.

What about sticking something in there while the system in on?
Because now when I turn on my computer it doesn't make any noise. (I
typically leave my computer on all the time, but the last few days
I've turned it off when I'm done using it.) But it might start to
make noise after it's on for a while.

I'm just wondering because if it's spinning ok, not
drastically low RPM, those dual ball bearing fans tend to
get really whiney before they fail, a quite different sound
than most PSU fans make from running dry and chewing up
their bearing. I wouldn't call the whiney noise loud, just
annoying due to the pitch of it.

Is it the case that once they start making a noise, they will continue
to make a noise until they die? Because my noise seems to be off and
on.

That's surprising, those PSU usually use dual ball bearing
Nidec fans which are about as good as it gets for a PC PSU.
Maybe you're just unlucky enough to have the one in a
million that's bad, I routinely replace crap fans in PSU
WITH the Nidecs and haven't had any fail.

Is it possible the system is just running so cool that the
PSU is trying to throttle the fan down more than (the
designers) could have anticipated? Some fans just don't
throttle down as well as others, unfortunately it's usually
the better ones that do worse at this as they have higher
torque motors and better magnets.

Hmm, I'll see if it makes any noises the next few days (it's supposed
to get colder).

 

[kony]

I suspect it is, because my computer didn't start making these noises
until after I installed that PSU.

Well... that doesn't necessarily tell US anything though it
might tell you. We would have to wonder if there is some
possiblity a wire is hanging down into a fan blade path or
something like this, particularly when installing a PSU the
wiring is quite subject to routing and securement issues.

But I can turn off some of the case fans to make sure one of them
isn't doing it.

I'm mainly asking because that fan was likely the highest
quality one in the system, playing odds it is higher quality
than just about any other fan sold in "PC" parts... but that
doesn't make 100% of them last forever of course, somebody
is bound to get one with a defect.

My motherboard monitoring software also reports the CPU fan speed, and
it has been spinning at its normal RPM whenever the noise is being
made.



What about sticking something in there while the system in on?

That is best avoided, as it can gouge the blade or jar the
bearing. You "might" be able to do it without harm, but the
default answer is: Blocking a fan from spinning in the
first place is a lot safer for it than trying to stop it,
particularly when you only have access to the back so
pressure can't be applied to the middle of the hub.

Because now when I turn on my computer it doesn't make any noise. (I
typically leave my computer on all the time, but the last few days
I've turned it off when I'm done using it.) But it might start to
make noise after it's on for a while.

If you haven't checked your fans recently... it's time,
because a failing fan will often just refuse to start
spinning after a cooldown.

Is it the case that once they start making a noise, they will continue
to make a noise until they die? Because my noise seems to be off and
on.

Generally with a ball bearing fan, it begins to get whiney
and then runs for a few more months with this mild (yet
annoying because of the high pitch) noise. Seldom will one
stop making noise once it starts, but sleeve bearing fans
can be the opposite, that they may make more noise when cold
but better when warmed up (till they get really badly worn).
That doesn't make it impossible to stop by any stretch, if
your Nidec is failing then you do have an unusual case
unless it's what I described previously, that the PSU is
trottling it down too far, but I can't hear what the noise
sounds like, how loud it is. You could even have a PSU
defect or failure besides the fan, though it seems far less
likely.

 

[Guest]

Tonight it has been making a faint noise or no noise at all.

But I just noticed something strange. If I push my computer so that
it is leaning back a bit (so that the front is raised off the floor
and only the back is touching the floor), then the noise becomes
louder. But then when I let go and the computer returns to its normal
position, the noise either gets quieter or disappears.

What do you think that means?

 

[kony]

Tonight it has been making a faint noise or no noise at all.

But I just noticed something strange. If I push my computer so that
it is leaning back a bit (so that the front is raised off the floor
and only the back is touching the floor), then the noise becomes
louder. But then when I let go and the computer returns to its normal
position, the noise either gets quieter or disappears.

What do you think that means?

That you need to open the system up and get a positive
confirmation of exactly where this noise is coming from.