On 2009-06-26, Andrew Smallshaw <> wrote:
> On 2009-06-24, DoN. Nichols <> wrote:
>> On 2009-06-23, Andrew Smallshaw <> wrote:
>>>
>>> The PSU fans on an SS5 I have here need replacing - they are making
>>> a very annoying growling noise so I assume they are simply worn
>>> out. Does anyone know what fans to replace them with? What
>>> diameter/thickness/voltage? I would guess that they are 12V 60mm
>>> fans but need something more concrete to actually order parts.
>>
>> Bearings of that size would either whine at a very height pitch
>> (ball bearings), or simply freeze up (sintered bronze bearings), while
>> the noise which you describe is the blades dragging on something.
>
> You're right. I opened up that machine today and the "growling"
> aspect of things is actually the CPU fan. I think I'm going to
> replace the PSU fans anyway since it is still those responsible
> for most of the racket if not the growling.
Hmm ... quieter fans will likely not cool as well. There is a
tradeoff between silence and air-moving capability.
What is the noise source in the current fans? There is cyclic
noise from the fan blade's air pulses against the grating through which
the air flows, and the noise of the bearings (if it has ball bearings.
Sintered bronze sleeve bearings stay quiet until just before things lock
up.
Some fan blades have V-notches in one edge of the blade (I
forget whether it is the leading or the trailing edge, but the purpose
is to generate less noise in operation.) O.K. Looking at an example,
it is the trailing edge which has the V-notches.
>> Once I reached this point, I saw a lot of dust still covering
>> some parts, even though I had used compressed air though the vents to
>> clean it out when I removed it from the otherwise dead system. So, I
>> went down to the shop and fired up the compressor to blow the remaining
>> dust away.
>
> I was actually surprised how clear the PSU was when I opened it
> up. Don't get me wrong, it was filthy inside, but it was in no
> danger of obstructing anything.
Great! I've found massive layers of dust between the PSU and
the side grilles -- but we have two cats in the house, so there is
plenty of lint generation going on. :-)
> The dust coating wasn't even
> particularly thick. You seem to get quite good airflow from those
> fans which may help prevent things getting too bad.
I know that I've taken compressed air to entire machines before,
but I don't know whether this particular power supply has benefited from
such attention. It was not particularly bad until you got into the
areas which were covered by the case and shielded from airflow.
> The date on
> the PSU was actually more recent than I thought (1997) but it is
> still a fairly old machine.
>
>> Next, the fans appear to be attached to hooks soldered at the
>> board level, way out of reach until you pull the board.
>
> This is why I opened the machine up today since I wasn't clear what
> you meant there. I thought this implied that the fans were something
> other than standard equipment fans but I understand what you mean
> now. The hooks are not part of the fan but rest in the regular
> screw holes. The arrangement also means you _need_ 25mm fans - you
> can't substitute slimmer ones.
Well ... you *can* -- if you have access to a machine shop. You
can machine up some plastic spacers which duplicate both the outside
dimensions of the fan and the bore needed to clear the airflow. Drill
and tap holes for small screws where the screw holes line up, and make
it just thick enough to make the full 25mm thickness when assembled.
Put the spacer on the outside side of the fan so the holes for the hooks
remain clear.
Hmm ... if you secure the spacer only by two upper screws, you
can remove those screws, slide the spacer out, and unhook the fan from
the hooks without as complete a disassembly -- unless there is a lip
bent above that area. It is back together, and I'm not going to open it
again. But I would not go for thinner fans, as they probably can't move
as much air, so they can't cool as well. My SS-5s kept running when the
air conditioning died and the room with my computers was at about 98
degrees F. The only thing which died was a Sun LX which was on top of a
six-foot rack, and thus into the hotter air nearer to the ceiling.
> So, it looks like they'll be easy enough to replace. The list
> price on a pair of suitable Papst fans was probably more than the
> machine is worth but I tracked them down on ebay for a much better
> price. I did consider simply swapping the PSU but it is difficult
> to establish if repalcements are genuinelly new - used PSUs are
> obviously going to have fans that are themselves worn.
Well ... the only fan which I have ever had fail on a Sun was
the CPU fan on an Ultra-2 (one of the earlier ones which still had an
individual CPU fan instead of heat sinks to benefit from the airflow
from the PSU.)
I replaced it with a ball bearing equivalent, and that lasted
until I retired that machine and replaced it with an Ultra-5 which draws
noticeably less power and runs cooler. (They also lived on top of the
rack, FWIW, and the Ultra-2 with the replaced fan kept going with no
problems in the heat wave.)
> Thanks for the help. I've now got a couple of fans on order and
> hopefully that will quieten the machine down a bit.
O.K. But as I said -- I consider proper airflow to be more
important than quiet. Of course, I currently have eight computers and
three trays of disk drives running in here, and among those is one Sun
Fire 280R (made for a computer room as a server, so reliability was
considered more important than quiet). The drive trays are next in
noise level (seven drives and dual power supplies in each), and the Sun
Blade 2000 and Sun Blade 1000 are next. The SS-5s fade to
insignificance in terms of noise output.
Good Luck,
DoN.
--
Email: <> | Voice (all times): (703) 938-4564
(too) near Washington D.C. |
http://www.d-and-d.com/dnichols/DoN.html
--- Black Holes are where God is dividing by zero ---
Similar Threads
Linear Mode
