Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

HP C200/C240/C360 Power Supply problems - solved

65 views
Skip to first unread message

Michael Kukat

unread,
Nov 27, 2004, 11:09:35 AM11/27/04
to
Hello,

i just found the problem in one of my C360s, which don't want to start up the
PSU any longer. This seems to be a common problem with those PSUs, as i could
repair one of 2 C240 PSUs the same way, then other of those 2 seems to have a
different problem on top, but the problem i'm talking about was also present.

PSU types:
Artesyn NFS550-9630E
Computer Products NFS550-9630E

Used in: C200, C240, C360

Problem: After power outage, PSU doesn't start any longer or very difficult.
May need hundreds of power cycles or several hours/days to get up running
again. Standby fans noise not present.

Solution as i found: replace R19. This resistor should have 180 KOhms and was
open in all 3 PSUs i checked. The resistor can be found relatively easy at the
rear of the PSU, opposite of the power inlet, near the screw to fix it into the
enclosure. There are 2 resistors (R19, R28) of same type, maybe R28 also might
suffer this problem. Check both, replace defective, try again.

I didn't have a matching resistor here (seems to be 1/2 Watts or so), so i used
4 680KOhms in parallel, the machine survived 5 test power outages without any
startup problems.

Hope it helps. At least you can save about USD 700 for a replacement PSU.

...Michael

--
http://www.unixiron.org/ Home Powered by: (Net|Open|Free)BSD IRIX NonStop-UX
Solaris AIX HP-UX Tru64 MUNIX Ultrix VMS SINIX Dolphin_Unix OpenStep MacOS A/UX

Martyn_B

unread,
Jan 13, 2005, 4:28:48 PM1/13/05
to
Hi there,

I am a C200 owner, and probably not the only one messing around with
graphics cards trying to get debian to work on it. Today I inserted a
Belkin USB 2.0 32-bit card into PA-RISC system. I have found from past
experience that system can be fussy starting up and gives fault indication
on front bezel. This can be immediately corrected by sliding out the m/b
and relocking it closed so the centronics get good conduction.

This time however as Michael said, there is no fan noise, although kettle
lead arcs a little when connecting.

There is no front bezel lights on power on either.

Just got out my multimeter and found R19 is damaged, R28 is completely
fine... R19 is just stone dead and acting as an insulator. And as Michael
says they should give the same multimeter reading, since band colours are
identical.

Obviously, it goes without saying that adding x86 pci cards to the system
was the straw that broke the donkey's back, but what i am unsure of is if
x86 are hardware compatable with runway bus?... Did the resistor fail due
to too high load, short circuit from pci card or just too much heat build
up from all the dust gathered in the PSU?

If anyone knows the answer to the PCI question, please get back to me so I
can know for next time... ;)

As soon as i know it is safe to use PCI cards in there, I will continue my
quest to get x-windows working again...

I can definitely tell you that none of the xfree86-4.conf options support
SiS6326 and xwindows int10 causes immediate HPMC system halt on a C200...


dipa

unread,
Jan 15, 2006, 8:07:20 AM1/15/06
to

"Martyn_B" <mart...@hotmail.com> wrote in message
news:f6c72ae96d6945bd...@localhost.talkaboutcomputing.com...

> Hi there,
>
> I am a C200 owner, and probably not the only one messing around with
> graphics cards trying to get debian to work on it. Today I inserted a
> Belkin USB 2.0 32-bit card into PA-RISC system. I have found from past
> experience that system can be fussy starting up and gives fault indication
> on front bezel. This can be immediately corrected by sliding out the m/b
> and relocking it closed so the centronics get good conduction.

I've been testing PCI FireWire on C360 and found silence after inserting a
card.


>
> This time however as Michael said, there is no fan noise, although kettle
> lead arcs a little when connecting.
>
> There is no front bezel lights on power on either.
>
> Just got out my multimeter and found R19 is damaged, R28 is completely
> fine... R19 is just stone dead and acting as an insulator. And as Michael
> says they should give the same multimeter reading, since band colours are
> identical.

In my case R28 was damaged.
Replaced with similar one and it works!
Thanks a lot Michael!

> Obviously, it goes without saying that adding x86 pci cards to the system
> was the straw that broke the donkey's back, but what i am unsure of is if
> x86 are hardware compatable with runway bus?... Did the resistor fail due
> to too high load, short circuit from pci card or just too much heat build
> up from all the dust gathered in the PSU?
>
> If anyone knows the answer to the PCI question, please get back to me so I
> can know for next time... ;)

:)))

0 new messages