[parisc-linux] Need Replacement was: Matrox graphics card and C360

[M. Grabert]

On 24 Jun 2003, Martin K. Petersen wrote:

> M> BTW, I have some hardware problems with my C240 lately. When it is
> M> cut of from power (i.e. when there is a power outage or I pull the
> M> power plug), I can turn it on again! I have to wait for some hours
> M> until and suddenly the power button works again. This started to
> M> happen a couple of months ago. I hardly turn the machine of, so I
> M> don't care that much, but it's annoying nevertheless!  Even more
> M> strangely the SCSI bay doesn't get any power anymore, so I have to
> M> use an external SCSI disk. This happened about a month ago after I
> M> wanted to install a new SCSI drive. I suppose a fuse or two are
> M> broken ...  The power botton works, but none of the LEDs!
>
> I've experienced similar things in the past.  One of my C160s is a
> notoriously bad booter.
>
> The trick is to make sure that both the disk tray and the motherboard
> tray are properly seated and screws tightened.

You can be sure it is properly seated! I've opened the C240 quite often
and installed/removed many PCI cards and hard drives in the past.
I guess I opened the case at least 30 times in the last years.

You're right, whenever there was a small gap between the motherboard and
the tray/power supply, I didn't work. However it is getting worse each
time I open the C240 to replace something. I think It's not supposed to be
opened so frequently ...

> I tend to put my Cxxx boxes down on the floor - face side down - and
> let the motherboard tray fall down the last couple of inches.  That
> usually seats things nicely.

Hehe, the same idea as me. The machine is really heavy, and this is also
the only way to open/close the machine without breaking your back ;)

> Also, make sure your disk tray screw is properly inserted.  The
> power/LED assy. as well as disk device power and SCSI data pins are
> all hooked up through a single connector at the back of the disk tray.

I thought of getting a multimeter and investigate which of the pins
are on power and which not and whether some capatitors/resitors are
broken. I gave up the idea since it is really difficult to get access
to the inside of the workstation while it is running :(

> Give that box a big, tight hug.  That usually fixes it *).

This one is a new idea, will try it ;)

> *) Make sure nobody can see you when you are doing this.  More than
>  once I have been accused of humping my PA-RISC hardware while
>  performing this delicate repair procedure.

Unfortunately I tried all other good ideas/hints of you before,

Thanks nonetheless
    Max