On Mon, 19 Feb 2007 11:52:41 GMT, "Pete" <not@this.address>
wrote:

>Hi
>
>I recently set up a system with an Asus A7A266 motherboard. I installed
>windows ME and all was fine, then the next day i started the machine up and
>got a boot disk failure message.
>
>I used a different power connector and a different 80 wire IDE ribbon cable.
>All seemed ok for a few days till I got another boot disk failure message
>when i booted up. I switched off and on a couple of times and the message
>went away and the computer loaded without problems.
>
>I got anothe error message a few days later so i did a full surface scan on
>the 20GB disc but no problems were found. I changed the hard disk for
>another 20GB one, so now I had changed the drive, the power cable and the
>ribbon cable and still I get intermittent boot disk failures messages at
>boot up. Both drives show no problems at all when scanned.
>
>I am assuming that the motherboard IDE is knackered or could this be caused
>by something else? Wrong master slave designation, incorrect bios setting,
>drive connected via wrong cable connector (grey instead of black)?????
>
>Everything else seems fine and there are no problems when the machine has
>booted up.


Have you ran any stress tests while the system is running
(seemingly ok) ?

Does it only fail to boot from a cold/off start, or does it
do the same if you reset, restart the computer from an on
state? If only when off, the temp difference could cause an
intermittent connection from a bad solder joint or crack in
the board, or a bad PSU can prevent the drive from spinning
up fast enough.

It could be a bad jumper connection, even an instable CPU
can cause it and an instable CPU could come from
overheating, overclocking, bad capacitors, etc. I'd
definitely examine the capacitors for signs of venting,
though one picture I saw of that board suggested it had some
Fuji polymer capacitors (short yellow jacketed) which due to
being solid, may just slightly bulge at the top, not as
visible a sign of failure as some of the electrolytics show.
Those polymer caps are pretty reliable though unless they're
running quite hot, the next time you get your system up &
running for awhile, you might touch-test the tops of those
caps to see if they feel excessively warm->hot.

I would think it least likely that only the motherboard IDE
"went out", unless it happened to be a physical damage to
the board, perhaps something old like a trace with the
lacquer scraped off and it took years for the copper to
oxidize away, or if it was moved out of a case then later
handling, or flexing the board to cause cold solder joints
or hairline crack(s).

If the system otherwise tests stable you might try using a
PCI IDE card.