[RiTG]

Hi,

Brief background - I'm an electronic engineer and I work for an elevator component company. I've been landed with a problematic bunch of elevator indicators to fix that are abroad. They are in fact mini PCs running Windows XP mounted behind a small TFT screen. They were designed specifically for this particular site two years ago (not by myself, I've only been with the company 12 months!) and they've been faulty since they were installed. I'm flying over to the site on Monday to try and diagnose and hopefully fix them.

They're reportedly not always booting up when they're powered on - sometimes showing a black screen with "Boot Failure. Insert BOOT diskette in A:". There is no A: drive - there's physically no floppy drive connected, and the A: drive is disabled in BIOS. There's only a mini IDE hard drive which contains the Windows XP installation.

Now, they sent back 4 of them for us to fix. When we got hold of them, they worked absolutely fine in the lab. We tried all sorts to recreate the problem but to no avail. We sent them back and they didn't work again! So we suspected the power supplies on site. They've been replaced with different models twice. They currently have supplies rated at 8A, and from my tests they consume 3.3A peak when they initially power up.

We have one spare unit that I've been running tests on - tested various power supplies, and choked the current of the supply. The only faults I can produce which are similar all involve preventing the hard drive from working (unplugging it, limiting the current so it wont spin up, pressure on the back of the drive so it cant spin, etc, etc). If the hard drive doesnt work, the unit shows the error "Boot Failure. Reboot and Select proper Boot device or Insert Boot Media in selected Boot device". Otherwise, they've been totally reliable in the lab.

So my question is, if there's no floppy drive, why would a computer not boot up with the error "Boot Failure. Insert BOOT diskette in A:", considering any type of hard drive failure seems to show a different message. Personally, I'm wondering if it's a noisy environment or mains that's causing corruption somewhere, but I'm just not sure - I'm not really a total computer pro! Since I'm flying over there, I want to be ready to try as many things as possible - hence my asking for suggestions before I go.


For reference, the mini PC is a Vega86-6270 with AMI BIOS, and it has "VIA Mark CoreFusion 533MHz 0518 0.90V EFT60 F0567G E5" printed on it. We've tried for months to get support from the company that supplied us with them, and they're just fobbing us off basically. We've had a little support from the manufacturer in Taiwan, but they seem to be saying it's a hard drive problem (despite the fact that hard drive problems dont show this error, and they worked when I tried them in the lab). The hard drives are 40GB Seagate Momentus 5400 with "ST940815A" printed on them. The indicators are mounted in holes in walls, but the faceplates have vents next to the screens with fans behind them to keep them cool.


I'd be hugely grateful for any suggestions on this - its had me scratching my head for a while now.

Thanks in advance!

[fgdn17]

does the aim bios have a delayed boot?? ie maybe 4 sec, so to insure the drive is really powered up before trying to boot from it???...just a thought...

[RiTG]

There are no options for a delayed boot unfortunately.

As something to try, I'm going to take a separate full size IDE drive that I've wired up to a seperate laptop style power supply. I can plug it in in place of the mini IDE drive, power the drive up before the PC board, and see if it boots then...

[fgdn17]

that's a good option to try and isolate it...but it sure does sound
like a power issue somewhere...as you've concluded...good luck...

and inquiring minds would like to know the outcome if you
got time...thanks

[Frank4d]

Acording to the manual the board uses 5 VDC @ 2.2 amps and it is on two 0.025 inch square pins. How is the power input wired? (wire guage and length from power supply). The board may not be getting enough juice.

I would also look in the BIOS setup to verify the HDD is set as the first boot device. And disable "floppy seek" at bootup if it has that option. Also see if there is an option to turn off fast memory test (slows bootup).

[JohnWill]

That quite a bit of current for a couple of .025" pins! That's not a very good design...