Strange floppy problem

[DanPenny]

This is the closest forum I could find in relation to this problem. If this should be somewhere else, please redirect me.

Unit being worked on;

System: IBM PC 300GL (Desktop)
Model: 6268-43U
Ser. #: 78-W4261
Celeron Proc
256MB Ram

I've never run across something this strange before. I'm hoping someone out there has, and has a solution. The floppy drive in this system doesn't work ~while it's in the system~. Take it out and it works. In any orientation. Switched floppy drives and cables (one at a time of course) with known good units. Everything works until it's placed back in it's bay. Then it doesn't work again. (The bay can't be changed due to the molded front panel.)

I've been to the IBM site, looked up the system (out of warranty of course), and gone through the troubleshooting procedures. These procedures cover seeing if it (floppy drive) works on it's side, then removed from the system. They've got the symptom covered in the procedures so they must know about it. Any links (at IBM) to follow up on this are dead links.

Has anyone experienced such behaviour? Specifically for this system, but any insight will do. ;>)

Thanks Much

It's a good day when you learn something

 

[fluffybunny]

Something may be shorting out your floppy. Check to see that your system is grounded, and no other wires or jumpers are touching the floppy unit.

Regards.

 

[DanPenny]

Earth grounds are good to the machine.
There are no other wires/components close to the drive.

Thanks

It's a good day when you learn something

 

[technome]

Definitely a short to ground either on the power legs or signal circuit, possibly something conductive and small found it's way into the drive. The ground might be after a resistor, which lower the power draw, as result the power supply does not use it's power protection circuit. After having the drive work outside the bay, have you held the drive against the case, and started the machine.. possible screw causing it once mounted.

 

[DanPenny]

More clarifying info;
The boot touches the drive on power up. The drive light flashes for this and it's recognized in the bios AND in 98SE. No problems etc reported in windows ~UNTIL~ I click on it or send something to it. "A: is not accessable. The device is not ready".

The (IBM) bios has no setting that I can find for a boot sequence. (I checked every single setting in every setup window).

On a boot, the drive is touched but ignored. Boots straight to the hard disk. I remove the drive from the bay and the same prodeure works. It boots from the floppy.

I've left the drive loose in the bay (thinking the button in the molded front may be holding the diskette out slightly) to ensure the diskette is fully engaged in the drive. No difference. I can leave the diskette in the drive, remove the drive from the system, and it works.

technome, I had thought about maybe some electrical anomaly of the drive metal frame/housing being in contact with the main chassis metal (the bay). Yet the power and data cable carry all (and multiple) grounds etc. I'll try grounding the floppy drive frame/housing to the main chassis while the drive is out of the machine.





It's a good day when you learn something

 

[technome]

Yes the connector black leads are grounded but what if a data signal path or power path on the floppy circuit board is shunted to ground, this greates a different situation; on the circuit board, if it is grounded to an area where it should not be, voltages to circuits would be different, due to resistors. I saw a similar situation years ago, when a kid inserted a very tiny metal toy into the floppy. Toy was wegged against the pcb board and the metal floppy case

 

[DanPenny]

Based upon what I have tried with this system, and the input I've received from various sources, I'm concluding that a Spectre exists in the enclosure. I have grounded the floppy enclosure to the floppy/main drive bay chassis (while outside of the machine) and the floppy still works. In any orientation. It's only while in the bay that it doesn't work. The last thing to try is the mainboard in another ATX tower. We'll see...........

Thanks to all who contributed.

It's a good day when you learn something

 

[technome]

Screw shorting out the unit once it is mounted in the case?
Time to get the sledge hammer out.

 

[DanPenny]

Screw shorting out the unit once it is mounted in the case?"

No. Screws, cables (power and data), drives, grounding, controllers, etc were all verified and ruled out.

It's a good day when you learn something

 

[technome]

Time for the hammer !

 

[fluffybunny]

Do let us know what the issue was, if you ever find out.

Regards.