I'm posting this with hopefully enough keywords so that other people
can Google and find it, and so save themselves some time if they run
into the same issue.

We have 8 identical workstations, each with a Lite-On SHM-165PS
CD/DVD optical drive. One of these drives had the following symptoms:

1. At BIOS (or in linux, or in Windows), pressing the front panel eject
button would result in a couple of "clunks", and the tray might move
a millimeter out, but the tray would not open.

2. If the front panel button was pushed enough times, or linux "eject"
was used, it might open eventually. If it did then pressing the button
again and again eventually resulted in it working every time. Except
that giving it even 10 seconds rest would put it back to the beginning.

3. With no power, pushing a paper clip in the release hole would
release the tray.

4. The motor could be heard whirring when it was active.

5. With front bezel and tray cover removed (and being careful to look
at an angle, so that no reflected laser light would go directly into
my eyes) I observed that during the clunking the emergency release was
retracting, and then eventually returning to its original position just
before the drive gave up.

The drive was removed from the machine and the top and bottom
of the case also removed. The rubber band was intact and seemed
tight enough. Even so, I tried replacing it with an O ring of
the same size, as this solves CD/DVD tray problems 90% of the time.
Unfortunately, it wasn't the solution this time.

To understand the repair one must visualize the tray mechanism, so
please bear with me.

Picture the drive in the normal upright orientation on the table
in front of you, with the tray opening towards your chest. There
is a "locking rack" which runs left to right across the top/front
of the drive just below the tray. It is easiest to see it
with the tray removed. The left hand side of this rack contains the
"paper clip" tray release mechanism. This consists of about an inch of
linear gear teeth on the rack mated to 1/4 of a circle of teeth on the
release lever/gear. On the right hand side of the rack there is a
shorter run of similar teeth which (can) mesh with a gear driven through
the rubber band by the motor. This driven gear meshes both with the
teeth on the rack and with the teeth on the bottom of the tray.
Slightly to the right of these rack teeth there is a small plastic
pin which sticks up vertically out of the rack. This pin fits into
a groove on the bottom of the tray, on the right side from the
specified viewing position. There is an outlet in this groove which
allows the pin to leave the groove when the tray is (nearly) closed.

Now then, when the tray is closed the rack is driven by the motor all
the way to the left. The pin is out of the groove. Before the tray
opens the rack moves to the right, driven by the gear. This also
retracts the paper clip emergency release lever. The pin moves with the
rack, engages into the tray slot lead in, by hitting a wall of that at a
45 degree angle, and this pushes the tray out slightly so that the tray
teeth will engage the gear as well. As the tray moves the pin in the
tray groove pulls the rack slightly farther to the right, which
disengages the rack teeth from the drive gear. The drive gear
stays engaged with the teeth on the bottom of the tray to
open the tray.

Careful inspection showed a series of very fine scratches on the tray
groove where this pin engages/disengages. Figuring that scratches
indicated something other than a smooth motion, a small amount
of white grease (from elsewhere in the mechanism) was applied to the top
and sides of the rack pin, and also to the 45 degree section of the lead
in groove on the underside of the tray. This tiny amount of lubrication
was all that was required to fix this problem.

Note that there is no evidence of grease on the lead in groove of the
trays on any of the other drives. So I suspect the design intended that
the interaction between the pin and the lead in groove 45 degree wall
not require lubrication. However, apparently sometimes it does. The
45 degree wall in the lead in groove on the bottom of the tray
can be (lightly!) greased without disassembling the drive if the tray is
first opened with a paper clip.

Regards,

David Mathog

Meat Plow wrote:
On Mon, 24 Sep 2007 15:22:53 -0700, David Mathog wrote:

1. At BIOS (or in linux, or in Windows), pressing the front panel eject
button would result in a couple of "clunks", and the tray might move
a millimeter out, but the tray would not open.

I just repaired a Lite On drive that had identical
problems. Turned out to the be mechanism belt.

That was my first thought but changing to a new O ring didn't fix it.

The drive here didn't open reliably, whether or not there was a disk in
the tray.

It's on open question as to why this one drive needed that point
lubricated, but seven other drives didn't.

Regards,

David Mathog