A few minutes ago, the picture on my monitor suddenly became very small
(moving from the whole screen to the middle of the screen), very flickery,
and very wavy.  I tried turning the monitor off and back on, to see if that
would help.  When I turned it back on, it had the same problem for maybe 10
more seconds, and then, with a loud clicking sound, the picture disappeared
completely.  After that, it was making a faint clicking noise a little more
than once a second.  I also started noticing an electrical burning smell.
I turned it off and unplugged it, and plugged the monitor from the console
of my FreeBSD box (which almost never gets used) into the Windows box.  That
works, but is constantly reminding me of why I bought the nicer monitor for
the Windows box a few years ago.

Is my monitor dead, or is there something fixable that would be likely to
cause this problem?  It's close to three years old, so I'm assuming it's not
still under warranty.

64 responses total.

If it's a 17" or bigger, it's probably worth fixing.  Most likely it is a
problem with the support electronics, rather than the tube itself.

The faint clicking noise you hear is the power supply going into overload
protection.  It usually means that something downstream shorted out.
Perhaps a capacitor or the flyback transistor and / or transformer.  I'd
guess about $100 or more to get it fixed.

before you spend any money, open the case and give it a good cleaning.
I've resurrected monitors which exhibit a regular 'clicking', discovering
that the accumulated dust was making a fine path to ground for the HV,
bypassing the tube.  Clean well, after carefully discharging any residual
charge on the CRT anode, and try it again.

But as you clean it look for smoking guns.  If you can find a heat
damaged area, you have a pretty good clue what died.  I did this to fix
my mono monitor.  It worked great for several months until a different
component blew.  I have yet to look at it this time.

Thanks.  I've got a monitor I borrowed from work, that I'm using for the
moment.  It will be wanted back eventually, so I do need to do something. 
The one that died was a 15" monitor, that was out of warranty by about nine
or ten months.  If it's likely to cost $100 or more to fix it, I should
probably spend $300 on a new monitor, with a warranty.

I may also take it apart and look at it.  I should figure out what the parts
that need to be carefully discharged are before I do that.

You really don't have to discharge anything unless you go to pull the tube
of the flyback transformer.  The flyback is the plastic tower with a thick
wire running to the tube, terminating under a suction cup on the side of
the tube.  When running, it will have 20,000 to 40,000 volts on it.  Not
enough current to kill you, most likely, however, enough to knock you on
your ass and remind you that electricity can hurt.

The picture tube acts as a capacitor that can hold this charge for a 
while.  If you are going to poke around this second anode, it is a good
idea to discharge it.  Simply clip one end of a wire to the bare wire
running around the back of the tube and the other end to a long, skinny,
screwdriver.  Then slide the end of the screwdriver under the suction
cup thingy, holding the screwdriver by its plastic handle, and make
contact with the metal clip.  If you hear a snap, it's discharged.  If
you don't it's probably discharged or your screwdriver isn't grounded.
If the monitor has been off for a while, it will discharge on its own
and you won't hear the snap.\

You only have to discharge the tube if you need to break the second
anode connection.  Otherwise don't worry about it.

My Apple Multiple Scan 15" monitor has recently started to have intermittent
wobblies - some darker waves move rapidly down the display, which also
twists slightly. These events last a fraction of a second, and occur
several times per hour. No sound accompanies the events. Am I on the
verge of losing it? It is nearly 3 years since purchase, so I presume
it is out of warranty, and I can safely (?!) fiddle inside: suggestions?

More than the tube and the flyback have high voltage on them.  If you don't
know which does, do not go messing around inside the monitor.  There
definitely is enough stored electricity inside a monitor to kill you under
the right conditions.  If you don't know what the right conditions are, you
have no business inside a monitor.

Re #7: Have you moved it recently?  In my dorm I get that effect when
someone turns on a fluorescent light that's directly on the other side of
the wall -- the ballest's magnetic field interacts with the one from the
CRT.  Maybe a similar thing is going on with yours.

No, it is in the same location as ever. I did install a new UPS, and
vaguely wondered if that could be giving me wobbly voltage. Is a
UPS "straight through" when it isn't UPSing, or is the power always
going through the power conversion?

Depends.  The cheap ones are usually "straight through" - they're referred
to as "off line" UPS's.  Big, expensive ones are sometimes "on line" UPS's,
which means they're always doing power conversion.  This gives better spike
protection and voltage regulation, since the UPS is basically generating its
own AC independent of what the power company sends you.

A thought -- you didn't put the UPS next to the monitor, did you?  If you
did, move it and see if your problem goes away.  I suppose most UPS units
probably have switching power supplies now, but I've seen some that had
regular transformers and would kick out a pretty good magnetic field when
they decided to switch into charge or self-test mode.

The UPS is about 2 feet from the monitor (center-to-center). It is a
Tripp-Lite 625. The wobblies did not occur yesterday or (so far) today.

The wobblies receded, but a new problem has appeared. Fibrillation on
changing resolution. This occurred just after I had installed MacOS 6 so I
thought I had a problem with an old driver, so I downloaded the new Apple
Display, but it didn't work: fibrillation consists of the picture breaking
into three pieces that flash on and off very rapidly, when I change
resolution. I was in 640x480 and switched to 832x624, but when I swtiched
back it fibrillated. I can now only get it into 832x624 and 512x384, while
trying to get to 640x480 and 800x600 is no longer possible. 

When one switches resolution the monitor clicks as though a relay is
working. What is that? It no longer clicks when I attempt to get into the
now "forbidden" resolutions. I had a chat with an Apple Care person,
wholed led me through all of the protocols he could think of for fixing
this, but nothing worked. He suggested I take the monitor in to be looked
at. 

Is this a home-repairable fault? In fact, is it in software or in the
hardware? 

(I may pick up a replacement monitor at UM PD - is there a way to
test a monitor there without bringing in a CPU?)

Get lucky and use one you find there.

Right - I can always donate it to Kiwanis if it doesn't work. 8^}

I took Klaus' advice and bought another Apple MultiScan 15 from
PD. I only had a few minutes to check it, but did connect it to
a computer they had on their workbench and it did work and show
640x480 resolution. I replaced the troubled one with it....and
it exhibits the same problem! Fibrillation started after it was on
for a bit. So, this is a generic old-age problem. Where do you
think I should look in the 'guts', Klaus? At least I have an operable
monitor (at 832x624) while I look into the problem on the 'old' one. 

Ar you sure it's the monitor?  Could be the video driver.

That has occurred to me, but I'm not sure what evidence I have to
distinguish a monitor problem from a video driver problem. What and where
is the video driver - hardware or software? I replaced the software driver
with the most recent version, with no effect. Is there a video chip? The
problems shown by the two monitors are similar but significantly different
- e.g., the old monitor went into fibrillation immediately at 640x480. The
'new' (old) monitor showed a distortion of X-Y size at 640x480 that could
not be fully corrected, and then started to "wobble"  as the old one had
before fibrillation started. What is completely common is that both won't
work in 640x480 resolution.

Oh - another thing that is peculiar is that there is no problem in 640x480
when the computer is booted up from the system installer CD-ROM. That
would suggest it is NOT the monitor. I have a video PCI card installed and
I've wondered about that - an Xclaim VR accelerator, video capture, and
other goodies card. The monitor plugs into that card, not the CPU monitor
port. I also reinstalled the ATI software (also updated) when this problem
began. Hmmm...maybe I should plug the monitor into the original port and
see what happens.... 

....so did that, and lo-and-behold, no wobblies, fibrillation, or
other problems at 640x480. So I cleaned and reseated the video card,
but that did not improve matters. But it appears the problem is in
the ATI video card. I've sent them a description of the problem - might
get a new card, since it has a 5 year warranty. And I have a spare
monitor, which is probably OK!  This has been quite an exercise in
both frustration and detection. I thank those, especially Klaus, that
submitted suggestions: it was his suggestion to look at the *driver*
that fortified my checking the video card. 

...or maybe you can plug both in, and double your screen space?

How about stereo? :)

My MultiScan 15 monitor problem appears to be solved. It isn't the
monitor: it was the ATI video board *driver* (one of Klaus' guesses).
I had reinstalled the driver (with extensions off), which didn't help,
but ATI said to first trash every ATI* extension and then reinstall
the driver. This seems to have done the trick. The 'owners manual' for
the board says nothing about *re*installing the driver, nor does the
installer have an uninstall mode. You just have to know (or ask, or
try things out at random.....). 

Now this same Apple MultiScan 15 monitor has come up with a new trick.  On
Saturday morning it was fine. Saturday afternoon my wife used it, and the
desktop is now in mottled pastsel colors. Icons are reddish near the
corners and greenish near the edge, and the central region is closest to
true colors and brightest. The color gradations are all....gradual. It
seems stable in this pattern now - rechoosing the background has no
effect, nor did a night off. (In a way, it is kinda attractive, though
distracting;  I've described it as psychadelic....) 

I haven't tried switching monitor ports (CPU vs ATI card - it is on the
ATI card, but reinstalling its software didn't help), nor swapping
monitors. Anyone have any suggestions before I start doing those things? 

Sounds to me like the mask has been bent or come loose from the front of
the tube, or the electron gun has somehow gotten dislodged.  You might
experiment with gently physically moving or jiggling the monitor.  You
might also consult with your family members to see if any have been
experimenting with powerful magnets near the monitor - a sufficiently
powerful one will bend the mask and can cause permament damage.

A degaussing might also be in order.

The focus is still OK - wouldn't some of those things defocus the guns?
Is the mask the screen that prevents each gun from illuminating the
adjacent color pixels of the wrong color? How is that set into the tube?
Can it be dislodged by a mechanical shock? I did whack the monitor
gently and saw no effect. 

I did confirm it wasn't the driver, as it was the same when I switched
from the ATI board to the CPU driver. Also, it is the monitor, as my
spare is now working OK (this is the spare from PropControl that has
an intermittantly bad red gun - momentarily OK). 

Does a magnet cause damage by magnetizing the screen, and hence bending
the electron beams? The color pattern is not precisely symmeteric, but
nearly so, and the effect is mostly around the perimeter, with smooth
blending of transitions from reds to greens blues and back again. I would
expect an applied magnet would show the effect just where it was applied.
However a rotated mask, might cause the effect. 

I think you may be right that it magnetizes the mask instead of bending
it.  Regardless, a magnet that was dragged around the edge *might*
produce an affect similar to what you are seeing.  I suppose a weird
failure of the electronics might also product the effect you see, but
that seems a bit unlikely to me - in principle, for an RGB monitor, the
logic for the 3 guns ought to be almost entirely separate, and should
also be independent of the position of the electron beam on the front of
the screen.

Yup, the mask is the thing that ensures each color electron can only
illuminate the appropriate color on the front of the tube.  I don't know
what exactly is in a modern trinitron, other than it must be steel.  The
only other thing I know is that you can tell a genuine trinitron by the
fact it has two very faint lines that go almost, but not quite
horizontally across the front of the screen, at about 30% of the screen
& about 70%.  I presume it's part of the mask, but I don't know why
exactly they're there.

The front of your monitor may be as thick as 1/2" - wacking it gently is
unlikely to pass any noticeable jiggle to the mask or the rest of the
tube.  I'd try picking the whole monitor up and giving it a firm shake,
or perhaps tilting it on its side or even upside down.  Be careful doing
this - it's possible there's some loose electronics inside, in which
case you might get some spectacular sparks.  Better yet, before trying
this, you might want to unplug it, wait a few minutes, and try this with
the thing unplugged, to make sure there isn't anything obviously loose
and flapping around inside the monitor.

I have a 13" Sony Trinitron monitor for the Mac (DB-15 connector)
for sale :-)

Yes, I noticed...I'm debating going to PropDis for another MultiScan 15 :).

Now that I'm back on my recent PropDis monitor, I've noticed more about
its intermittant lack of red. This seems to occur more often on startup
and then improve while in use. The symptom is that the red gun simplyuu
cuts out, and then comes back on intermittantly, until it is mostly
OK for long periods. What would cause the red gun to be intermittant
more frequently in earlier periods of use than later?

Temperature.  It's probably a loose connection, somewhere.

So, I have to open it up and check/shake all the connections, which
brings me back to - how does one open an Apple MS-15 monitor? Where
should I start prying (hmmm...I have the one with the probably
slipped mask to practice on...).

The Trinitron shadow mask is made of fine vertical wires held under tension. 
That's why Trinitron screens are curved horizontally, but are always
straight vertically.  The faint horizontal lines are support wires to help
keep the vertical wires spaced properly.

*Usually* a mask only gets bent if the monitor experiences some kind of
severe trauma...being dropped on its face, for example, or having a really
strong magnet stuck to the front of it.  The monitor may just need
degaussing, in which case the problem my gradually heal itself...most
monitors have a degaussing coil that operates briefly when the monitor's
turned on.  Actually, I think some Apple monitors have a degauss button on
the back; look for one with a symbol of a horseshoe magnet with a line
through it.

Here's another thought.  Have you rearranged stuff around the monitor
lately?  Putting a set of unshielded speakers next to the monitor, or
anything else that uses magnets, can cause exactly the symptoms you're
seeing.  So can dropping a "shielded" speaker and knocking loose the second
magnet that cancels out the magnetic field of the main one.

Nothing has been jarred, much less dropped, and no changes have been
made in the local setup (and, no separate speakers). It just "happened"
(unless my wife or daughter are covering up... 8^/).

I think Mr. Watts is onto something.  A solder joint gone bad or some-such.
Often difficult to find but resoldering heavy items or items with large
pins will often "find" it.

My SawsAll will open any monitor ;-)  Apple engineers LOVE self-locking
mating plastic parts.  They usually have a secret place that you need to
push or pry at to open things up.  It's always a challenge.

In the choices for my monitor in the control panel are screen resolutions
and *frequency*, such as 640x480 75Hz, 640x480 85 Hz, 832x624 75Hz,
etc. What exactly is changed when these are changed? That is, what does
each number mean and where in the system is each generated, and used?

640x480 75Hz.  That's 640 pixels across the screen, 480 scan lines down.
307200 pixels total.  The screen is redrawn 75 times a second.  The
screen refresh rate affects "flickeriness" -- generally most noticeable
at a distance, or when moving the eyes across the screen rapidly.  Some
people are bothered by this more than others.  Below 60 Hz is where more
people start to notice it.

This does not answer any of my questions. Where are the Hz generated
and where are they used? If I set Hz to other values at the same
resolution, I get very weird effects, so some kind of synchronization
of a whatsis with a whosis is occurring. Then, pixels: I thought the
pixel count was determined by a performated screen (or mesh). What, then,
is what doing to create pixel counts that may or may not be (in)compatible
with the screen mesh?

- Backtalk version 1.3.30 - Copyright 1996-2006, Jan Wolter and Steve Weiss