Intermittent fault in Eizo 19" T766 CRT

[Wiebe Cazemier]

Hi,

My 4 or 5 year old Eizo T766 19" CRT began showing an intermittant fault
yesterday. Sometimes the screen contracts a bit and comes back with wrong
colors. An example is that the screen turns very blue, and you can see the
diagonal vertical retrace lines in blue running over the screen (at least, I
assume they are the retrace lines). When the screen snaps back, it's often too
greenish, but turns normal over the course of a few minutes.

I had it open yesterday, and a couple of times, I could reproduce the fault by
touching the cathode assembly, so I thought it was a physical error, like a
loose contact. And just now, when it did it again, the monitor responded to me
hitting it (gently...). But, I couldn't reproduce it consistently, yesterday
or now.

I have a couple of questions.

1) A couple of times, the fault caused the monitor to go into off mode. At
least, the power led turned yellow, which it normally does in offmode. Is it
likely that when the fault is somewhere in a late stage as the cathode
assembly, that the control electronics is aware of it?

2) Should I want to refit the plugs connected to the cathode assembly (three
big wires; are those the RGB line voltage wires?), I would need to discharge
the CRT. I've seen a lot of videos of people just sticking a screwdriver
connected to the metal chassis under the anode plug, but is that safe? I've
never discharged a CRT before, so I'm kind of apprehensive.

3) How vital is it to discharge the power supply caps? If so, how does one do
that?

Any other insight is welcome, of course.

Thanks in advance,

Wiebe Cazemier

 

[James Sweet]

Hi,

My 4 or 5 year old Eizo T766 19" CRT began showing an intermittant fault
yesterday. Sometimes the screen contracts a bit and comes back with wrong
colors. An example is that the screen turns very blue, and you can see the
diagonal vertical retrace lines in blue running over the screen (at least,
I
assume they are the retrace lines). When the screen snaps back, it's often
too
greenish, but turns normal over the course of a few minutes.

I had it open yesterday, and a couple of times, I could reproduce the
fault by
touching the cathode assembly, so I thought it was a physical error, like
a
loose contact. And just now, when it did it again, the monitor responded
to me
hitting it (gently...). But, I couldn't reproduce it consistently,
yesterday
or now.

I have a couple of questions.

1) A couple of times, the fault caused the monitor to go into off mode. At
least, the power led turned yellow, which it normally does in offmode. Is
it
likely that when the fault is somewhere in a late stage as the cathode
assembly, that the control electronics is aware of it?

2) Should I want to refit the plugs connected to the cathode assembly
(three
big wires; are those the RGB line voltage wires?), I would need to
discharge
the CRT. I've seen a lot of videos of people just sticking a screwdriver
connected to the metal chassis under the anode plug, but is that safe?
I've
never discharged a CRT before, so I'm kind of apprehensive.

3) How vital is it to discharge the power supply caps? If so, how does one
do
that?

Any other insight is welcome, of course.

Thanks in advance,

Wiebe Cazemier

I would suspect bad solder joints in the area where you can induce the fault
by tapping. It *may* be a short inside the CRT, but don't condemn that yet.
Don't worry about discharging the capacitors, just unplug it for 10 minutes
or so and it should be safe. Even when you're sure everything is dead, treat
it as live just in case.

 

[Wiebe Cazemier]

I would suspect bad solder joints in the area where you can induce the fault
by tapping. It *may* be a short inside the CRT, but don't condemn that yet.
Don't worry about discharging the capacitors, just unplug it for 10 minutes
or so and it should be safe. Even when you're sure everything is dead, treat
it as live just in case.

And what about the CRT? Is discharging as I described safe?

When the fault is in the CRT, I assume nothing can be done? Well, let's just
hope it's not that; those Sony Trinitron tubes are pretty good, so I guess the
chance is small. But on the other hand, because after a fault, the screen is
greenish, which turns OK over the course of about 30-60 minutes, heating
cathodes may very well be it. The cathode assembly itself doesn't heat up that
fast.

 

[James Sweet]

And what about the CRT? Is discharging as I described safe?

When the fault is in the CRT, I assume nothing can be done? Well, let's
just
hope it's not that; those Sony Trinitron tubes are pretty good, so I guess
the
chance is small. But on the other hand, because after a fault, the screen
is
greenish, which turns OK over the course of about 30-60 minutes, heating
cathodes may very well be it. The cathode assembly itself doesn't heat up
that
fast.

You shouldn't have to discharge the CRT, that's only necessary if you have
to remove the anode lead, even then it usually discharges itself through the
focus divider on color monitors.

The 22" flat trinitron tubes commonly have intermittant focus issues, mine
acts up occasionally but has behaved pretty well lately. Haven't run into
that with the 19" tubes.

 

[bz]

You shouldn't have to discharge the CRT, that's only necessary if you
have to remove the anode lead, even then it usually discharges itself
through the focus divider on color monitors.

Caution: even if you discharge the CRT, if the anode lead is disconnected,
as the dielectric (glass in this case) 'relaxes from the stresses of being
charged', a charge can build back up.

I can remember drawing a nice arc many minutes after discharging the tube
the first time.

This can be a bit of a surprise, especially if you are removing the CRT to
replace it. Shocking, one might say.

Also, the comment about letting it sit for 10 minutes so all the capacitors
will be discharged is good, assuming that there are no open bleeder
resistors, etc. But, it might NOT be good to 'bet your life' on the
bleeders doing their job. Occasionally, they do open, so your caution to
treat them as if they might be 'live' is a good one for those that want a
long life.

--
bz 73 de N5BZ k

please pardon my infinite ignorance, the set-of-things-I-do-not-know is an
infinite set.

[email protected] remove ch100-5 to avoid spam trap

 

[James Sweet]

Caution: even if you discharge the CRT, if the anode lead is disconnected,
as the dielectric (glass in this case) 'relaxes from the stresses of being
charged', a charge can build back up.

I can remember drawing a nice arc many minutes after discharging the tube
the first time.

This can be a bit of a surprise, especially if you are removing the CRT to
replace it. Shocking, one might say.

Also, the comment about letting it sit for 10 minutes so all the
capacitors
will be discharged is good, assuming that there are no open bleeder
resistors, etc. But, it might NOT be good to 'bet your life' on the
bleeders doing their job. Occasionally, they do open, so your caution to
treat them as if they might be 'live' is a good one for those that want a
long life.

Yes if you disconnect the anode, you should definitely discharge it again
before you touch it, I usually leave a clip lead connecting it to the ground
strap so that doesn't happen. The zap won't hurt you, but it's enough of a
surprise to drop the thing or cut your hand open. In this case he shouldn't
have to disconnect the anode though.

 

[Wiebe Cazemier]

Also, the comment about letting it sit for 10 minutes so all the capacitors
will be discharged is good, assuming that there are no open bleeder
resistors, etc. But, it might NOT be good to 'bet your life' on the
bleeders doing their job. Occasionally, they do open, so your caution to
treat them as if they might be 'live' is a good one for those that want a
long life.

OK, but since I will need to solder and/or refit plugs, I can't consider it to
be 'live' at all time. Will it be sufficient to measure the voltage between
the RGB line wires, for example, and ground, to give me an idea of whether
it's safe?

 

[Wiebe Cazemier]

You shouldn't have to discharge the CRT, that's only necessary if you have
to remove the anode lead, even then it usually discharges itself through the
focus divider on color monitors.

So discharge through the cathode assembly won't happen?

The 22" flat trinitron tubes commonly have intermittant focus issues, mine
acts up occasionally but has behaved pretty well lately. Haven't run into
that with the 19" tubes.

Focus issues are not my problem. When there is a fault, the screen flashes a
couple of times, contracts, expands (like an old TV) and sometimes it shuts
itself off.

 

[bz]

OK, but since I will need to solder and/or refit plugs, I can't consider
it to be 'live' at all time. Will it be sufficient to measure the
voltage between the RGB line wires, for example, and ground, to give me
an idea of whether it's safe?

Yes.




--
bz

please pardon my infinite ignorance, the set-of-things-I-do-not-know is an
infinite set.

[email protected] remove ch100-5 to avoid spam trap

 

[bz]

So discharge through the cathode assembly won't happen?


Focus issues are not my problem. When there is a fault, the screen
flashes a couple of times, contracts, expands (like an old TV) and
sometimes it shuts itself off.

Sounds like you might have an intermittent short to ground on a power
supply line, perhaps inside a bad capacitor.
Or an intermittent connection in the voltage sensing circuits, or the AC
line into the power supply.

Intermittents are difficult to localize and fix.

Perhaps you can narrow down the problem by monitoring voltages at various
points while it is acting up.



--
bz 73 de N5BZ k

please pardon my infinite ignorance, the set-of-things-I-do-not-know is an
infinite set.

[email protected] remove ch100-5 to avoid spam trap

 

[Wiebe Cazemier]

Sounds like you might have an intermittent short to ground on a power
supply line, perhaps inside a bad capacitor.
Or an intermittent connection in the voltage sensing circuits, or the AC
line into the power supply.

The problem here is, that I'm not too familiar with CRT screen electronics. I'm
pretty well versed in audio electronics and other more common, low voltage,
stuff, but I've never really had to deal with CRTs, so I don't know too much
about them, or how to localize the circuits you mentioned (except perhaps the
PSU).

Would you happen to know of a good book/guide about CRT repair, or at least an
explanation of the insides?

Intermittents are difficult to localize and fix.

Tell me about it... This fault only happens when it's just turned on, thus far,
and very hard to reproduce.

Perhaps you can narrow down the problem by monitoring voltages at various
points while it is acting up.

I also have a transient recorder, although only a single channel. Still, it
should be useful. However, come to think of it, the voltages inside the
monitor are probably too high for it...

 

[bz]

The problem here is, that I'm not too familiar with CRT screen
electronics. I'm pretty well versed in audio electronics and other more
common, low voltage, stuff, but I've never really had to deal with CRTs,
so I don't know too much about them, or how to localize the circuits you
mentioned (except perhaps the PSU).

Would you happen to know of a good book/guide about CRT repair, or at
least an explanation of the insides?


Tell me about it... This fault only happens when it's just turned on,
thus far, and very hard to reproduce.

So, what is different about 'just turned on'?
Components are cool.
High transient currents occur.

Look for places that those can make a difference.

I also have a transient recorder, although only a single channel. Still,
it should be useful. However, come to think of it, the voltages inside
the monitor are probably too high for it...

From the symptoms, I would start by looking at the PSU outputs.
Troubleshoot by 'divide and conquer'.

Divide the device logically into two halves, and localize the problem to
one of the two.

Continue until you reach the bad part.






--
bz 73 de N5BZ k

please pardon my infinite ignorance, the set-of-things-I-do-not-know is an
infinite set.

[email protected] remove ch100-5 to avoid spam trap

 

[Wiebe Cazemier]

So, what is different about 'just turned on'?
Components are cool.
High transient currents occur.

Look for places that those can make a difference.

(snip)

From the symptoms, I would start by looking at the PSU outputs.
Troubleshoot by 'divide and conquer'.

Divide the device logically into two halves, and localize the problem to
one of the two.

Continue until you reach the bad part.

I've been reading the repair faq at http://www.repairfaq.org/sam/crtfaq.htm. At
some point it says this:

Occasionally, small conductive flakes or whiskers present since the day of
manufacture manage to make their way into a location where they short out
adjacent elements in the CRT electron guns. Symptoms may be intermittent or
only show up when the TV or monitor is cold or warm or in-between. Some
possible locations are listed below:

One of them is: Heater to cathode (H-K). The cathode for the affected gun will
be pulled to the heater (filament) bias voltage - most often 0 V (signal
ground). In this case, one color will be full on with retrace lines. Where the
heater is biased at some other voltage, other symptoms are possible like
reduced brightness and/or contrast for that color. This is probably the most
common location for a short to occur.

That appears to be exactly what's happening. The blue gun is turned on
completely, also showing the retrace lines. Also, remember that I have been
able to reproduce the fault a couple of times by tapping the cathode assembly.

I will try the "put the monitor on it's face and try tapping it to dislodge the
short" method first. If it works, it would be a very low tech, but effective
solution

The "blow out the short with a capacitor" method also seems fun, but also a
little risky

 

[bz]

I've been reading the repair faq at
http://www.repairfaq.org/sam/crtfaq.htm. At some point it says this:

Occasionally, small conductive flakes or whiskers present since the day
of manufacture manage to make their way into a location where they short
out adjacent elements in the CRT electron guns. Symptoms may be
intermittent or only show up when the TV or monitor is cold or warm or
in-between. Some possible locations are listed below:

One of them is: Heater to cathode (H-K). The cathode for the affected
gun will be pulled to the heater (filament) bias voltage - most often 0
V (signal ground). In this case, one color will be full on with retrace
lines. Where the heater is biased at some other voltage, other symptoms
are possible like reduced brightness and/or contrast for that color.
This is probably the most common location for a short to occur.

That appears to be exactly what's happening. The blue gun is turned on
completely, also showing the retrace lines. Also, remember that I have
been able to reproduce the fault a couple of times by tapping the
cathode assembly.

If you are SURE that no vibrations from your tapping traveled elsewhere,
that is a strong clue.
Otherwise it can be a misleading clue.

I have played the 'taps' game and won. I have played it and lost.
It is best when you can make smaller and smaller taps while getting closer
and closer to the problem.

If the tap will only occasionally trigger/fix the problem, then it is easy
to be misled.

I will try the "put the monitor on it's face and try tapping it to
dislodge the short" method first. If it works, it would be a very low
tech, but effective solution

The "blow out the short with a capacitor" method also seems fun, but
also a little risky

When I had a TV repair shop, in the early 70's, we had a tester for picture
tubes.
It had a short indicator and a 'remove short' button that discharged a
capacitor through the short.
Sometimes it would fix things. Sometimes it would make things worse.

Good luck with your experiment. Hopefully, it won't drop something into the
center of the shadow mask.

Be aware that with 15 lb of air pressing on each square inch of the CRT,
you are playing with a live bomb.

I once took a picture tube out, laid it on its faceplate, got about 50 feet
away and tossed rocks at it.

Woomp. Dust and dirt and glass flew everywhere.

Chunks of the face plate (glass about 1 1/2 inches thick) landed about 50
feet BEHIND me.

Now, I would want a thick sheet of lexan between me and any CRT that I was
tapping upon, and safety goggles, gloves.




--
bz

please pardon my infinite ignorance, the set-of-things-I-do-not-know is an
infinite set.

[email protected] remove ch100-5 to avoid spam trap

 

[James Sweet]

I've been reading the repair faq at
http://www.repairfaq.org/sam/crtfaq.htm. At
some point it says this:

Occasionally, small conductive flakes or whiskers present since the day of
manufacture manage to make their way into a location where they short out
adjacent elements in the CRT electron guns. Symptoms may be intermittent
or
only show up when the TV or monitor is cold or warm or in-between. Some
possible locations are listed below:

One of them is: Heater to cathode (H-K). The cathode for the affected gun
will
be pulled to the heater (filament) bias voltage - most often 0 V (signal
ground). In this case, one color will be full on with retrace lines. Where
the
heater is biased at some other voltage, other symptoms are possible like
reduced brightness and/or contrast for that color. This is probably the
most
common location for a short to occur.

That appears to be exactly what's happening. The blue gun is turned on
completely, also showing the retrace lines. Also, remember that I have
been
able to reproduce the fault a couple of times by tapping the cathode
assembly.

I will try the "put the monitor on it's face and try tapping it to
dislodge the
short" method first. If it works, it would be a very low tech, but
effective
solution

The "blow out the short with a capacitor" method also seems fun, but also
a
little risky

Don't be so quick to condemn the CRT, sure it could be at fault, but
eliminate the external stuff first! A bad connection on the neck board can
just as easily cause the same symptom.

 

[Wiebe Cazemier]

Don't be so quick to condemn the CRT, sure it could be at fault, but
eliminate the external stuff first! A bad connection on the neck board can
just as easily cause the same symptom.

But won't a bad connection give me a black screen, because when the cathode is
disconnected from it's line, no current can flow. How can the blue cathode be
shorted to ground on the neck board? I can't see anything (obvious) on the
board that could cause a short.

BTW, I thought the RGB guns had separate line wires, but besides the focus and
grid voltage wires, only one extra thick wire runs to the neck board. That has
to be the cathode line voltage, right? And if so, doesn't that mean that when
it's shorted to ground, all the guns are fully turned on, and not just the
blue one (as is the case here)?

 

[Wiebe Cazemier]

If you are SURE that no vibrations from your tapping traveled elsewhere,
that is a strong clue.
Otherwise it can be a misleading clue.

I have played the 'taps' game and won. I have played it and lost.
It is best when you can make smaller and smaller taps while getting closer
and closer to the problem.

If the tap will only occasionally trigger/fix the problem, then it is easy
to be misled.

I know what you mean. Just yesterday, I had another CRT open, which often has
one or more of it's colors failing. You can fix it by smashing it, but when I
had it open, I could fix it by very gently touching the neck board. In that
case, it was clear where the fault was. In the case with this Eizo, I can tap
the neck board a lot harder without it doing anything, so you're right in that
it could very well be something else.

When I had a TV repair shop, in the early 70's, we had a tester for picture
tubes.
It had a short indicator and a 'remove short' button that discharged a
capacitor through the short.
Sometimes it would fix things. Sometimes it would make things worse.

I assume that would connect between chassis ground and cathode? As I mentioned
in my other post, I thought there were three line wires, but I can only
discern one; is that the one?

And the cathode line capacitor BTW, where is that likely to be located? Near
the flyback on the mainboard?

Good luck with your experiment. Hopefully, it won't drop something into the
center of the shadow mask.

Aperture grill mask, actually

Be aware that with 15 lb of air pressing on each square inch of the CRT,
you are playing with a live bomb.

I once took a picture tube out, laid it on its faceplate, got about 50 feet
away and tossed rocks at it.

Woomp. Dust and dirt and glass flew everywhere.

Chunks of the face plate (glass about 1 1/2 inches thick) landed about 50
feet BEHIND me.

Now, I would want a thick sheet of lexan between me and any CRT that I was
tapping upon, and safety goggles, gloves.

I was very careful, because I really didn't want to break anything. But anyway,
I wasn't able to fix the problem. I'm looking at the screen right now, and as
I'm typing this, it just flashed again. I don't know if I can or should tap
harder, but I wouldn't feel too comfortable trying it.

It's really annoying that I can't reproduce the fault reliably... Even if it is
a short between the blue gun and ground, I couldn't even short it out with a
cap, because the problem only every occurs for a fraction of second, very
sporadically...

 

[Wiebe Cazemier]

But won't a bad connection give me a black screen, because when the cathode
is disconnected from it's line, no current can flow. How can the blue cathode
be shorted to ground on the neck board? I can't see anything (obvious) on the
board that could cause a short.

BTW, I thought the RGB guns had separate line wires, but besides the focus
and grid voltage wires, only one extra thick wire runs to the neck board.
That has to be the cathode line voltage, right? And if so, doesn't that mean
that when it's shorted to ground, all the guns are fully turned on, and not
just the blue one (as is the case here)?

Wait, I've been thinking. An intermittent short in a capacitor between the blue
cathode line voltage and ground could of course also cause this (and there are
some of them, ceramics, on the neck board). But one question about that, which
I guess also applies to when the short is inside the CRT:

Provided there is one cathode driving voltage for all guns, why aren't all guns
fully turned on when one of the cathodes is shorted to ground? Because if one
is shorted to ground, there is no more line voltage for the other guns either,
right?

 

[bz]

I know what you mean. Just yesterday, I had another CRT open, which
often has one or more of it's colors failing. You can fix it by smashing
it, but when I had it open, I could fix it by very gently touching the
neck board. In that case, it was clear where the fault was. In the case
with this Eizo, I can tap the neck board a lot harder without it doing
anything, so you're right in that it could very well be something else.


I assume that would connect between chassis ground and cathode?

If I remember correctly, the short was located via the earlier testing,
rotating a switch to localize the short.
You would then charge up a capacitor that was built into the tester and
then discharge the capacitor through the short via a pushbutton switch.

As I
mentioned in my other post, I thought there were three line wires, but I
can only discern one; is that the one?

[http://en.wikipedia.org/wiki/Trinitron]
Another feature was a three cathode single electron gun arrangement, in
contrast to the then-dominant (and still common) three gun arrangement.
Single gun systems tend to be easier to manufacture reliably, simplify beam
focus and control, and are less prone to inter-electrode short circuits.
[unquote]

And the cathode line capacitor BTW, where is that likely to be located?
Near the flyback on the mainboard?

Sorry, no idea. But I think you are mistaking which capacitor is going to
be used to clear the short.

Aperture grill mask, actually
Correct.


I was very careful, because I really didn't want to break anything. But
anyway, I wasn't able to fix the problem. I'm looking at the screen
right now, and as I'm typing this, it just flashed again. I don't know
if I can or should tap harder, but I wouldn't feel too comfortable
trying it.

It's really annoying that I can't reproduce the fault reliably... Even
if it is a short between the blue gun and ground, I couldn't even short
it out with a cap, because the problem only every occurs for a fraction
of second, very sporadically...

Intermittents are very problematic.



--
bz

please pardon my infinite ignorance, the set-of-things-I-do-not-know is an
infinite set.

[email protected] remove ch100-5 to avoid spam trap

 

[James Sweet]

Wait, I've been thinking. An intermittent short in a capacitor between the
blue
cathode line voltage and ground could of course also cause this (and there
are
some of them, ceramics, on the neck board). But one question about that,
which
I guess also applies to when the short is inside the CRT:

Provided there is one cathode driving voltage for all guns, why aren't all
guns
fully turned on when one of the cathodes is shorted to ground? Because if
one
is shorted to ground, there is no more line voltage for the other guns
either,
right?

It appears you need to look at the schematic for a CRT monitor or TV to get
a better understanding of how this works. A disconnected cathode will float,
you have to pull it down towards ground to turn it on. The transistors
driving the cathodes isolate them from one another.


It depends on where the bad connection is, and the design of the cathode
driver circuits. It's also possible for one of the driver transistors to
have an intermittant short. I would desolder the blue cathode pin on the CRT
socket and tap the neck with the monitor on. If you can still repro the
symptoms, then the problem is probably in the tube.