Hello, this was going to go to sci.electronics.repair but it
seems that Aioe users are blocked from there, so I'm posting
it here instead.

I recently bought a broken "cocktail" style arcade game
machine made in 1980. Initial repairs were easy. I got the
machine to start by playing with the connections to the CPU
and logic boards and fixed the lack of green on the monitor's
display by soldering the green lead of the RGB signal back
onto the monitor board.

The problem I'm left with is that after about three minutes
(less if it has been recently used), the picture on the
monitor is reduced to some coloured vertical lines and a
whining noise is emitted. Before this, the display bends
inwards at the bottom (and a bit at the top) of the screen. My
assumption has been that this is a sync issue so the first
course of action was to look at the sync signal that goes from
the CPU board to the monitor with my Oscilloscope. I did this
expecting to see a slowly worsening waveform, but instead I
saw what seemed to my eye a quite nice saw tooth.

However what did catch the attention of my untrained eye was
that this waveform varied by less than a volt between 4 and
5V, except for a momentary drop to 0V at the end of the wave.
Period of the waveform was 90uS.

I only have theoretical knowledge of monitor sync signals, so
I didn't know what to expect, but it seems to me that this is
an oddly small voltage change for such a signal. I tried to
find something on the web that would tell me if this is the
case, but failed. Hence I ask here.

Thank you for reading.

--
__ __
#_ |\| | _#

"Computer Nerd Kev" schreef in bericht
. ..
Hello, this was going to go to sci.electronics.repair but it
seems that Aioe users are blocked from there, so I'm posting
it here instead.

I recently bought a broken "cocktail" style arcade game
machine made in 1980. Initial repairs were easy. I got the
machine to start by playing with the connections to the CPU
and logic boards and fixed the lack of green on the monitor's
display by soldering the green lead of the RGB signal back
onto the monitor board.

The problem I'm left with is that after about three minutes
(less if it has been recently used), the picture on the
monitor is reduced to some coloured vertical lines and a
whining noise is emitted. Before this, the display bends
inwards at the bottom (and a bit at the top) of the screen. My
assumption has been that this is a sync issue so the first
course of action was to look at the sync signal that goes from
the CPU board to the monitor with my Oscilloscope. I did this
expecting to see a slowly worsening waveform, but instead I
saw what seemed to my eye a quite nice saw tooth.

However what did catch the attention of my untrained eye was
that this waveform varied by less than a volt between 4 and
5V, except for a momentary drop to 0V at the end of the wave.
Period of the waveform was 90uS.

I only have theoretical knowledge of monitor sync signals, so
I didn't know what to expect, but it seems to me that this is
an oddly small voltage change for such a signal. I tried to
find something on the web that would tell me if this is the
case, but failed. Hence I ask here.

Thank you for reading.

--
__ __
#_ |\| | _#

Apparently something wrong in your horizontal drive signal. Most likely the
line output transformer, maybe the horizontal coils but there are still some
other possibilities. Be aware that the coils are current driven.

petrus bitbyter

petrus bitbyter wrote:

Apparently something wrong in your horizontal drive signal.

OK, so the sync signal going in is fine then?

Most likely the
line output transformer, maybe the horizontal coils but there are still some
other possibilities. Be aware that the coils are current driven.

Forgive my ignorance, but if a coil or transformer failed, would this be
possible in such a way as for the monitor to function for the three minutes
it usually runs before failure. Or did you mean the components associated
with the the coil or transformer?

Also, if I go to repair this part of the circuit (and let's assume I'm not
replacing the Line Output Transformer, I don't suppose I could find a spare
available), I know I'll need to discharge some capacitors, but should I
discharge the high voltage output of the transformer as well?

Thank's for your help.

--
__ __
#_ |\| | _#

"Computer Nerd Kev" schreef in bericht
...
petrus bitbyter wrote:

Apparently something wrong in your horizontal drive signal.

OK, so the sync signal going in is fine then?

Most likely the
line output transformer, maybe the horizontal coils but there are still
some
other possibilities. Be aware that the coils are current driven.

Forgive my ignorance, but if a coil or transformer failed, would this be
possible in such a way as for the monitor to function for the three
minutes
it usually runs before failure. Or did you mean the components associated
with the the coil or transformer?

Also, if I go to repair this part of the circuit (and let's assume I'm not
replacing the Line Output Transformer, I don't suppose I could find a
spare
available), I know I'll need to discharge some capacitors, but should I
discharge the high voltage output of the transformer as well?

Thank's for your help.

--
__ __
#_ |\| | _#

This kind of failure is pretty common especially with coils and
transformers. In this components insulation tends to degrade partially over
time, usually depending on the temperature. So a transformer may look liko
te perform excellent while cold but fail when it grows warmer. Off course,
other components may fail the same way sometimes. I'm afraid you to lack
both skills and equipment to find out what component(s) are actually
failing. Even well trianed and equipped professionals usually dislike this
type of failures. And, you're right about the line output transformer.
Hardly a chance to find a good replacement.

petrus bitbyter

petrus bitbyter wrote:

This kind of failure is pretty common especially with coils and
transformers. In this components insulation tends to degrade partially over
time, usually depending on the temperature. So a transformer may look liko
te perform excellent while cold but fail when it grows warmer. Off course,
other components may fail the same way sometimes. I'm afraid you to lack
both skills and equipment to find out what component(s) are actually
failing. Even well trianed and equipped professionals usually dislike this
type of failures. And, you're right about the line output transformer.
Hardly a chance to find a good replacement.

OK, I'll pull the monitor unit out again to at least give the main board a
close visual inspection. Just my luck that the monitor is the only part of
the machine that I haven't got a schematic for.

Thank you for your help. It probably saved me a lot of messing about with
sync signals.

--
__ __
#_ |\| | _#

petrus bitbyter wrote:
"Computer Nerd Kev" schreef in bericht
...

Snip

This kind of failure is pretty common especially with coils and
transformers. In this components insulation tends to degrade partially over
time, usually depending on the temperature. So a transformer may look liko
te perform excellent while cold but fail when it grows warmer. Off course,
other components may fail the same way sometimes. I'm afraid you to lack
both skills and equipment to find out what component(s) are actually
failing. Even well trianed and equipped professionals usually dislike this
type of failures. And, you're right about the line output transformer.
Hardly a chance to find a good replacement.

petrus bitbyter


Quote
"Even well trianed and equipped professionals usually dislike this type
of failures."
End Quote

"usually"..................I would think "always" would be closer to the
mark, Petrus!! '-)

Daniel

"Computer Nerd Kev" wrote in message
...
petrus bitbyter wrote:

This kind of failure is pretty common especially with coils and
transformers. In this components insulation tends to degrade partially
over
time, usually depending on the temperature. So a transformer may look
liko
te perform excellent while cold but fail when it grows warmer. Off
course,
other components may fail the same way sometimes. I'm afraid you to lack
both skills and equipment to find out what component(s) are actually
failing. Even well trianed and equipped professionals usually dislike
this
type of failures. And, you're right about the line output transformer.
Hardly a chance to find a good replacement.

OK, I'll pull the monitor unit out again to at least give the main board a
close visual inspection. Just my luck that the monitor is the only part of
the machine that I haven't got a schematic for.

Thank you for your help. It probably saved me a lot of messing about with
sync signals.


If its old, there might be issues with dried out electrolytic capacitors,
but these tend to improve as they warm up, but check whether there's a
coupling electrolytic between the horizontal driver transformer and the base
of the power transistor.

If the E/W correction starts to go awry at the start of the problem, the
fault may not be confined to the sync system - observe all the symptoms as a
whole and consider which part of the display unit as a whole could affect
all of them.

On 20 Feb 2013, Ian Field wrote:

If its old, there might be issues with dried out
electrolytic capacitors, but these tend to improve as they
warm up, but check whether there's a coupling electrolytic
between the horizontal driver transformer and the base of
the power transistor.

If the E/W correction starts to go awry at the start of the
problem, the fault may not be confined to the sync system -
observe all the symptoms as a whole and consider which part
of the display unit as a whole could affect all of them.

Thank's, I'll keep that in mind when I finally get around to
looking at the thing. Unfortunately it's something of a
process getting the monitor unit into a visible position.

--
__ __
#_ |\| | _#

"Computer Nerd Kev" wrote in message
. ..
On 20 Feb 2013, Ian Field wrote:

If its old, there might be issues with dried out
electrolytic capacitors, but these tend to improve as they
warm up, but check whether there's a coupling electrolytic
between the horizontal driver transformer and the base of
the power transistor.

If the E/W correction starts to go awry at the start of the
problem, the fault may not be confined to the sync system -
observe all the symptoms as a whole and consider which part
of the display unit as a whole could affect all of them.

Thank's, I'll keep that in mind when I finally get around to
looking at the thing. Unfortunately it's something of a
process getting the monitor unit into a visible position.


In equipment like a VDU, where there are pulsed currents above the hearing
frequency - electrolytics suffer a lot of ripple current, the weak ones will
get noticeably warm - or even hot.

Your sense of touch could save you the cost of an ESR meter - but the
aluminium can isn't isolated, so rail decouplers on negative rails will be
live.

Beware the mains in rectifier reservoir can has negative pulsating DC equal
to the peak value of the mains waveform.

On 21 Feb 2013, Ian Field wrote:

In equipment like a VDU, where there are pulsed currents
above the hearing frequency - electrolytics suffer a lot of
ripple current, the weak ones will get noticeably warm - or
even hot.

Your sense of touch could save you the cost of an ESR meter
- but the aluminium can isn't isolated, so rail decouplers
on negative rails will be live.

I presume I could use an infared temperature sensor to do the
same job while keeping my fingers out of zapping range?

--
__ __
#_ |\| | _#

"Computer Nerd Kev" wrote in message
. ..
On 21 Feb 2013, Ian Field wrote:

In equipment like a VDU, where there are pulsed currents
above the hearing frequency - electrolytics suffer a lot of
ripple current, the weak ones will get noticeably warm - or
even hot.

Your sense of touch could save you the cost of an ESR meter
- but the aluminium can isn't isolated, so rail decouplers
on negative rails will be live.

I presume I could use an infared temperature sensor to do the
same job while keeping my fingers out of zapping range?


In theory yes - but accuracy is very dependant on the reflectivity of the
surface.

Another trick - any capacitor that can melt a hot glue stick is in serious
trouble!