Hi Folks,

I have a powersupply inside a Samsung LCD monitor that turns itself off
without fail after 20 minutes.
All the Switchmode DC rail voltages such as 12V, 5V and 3.3volts all
completely disappear.
When you disconnect the mains for about 5 minutes and then plug it in again
the power LCD monitor goes for another 20 minutes
before completely losing it's power i.e. No Green power LED and no Picture.
Even disconnected the load from the power supply, but it still has the same
symptons.
Sprayed Rapid Freeze on all the components, but it did make the power supply
come back on-
Even changed some Electrolytic caps due to higher than normal ESR readings
but these have still made no difference.
If you leave the power supply in standby mode with the green led flashing,
it will stay on indefinately for days without
losing the 5V DC rail.
So if there are some Switchmode power supply experts out there, it would be
nice to get
a tip to solve this strange peculiar problem.

Kind regards

Paul Bellett

On Sat, 6 May 2006 12:00:20 +1000, "Paul & Fiona"
put finger to keyboard and composed:

Hi Folks,

I have a powersupply inside a Samsung LCD monitor that turns itself off
without fail after 20 minutes.
All the Switchmode DC rail voltages such as 12V, 5V and 3.3volts all
completely disappear.
When you disconnect the mains for about 5 minutes and then plug it in again
the power LCD monitor goes for another 20 minutes
before completely losing it's power i.e. No Green power LED and no Picture.
Even disconnected the load from the power supply, but it still has the same
symptons.
Sprayed Rapid Freeze on all the components, but it did make the power supply
come back on-
Even changed some Electrolytic caps due to higher than normal ESR readings
but these have still made no difference.
If you leave the power supply in standby mode with the green led flashing,
it will stay on indefinately for days without
losing the 5V DC rail.
So if there are some Switchmode power supply experts out there, it would be
nice to get
a tip to solve this strange peculiar problem.

Kind regards

Paul Bellett

I'd be doing a little circuit tracing. For example, locate the PWM
controller IC, find its datasheet, and determine if it has a latching
shutdown pin. Then work backwards from this pin and determine whether
it is connected to an overvoltage circuit, a current sense circuit, or
a on/off pin on the mainboard.

- Franc Zabkar
--
Please remove one 'i' from my address when replying by email.

Thanks Franc,

Sounds like you are right on to it - A good trouble shooting technique for
analysis.
I guess, if the PWM controller is anything like a common garden variety of
UC3842 PWM IC , there are a couple
methods for shutdown that are accomplished by, either raising pin 3 above 1
V or pull pin 1 below 1 V,
which would localize as to where the external fault to the PWM would
probably lie.
I'll keep you posted on what particular component is causing the PWM IC a
bit of grief!

Cheers

Paul


"Franc Zabkar" wrote in message
...
On Sat, 6 May 2006 12:00:20 +1000, "Paul & Fiona"
put finger to keyboard and composed:

Hi Folks,

I have a powersupply inside a Samsung LCD monitor that turns itself off
without fail after 20 minutes.
All the Switchmode DC rail voltages such as 12V, 5V and 3.3volts all
completely disappear.
When you disconnect the mains for about 5 minutes and then plug it in
again
the power LCD monitor goes for another 20 minutes
before completely losing it's power i.e. No Green power LED and no
Picture.
Even disconnected the load from the power supply, but it still has the
same
symptons.
Sprayed Rapid Freeze on all the components, but it did make the power
supply
come back on-
Even changed some Electrolytic caps due to higher than normal ESR readings
but these have still made no difference.
If you leave the power supply in standby mode with the green led flashing,
it will stay on indefinately for days without
losing the 5V DC rail.
So if there are some Switchmode power supply experts out there, it would
be
nice to get
a tip to solve this strange peculiar problem.

Kind regards

Paul Bellett

I'd be doing a little circuit tracing. For example, locate the PWM
controller IC, find its datasheet, and determine if it has a latching
shutdown pin. Then work backwards from this pin and determine whether
it is connected to an overvoltage circuit, a current sense circuit, or
a on/off pin on the mainboard.

- Franc Zabkar
--
Please remove one 'i' from my address when replying by email.

Hi Frank,

Problem solved - Found that after around 20 minutes, an obsure 5 pin SMPS
supervisory IC with the only markings
on it (S0202 / DP502T) had made the heatsink it was mounted on, rather
hot to touch. In fact even the PCB had some brown heat marks close to where
the actually IC resided.
However Samsung did not fool me with their cyptic identification of the
device! I thought to myself, it is probably a common SMPS IC under a
different manufacturers name.Yes:-, In fact it was a Fairchild
KA5Q12656RTYDU IC, and I confirmed this by Fairchild's Semiconductors web
application note on the KA5Q series, for which Samsung have copied the
identical circuit including surrounding components for 'their' LCD monitor
power supply!
This IC has a built-in thermal shutdown latching circuit for when it is
operating abnormally.
Anyway, I replaced it with the Fairchild KA5Q12656RTYDU IC from Mouser
Electronics ($3.80 USD) and it is now running almost as cool as a cucumber,
and the fault has of course entirely disappeared.

Cheers for now

Paul
"Paul & Fiona" wrote in message
u...
Thanks Franc,

Sounds like you are right on to it - A good trouble shooting technique for
analysis.
I guess, if the PWM controller is anything like a common garden variety of
UC3842 PWM IC , there are a couple
methods for shutdown that are accomplished by, either raising pin 3 above
1 V or pull pin 1 below 1 V,
which would localize as to where the external fault to the PWM would
probably lie.
I'll keep you posted on what particular component is causing the PWM IC a
bit of grief!

Cheers

Paul


"Franc Zabkar" wrote in message
...
On Sat, 6 May 2006 12:00:20 +1000, "Paul & Fiona"
put finger to keyboard and composed:

Hi Folks,

I have a powersupply inside a Samsung LCD monitor that turns itself off
without fail after 20 minutes.
All the Switchmode DC rail voltages such as 12V, 5V and 3.3volts all
completely disappear.
When you disconnect the mains for about 5 minutes and then plug it in
again
the power LCD monitor goes for another 20 minutes
before completely losing it's power i.e. No Green power LED and no
Picture.
Even disconnected the load from the power supply, but it still has the
same
symptons.
Sprayed Rapid Freeze on all the components, but it did make the power
supply
come back on-
Even changed some Electrolytic caps due to higher than normal ESR
readings
but these have still made no difference.
If you leave the power supply in standby mode with the green led
flashing,
it will stay on indefinately for days without
losing the 5V DC rail.
So if there are some Switchmode power supply experts out there, it would
be
nice to get
a tip to solve this strange peculiar problem.

Kind regards

Paul Bellett

I'd be doing a little circuit tracing. For example, locate the PWM
controller IC, find its datasheet, and determine if it has a latching
shutdown pin. Then work backwards from this pin and determine whether
it is connected to an overvoltage circuit, a current sense circuit, or
a on/off pin on the mainboard.

- Franc Zabkar
--
Please remove one 'i' from my address when replying by email.

On Thu, 29 Jun 2006 19:43:26 +1000, "Paul & Fiona"
put finger to keyboard and composed:

Hi Frank,

Problem solved - Found that after around 20 minutes, an obsure 5 pin SMPS
supervisory IC with the only markings
on it (S0202 / DP502T) had made the heatsink it was mounted on, rather
hot to touch. In fact even the PCB had some brown heat marks close to where
the actually IC resided.
However Samsung did not fool me with their cyptic identification of the
device! I thought to myself, it is probably a common SMPS IC under a
different manufacturers name.Yes:-, In fact it was a Fairchild
KA5Q12656RTYDU IC, and I confirmed this by Fairchild's Semiconductors web
application note on the KA5Q series, for which Samsung have copied the
identical circuit including surrounding components for 'their' LCD monitor
power supply!

"KA" is/was a prefix used by Samsung. The reason Fairchild also use it
is because Samsung sold off its power electronics division to them.

http://query.nytimes.com/gst/fullpag...C1A96E9582 60

Congratulations on your clever sleuthing, BTW.

This IC has a built-in thermal shutdown latching circuit for when it is
operating abnormally.
Anyway, I replaced it with the Fairchild KA5Q12656RTYDU IC from Mouser
Electronics ($3.80 USD) and it is now running almost as cool as a cucumber,
and the fault has of course entirely disappeared.

Cheers for now

Paul
"Paul & Fiona" wrote in message
. au...
Thanks Franc,

Sounds like you are right on to it - A good trouble shooting technique for
analysis.
I guess, if the PWM controller is anything like a common garden variety of
UC3842 PWM IC , there are a couple
methods for shutdown that are accomplished by, either raising pin 3 above
1 V or pull pin 1 below 1 V,
which would localize as to where the external fault to the PWM would
probably lie.
I'll keep you posted on what particular component is causing the PWM IC a
bit of grief!

Cheers

Paul


"Franc Zabkar" wrote in message
...
On Sat, 6 May 2006 12:00:20 +1000, "Paul & Fiona"
put finger to keyboard and composed:

Hi Folks,

I have a powersupply inside a Samsung LCD monitor that turns itself off
without fail after 20 minutes.
All the Switchmode DC rail voltages such as 12V, 5V and 3.3volts all
completely disappear.
When you disconnect the mains for about 5 minutes and then plug it in
again
the power LCD monitor goes for another 20 minutes
before completely losing it's power i.e. No Green power LED and no
Picture.
Even disconnected the load from the power supply, but it still has the
same
symptons.
Sprayed Rapid Freeze on all the components, but it did make the power
supply
come back on-
Even changed some Electrolytic caps due to higher than normal ESR
readings
but these have still made no difference.
If you leave the power supply in standby mode with the green led
flashing,
it will stay on indefinately for days without
losing the 5V DC rail.
So if there are some Switchmode power supply experts out there, it would
be
nice to get
a tip to solve this strange peculiar problem.

Kind regards

Paul Bellett

I'd be doing a little circuit tracing. For example, locate the PWM
controller IC, find its datasheet, and determine if it has a latching
shutdown pin. Then work backwards from this pin and determine whether
it is connected to an overvoltage circuit, a current sense circuit, or
a on/off pin on the mainboard.

- Franc Zabkar
--
Please remove one 'i' from my address when replying by email.




- Franc Zabkar
--
Please remove one 'i' from my address when replying by email.

Paul & Fiona wrote:
Problem solved - Found that after around 20 minutes, an obsure 5 pin SMPS
supervisory IC with the only markings
on it (S0202 / DP502T) had made the heatsink it was mounted on, rather
hot to touch. In fact even the PCB had some brown heat marks close to where
the actually IC resided.
However Samsung did not fool me with their cyptic identification of the
device! I thought to myself, it is probably a common SMPS IC under a
different manufacturers name.Yes:-, In fact it was a Fairchild
KA5Q12656RTYDU IC, and I confirmed this by Fairchild's Semiconductors web
application note on the KA5Q series, for which Samsung have copied the
identical circuit including surrounding components for 'their' LCD monitor
power supply!

Actually this is a Samsung component. Probably Fairchild took over the
Samsung semiconductor division, or the other way around??

--
Met vriendelijke groet,

Maarten Bakker.