Samsung Syncmaster 245BW

[cloepula]

Originally Posted by stewart710

After seeing the success of a few people on here, I finally pulled the LCD monitor out again that I started this thread for. Cloepula, I checked the voltage you asked about and it was 15.61v. I bought a f9222l and installed it with no change. I checked my transistor 'BR' in place and I could only get a reading in one direction with my ohmmeter.

Speaking of the shotgun approach, I thought maybe I would get lucky replacing the f9222l, but I still only get the standby voltage. Still just the blue light special and no picture.

Welcome back Bill!

So, then you must change this br because in my case was the same thing, if you read with care my post you will find thet me alsow found it conductive in only one position, when change with replacment piece, there is no reading with dmm.

[stewart710]

You are correct. I misread that. Did you test with it in place or out of circuit?

Thanks

[stewart710]

Originally Posted by cloepula

So, then you must change this br because in my case was the same thing, if you read with care my post you will find thet me alsow found it conductive in only one position, when change with replacment piece, there is no reading with dmm.

Would this be the correct part to order: http://search.digikey.com/scripts/Dk...16794095449691

[cloepula]

Tested in and out of circuit, result the same.

[cloepula]

You can try this transistor, but maybe your controler is burned again. I first changed this br and then changed f9222L.

[stewart710]

Originally Posted by cloepula

You can try this transistor, but maybe your controler is burned again. I first changed this br and then changed f9222L.

Hopefully not. I checked the voltages and while they weren't correct (probably because of the 'br' transistor), it was outputting different voltages. If it was burnt out, wouldn't there be nothing to read coming off the f9222l?

[cloepula]

I dont know this.

[Aruba]

Originally Posted by cloepula

I dont know this.

Cloepula,

I finally got my F9222L parts in and replacing it did not help with the unstable
24V when all six inverters are connected to the backlights. There is something
else that is the root cause and I will need to dig in deeper to find it.
The "BR" transistor you mentioned on your mail is actually "BQ" on mine and I
believe someone else mentioned that. I was able to find a partial schematic for this power supply so it should make things a little bit easier in trying to
troubleshoot this. In the schematic this "BQ" transistor is marked as a
2SC2412 NPN transistor. It does indeed come up with this transistor when
searching for "BQ sot-23" on the net. Unfortunately the one on this board
measures ok as a transistor. I was going to use a resistive load to load one
of the inverter outputs such that it will the power supply will stay on. Do you
know how much current a typical 19 or 20 inch length backlight consumes ?
The 24V output is rated for 2.5A per the silkscreen on the power supply
board. I'm estimating each of the backlights is drawing about 300-350mA. Using 3X 1K ohm in series should get me in the ballpark. I'm expecting the
24V output to be stable and continue to run. We will see.

[technicalspecialist]

How did you manage to find a schematic of the power supply? I'm pretty sure samsung wouldn't have just emailed you a PDF if you contacted them. Or maybe they did?
Could you post a link?

[cloepula]

Originally Posted by Aruba

Cloepula,

I finally got my F9222L parts in and replacing it did not help with the unstable
24V when all six inverters are connected to the backlights. There is something
else that is the root cause and I will need to dig in deeper to find it.
The "BR" transistor you mentioned on your mail is actually "BQ" on mine and I
believe someone else mentioned that. I was able to find a partial schematic for this power supply so it should make things a little bit easier in trying to
troubleshoot this. In the schematic this "BQ" transistor is marked as a
2SC2412 NPN transistor. It does indeed come up with this transistor when
searching for "BQ sot-23" on the net. Unfortunately the one on this board
measures ok as a transistor. I was going to use a resistive load to load one
of the inverter outputs such that it will the power supply will stay on. Do you
know how much current a typical 19 or 20 inch length backlight consumes ?
The 24V output is rated for 2.5A per the silkscreen on the power supply
board. I'm estimating each of the backlights is drawing about 300-350mA. Using 3X 1K ohm in series should get me in the ballpark. I'm expecting the
24V output to be stable and continue to run. We will see.

This br part if is ok will not give you any reading with dmm. Will always seam like completly open

[cloepula]

Or maybe Aruba your board is not the same like mine, and have somme other diferences. And for flickering, maybe is somme zener become a little bit weak and releasing on lower voltage then is normal to release. And check all caps with esr tester for hi esr, and with dmm for shorts.

[fattball]

I have 2 of these 245bw's 1 has the blue light only problem, the other is a flickerer, just found this forum, I will try to test them out

[Aruba]

Originally Posted by cloepula

Or maybe Aruba your board is not the same like mine, and have somme other diferences. And for flickering, maybe is somme zener become a little bit weak and releasing on lower voltage then is normal to release. And check all caps with esr tester for hi esr, and with dmm for shorts.

The experiment with the load worked and kept the screen on but it was still flickering. Found out that by adding a load of 0.7A or more on the 24V output
the flickering goes away and the 24V is stable. Once it sits like this and all
warmed up I can lower this current lower to about 0.2A before it will start to
flicker again. Some components are much more sensitive than others to the
freeze spray I apply to them but the area of where the 5.6V zener diode
(DZM803) is the most sensitive. I have tested this zener and it does test out
as a 5.6V zener. I do have some other 5.6V zener I can try to replace with
but don't have a good feeling that it will fix the problem.
I measure a slightly fluctuating voltage on CP803 82uF, 450V bulk cap of
between 356-375V and the power factor measured is only 0.65. What do you
measure for voltage on the bulk cap and have you measured the power
factor ?

[cloepula]

Originally Posted by Aruba

The experiment with the load worked and kept the screen on but it was still flickering. Found out that by adding a load of 0.7A or more on the 24V output
the flickering goes away and the 24V is stable. Once it sits like this and all
warmed up I can lower this current lower to about 0.2A before it will start to
flicker again. Some components are much more sensitive than others to the
freeze spray I apply to them but the area of where the 5.6V zener diode
(DZM803) is the most sensitive. I have tested this zener and it does test out
as a 5.6V zener. I do have some other 5.6V zener I can try to replace with
but don't have a good feeling that it will fix the problem.
I measure a slightly fluctuating voltage on CP803 82uF, 450V bulk cap of
between 356-375V and the power factor measured is only 0.65. What do you
measure for voltage on the bulk cap and have you measured the power
factor ?

If you check the diode and is ok, then is no need to change. How you measure power factor? And this 0.65 what is this? My monitor is returned to friend, and is in use by him, so i canot anymore make measurment on the supply board. About the unstable voltage, my board alsow have alot of unstable voltages on it before was fixed, to be onest when everithing start to work i measure just somme voltages on 5.2v ic controler, and was stable, output 5.3, 24v, and 5.2v was stable to. On this board i change a capacitor i think was 18nf 630v hi frequ, mkp. Changed it with phillips 1500v 15nf.

[Aruba]

Originally Posted by cloepula

If you check the diode and is ok, then is no need to change. How you measure power factor? And this 0.65 what is this? My monitor is returned to friend, and is in use by him, so i canot anymore make measurment on the supply board. About the unstable voltage, my board alsow have alot of unstable voltages on it before was fixed, to be onest when everithing start to work i measure just somme voltages on 5.2v ic controler, and was stable, output 5.3, 24v, and 5.2v was stable to. On this board i change a capacitor i think was 18nf 630v hi frequ, mkp. Changed it with phillips 1500v 15nf.

When using the DVM I mentioned I measure 356-375VDC on the bulk cap
which when working properly should be a solid voltage. I was expecting
somewhere between 370-400V. Today using a scope I measured this voltage
again and I got something totally unexpected. You can see the waveform on
the included picture. As you may know this monitor has a PFC (power factor
corrected) front end that boost the line voltage up to around 370-400VDC.
A TDA4863 IC from Infineon is used to accomplish this feature which when
working properly should present a power factor that is close to 1. Practically
though the achievable power factor is between 0.90 - 0.99. As when I
measure the power factor using a power analyzer I measured only a power
factor of 0.65 which tells me that the power factor correction is not
happening. I was not able to find anything wrong with any of the components
surrounding the IC so maybe the controller is at fault. Again I don't have any
of these IC so need to make another order.
For those that want to check their flickering monitors if they have the same
symptoms I have they can:
- measure the voltage across the 82uF, 450V cap and see if they also have a
fluctuating voltage between 356-375V.
- measure the 24V output and see if that fluctuates between 20-24V.
- use a hair dryer and heat up the area of pin7-14 of the F9222L IC and see
if the flickering goes away and when it cools down that the flickering
returns.

[cloepula]

Originally Posted by Aruba

When using the DVM I mentioned I measure 356-375VDC on the bulk cap
which when working properly should be a solid voltage. I was expecting
somewhere between 370-400V. Today using a scope I measured this voltage
again and I got something totally unexpected. You can see the waveform on
the included picture. As you may know this monitor has a PFC (power factor
corrected) front end that boost the line voltage up to around 370-400VDC.
A TDA4863 IC from Infineon is used to accomplish this feature which when
working properly should present a power factor that is close to 1. Practically
though the achievable power factor is between 0.90 - 0.99. As when I
measure the power factor using a power analyzer I measured only a power
factor of 0.65 which tells me that the power factor correction is not
happening. I was not able to find anything wrong with any of the components
surrounding the IC so maybe the controller is at fault. Again I don't have any
of these IC so need to make another order.
For those that want to check their flickering monitors if they have the same
symptoms I have they can:
- measure the voltage across the 82uF, 450V cap and see if they also have a
fluctuating voltage between 356-375V.
- measure the 24V output and see if that fluctuates between 20-24V.
- use a hair dryer and heat up the area of pin7-14 of the F9222L IC and see
if the flickering goes away and when it cools down that the flickering
returns.

Why to use hairdryer and worm the space around ic?

[Aruba]

Originally Posted by cloepula

Why to use hairdryer and warm the space around ic?

In many cases you may find that the product you are working on works
better when it is "cold" meaning the components within the product
has not warmed up to their usual working temperature. The time to failure could be hours before the product starts to show the failing symptom(s).
A heat gun or a hair dryer will do in a pinch to warm up the area you think has a temperature sensitivity. This method is just to accelerate the heating process where otherwise you may have to wait for many hours before the
symptoms appear before you can narrow the problem area. Products that
works correctly do not or should I say should not have a temperature
sensitivity in the typical operating range of 0-40 deg C. On the other hand
for a product that has a problem only when it is "cold" you can do the opposite and try cooling the area down that has a temp sensitivity
with freeze spray. These usually come in a can form and you can attach a
tiny tube to it so you can direct the cold spray only to the area of concern.
Using both heat gun and freeze spray in conjuction with each other you can
sometimes see the symptom come and go and therfor hopefully narrow the
problem down to the component(s) that is no longer operating correctly within the typical 0-40 C temp range.

[cloepula]

Originally Posted by Aruba

In many cases you may find that the product you are working on works
better when it is "cold" meaning the components within the product
has not warmed up to their usual working temperature. The time to failure could be hours before the product starts to show the failing symptom(s).
A heat gun or a hair dryer will do in a pinch to warm up the area you think has a temperature sensitivity. This method is just to accelerate the heating process where otherwise you may have to wait for many hours before the
symptoms appear before you can narrow the problem area. Products that
works correctly do not or should I say should not have a temperature
sensitivity in the typical operating range of 0-40 deg C. On the other hand
for a product that has a problem only when it is "cold" you can do the opposite and try cooling the area down that has a temp sensitivity
with freeze spray. These usually come in a can form and you can attach a
tiny tube to it so you can direct the cold spray only to the area of concern.
Using both heat gun and freeze spray in conjuction with each other you can
sometimes see the symptom come and go and therfor hopefully narrow the
problem down to the component(s) that is no longer operating correctly within the typical 0-40 C temp range.

Ok. i never use this metod, but now i will try when have similar problem. Ty.
I wait for you to solve this flickering problem.

[cloepula]

Aruba??
Flickering solved or you gived up on this?

[cloepula]

I see you have an oscilloscope.
This days me to bouth my first scope, is a cheap one, and second hand to, and have only one probe 10x for now. Starting to make some measurment with scope, but must buy a 100x probe, because my scope is only 20v. Picoscope adc200/50. 25mhz. For starting and investigating on 100khz smps power supply i think will be good anough. Can you give me somme advice on this?