Samsung SyncMaster T240 - 2 seconds then black problem

[PlainBill]

Re: Samsung SyncMaster T240 - 2 seconds then black problem

Thanks Plain Bill, sorry I haven't been quicker on my follow ups. Been a week of broken cars, lost hamsters, and general mayhem. Being a single parent aint for wussies.

I re-soldered that jumper as you advised and there is still no output on 'B' connector. So this means inverter/transformer?

I contacted Samsung for a little info and that was a dead end (no surprise really). They were nice enough to get back in 24 hrs though. So no service manual, no part specs, no comparable board listing, nada.

The other poster did solve this by going to an E-pay seller and I am considering the same. One beef I have with that is: is this a faulty design or a faulty part? Fixing the board if possible is better if its a weak design or under speced. If I just have a bad apple then another board may be a whole lot easier than finding compatible parts.

Thanks again to PlainBill and the latest poster.

Fire away folks, I am all ears!

Bruce

One of the laws of electronics is 'Components fail'. The corollary is that overstressed components fail sooner. Some circuits can be designed can be designed conservatively, others it is more difficult. Designing a reliable high voltage transformer is relatively easy.

Ensuring they are built to the design is difficult. Far and away the most difficult part of the process is ensuring the wire used has an adequate insulation thickness. Even sawing a transformer apart, potting it, and examining the windings under a microscope won't ensure it. All it takes is a thin spot where two wires are in contact and it will be more likely to fail.

PlainBill

[DUCE]

Re: Samsung SyncMaster T240 - 2 seconds then black problem

Thanks Plain Bill, sorry I haven't been quicker on my follow ups. Been a week of broken cars, lost hamsters, and general mayhem. Being a single parent aint for wussies.

I re-soldered that jumper as you advised and there is still no output on 'B' connector. So this means inverter/transformer?

I contacted Samsung for a little info and that was a dead end (no surprise really). They were nice enough to get back in 24 hrs though. So no service manual, no part specs, no comparable board listing, nada.

The other poster did solve this by going to an E-pay seller and I am considering the same. One beef I have with that is: is this a faulty design or a faulty part? Fixing the board if possible is better if its a weak design or under speced. If I just have a bad apple then another board may be a whole lot easier than finding compatible parts.

Thanks again to PlainBill and the latest poster.

Fire away folks, I am all ears!

Bruce

[severach]

Re: Samsung SyncMaster T240 - 2 seconds then black problem

Samsung Touch of Color T240
Panel: Samsung LTM240CT04 TN Film
Power Supply: Samsung IP-54155A BN44-00226D SEM2005 AP4506GEH
Video: BN41-01030A LST24TWHSUV/ZA MST97889CLD-LF USB2514-AEZG TDA7266D PI3HDMI
Button Board: Too hard to remove. The wires are easily broken.

This Samsung T240 powers on with no backlight. No image can be seen with a bright light. This case is odd because as you take it apart it appears the previous owner forgot to put the screws back in. They didn't. This almost screwless case only has two screws to hold the case together. The rest comes apart with a putty knife. It's tough to figure out how to get it apart until you pop the stand mount out of the bottom of the monitor to reveal the screws. It just snaps in there and the rest of the case just snaps together.

All caps look good on the power supply. I pull and test Samwha 820uf 25v C316 and C317, all good. No shorts on complementary MOSFET pair U202 4506GEH. I resolder the transformer with no success.

Each lamp connector has two lamp outputs and the plugs can be shifted to test each lamp separately. With 4 lamps plugged in the backlight comes only for an instant and turns off. With a either lamp pair plugged into CN303 CN304 and CN301 CN302 empty the back light stays on and the transformer hisses for about 2 seconds then turns off. Both lamp pairs work equally well in CN303 CN304. With lamps plugged into CN303 CN304 and only one of CN301 or CN302 the back light stays on for 2 seconds. No help there so I unplug CN303 and CN304 and try CN301 and CN302 individually. The instant flash brightness appears the same. When all lamps are unplugged there is no hiss so the backlight is probably shut off in an instant.

No lamps: Unknown, likely to be only on for an instant
CN301+CN302+CN303+CN304: Instant
CN303+CN304: 2 seconds + hiss
CN302+CN303+CN304: 2 seconds + hiss
CN301+CN303+CN304: 2 seconds + hiss
CN301: Instant
CN302: Instant
CN303: 2 seconds no hiss
CN304: 2 seconds no hiss
CN301+CN302: 2 seconds no hiss
CN301+CN302+CN304: 2 seconds no hiss
CN301+CN302+CN303: 2 seconds no hiss
CN302+CN304: 2 seconds no hiss
CN302+CN303: Instant
CN303+CN301: 2 seconds no hiss
CN304+CN301: Instant

All this tells me is that some faults take longer to detect than others which means that users may report no back light, a back light for an instant, or a 1 second back light rounded up to 2 seconds.

With the lights off I see arcs inside the transformer. One side has major arcing to create the hiss and the other side makes a small arc occasionally. One side is dead and the other is close. The TM-09175 transformer isn't available anywhere so I order the power supply inverter board from eBay. There are at least two versions because none for sale look like mine with the extra two transformers.

SEM2005 and AP4506GEH make quite a nice inverter driver duo.

SAMSUNG ELECTRO-MECHANICS
Halfbridge Inverter Driver IC (SEM2005)
FEATURES
- High Frequency Voltage Mode PWM Control
- Latched Mode Protection
- Totem Pole Output
- Built-in Low Frequency PWM Generator
- Built-in UVLO function
- Adjustable Shut-down time
- Easy Burst Dimming Control
- Variable Protection function

PIN DESCRIPTION

Pin No Symbol I/O Description
1 SDT I Shut Down Timing capacitor 03 07}
2 OVP I Over Voltage Detection Input
3 OLP I Open Lamp Detection Input
4 FB I Error Amplifiers Inverting Input
5 ERO O Error Amplifiers Output
6 SGND I/O Signal Ground
7 PGND I/O Power Ground
8 NOUT O N-type MOSFET Driver Output
9 POUT O P-type MOSFET Driver Output
10 VCC I Power Supply
11 EN I On/Off Control Er Ir
12 CT I/O Oscillator Timing Capacitor 03 07}
13 BCT I/O Burst Dimming® OSC Timing Cap. 03 07}
14 BDIM I Burst Dimming Input
15 RT I Oscillator Timing Resistor 0307}
16 Vref o 5V/10mA Reference Voltage Output

SEM2005 Rev 1.0

The new board is equipped with a TM-10160 transformer. The new board doesn't work and the transformer doesn't hiss. This board has heat damage. CN303 CN304 doesn't light the lamps. A little continuity testing shows the transformer is a single input two output transformer with no feedback windings. The non paired edge contacts on the input side are not connected. My meter shows >2KV on the transformer output that feed CN301 CN302 and 1600v on the transformer output that feed CN303 CN304. The transformer must be shorted across some windings internally on one side and doesn't produce enough voltage to start the lamps.

I return that board and order a second again with a TM-10160. This board works. These power boards look good with a low part count and great labeling. Unfortunately the transformers are junk.

For a TN Film this T240 has a very nice picture and a decent vertical view angle.

[PlainBill]

Re: Samsung SyncMaster T240 - 2 seconds then black problem

Sorry I didnt get back with my test results sooner. Fixing broken things has become a full time job here lately.

So as PlainBill suggested I swapped the connections and sure enough only one connector will make the CCFL's do the intermittent flash. I provided a modified pic to explain although its not a complicated concept.

So this means bad inverter am I right? I attached a pic of what I figured the inverter to be and its markings are:
TM-10160
TNE 0852 DT

If I am right an inverter replacement is the next course of action. Any hint on a good source for this? If it is hard to source out a direct placement maybe I can figure out its specs and find a suitable substitute.

Any advice welcome here folks!

DUCE

Not so fast!!! It COULD be a bad inverter transformer. Or it could be something else in the same area.

First step: On the second picture, there is a jumper that runs from just to the left of the transformer to up under the two poorly installed caps. There is a scorch mark at the lower end of this jumper. Resolder the jumper, then retest.

Second step (only if step 1 fails), I need a good quality picture of the bottom of the board; the inverter area is the primary interest.

PlainBill

[DUCE]

Re: Samsung SyncMaster T240 - 2 seconds then black problem

Sorry I didnt get back with my test results sooner. Fixing broken things has become a full time job here lately.

So as PlainBill suggested I swapped the connections and sure enough only one connector will make the CCFL's do the intermittent flash. I provided a modified pic to explain although its not a complicated concept.

So this means bad inverter am I right? I attached a pic of what I figured the inverter to be and its markings are:
TM-10160
TNE 0852 DT

If I am right an inverter replacement is the next course of action. Any hint on a good source for this? If it is hard to source out a direct placement maybe I can figure out its specs and find a suitable substitute.

Any advice welcome here folks!

DUCE

[PlainBill]

Re: Samsung SyncMaster T240 - 2 seconds then black problem

It was supposed to be a safety margin in case it ever encountered an inverter that was putting out far too much output voltage. A lot of them seem to drive the output voltage directly off a transformer, where a short on the input side could change the transformer ratio and jack the output voltage way up. Which brings up another question - would a typical DVM's AC measurement of the output voltage be accurate at the high frequencies used by the inverters?

I believe a shorted turn on the input would REDUCE the output voltage. While the turns ratio is increased, the shorted turn acts as a shunt and damps the magnetic field.


It depends on the meter. I would say it is pushing the voltage and frequency limits of many DMMs.

One of the ideas I had was to build a test load. Say a 90K ohm resistor in series with a 10K. The free end of the 90K goes to the inverter output, the free end of the 10K goes to the return line. That means the junction is at 70 VAC. Tie the anode of a schottky diode to the junction, the cathode to a .1uF cap, the other end of the cap is tied to the return line. The DC voltage across the cap will be 10% of the peak AC voltage. Or tweak the resistances a bit and the DC voltage will 10% of the average AC voltage.

It makes a quick way to compare inverter outputs.

PlainBill

[mathog]

Re: Samsung SyncMaster T240 - 2 seconds then black problem

I think the 20 watt rating is overkill. Do the math. The current of most CCFLs is in the 4-6 ma range. 600 volts, 100K ohm, 6 ma, 3.6 watts. Also, if I recall correctly, Wizard suggested 120 K ohm.

I'd think PPC120KW-3JCT-ND would do the job.

It was supposed to be a safety margin in case it ever encountered an inverter that was putting out far too much output voltage. A lot of them seem to drive the output voltage directly off a transformer, where a short on the input side could change the transformer ratio and jack the output voltage way up. Which brings up another question - would a typical DVM's AC measurement of the output voltage be accurate at the high frequencies used by the inverters?

[PlainBill]

Re: Samsung SyncMaster T240 - 2 seconds then black problem

The jumper is probably an innocent bystander. It looks thick enough to handle the expected currents without heating up enough to darken the board. Hard to tell from the pictures, but it looks like the transformer might have blown smoke, or maybe even an arc, out along the general direction of the core. Since there were apparently previous repairs, it might have been the previous transformer that occupied that position which did this.

It would help to have better pictures, or even better, the actual board to see, but I was thinking along the lines of a bad solder joint that started off with a high resistance and eventually went open.

PlainBill

[PlainBill]

Re: Samsung SyncMaster T240 - 2 seconds then black problem

For the other half of the equation, would one of these be OK for a dummy load? The 20W part shouldn't overheat even if the voltage is a bit too high, and they are less than $5 each.

http://search.digikey.com/scripts/Dk...e=HLWB-100K-ND

I keep trying to think of a really cheap and clever 100K load that could be put together from whatever one would have about the house. The only thing I have been able to come up with so far is a bucket of lightly salted water, in which is submerged a board with two nails through it, one test lead to each nail. The resistance could be adjusted to 100K by adding salt (titrating down from distilled water if the tap water conducts too much). The water could handle the heat produced, as long as the test wasn't run forever. But wiring hundreds of volts into an open container of conducting fluid just seems like an accident waiting to happen.

I think the 20 watt rating is overkill. Do the math. The current of most CCFLs is in the 4-6 ma range. 600 volts, 100K ohm, 6 ma, 3.6 watts. Also, if I recall correctly, Wizard suggested 120 K ohm.

I'd think PPC120KW-3JCT-ND would do the job.

PlainBill

[mathog]

Re: Samsung SyncMaster T240 - 2 seconds then black problem

The transformer is definitely a suspect, but it appears the dark spot coincides with the jumper!

The jumper is probably an innocent bystander. It looks thick enough to handle the expected currents without heating up enough to darken the board. Hard to tell from the pictures, but it looks like the transformer might have blown smoke, or maybe even an arc, out along the general direction of the core. Since there were apparently previous repairs, it might have been the previous transformer that occupied that position which did this.

[mathog]

Re: Samsung SyncMaster T240 - 2 seconds then black problem

The ideal would be to apply the mythical "cheap and effective CCFL testing circuit" directly to the lamps, and so take the inverter completely out of the equation. Unfortunately said circuit has not yet been spotted on these forums.

For the other half of the equation, would one of these be OK for a dummy load? The 20W part shouldn't overheat even if the voltage is a bit too high, and they are less than $5 each.

http://search.digikey.com/scripts/Dk...e=HLWB-100K-ND

I keep trying to think of a really cheap and clever 100K load that could be put together from whatever one would have about the house. The only thing I have been able to come up with so far is a bucket of lightly salted water, in which is submerged a board with two nails through it, one test lead to each nail. The resistance could be adjusted to 100K by adding salt (titrating down from distilled water if the tap water conducts too much). The water could handle the heat produced, as long as the test wasn't run forever. But wiring hundreds of volts into an open container of conducting fluid just seems like an accident waiting to happen.

[PlainBill]

Re: Samsung SyncMaster T240 - 2 seconds then black problem

I think that it was a fluke leaving half the inverter unplugged worked on the Sharp TV.

As for the brown spot, isn't that in the same position as the brown mark on one side of the transformer on the other side of the board? In your marked up picture of the other side it is just below the "'A; connector" label. At least, that's the way it looks to me, triangulating on the blue capacitors visible through the holes in the board. There doesn't seem to be anything active there other than the transformer, so as you said earlier in this thread somewhere, the transformer is suspect.

The transformer is definitely a suspect, but it appears the dark spot coincides with the jumper! A lot of these transformers are actually two transformers in a single package. I'd definitely try resoldering the jumper as the first step!

PlainBill

[mathog]

Re: Samsung SyncMaster T240 - 2 seconds then black problem

Upon closer examination, I think there is a pretty obvious explanation of why the upper CCFLs don't work. I've marked it 'Overheating, no obvious cause'. What do you think?

I think that it was a fluke leaving half the inverter unplugged worked on the Sharp TV.

As for the brown spot, isn't that in the same position as the brown mark on one side of the transformer on the other side of the board? In your marked up picture of the other side it is just below the "'A; connector" label. At least, that's the way it looks to me, triangulating on the blue capacitors visible through the holes in the board. There doesn't seem to be anything active there other than the transformer, so as you said earlier in this thread somewhere, the transformer is suspect.

[Wrog]

Re: Samsung SyncMaster T240 - 2 seconds then black problem

I did learn somthing useful today while attempting to swap the lamp output lines...the housing around the power plug is hot!!

Just in case it hasn't been said (or maybe this is reinforcement) -- only swap plugs with the power disconnected!

[DUCE]

Re: Samsung SyncMaster T240 - 2 seconds then black problem

I am sorry for contributing to any male pattern baldness PlainBill, being a senior member must be tough but I am glad you have taken the time on this because your comments are spot on.

I didn't know that an inverter was designed to shut down when the lamps are not drawing current.

I did learn somthing useful today while attempting to swap the lamp output lines...the housing around the power plug is hot!!



I will take another stab at it later after the buzzy feeling has stopped.

[PlainBill]

Re: Samsung SyncMaster T240 - 2 seconds then black problem

Are you sure of that? I used exactly this method on a Sharp TV and in that instance one side of the inverter would drive lamps when the other wasn't plugged in. As it turned out on that machine both the lamps were working but one half of the inverter was out due to a bad capacitor. Either that inverter was not designed to shut down if unloaded, or perhaps that bad capacitor was leaking enough current internally that both sides of the inverter appeared to be loaded, even though they were not. I did try switching the lamps on this board and it immediately shut down in both positions. Probably because leakage current + CCFL current was too high for the inverter.

The ideal would be to apply the mythical "cheap and effective CCFL testing circuit" directly to the lamps, and so take the inverter completely out of the equation. Unfortunately said circuit has not yet been spotted on these forums.

The OP's photos leave something to be desired, but the voltage and current sense components are clearly visible. I've marked them on the attached picture. Upon closer examination, I think there is a pretty obvious explanation of why the upper CCFLs don't work. I've marked it 'Overheating, no obvious cause'. What do you think?

Again, what remains to be established is if the upper CCFLs are also defective.

PlainBill

[mathog]

Re: Samsung SyncMaster T240 - 2 seconds then black problem

This is one of the things that makes me want to tear out what little non-gray hair remains. By design, an inverter will shut down as soon as it senses a lamp that is not drawing any current.

Are you sure of that? I used exactly this method on a Sharp TV and in that instance one side of the inverter would drive lamps when the other wasn't plugged in. As it turned out on that machine both the lamps were working but one half of the inverter was out due to a bad capacitor. Either that inverter was not designed to shut down if unloaded, or perhaps that bad capacitor was leaking enough current internally that both sides of the inverter appeared to be loaded, even though they were not. I did try switching the lamps on this board and it immediately shut down in both positions. Probably because leakage current + CCFL current was too high for the inverter.

The ideal would be to apply the mythical "cheap and effective CCFL testing circuit" directly to the lamps, and so take the inverter completely out of the equation. Unfortunately said circuit has not yet been spotted on these forums.

[PlainBill]

Re: Samsung SyncMaster T240 - 2 seconds then black problem

What happens if you leave the bad set unplugged, and the good set plugged in? Will it stay on then? If yes, how about with the other side of the PS driving the good lamps, and the bad lamps not plugged in? Most PS will run happily with no load, but they shut down really fast if they see something that looks like a short. So if you are lucky, what you will find is that both sides of the PS can drive the good lamps, and all you have is bad lamps. That said, even in that case you might want to change all of the lamps if you change any, since the good ones may be very nearly at their end of life.

This is one of the things that makes me want to tear out what little non-gray hair remains. By design, an inverter will shut down as soon as it senses a lamp that is not drawing any current.

You are right, he has to figure out if he has bad lamps or an inverter problem. You don't do this by leaving one set of lamps unplugged. And this is the reason I asked what happens if he switches lamps and inverter outputs. By his description, he seems to be saying if the connections are swapped NEITHER set of lamps will work. What does that mean?

PlainBill

[mathog]

Re: Samsung SyncMaster T240 - 2 seconds then black problem

Powering it up while having the backlights exposed showed that the power line from board to lights was not providing anything to the top set of lights. I tested this by having both lines connected. The bottom set did the classic pulse with 2 second delay but the top did nothing. I swapped the power lines over so the "good" line was feeding to the dead light area and nothing at all.

What happens if you leave the bad set unplugged, and the good set plugged in? Will it stay on then? If yes, how about with the other side of the PS driving the good lamps, and the bad lamps not plugged in? Most PS will run happily with no load, but they shut down really fast if they see something that looks like a short. So if you are lucky, what you will find is that both sides of the PS can drive the good lamps, and all you have is bad lamps. That said, even in that case you might want to change all of the lamps if you change any, since the good ones may be very nearly at their end of life.

[PlainBill]

Re: Samsung SyncMaster T240 - 2 seconds then black problem

I thought those caps looked a little suspect. There were other signs that the case had been cracked after factory like masking tape on the back plate to hold down wires. I hope that I have not purchased a total dud that I just cant fix but it seems that the last person was a tad careless so maybe they just missed something simple (I hope).

The 2 connections from power board to lights are shown here connected. Basically its just 2 cables that have 4 wires each and they connect to 4 CCFL lights in the screen back plate. Hope that pic and explanation explains what I meant better.

What bothers me and is the same as the OP is the burnt area around the caps and inverter area. My pics dont show it very well but I think both me and the OP have found a weak spot on this board. In my case the caps were obviously replaced but that may or may not have solved the original prob. The blinking on the lower section of lights indicates inverter (as I have learned lately). The scorching around the cap area makes me think that a cap failure could have aided in an inverter failure. Not sure if that is a plausible theory because of my low experience level but the burnt area and blinking CCFL plus the replaced caps seem to leave a bit of a bread trail.

I would like to test the CCFL's to see if they have failed and test my power lines connecting the CCFL's.

The specs for the replacement CCFL:
Strike Voltage 1600
Working Voltage 920
Watts 5.5

Strange voltages there (IMHO) maybe I missed somthing...
http://www.lcdparts.net/ccfldetail3D...ProductID=1600

I have no idea how to test the inverter yet. Thats one I will look into more but any hints or ideas are welcome.

I have scoured the net for a service manual but only found the older model 240T which is quite different so there is no correlation on part numbers and specs.

Basically it seems we are on our own with this monitor for reliable data to use. Maybe somthing will turn up in later searches. At least we have 2 people with similar issues posting here, hopefully the folks with more experience can help us deduce the problem.

OK, I'm trying to make this very simple. First of all, the discoloration isn't unusual. Yes, it MIGHT indicate a failed transformer. Or it might be the result of normal heating from the transformer drivers.

I have marked up your picture. Notice I have marked A and B cables, A and B connectors. You have stated that with the present connections the "top" CCFLs (presumably the 'A' set) does not light. What happens if you plug the 'A' cable into the 'B' connector and the 'B' cable into the 'A' connector, then apply power?

PlainBill