Monitor repair advice

[kbs1]

Hello, some time ago my iiyama X486S 19" LCD monitor backlight stopped working completely. I can see the picture when I shine a flashlight on it. I don't know if this is a right place to ask, but I feel there might be some very good help around here.

I did a lot of short circuit troubleshooting on the inverter board only to find out nothing I checked is shorted - that includes diodes, mosfets, other transistors, random resistors and so on.

Now I have bought / did previously have this equipment:
1) two cheap DMMs (ohms, amps, voltage, diodes, continutiy)
2) 20MHZ analog TESLA oscilloscope
3) a good LCR meter (to test capacitors)
4) soldering iron

The backlight does not come on at all, never. I have even measured voltage coming directly to CCFLs when the monitor is ON (even accepting the fact that this could ruin my cheapo multimeter / whole monitor) and there are 0 volts AC and 0 volts DC measuring any possible combination of the 4 inverter output pins.

That possibly ruled out bad CCFLs, which leaves the the fault to be on the inverter board itself.

The board is visually undamaged, unburnt, detailed images are in the attachment. The PSU part number is 715G1236-3-AS ver:A

I have just recently downloaded all required datasheets and service manuals for monitors using the same or similar power supply, which are also in attachment. Such a monitor is TOSHIBA LCD-AD174TW.

There are service instructions to troubleshoot the powerboard itself (section 8.2.3 POWER/INVERTER BOARD). At page 41 titled "2.) W / LED , No Backlight" there are steps to check if there is no backlight. I do not know what "W/LED" means but I proceeded.

Procedure is as follows:
1) Check C928(+) = 12V (NG Change F902)
- CHECK (F902 also OK)
2) Check ON/OFF signal (NG Check Main board)
- don't know how to do this
3) Check U201 pin5=6V (NG Change Q205 or Q207)
- on my board it's IC201, datasheet is in IC201.pdf, and pin5 VAROUT_L is only 4.80V, not 6. This is measured using my DMM and my oscilloscope.
4) Check the pin1 of U201 have saw tooth wave (NG Change U201/R217/R215)
- pin1 is ground according to datasheet, so this does not make a lot of sense to me. There are 0 volts, no ripple, no saw tooth wave. I measured R217 / R215 and they came out low, R217 has code 6192 (so 61900 ohms) but measured only 58200 ohms. R215 has code 431 but measured only 362 ohms. This result is the same using my two cheapo DMMs.

I don't know whether this is OK or not, whether my DMMs are faulty or the resistors are. I measured in circuit. Should I desolder the resistors to measure them properly? Is IC201 dead?

Any advice on how to troubleshoot this board would be greatly appreciated. Attachment was too big, so it is downloadable at http://adina.sk/MONITOR.zip

Thanks!

[Psi]

The CFL backlight is high voltage, i think they're like 300-600V. You maybe unable to measure this with a DMM, the frequency could be quite high.

What normally happens is a bad capacitor fries the driver.
Or the lid hinge traps a wire and breaks it after repeated open/shut cycles.

I'd check the input voltage to the inverter module, see if its actually getting any power.

[kbs1]

Where can i find the inverter module? I don't know exactly because the power supply and inverter are one board in this case. I have checked the CCFL wiring and it seems physically OK.

[Psi]

Most laptops have the inverter module inside the laptop lid right next to the LCD.
It's usually a separate module kept as close to the LCD as possible to avoid running high voltages around the laptop.

[mariush]

The lamps are actually 700v or higher. Sometimes they're even coupled together and the inverter outputs 1400v so you might break your multimeters.

NG check main signal may refer to the power good wire that comes from the board to the power supply and enables the 12v or whatever gets sent to the inverter section. The inverter section takes 12v-24v (varies from monitor to monitor design) and generates 700-1400v @ 1-2mA or whereabouts. In your case looking at the pictures, it's probably 12v because the capacitors are rated for 16v.

First thing you should do is to double check the soldering around the inverters, there's quite a lot of solders spots that look horrible, like they're dry joints. add some leaded solder to those pads or remove the lead free solder and resolder them with leaded solder.
Measure the resistance of the coils of the inverters to make sure they're still good and not open.

Tests the resistor looking parts near the lamp connectors, one is right near the connectors at the corner, one to the right of an inverter. They may be fuses to protect the inverters in case the lamps go bad and pull more current than what the inverters can supply. They don't look bad from here but who knows....

Since you can see image on screen when lighting the screen, the video board is most likely OK - the problem is inverter related or the power board no longer receives the signal from the video board. Probe the wires in the connector with wires that connect the video board to the power board...

I'm not really experienced in fixing monitors more than the obvious problems so that's about as much as I can tell you but I suggest you create an account on the Badcaps.net forum and repost these pictures there along with your questions. There's lots of people with experience in fixing monitors there and they'll be glad to give you tips.


PS.  Search on eBay if you want for the complete power board using the codes on it - in your case the code on the right of the cable ... that one on the left of Ver: A at the end
But before ordering a new board, you should also try to check the lamps - if you can unscrew the metal lid and see if the lamps are still good - if they're quite dark at the ends they're probably gone bad.

[kbs1]

Hello, thank you all for your advice. I did measure the inverter transformers and they are feeded 0 volts each. This was checked with lamps connected because I could measure the pins from the top side of the board.

There are four mosfets controlling the input to each inverter transformer, two in one 8-pin package, namely Q204, Q208 for PT201 and Q209, Q210 for PT202. Datasheet for those mosfets is in Q210.pdf in atachment posted earlier.

Pin 1 [Source (NMOS)] is always connected to ground.
Pin 2 [Gate (NMOS)] is wired as follows:
- Q204 gate connected with Q209 gate
- Q208 gate connected with Q210 gate
Pin 3 [Source (PMOS)] is always connected to CN102 (connector to monitor logic board) pin 12. I assume that could be the on/off signal. It always reads 13 volts even in standby.
Pin 4 [Gate (PMOS)] iterconnected exactly as Pin2 (Q204 and Q209, Q208 and Q210)

Drains (pins 5-8) are always connected to inverter transistors input.

So if pin 1 is ground, pins 2 and 4 are interconnected and pin 3 has 13V, why is there no output on the drain side? Is one of the mosfets faulty?

I double checked the soldering and resoldered bad looking dry joints with new leaded solder. Results are still the same.

[T4P]

W/LED is the designation for the model with the LED backlight and not your model since you have a CCFL backlight

[tec5c]

This is somewhat of a common problem with certain LCD TVs/Monitors, and it's good that you have already had a look at the display with a flash light as this would have been the first step if you're getting power and sound with no picture. Are you able to take some photos of inside the monitor? This will make it easier for us to all help you fault find and repair the PCB. With a picture I can show you where to probe with your DMM on the power supply board to check your voltage rails.

My apologies, I did not read the first post entirely. The file is currently downloading....

[tec5c]

Okay I just had a look through the photos. The first things I'd check on the SMPS/Inverter board would be that fusible resistor which is visible in image DSC01019.jpg which is labelled as FB901. Set your DMM to the ohms/continuity range and then with the monitor switched off from mains, place a lead to each leg of the component. If it is good then you will get some resistance/continuity feedback on your DMM. If you get 0 Ohms or no continuity beep then it is faulty.

After that I would look at the 4 MOSFETs. You mentioned that you had 0v on the drain of one of the FETs, does this happen to be one of the two near the fusible resistor?
A simple in circuit diode test can show a short across a FET though always remove the component from circuit for a more accurate test.

[kbs1]

Hello, thank you for your input. I did measure FB901 an I get a continuity beep & 0 ohms, so that it's shorted. On my meter the continuity / diode tester displays in the 2000 ohm range, so I did try to measure FB901 on 200 ohm range. I get 0.5ohm this way, but that's the resistance of my test leads, so FB901 is 0 ohms. Is it faulty?

I did also measure all SMD resistors in the "Q903" area (the mosfet on the heatsink) [image DSC01026.JPG lowe left corner circuitry), and they were all spot on (including 1 MOhm R906 & R907).

All regular resistors were also spot on (if I subtract resistance of my test leads, it turns out 0.39ohm R917 is also spot on).

I also measured all diodes in the area (without desoldering anything) and they show 400-800 ohms in one direction and out of range (closed) in other direction.

Should I check the drain voltage of Q903 going to main transformer T901?

> After that I would look at the 4 MOSFETs. You mentioned that you had 0v on the drain of one of the FETs, does this happen to be one of the two near the fusible resistor?
A simple in circuit diode test can show a short across a FET though always remove the component from circuit for a more accurate test.
- I was testing direct input to the inverter thansformers at that point, which is drain output of 4x 8-pin mosfets right in the center of the board (DSC01025.JPG in center of the board).

Where should I place my negative multimeter lead when measuring voltages? For this PCB region I was thinking lower side of the board near the white sticker (DSC01026.JPG bottom - ground point)

Voltage accross main rubycon capacitor is 303volts DC.

[T4P]

Always keep your negative (NOT GROUND! Matters in analog's ...) at the ground point, so you don't have to wave your ground probe everywhere

[kbs1]

Somebody told me that every circuit has a hot side and a cold side and that those are usually separated by a white line (also on this board). So does it not matter to which side's negative point I place my negative multimeter lead when measuring various circuitry? Also I tried googling but did not find any answers on what those sides are. I presume it's the side before and after main transformer?

[kbs1]

I'll try to desolder all standard diodes, I'm not too good at soldering SMD stuff. Should I start around "Q903" area? Also, do you think I should measure output of Q903 somehow? I don't want to blow anything up.

[kbs1]

I did try to measure IC901 PWM output to Q903 mosfet using my oscilloscope. I never understand where to place the oscilloscope negative lead so I figured I'll place it to IC901 pin1 while probing pin8. This of course blew a hole in the PCB at the touch point which I had to patch. Looks horrible but continuity is good. Q903 did not blow up but main fuse did let go (fortunately).

This had happened in the past, while probing the board with my DMM at that time, I did accidentaly connect IC901 pins 7 and 8 and blew up the mosfet Q903 and main fuse.

From that time (~ 2 months ago), I replaced Q903 with equivalent and F901 (main fuse) with an easily interchangable one and the monitor did once again "work" like before, so no backlight but good picture.

All current datasheets are in archive at http://adina.sk/MONITOR.zip

This time, I do not have a replacement fuse and will buy one tomorrow. I hope nothing blew this time. Mosfet is not shorted.

Just to update the thread I am posting images of the current looks of the board with patch applied, Q903 changed to equivalent one and main fuse changed to easily changable one.

Well so this once again proves I do not really know much about electronics, can someone please tell me where to place my oscilloscope probe ground when measuring PWM output of IC901? Assuming the monitor still works, we'll find out tomorrow with new fuse.

New board images are still in archive at http://adina.sk/MONITOR.zip, I've updated it. I'll still desolder the diodes and check them, I just want to first look for an obvious cause, IC901 is PWM controller of Q903 so I assume it controls *brightness* of the backlight, and the monitor broke after ~5 days of usage after I lowered it's brightness for the first time in it's history. So I was thinking IC901 might have gone bad.

[kbs1]

Well guys good news the monitor still works! Nothing blew up this time except the fuse and the PCB patch is also working correctly. The hunt continues! There is still no backlight but there is picture.

My oscilloscope is a two channel analog type and there is an option to show both channels labeled A±B, that seems to merge the channels together, so I'll use that. I did not try it but it looks it works.

>> I never really use a ground when just sniffing for signals. Speeds things up and gets rid of the annoying ground cable in the way.

Yesterday when I was measuring pin8 of IC901 I did see something on the oscilloscope, but thought that was not a correct display since I did not connect the ground. Do I need to connect the ground while probing around in single channel mode? If not, what is the ground cable for?

I'd like to set up my scope in a way that there is a minimal chance of blowing things up.

>> (1) Check on your DMM for the forward voltage drop and compare it with the datasheet to make sure it's about the same.
Is this measured in ohms? My DMM shows values around 450-800 ohms when measuring diodes in circuit in forward direction. Does this value indicate the 0.45 - 0.8v forward voltage drop on the diode?

>> (2) Check in continuity mode 'both' ways around. One way should beep and the other way should not. Usually you'll find a diode that beeps both ways, so it's a definite dud and need replacing.
My meter has a combined position for continuity tester and diode tester. When measuring in circuit, the meter always shows 400-850 (ohms) for diodes in one direction and "1" meaning no reading / over range for other direction. I'll try to desolder one random diode and see if the reading changes on the DMM. I'm not sure if it tests diodes correctly. Thanks a lot for your help so far.

[kbs1]

Thank you for all your good help. I just tried to measure a brand new 6 amp diode using my DMM. It has combined position for continuity and diode tester. This position has 1999 counts and displays ohms, beeps when ohms < 50. It had shown 489 ohms on the diode in permissible mode (forward way), and "1" or out of range in opposite direction. It did not work at all way when set to 2000 ohms, because I guess the voltage on the ohms range is not enough to turn on a diode. So the diode / continuity tester works, just not very well. I'll see if I can find any shorted diodes after desoldering them.

Regarding the scope I'll use it in the way you described. I'll touch the ground on opposite side of the white line while measuring random circuitry. I'll update the thread when I (hopefully) discover some flaw in the circuit.

[kbs1]

I have a few questions about the board, please first see the attachment.

I've found even better manual for very similar monitor (using presumably the same power supply) and have found diagram for CN102 (connector to logic board). I have measured it and it is OK, I have also maxed out brightness to 100% in software while shining a flashlight on monitor.

Attachments:
1) clear picture of the back side of the board with a couple of drawings
2) similar monitor service manual (containing diagram of CN102)

Questions:
1) T901 (main transformer) is shorted as drawn in the picture. Is this ok? Especially the other side which is completely shorted alltogether. I was tracing the PCB but could not find any definitive short between the traces. I'm thinking of desoldering T901 to check it, can it be shorted like that? I can't find any datasheet for it (80ll17t-2-t)

... and that's really it for the moment. Next I'm going to check on/off signal path if T901 is OK. I've also desoldered two standard zener diodes and they checked OK. Soldered back in exactly the same direction.

Attached is another as clear as I can make picture of the board. Assumption: If T901 is bad, the main board would have no power, then there would be no picture. So T901 can't be bad. Correct?

[kbs1]

Thanks for your help and a nice picture. I have checked all the electrolytic caps on the board. They all tested good.

The filter ones check around 940-975 uF, rated for 1000 uF. Small ones near the bottom are 22uF and 10uF and checked spot on. My LC meter is LC100-A (http://www.ebay.com/itm/x/150805576248)

All the diodes also check out good, the 12V schottky diode had 408ohms in the permissible direction from both pins to the center one, almost infinity ohms from other direction. None of the pins were shorted. The same applies to the 5V diode, only it had 104ohms in the permissible direction.

12V schottky diode has a tiny mark above the middle pin, I tried to photograph it, but it just seems to be some kind of production mark and not a deep hole (see picture). Maybe it denotes the output pin.

Q901 - it was just a bad lighting on the pic, also it is not shorted when measured in circuit.

High voltage outputs have 0 volts when measured in any combination of the 4 pins for one output pair. Both inverter transformers also have 0 volts AC on the input, this is with the lamps connected. I measured this by placing one lead of my multimeter to pin1 of the transformer and the other on pin6.

Tracing back from inverter transformers on the board the first thing that stands in the way are the 8 mosfets in 4 packages that probably switch the transformers on and off. How do I check them? Their sources are connected all together and their gates are connected in pairs. I have drawn their interconnection for better view in a picture ( http://adina.sk/monitor/mosfets.png )

Attached is also datasheet for the mosfets. There are 0 volts on the drain side of the mosfets, am I measuring the inverter transformers inputs correctly?

Last attachment is newest top view image with marked components that were desoldered and tested good.

[kbs1]

Hello, the fuse is not blown. It shorts to 0 ohms (without desoldering from circuit). It is connected do pin 3 (S2) of all the 4 mosfet packages and also through a zener diode to pin 4 (G2). I've drawn the schematic in a picture (http://adina.sk/monitor/mosfets_and_f902.png). Should I check pin 3 (S2) voltages at any of the 4 mosfet packages with negative multimeter ground tapped to ground on the top side of the board? (near the screw hole)

[kbs1]

I just measured the circuit and there is 12.95V on both sides of the fuse and this power is also at "Source 2" pin of all the mosfets. Test was done with backlight disconnected so that I can measure the fuse and the mosfets. Mosfet #2 is PMOS according to the documentation (in attachment). I also measured gate-source voltage of Q210 (top one; pins 3 and 4) and it is -1.55V when backlight should be ON (0 volts when in standby).

Next I measured
Q209 S2G2 voltage = -1.53V with backlight on
Q208 S2G2 voltage = -1.54V with backlight on
Q204 S2G2 voltage = -1.53V with backlight on

So according to the datasheet, this P-mosfet should be on? Then why is there 0 volts on the drain?

[kbs1]

I just realised there are 13.18 volts DC on all 6 pins of all the 2 inverter transformers when the backlight should be on (with lamps disconnected). Previously I was measuring by probing pin1 and pin6 of the transformers with my DMM, hence 0 volts difference. I'm sorry if that measurement was misleading.

[kbs1]

I've just measured S1G1 voltage of all the N-channel mosfets (pins 1 and 2) and it is always 0 volts DC.

Source of the N-channel mosfets (pin1 "Source 1") is always connected directly to ground. Pin 2 ("Gate 1") of Q204 and Q209 is connected through a 22ohm (checked spot on) resistor directly to pin20 of IC201 (labeled GND in the datasheet - see attachment).

Pin 2 ("Gate 1") of Q208 and Q210 is also connected to IC201 through a 22ohm resistor (spot on) but to pin11 (STBY). I've drawn a picture of the connection (http://adina.sk/monitor/Gate_1_and_IC201.png).

Attached are datasheets for IC201 and Q2xx mosfets. Maybe there is a problem with IC201 or it's supporting circuitry, but how do I check it out?

[kbs1]

Sorry I had a wrong datasheet for IC201. Here is the correct one.

[kbs1]

IC201 measurements when BL should be on:

pin 3 (ENA - Enable input) = 4.83V
pin 4 (SST) = 4.63V
pin 5 (VDDA - Supply voltage) = 4.83V
pin 6 (GNDA  - Signal ground) -> OK (continuity)
pin 7 (REF - Reference voltage output) = 3.30V
pin 14 (DIM) = 2.03V; 0.60V in standby (CCFL should be off)

Everything looks in order here. Lamps were disconnected during every test.

[kbs1]

IC201

pin 1  = 0V
pin 2 = 0V

Althrough I don't know if I can measure these using an ordinary DMM and with the lamps disconnected. End of spam for tonight, I'll wait for some advice now.