Author Topic: monitor fail (Read 8262 times)

christos · « **on:** July 01, 2014, 03:18:25 pm »

hello..i have lg monitor that wont start...its dead..no lights ..nothing..the power supply works but the smaller board gets very hot in the voltage regulator (the kia 1117f) and the other one ...can that be the problem?

christos · « **Reply #1 on:** July 01, 2014, 03:26:18 pm »

http://wenku.baidu.com/view/4eca61956bec0975f465e28c

i think its the schematics

mariush · « **Reply #2 on:** July 01, 2014, 03:40:02 pm »

Here's the service manual (pdf, 2.5 MB) : http://savedonthe.net/download/565/lg_l204wt-s-bfq.html

Get a multimeter and check the voltages.

On the large capacitors you should have about AC x 1.414 ... if you're on US, expect about 170v , in Europe you should measure about 350v. If you don't have voltage there, the fuse is probably blown, or one of the main transistors is dead.

On that connector (CM201) that goes to the other board (the one with the vga connector), measure the voltages. Near the connector or on the other side of the big board, it should say what voltages are there. There should be at least 5v, there could be also 12v and maybe 3.3v.
The two chips on the second picture you posted are linear regulators. They take in some voltage (5v or 3.3v) and output a smaller voltage (for example 3.3v, 2.5v or 1.8v). The first kia 1117 may be a 1.8v output regulator (due to the code starting with 18 on it) but the input voltage could be 3.3v or 5v, i have no way of knowing from pictures.
If the input is 5v and output is 1.8v it could explain the heat. It's also possible something is shorted on the output of that linear regulator (or the regulator itself is shorted) making it output up to its maximum rating (1.8v at 0.8-1A) therefore producing (5v-1.8v)x1A or (3.3v - 1.8v)x1A watts as heat, more than the board was meant to use.

Check the bottoms of those capacitors, if they rubber bottom looks swollen, the capacitor is bad and must be replaced.

christos · « **Reply #3 on:** July 01, 2014, 04:20:02 pm »

i measured it and im getting at the input 3.3v and out 1v from the kia117 regulator and form the smaller one 4.75v input and output 0.8v...problem?..also..the resistor in the picture,gets hot too..see the picture for the pins .

the smaller one gets VERY hot..cant touch it

mariush · « **Reply #4 on:** July 01, 2014, 05:57:56 pm »

So you definitely have a problem there.

EH16A code is used by this 1117 regulator, and the notation is reserved for the 3.3v output version : http://www.taitroncomponents.com/catalog/Datasheet/LD1117A12.pdf

From left to right, you have Ground/Adjust , Output and Input so on the center pin (and I think the tab as well) you should have 3.3v). You have 4.75v at the input which is a bit low but if it's stable it's fine so you either have a dead regulator, or there's something using so much power at 3.3v that the regulator is going into some sort of current limit mode and reduces the output voltage to the minimum to keep itself from blowing up.
You could measure the resistance between the ground pin and the output pin, to see if there's something shorted. Low resistance or continuity will tell you the regulator is dead or there's something shorted after the regulator.

The regulator on the left in DPAK package is made by KEC semiconductor, you can get the datasheet here: http://doc.chipfind.ru/kec/kia1117f18.htm
It's still a 1117 so the pinout is the same (from left to right) you have ground/adjust , output (and tab) , input ... so you probably have 5v on the input and some output voltage. It's not normal to have 1v on the output, since it's a fixed regulator (you'd see two resistors near the ground/adj pin otherwise), the only possible values would be 1.5v, 1.8v - the regulator has a 1.1v voltage drop so since you have 3.3v at the input, the maximum possible would be 2.2v which is lower than 2.5v so only 1.5v and 1.8v output is possible. Since the first two digits below the kia1117 text say 18, I'm leaning on calling it a 1.8v regulator.
Again, you may see only 1v because the regulator is in current limiting mode (maybe something pulls a lot of energy from the output), check resistance between output and ground pin. alternatively, desolder it and put 3v or more between input and ground and measure the output voltage, if it shows 1.8v or 1.5v it's good.

The big resistor on the left (R602 or R682) seems like it's between the inputs of those regulators, you can look at the circuit traces and you can see they go to the resistor. I think it's either acting as a fuse or it's simply designed to drop a volt or so so that the regulator won't have to dissipate so much power.

rob77 · « **Reply #5 on:** July 01, 2014, 06:00:01 pm »

something is shorting out the regulators then - but i'm afraid that the source of the short is the chip on the board.

do you know how that monitor died ?

christos · « **Reply #6 on:** July 01, 2014, 06:51:31 pm »

they just gave me the monitor...they didnt tell me how it died.., lets say that i change the regulators..will it power up?
are these regulators the power source for the whole board?

rob77 · « **Reply #7 on:** July 01, 2014, 07:10:30 pm »

yes they are powering the board. but most probably the chip behind them is broken, not the regulators only. both rails (1.8V and 3.3V) seems to be overloaded according to your measurement - so i'm afraid the big chip is dead.

christos · « **Reply #8 on:** July 01, 2014, 07:21:31 pm »

shoud i replace these regulators? and..the big chip gets warm..not like the regulators

mariush · « **Reply #9 on:** July 01, 2014, 07:41:47 pm »

Dude, we've told you already.

Desolder the regulators and TEST them. You know the ground and input pin, you know the output pin. Get 5v from somewhere ( your computer's power supply for example) put +5v on input, put ground on ground, measure the output voltage.
If they work, you don't have to replace them, you can reuse them.

If they don't work, there's no guarantee replacing them will solve the problem but it's a start. In this case, the problem that caused the regulators to fail may still exist.

Desolder the regulators and measure resistance between output and ground tab, where the ground pin used to be. If you have a short, you know placing new regulator there will not solve the problem.

Follow the output trace on both regulators and see what components are there connecting to it : one of those components may be faulty and cause a short. Even those two electrolytic capacitors may be shorted... I told you already to TEST it... is it so hard to desolder one leg of those capacitors and check them to see if they're shorted? You have nothing to lose right now, it's dead anyway.

If you don't find anything obvious, it may just be that big chip in the center of the board that's dead, faulty. Carefully remove the sticker and see what it says on it.
The service manual says it's a Genesis GM5726-LF but I don't think your board has the same chip, if it's the same then you can buy one on ebay: http://www.ebay.com/itm/GM5726-LF-Manu-GENESIS-Encapsulation-QFP-/380549587448?pt=LH_DefaultDomain_0&hash=item589a85e1f8
But I doubt you have the experience and tools too solder something with legs so tiny, so it's probably better to just leave it alone and think of buying/getting another display board or just get some other faulty monitor from eBay that uses the same connector

w2aew · « **Reply #10 on:** July 01, 2014, 07:43:40 pm »

One of the most common failure modes is bad or shorted capacitors, or possibly a shorted MOSFET in a switching supply. These can place a heavy load on the regulators driving them, making them run hot. So, a "hot" regulator usually indicates that something downstream is dragging the output down, causing a large current draw - thus a lot of dissipation in the regulator = heat. My first guess would be that the regulator is OK, but something after it is bad. Remember, a perfectly shorted component will be "cool" to the touch because it's not dissipating any power, so you can't always use "heat" as an indicator for bad parts.

christos · « **Reply #11 on:** July 02, 2014, 02:06:32 pm »

well..first i tested the c604 and found that its been shorted..then i folowed the traces and found that c604,c607,c610,c516 and some other caps are shorted..i think its the big chip..if you see the schematic of the scaler,,the problem starts from pin 47 and goes on..pin 25 goes to one cap (c512)and pin 2 to (c516)[on the board,these two caps are c512 and c513] that may be shorted and connects the pin 47 to ground..follow the schematic and tell me if there is hope for this monitor

what do you think?

w2aew · « **Reply #12 on:** July 02, 2014, 02:17:07 pm »

How did you test the caps? Did you lift one leg of each and test them individually?

mariush · « **Reply #13 on:** July 02, 2014, 02:42:14 pm »

Christos, if you have several components that are connected to two wires (or traces) and one of them is shorted, no matter where you test you will get short.

----A1----B1----C1----D1----

----A2----B2----C2----D2----

For example, let's say there's a capacitor (or any other component that's not normally short) between A1 and A2, another capacitor between B1 and B2 and so on.

If there's a short between C1 and C2 and you put the multimeter probes on A1 and A2 your multimeter will say short even though there's no short directly between A1 and A2, because the signal from the multimeter will go A1 to B1 to C1 then pass on to C2 and move to B2 and finally reach A2 ... so your multimeter will say there is a short between A1 and A2 even though there is no direct short.

If you want to truly test one component, you have to pull out at least one of its legs so that the signal from your multimeter won't go through the traces on the pcb through another component that's shorted.

Now if you have a very good multimeter (for example on with 40000 counts or more) that can show very small differences in resistance, you could actually determine where the short is because the actual resistance changes with the distance the electric signal has to go through.
For example, if there's short between C1 and C2, when you put probes on C1 and C2 you would get 0 ohms.
But, if you put the probes on B1 and B2, now the electric signal has to go from B1 to C1, then from C1 to C2, then from C2 to B2, so there's a longer distance. Your multimeter may now show 0.0001 ohms.
If you put probes on A1 and A2, the distance is even longer, so the multimeter may show 0.0003 ohms.
If you see resistance increasing, you know you're moving away from a short.
But again, you need a very good multimeter for that, a cheap 5-10$ (well, often not even <50$ ones) will not be able to help you with something like this.

christos · « **Reply #14 on:** July 02, 2014, 05:20:53 pm »

the caps are fine..so ,,whats the problem? shoud i buy the big chip and replace it?or just give up?

or is there any way to power it up?like buy a new board?

mariush · « **Reply #15 on:** July 02, 2014, 06:18:47 pm »

If you had problems with the above (not knowing multiple components in parallel would make it hard to determine shorts) and considering your previous post history, I would say you're don't have enough experience to be able to desolder that big chip and solder a new one there. The pins of that chip are simply very small and ideally you'd have to use a hot air gun and/or a soldering iron with a fine tip.
A new chip is also kind of expensive, you will pay about 10$ with shipping included just for that chip. Not worth it.

My advice would be to get a replacement display board. Doesn't have to be one from exactly the same monitor model, but the board must come from a monitor having the same resolution and it needs to have the same connector for the cable going to the flat panel.
The connector to the power supply doesn't really matter - most monitors use the same 3.3v, 5v, 12v, backlight adjust, backlight on/off so you can rearrange those wires if needed.

So for example, you could go to some computer service/repair shop or some stores that sell used/refurbished monitors and see maybe they have some monitor that they received with broken panel or bad fluorescent lights (which sometimes makes it not worth fixing it) and they may be willing to give (or sell) you just the display board from that monitor.
Or just go to a repair shop and ask them to replace the chip for you with a compatible one they may have from faulty monitors that went through their hands, they'll have the proper tools to solder that chip.

christos · « **Reply #16 on:** July 02, 2014, 06:37:29 pm »

ok thx for the info..ill try to find a simillar monitor

thx

Shock · « **Reply #17 on:** July 02, 2014, 07:05:16 pm »

I'll take a stab, I may be reading the schematic incorrectly but I think you are discounting IC503 and associated circuit and several other capacitors connecting to IC501.

If you can't isolate the short by low ohms or when powered by heat, freeze spray, or a current probe, then with that regulator output isolated and the circuit depowered keep components off the 3.3v rail until you find it doing continuity testing. You can also cheat sometimes by dividing the circuit in half and starting from a new test point and repair it later.

Then start lifting pins on the IC503 then IC501. After that you should have found if an IC is involved then start troubleshooting by testing and isolating the input/output and comparing the current draw to see if it's something else.

I copy the schematic, paste into MSpaint and mark the circuit I'm looking at. Similar to printing the schematic and using a highlighter.

I'm happy to be criticized on this, I'm no expert.

christos · « **Reply #18 on:** July 22, 2014, 04:30:28 pm »

Hello...i found a driver board but the Resolution is 1600×900 and mine is 1680x1050??
does it work?its for 20'' widescreen monitor like mine??

http://www.ebay.com/itm/LVDS-NTA93A-Universal-LCD-Display-Monitor-Driver-Board-For-20-Wide-Screen-/190753636734?pt=US_Laptop_Screens_LCD_Panels&hash=item2c69cd817e

http://www.ebay.com/sch/i.html?_trksid=p2047675.m570.l1313.TR0.TRC0.H1.X20+widescreen+driver+board&_nkw=20+widescreen+driver+board&_sacat=0&_from=R40


EEVblog Main Site	EEVblog on Youtube	EEVblog on Twitter	EEVblog on Facebook	EEVblog on Odysee

Author Topic: monitor fail (Read 8262 times)

Share me