LCD monitor repair - identifying rattling component?

[skyhawk]

So one of my beloved Samsung 204Ts quit a few weeks ago. Failure mode is the power light comes on but nobody's home, and the front panel controls do absolutely nothing. The only thing that changes the behaviour of the monitor is the physical power switch on the back.

Pulled it apart, replaced all the bulged caps on the power supply board, then realized something was rattling when I was putting it back together. Pulled the signal board out, and the damn thing rattles like a broken incandescent light bulb!

This is a high-density multilayer SMD board - Pretty much completely beyond my ability to do much more than look at it. For curiosity's sake though I would love to identify which component decided to be in multiple pieces inside. Does anyone have any tips for how to narrow down which component is rattling?

If it helps, it's this guy right here. I think the rattle is coming from the rightmost third of the board.

[Paul Moir]

Looking at it, I think it's got to be either something in one of the crystal cans or perhaps that inductor(?) on the far right of the board below that 8 pin SOICs.
Nothing else really has an "inside" with air in it.  Your description makes me think "something in a can" rather than a loose core in the inductor.

[wblock]

Is it made to pivot?  A lot of the pivoting monitors have a gravity switch to determine the orientation.  There is a little metal ball in those that rattles.

[skyhawk]

Yes, it is a pivoting monitor - one of the reasons I like it so much. I've been wondering what that thing is below the 8-pin SOIC pointed out by Paul, it very well could be the gravity switch you describe.

[Paul Moir]

That would make sense.  I thought I was seeing L0**  (* = some number) in it's silkscreen designation, but looking again it might be S*01?  (L is typically an inductor, S or SW is typically a switch).

[wblock]

It's probably working as designed, then.  Incidentally, it's not a bad idea to replace all the caps on the power board except the big one.  They don't always bulge when they fail, and when some fail, they put more stress on the remaining ones.  So you might find yourself opening it up again soon.  Don't forget the small startup cap for the controller chip.

[skyhawk]

Silkscreen designation on that weird thing is S301.

I've already replaced all 5 not-enormous electrolytics on the power board - I don't remember seeing a smaller cap for the control chip? I'll pull it apart in a few minutes and look at it again.

[skyhawk]

And my memory is crap. There's a little 50v 47uF all off by it's lonesone. Doesn't look distressed though.

This power supply does both the backlight and processor board power. I really need to find a way to get it powered up and see what the 13v rail is doing.

[wblock]

And my memory is crap. There's a little 50v 47uF all off by it's lonesone. Doesn't look distressed though.

This power supply does both the backlight and processor board power. I really need to find a way to get it powered up and see what the 13v rail is doing.

A plumber would say "That'll be fifty bucks." 

The smaller caps don't have vents to weaken the top, so it is rare to see them bulge.  I have seen general-purpose small-diameter caps bulge on processor boards in Gateway monitors, though.  It might be fine, could be checked with an ESR meter.

[skyhawk]

So I found somewhere to stick my meter, the supply is giving 13.1v to the processor board. Unfortunately the processor board remains my main suspect here.

That 50v 47uF cap actually does have a little vent scored in the top of it. I don't have an ESR meter to prove it with. If I had a suitable replacement on hand I'd replace it just for the heck of it, but I don't.

[amyk]

The monitors with 13V power supplies (for backlight) and 5V will run fine off a ATX PSU from a PC.

I've modded mine with molex connectors and plugged them into the PC so they turn on and off at the same time.

[skyhawk]

Not this one. Power supply has two outputs - 13v to the control board, and 750v to the backlights...

[amyk]

The inverters are usually part of the power board but are driven off the 13V rail, so just connect 12V there (with the mains unplugged) and you should be able to test their functioning.

[tautech]

So I found somewhere to stick my meter, the supply is giving 13.1v to the processor board. Unfortunately the processor board remains my main suspect here.

That 50v 47uF cap actually does have a little vent scored in the top of it. I don't have an ESR meter to prove it with. If I had a suitable replacement on hand I'd replace it just for the heck of it, but I don't.

You probably won't need an ESR meter for these, I've seen them go low uF values, less than half their nominal values so easily checked with any DMM with a capacitance range.

Have you done the normal "failed backlight" test with a bright torch shone at the screen to see IF there is an un-backlit image?

[tec5c]

My advice would be to pinpoint the location of the rattling before continuing with the troubleshooting/repair.

From the brief description you gave of the rattle, it is made to sound like one of the backlight bulbs is broken, thus if this is the case fixing any abnormal power rails is moot.

Is the rattle noise only apparent when the monitor is (lightly) shaken or is it audible when you try to power the monitor up?

[kjn4685]

Hi there
I was reading your post and I mite have a solution to your problem. This my sound crazy, but I had a heat sink came loss. I hope that will help you.
Lou