LCD TV Reliability

Status
Not open for further replies.
In this set, there are four rows consisting of eight LEDs. Each is driven by a differential pair (+-1, +-2, etc.)where the driving current is closely monitored by the LED driver ICs. If it detects a short or open, it will attempt to overcome it, raising or lowering the voltage as necessary. IOW, power is applied dynamically. I think in the above example, it raised the voltage so high, it damaged the trace at the far end by arcing. Odd it didn't occurr at an LED as all LEDs work.

When I turned the PS board over, I found each diff. pair isn't driven separately. MOF, there may be only a single driver ckt. I have yet to find a schematic to verify this however.
 
Originally Posted By: RTexasF
I've yet to have an LCD TV conk out but both of my high dollar plasma sets did.

The circuitry in the plasma is more demanding on the power supply which is where most failures occur.
 
Originally Posted By: Lubener
Originally Posted By: RTexasF
I've yet to have an LCD TV conk out but both of my high dollar plasma sets did.

The circuitry in the plasma is more demanding on the power supply which is where most failures occur.


Oh my yes, my Pani Plasma used over 750 watts of power!

The same size Samsung LED I replaced it with uses 175 watts!

With that said, the I never had any issue with the Pani, sold it to a neighbor.
 
Last edited:
Well this set may be Toes Up. I fixed the burned LED strip, powered up the set with a VARIAC and began meas. PS voltages, then I heard sparking again. Even with reduced VAC. Not good. This time a LED began to burn! I suspect the LED driver is defective. Low-V section to main board appears OK. Still unable to find a SM or schematic. Lots of teeny-tiny SMD's on the PSB's back-side, requiring a microscope to ID (which I don't have).

Oh well......
 
From my experience supporting family and friends, the LCDs have been pretty dependable - for the exception of a Toshiba with a dead CCFL inverter. Only problem I saw so far is with a Sharp Aquos from 2006-2007 showing ghosting and color shift. I do notice the cheaper TVs(Funai-made models mostly and earlier Vizios) have panel issues later in life with color quality.

Samsung, LG and non-Hisense Sharp are solid, so were Pioneer and Panasonic for plasmas. My parents still have their 42" Samsung 720p plasma in a room they don't use. I'd like to buy them a mid-level 4K 50" LCD and move the old-school Sammy to my mom's room with a Comcast HD uDTA. She doesn't want to pay Comcast $10 for an additional X3 box and I don't think she can figure out a Amazon Fire TV or Roku.

Rear projection TVs are troublesome - the old-school Sony Videoscopes are great but DLPs and LCOS-based TVs need bulbs and sometimes new color wheels. I remember fixing a Sammy DLP when the color wheel gave out(the wheel was shattered when a bushing failed) and I was able to get a Toshiba DLP going again with a generic bulb. I saw a Sony SXRD go belly up after 2 years.
 
Last edited:
I've done some more testing out of curiosity mostly. I keep ping-ponging between is the problem the backlights or the PS or both? They are quite dependant on each other for proper operation of both.

Using a bench supply set on 22Vdc and a DVM to monitor current draw, I measured each strip of 8 LEDs:

1. Top strip: I=97 mA. All LEDs bright.
2. 2nd from top: 102 mA. All bright.
3. 3rd from top: 1.9 mA. All very dim. Burned out section shown above.
4. Bottom strip: 127mA. All bright.

Next, I cut off the burned out section + a single LED on #3 that was on the end of the run. Shorted the two traces together as this is a series ckt, then retested it at 22V. Current drawn is 450 mA! Holy Cow! What a surprise. Evidently some of these LEDs in this strip are turning on far sooner than others. I turned the voltage down to 18V, then compared #3 against #1. At 18V, #1 wasn't drawing any current; #3 was drawing 5.4 mA. At 20V, #1: 1.1 mA; #3: 250 mA. At 22V, #1 97 mA; #3 450 mA. Looks like a short, but I don't see where. Too bad I don't have a FLIR camera.

At this point, this #3 string is making life for the PS very difficult, so it's acting up and quickly shutting off. The PSB data indicates the backlight driver rating at: 130V @ 750 mA. Thus my focus is away from the PSB and to strip #3 in particular. Perhaps adding a resistor in series with it will help. I could also add a few diodes to match the 3V drop of the missing LED.
 
I found a local TV repair shop (dying breed unfortunately) with two half-strips that were the right type, for a very reasonable amount. When installed, I now have all 4 strips continuously lit (32 LEDs) total! Major milestone event. Voltage across the backlight = 98Vdc, which is 3.06V/LED and well within the BL drivers rating of 130Vdc @ 750mA. Not sure of the loop current. V open-ckt is about 225V if I recall correctly. With the damaged half-strip installed the voltage across the backlight string meas. 190 Vdc!

The BL problem was this half-strip began drawing far more current over time than the others, the LED BL driver sensed this and raised the voltage which over time caused the burned area. As it burned, the trace resistance increased due to the rapidly shrinking strip width caused by arcing vaporizing the copper. The PS detected this increasing resistance, and raised the voltage even more. The LED BL driver behaves like a constant current source and therefore will raise the voltage to attempt to keep it so. However, it does have limits. So do the LEDs and copper within the strips. When V & I get too out-of-whack, the wheels come off, LEDs will burn, as will the copper foil, and the PS shuts down.

Problem here though is it didn't shut down soon enough to prevent arcing, which is a fire hazard!

I attempted to rig the defective strip with enough resistance to keep the PS on for testing, so I replaced the shorted end with about 30 Ohms. Problem is a resistor isn't a diode or LED. But I also didn't have several 3W diodes on hand. The damaged strip lit like all the others. The problem though it was drawing far too much current. Voltage testing sometimes won't find faults that current testing will reveal. Even with a 500 Ohm R across the end, it still drew too much current.

So the backlight is now fixed.

However there's another problem elsewhere, perhaps in the timing-control board. More on that later.
 
Am I understanding this correctly that you have found, at least pairs, if not more, of the LEDs are in parallel with each other? Seems like a very flawed design (Kirchhoff's laws) if this is the case.
 
Each strip is made up of a left & right-hand section. This set has 4 strips. Smaller/larger sets have fewer/more. All LEDs are in series.
 
sleddriver, no disrespect but I just had to post this drawing that I chuckle at often.
It so well describes much of my career as an el tech ...
lol.gif


 
No, not at all! I have that myself. Great laugh, just like Hazard Fraught Tools.

The only thing missing is the little box that contains the "Magic Smoke"...OH WAIT...I JUST SAW IT!

OK, we're good. Magic Smoke still contained. All is right with the ckt.
 
Yep, I think I mentioned that above.

BL "torch" mode setting is determental to life. I'm going to increase the R that senses BL current .
 
I'm now getting up to speed to troubleshoot the picture problem. During testing, I noticed in the lower, left-hand corner, that there are at least two station bugs burned in! Rather amazing to actually see that. So these sets will do this over time with the backlight cranked up in TORCH MODE. This set is barely two yrs old and it's previous owners spent 4-5 months in another country so it wasn't being used all year either.

I've long ago learned to control the white level (contrast) since I don't like glare period. This point was really driven home when I took the ISF calibration course and worked on CRT TV's.

Another point: If you watch TV late at night and have BIG screen and it's very bright, then you go to bed and have trouble sleeping there is a reason: The two are connected. I knew a guy with a bright 17" LT who used to look into the thing for a couple of hrs before bed, then complained of restless sleep.

There's a reason for that.....
 
I had good luck with all my LCD & LED TVs. My oldest is a Toshiba Aquos and probably 10-12 year's ($1743) they we new technology then for TV, it looks bulky looking at today's TVs. Have Samsung, Sharp, Vizio (2), they have stood up very well.
 
Originally Posted By: Malo83
Sony 60" HDTV going on 8yrs with no problems
Some...some...of the older sets are better made & feature CCFL backlights, which may last longer than LEDs.

Rather amazing how the newer sets have been de-parted/contented vs. the older. Avoiding the lowest tier line may get you a better power supply I've noticed while viewing many videos. Lowering the level of the BL will also help with set life.
 
Status
Not open for further replies.
Back
Top