Samsung LE32R41B foggy picture

[Th3_uN1Qu3]

Considering picking this one up. It still works, but picture is foggy. Attached is the photo in the ad. Someone who's seen more Samsung LCDs than i did, please chime in.

I have to decide between this one and a Hantarex 42" plasma, which shows some "stains" or "shadows" over the picture, but they aren't really apparent in the pics the seller provided, it appears to work (almost) normally. Could those "stains" be the sparklies? If so, which one would you pick? I've worked on Samsung TVs before so i know a thing or two about them, but the plasma is FullHD and larger in size.

[tom66]

Re: Samsung LE32R41B foggy picture

Gamma IC usually, on the T-con. Easy fix -- TQFP48 soldering. It's probably got a Chimei panel.

Are you sure that the plasma is 1080p? I say this because I know of -very- few off-brand plasmas that are 1080p, and only Panasonic and LG have produced 1080p 42" panels (which are produced in low quantities now.) Also, I do not know of any sparklies issues for 1080p sets. More likely, the seller is lying, and it is a 720p set.

[Th3_uN1Qu3]

Re: Samsung LE32R41B foggy picture

No, i'm not sure, but a quick google seems to suggest that. The plasma is a Hantarex PD42 Xframe.

What board is this gamma IC on, Tcon or main? I'd like to fix it of course, but boards aren't too expensive either. If you say it isn't too much work, i think i'll get the Samsung tomorrow. Does the chip need to be replaced or just reworked?

[tom66]

Re: Samsung LE32R41B foggy picture

T-con. The chip can be replaced, it's about $5 on eBay.

Regarding the plasma, the clue is (on one website):

Risoluzione orizzontale (pixel): 1024 Pixel
Risoluzione verticale (pixel): 1024 Pixel

It's a 1024x1024 ALiS panel. This makes it 720p capable (all 42" 720p panels are 1024x768). But it can't display 1080p content. This is a Fujitsu-Hitachi panel. They have poor black levels generally, but the panels cost a lot less than other manufacturer's panels.

[Th3_uN1Qu3]

Re: Samsung LE32R41B foggy picture

Thanks. I've done a few TQFP replacements, all successful, so i feel confident about handling this one as well. Samsung it is then.

[budm]

Re: Samsung LE32R41B foggy picture

See if it uses AS15 IC:
See example here:
https://www.badcaps.net/forum/showth...s15#post322816

https://www.badcaps.net/forum/showth...mma#post304275

[Th3_uN1Qu3]

Re: Samsung LE32R41B foggy picture

Cool, thanks guys. I'll get the TV and then see what gamma IC it uses. If i can find it at Utsource, that would be great, as they ship a bit faster than most other Chinese sellers i've dealt with until now. Minimum order is $10 IIRC, so if it's the AS15G which i see is $3 a pop, i'll grab 4 or 5 of them, they might come in handy in the future.

[Th3_uN1Qu3]

Re: Samsung LE32R41B foggy picture

Okay, i got it. And found out a bit of this set's history as well.

The guy i got it from was one unlucky chap this time. He doesn't have much electronics knowledge, he's just a board swapper. He bought the TV described as no power. However what he found out was that the set would power up and the backlights came on, but there was no picture.

He bought a replacement Tcon and swapped it, only to be greeted by the picture you see in the photo.

The good news: I also got the original Tcon from him, and i'd like to fix that one if possible, because it's got a bit of extra circuitry on it which he explained is for the ambient light sensor. The replacement Tcon does not have it (by his words, i haven't taken it apart yet). Since the function is present in the menu, i'd like to have that work as well. The original Tcon appears to have no interventions apart from 2 pins of an EEPROM that have been bridged together with solder. He said he found it like that, he didn't work on the board at all.

The gamma IC used is the ADD8709 btw.

What i'm worried about is that horizontal line in the lower quarter of the screen, which you can just make out in the photo, but is much more apparent in real life. However, it changes colors depending on the picture content, fading in and out slowly, so it's likely also due to the Tcon, and not the panel itself. At least, i've never seen bad tab bonds make horizontal lines.

We'll see in a couple days...

[tom66]

Re: Samsung LE32R41B foggy picture

So, we can deduce he's probably lying or clueless.

The T-con board doesn't have ANYTHING to do with the ambient light sensor; that would be on the main board.

He probably put a bridge across the eeprom to reset it in a failed attempt to get it going.

The horizontal line may be a bad panel, though I have seen tab bonds cause them too. The t-con cannot cause regional defects.

[Th3_uN1Qu3]

Re: Samsung LE32R41B foggy picture

Well, dunno. Near the parts that he told me are not populated on the Tcon he put in, there is a little header with a 2-pin cable, which judging by the silkscreen, would plug somewhere into the inverter. It's labeled INVSYNC.

If you're sure it's a bad panel, i may not even bother trying to fix it, and just part it out. Edit: The Tcon (and thus, panel) are made by AUO.

Edit 2: On the original Tcon (T315XW01 T-CON 04A05-1E), the pins that are bridged together on the EEPROM are WP and VCC. This means that the EEPROM is write protected. Doesn't make too much sense, write protecting it for WHAT reason? Anyway, i'll check voltages on the 2 Tcons against each other, dump both EEPROMs, compare, reflash, and whatever else i can think of.

[tom66]

Re: Samsung LE32R41B foggy picture

I'm not sure it's a bad panel. The line appears to span more than one row which would possibly indicate tab bonding issues. Strangely, it's often the top bonds that cause this.

INVSYNC is used to reduce flicker on lower frame rate sources for TVs with a 24Hz display mode (cinema playback) or sometimes for "black frame insertion" for display motion enhancement. (NOT the same as 100Hz etc nonsense.)

AUO T-cons often suffer from the gamma IC failure. Often, the IC will run very hot when faulty.

[Th3_uN1Qu3]

Re: Samsung LE32R41B foggy picture

I'm not sure it's a bad panel. The line appears to span more than one row which would possibly indicate tab bonding issues. Strangely, it's often the top bonds that cause this.

It does span more than one row. And it's transparent, not solid.

AUO T-cons often suffer from the gamma IC failure. Often, the IC will run very hot when faulty.

Could it by any chance be this? Go to page 3 and scroll all the way down... Of course, this is an entirely different panel, it's got the drivers at the bottom. But it's worth a shot before i get into component level repairs.

Either way, looks like i'll have to tear the whole panel down.

[tom66]

Re: Samsung LE32R41B foggy picture

Drivers at the bottom just mean that the panel is flipped vertically and the image processor sends it a reversed image so everything appears otherwise normal. This is done for power dissipation reasons, though the panels still seem to have tab bond issues regardless.

[Th3_uN1Qu3]

Re: Samsung LE32R41B foggy picture

I've looked at it today in more detail. The horizontal line spans exactly two rows. Also, there's a weird issue in the corners, about 2cm triangle in each corner, and thinner lines going from the corners along the extreme screen edges, with blurring and ghosting. It looks like it's got a frame around it, almost like an old CRT with a tube with rounded edges. There's severe ghosting and image retention in those areas.

In fact, there's pretty bad image retention all over. Hope it gets better once i fix the Tcon.

[tom66]

Re: Samsung LE32R41B foggy picture

It does sound more like tab bonds -- gamma failure doesn't usually cause motion blurring.

[Th3_uN1Qu3]

Re: Samsung LE32R41B foggy picture

I've just torn down the panel. Bad news. Someone's worked on the driver boards, and messy at that. A number of SMD resistors and caps appear to have been unsoldered and resoldered, and there's even a cap missing entirely on the upper left board. I'll poke and prod here and there, but i'm afraid this panel is beyond economical repair.

[Th3_uN1Qu3]

Re: Samsung LE32R41B foggy picture

I've deducted the missing resistors and caps and installed them, it wasn't rocket science actually. Unfortunately, i think i've successfully diagnosed a bad panel.

With all the parts back in place by the trimpots, the panel lit up blank, backlights only. Upon measurement, the 13.5v AVDD rail went to 5.11v. Hence why all the lifted and re-installed parts, the guy who attempted to service it wanted to trace the part that caused the voltage to drop.

I've disconnected the row driver (it has a ribbon with a ZIF cable), and the panel lit up normally with some lines, AVDD back where it should have been.

The only thing left was to attempt to power the thing externally and see if there's any smoke. I wired in 12v from a PC power supply, thru a 10 ohm resistor. Panel lit up with normal colors , but too dark. Voltage at the row driver board was roughly 10v. I parallelled another 10 ohm resistor, voltage now 10.9v, brightness got better, but ghosting got worse. Fiddled a bit with the trimpots, and the top and bottom one do nothing. The one in the middle has a huge effect on the brightness, but not on the ghosting unfortunately. And yes, that horizontal line is still there, and behaves the same.

Now i have it hooked directly to 12v, brightness is almost normal, but ghosting is terrible. The closer i get to its normal 13.5v, the better the colors get, but the worse the ghosting. I guess i'll just run it till it breaks. Edit: Used a laptop charger i made adjustable to go to 13.5v, and nothing bad happened, except it still ghosts like crazy. Will make a nice picture frame tho... built-in fade effect.

[tom66]

Re: Samsung LE32R41B foggy picture

As far as ghosting, look at the VON/TFTON and VOFF/TFTOFF. They are generated on the same supply as AVDD, and due to the charge-pump nature (based off the AVDD switching action), require that the panel be drawing current to maintain the outputs.

[Th3_uN1Qu3]

Re: Samsung LE32R41B foggy picture

Thanks, you gave me some hope. Here's some pics of the progress. And the blurring

I also figured out why the parts were removed - in an effort to diagnose what was pulling that voltage down. Makes sense now. I'll also try the original Tcon in the next few days, because i have a hunch it might be good as well.

[Th3_uN1Qu3]

Re: Samsung LE32R41B foggy picture

Many thanks tom66. Things started making sense. AVDD is made by a boost converter, and the 5.11-5.12v it was dropping to matches exactly Samsung's preference to use fractional voltages - the main "5v" rail is usually 5.1v. So, the boost converter was dropping out. This wasn't an overcurrent condition as otherwise the fuse on the Tcon would have blown and AVDD would be zero.

On this panel VON is marked as VGHC and VOFF as VGL. I googled some, and found some datasheets of LCD panel power supply ICs. This Tcon does not use a specialized IC for all three voltages, it's using regular SMPS controllers. Anyway. I found the ballpark those voltages usually resided in. VGL i found could be anywhere between -5v and -12v, and VGH is usually 30v.

Well... both of them were missing on this panel. With AVDD wired in externally, VGL was 0.3v, positive, and VGH was 5.40v.

I wired VGL in from a lithium battery in series with a 10 ohm resistor. Things got better, less bleeding, better brightness and colors, but still ghosting like crazy. Then i wired VGH too, from 2 12v SLA batteries in series, and a 100 ohm resistor, because as far as i understood, VGH is 50mA tops. And... the difference is like night and day. The ghosting is entirely gone, and the panel displays a clean, vivid picture. It's ALIVE!

Horizontal line's still there, but i'm not gonna bother. Time to put the frame back onto the panel, and focus on the Tcon.