Poll

Best meat:

Bacon
0 (0%)
BACON
3 (27.3%)
MORE BACON!
8 (72.7%)

Total Members Voted: 11

Author Topic: Samsung LED Monitor (S27B350H) -- Bizzare Failure!  (Read 8888 times)

0 Members and 1 Guest are viewing this topic.

Offline Obsidian_ShadowHawk

  • Contributor
  • Posts: 10
  • Country: us
Samsung LED Monitor (S27B350H) -- Bizzare Failure!
« on: September 02, 2016, 12:46:43 am »
Thank you in advance for reading if you have the time!

Ok, full disclosure:  this is another Goodwill find, so any repair short of buying a new, equivalent monitor is a win. 8) (Looks like equiv units go for about $200-250 new.)

First, see video of failure mode, as it is worth 10,000 words (sorry for the quality, not the best):
https://youtu.be/ztuk_H6ZwrY

(Display mode is "duplicate" in Windows, happened before and after OS upgrade.  Picture is being sent over HDMI at native display settings for the Samsung (1080p, 60-ish hz.  Failure is independent of what input is selected (monitor has 1 HDMI and 1 VGA).  OSD menu and overlay are also affected.)

Monitor worked flawlessly for about 6 months in my house.  Then the issue in the video and pictures attached started happening (probably why it went to Goodwill).  Happens at power-up for 30-60 seconds, then snaps to a good image.  At first, it only happened occasionally, maybe 1 in 10 power-ons, then became for frequent over a few months.  Now it happens every 99 in 100 power-ons, and can last for up to 15 minutes before snapping to a good image.  (By "snapping", I mean that there is no transition from failure to a good image, it corrects itself instantly, albeit at a random time.)  The monitor displays the top-left 1/4 of the input, stretched across the whole LCD.  It is green-tinted and each row of pixels wavers back-and-forth quickly.  Framerate seems to be normal.

Topology of monitor:
  • External power brick, inputs 14V, rated at 4A. Voltage under load is within .2V and stable.
  • Monitor has a main logic board with inputs and DC jack.  Capacitive front bezel controls and LED backlight run off this board.
  • Ribbon cable from main logic board goes up to LVDS board (correct term?) which is hot-barred to the LCD flat-flex.

Things I have tried so far:
  • Caps are good brand, well-rated.  Can't see any obvious failures there.  Looked under magnification for solder issues.
  • Thermals.  The two chips on the main logic board get pretty warm (~75°C).  Freezer spray had no effect on any part of the power/logic board, cables, or LVDS board/flex.
  • Early troubleshooting seemed to point to the LVDS ribbon cable, as it was starting to delaminate where it is folded over.  I replaced it with an official Samsung part from samsungparts.com.  Issue remains.  Oh well, only $13.
  • Again, to reiterate, other input sources do not fix the problem.  LCD menu is affected.
  • Reset display settings on the monitor, made sure refresh rates, etc were being detected correctly.
  • Several in-depth searches turned up no similar issues reported on other corners of the internet.
  • Putting pressure on or flexing the LCD flat flex where the chip-on-flex drivers are doesn't seem to have an effect.  Neither does messing with the hotbar areas or other connectors/caps/components.
  • UPDATE: Tried powering the montior from a universal DC supply set at 14V, got the same issues.

Official service manual attached, not very thorough, I'm afraid...

Can anyone who has a better understanding of LCDs hypothesize here?

More pics:

Dropbox link for the service manual!

Sorry for the long post.  :blah: TLDR is totally understandable.  ;)
« Last Edit: September 07, 2016, 02:47:35 pm by Obsidian_ShadowHawk »
 

Offline imidis

  • Frequent Contributor
  • **
  • Posts: 427
  • Country: ca
Re: Samsung LED Monitor (S27B350H) -- Bizzare Failure!
« Reply #1 on: September 02, 2016, 09:18:25 am »
Definitely interesting. I'm probably not too much helps, looks like a heck of a lot of noise. Bit different from the more common backlight or inverter going bad. I wish I could be of more help, that's not one I've come across.  :-\
Gone for good
 

Offline CJay

  • Super Contributor
  • ***
  • Posts: 3547
  • Country: gb
Re: Samsung LED Monitor (S27B350H) -- Bizzare Failure!
« Reply #2 on: September 02, 2016, 09:20:37 am »
Power supply, change those caps.
 

Offline RGB255_0_0

  • Frequent Contributor
  • **
  • Posts: 774
  • Country: gb
Re: Samsung LED Monitor (S27B350H) -- Bizzare Failure!
« Reply #3 on: September 02, 2016, 09:25:22 am »
Since it takes a while for it to snap into place, perhaps thermal expansion creates a good contact?

Anyway, it's a TN panel with a low res for 27" and diabolical contrast. Worth repairing? IMO no if you don't have the time.
Your toaster just set fire to an African child over TCP.
 

Offline Obsidian_ShadowHawk

  • Contributor
  • Posts: 10
  • Country: us
Re: Samsung LED Monitor (S27B350H) -- Bizzare Failure!
« Reply #4 on: September 02, 2016, 12:29:01 pm »
Since it takes a while for it to snap into place, perhaps thermal expansion creates a good contact?

Anyway, it's a TN panel with a low res for 27" and diabolical contrast. Worth repairing? IMO no if you don't have the time.

Yeah, it really feels like a thermal issue. It lasts longer if the monitor is left unpowered overnight, and has less chance of happening after a long day of use, if power-cycled quickly. Although, I wasted two cans of duster trying to make it fail, no dice.

Also, yes, I know there are better panels out there, but I really can't afford it right now. Im on a tight budget with alot of time. And I use this one alot, the large size and lowish resolution are very helpful, I don't have the best eyesight in the world.  Also, to be fair, not all TN panels are beyond deplorable. Samsung does some wiz-bang processing and I borrowed a colorimeter from a friend to match the two displays. Looks damn good now. ;)

Power supply, change those caps.

Thanks for the suggestion, but the external supply has been measured and is rock solid. The 12v and 5v rails inside are also not showing any issues.

Edit: Never trust spellcheck.
« Last Edit: September 02, 2016, 12:38:41 pm by Obsidian_ShadowHawk »
 

Offline Srbel

  • Frequent Contributor
  • **
  • Posts: 359
  • Country: cs
  • Electronics engineer
Re: Samsung LED Monitor (S27B350H) -- Bizzare Failure!
« Reply #5 on: September 02, 2016, 04:46:52 pm »
But it can have excess voltage ripple.

P.S. Is it and LED or LCD screen?
 

Offline Obsidian_ShadowHawk

  • Contributor
  • Posts: 10
  • Country: us
Re: Samsung LED Monitor (S27B350H) -- Bizzare Failure!
« Reply #6 on: September 02, 2016, 04:53:12 pm »
But it can have excess voltage ripple.

P.S. Is it and LED or LCD screen?

LCD panel, LED backlight. True, how could I check for that? I only have a fairly basic multimeter, no other equipment. Trying to find a.Pico scope on ebay just to get by until I can get a Rigol. Years away, though...
 

Offline amyk

  • Super Contributor
  • ***
  • Posts: 6848
Re: Samsung LED Monitor (S27B350H) -- Bizzare Failure!
« Reply #7 on: September 04, 2016, 01:12:57 am »
Then the cheaper alternative would be to replace the caps (or just tack new ones in parallel to the old if you want to avoid desoldering).
 

Offline Obsidian_ShadowHawk

  • Contributor
  • Posts: 10
  • Country: us
Re: Samsung LED Monitor (S27B350H) -- Bizzare Failure!
« Reply #8 on: September 04, 2016, 01:19:03 am »
Caps don't seem to be the problem here. ESR is normal, no bulging or leaking. Any other ideas? A problem with the column or row drivers?  How would the screen only show a quarter?
 

Offline amyk

  • Super Contributor
  • ***
  • Posts: 6848
Re: Samsung LED Monitor (S27B350H) -- Bizzare Failure!
« Reply #9 on: September 04, 2016, 01:28:11 am »
If you're testing the caps in-circuit, the readings may be misleading.

I don't think it's the panel, but the scaler/video processor. It looks like it's syncing to the wrong frequency.
 

Online wraper

  • Supporter
  • ****
  • Posts: 11697
  • Country: lv
Re: Samsung LED Monitor (S27B350H) -- Bizzare Failure!
« Reply #10 on: September 04, 2016, 01:31:20 am »
Power supply, change those caps.
unlikely, probably T-CON problem
 

Offline TerraHertz

  • Super Contributor
  • ***
  • Posts: 3712
  • Country: au
  • Why shouldn't we question everything?
    • It's not really a Blog
Re: Samsung LED Monitor (S27B350H) -- Bizzare Failure!
« Reply #11 on: September 04, 2016, 10:16:11 pm »
Oddly enough, I was involved in the design of one of the first commercial uses (in Australia) of LCD panels adapted to PC VGA outputs. Company was Keycorp, the screens were for the Commonwealth Bank.
And I recognize that problem. It's a PLL lock failure, in something that is supposed to be generating a pixel sampling (or resampling) clock.

Early screens taking VGA inputs, had to generate their own sampling clock from the VGA H sync, since there is no clock on the VGA connector. If the PLL generating the clock was poorly locked, that's the exact visual effect.
Why it would happen with HDMI, which includes a data clock, I don't know. I'm guessing that the screen is doing some resampling to a different clock rate due to different screen aspect ratio or resolution, and that's where the fault is.

Suggestions for causes:
 - Bad supply ripple or loss of decoupling caps around the PLL (which might be internal to a VLSI chip.)
 - A grounding impedance problem around the PLL.
 - Something wrong in the PLL feedback loop. (Thermal intermittent?) It's _trying_ to lock, just not very well.
 - The PLL may have its own noise-isolated power and ground - inductors from the main rail and ground, with a local electro and ceramic decoupling cap. If that electro is bad, or there's a short between the noise-isolated rails and the main ones, system power noise will kill the PLL's ability to lock.
 - Something wrong with the reference clock (or sync) used by the PLL.
« Last Edit: September 04, 2016, 10:23:35 pm by TerraHertz »
Collecting old scopes, logic analyzers, and unfinished projects. http://everist.org
 
The following users thanked this post: Srbel, Obsidian_ShadowHawk

Offline Rasz

  • Super Contributor
  • ***
  • Posts: 2396
  • Country: 00
    • My random blog.
Re: Samsung LED Monitor (S27B350H) -- Bizzare Failure!
« Reply #12 on: September 06, 2016, 02:39:00 am »

  • Caps are good brand, well-rated.  Can't see any obvious failures there.  Looked under magnification for solder issues.
it doesnt work like that
+ is your hearing impaired? this noise is from the smps with dead caps
Who logs in to gdm? Not I, said the duck.
My fireplace is on fire, but in all the wrong places.
 

Offline Obsidian_ShadowHawk

  • Contributor
  • Posts: 10
  • Country: us
Re: Samsung LED Monitor (S27B350H) -- Bizzare Failure!
« Reply #13 on: September 06, 2016, 04:08:23 am »
it doesnt work like that
+ is your hearing impaired? this noise is from the smps with dead caps

I bow to Lord Rasz, Knower of All Things. May He forgive my ignorance of all things related to SMPS.

Honestly dude, care to chill a bit? I'm still a noob. A little explanation would go a long way towards proving your theory. Needless insults do not.  High pitched whining noises are present in perhaps 60% of the power supplies in my home, all of which show no signs of failure (aside from the monitor, of course). That's not exactly something I would notice as out-of-the-ordinary.

Also, when narrowing down culprits, checking off "bulging/leaking caps" and "poor solder joints" is legit. I'm not saying a cap isn't an issue, just saying that it is not readily obvious, and needs more thourogh testing than I am equipped for.
 
The following users thanked this post: Srbel

Offline Obsidian_ShadowHawk

  • Contributor
  • Posts: 10
  • Country: us
Re: Samsung LED Monitor (S27B350H) -- Bizzare Failure!
« Reply #14 on: September 06, 2016, 04:16:59 am »
Oddly enough, I was involved in the design of one of the first commercial uses (in Australia) of LCD panels adapted to PC VGA outputs...

Dude, awesome info, thanks! It's pretty reassuring to finally find someone who has seen this before.  :phew:  I do have a few questions:

Where would the PLL circuit likely be? Integrated with the processor on the main boards, or after the LVDS cable on the panel board?

By VLSI chip, do you mean one of the two large chips on the man board? Tried to look up the term, but definitions were very broad.

How would I test the surface mount caps on the board for failure? Or even the solid metal can surface mount caps?
 

Offline Rasz

  • Super Contributor
  • ***
  • Posts: 2396
  • Country: 00
    • My random blog.
Re: Samsung LED Monitor (S27B350H) -- Bizzare Failure!
« Reply #15 on: September 06, 2016, 06:45:19 am »
I bow to Lord Rasz, Knower of All Things. May He forgive my ignorance of all things related to SMPS.

just change all electrolytic caps, whining power supplies are not normal contrary to what you might have assumed, it means you either have a  lot of faiiling gear, or a lot of cheap chinese garbage waiting to burn your house while you sleep :(
btw, not everyone has the same monitor you have, so its helpful to provide pictures of pcb

sorry for asshole reply, I got irritated by another 'hey guys my car has no traction but tires are fine because wheels are still round' :'(
Who logs in to gdm? Not I, said the duck.
My fireplace is on fire, but in all the wrong places.
 
The following users thanked this post: HackedFridgeMagnet

Offline amyk

  • Super Contributor
  • ***
  • Posts: 6848
Re: Samsung LED Monitor (S27B350H) -- Bizzare Failure!
« Reply #16 on: September 07, 2016, 10:50:48 am »
You haven't even told us what brand of caps are in the monitor, but past experience is what has given you multiple "change the caps!!" replies from everyone here. There's even an entire site dedicated to it: badcaps.net
 

Online MosherIV

  • Super Contributor
  • ***
  • Posts: 1373
  • Country: gb
Re: Samsung LED Monitor (S27B350H) -- Bizzare Failure!
« Reply #17 on: September 07, 2016, 11:49:33 am »
Quote
Dude, awesome info, thanks! It's pretty reassuring to finally find someone who has seen this before.  :phew:  I do have a few questions:

Where would the PLL circuit likely be? Integrated with the processor on the main boards, or after the LVDS cable on the panel board?
I suspect the circuitry you are looking is on the main boards, by the time the video signal is sent to the flexi cables, the video signal has already been processed down to pixel lines.

Quote
By VLSI chip, do you mean one of the two large chips on the man board?
Yes, most likely.

Quote
Tried to look up the term, but definitions were very broad.

How would I test the surface mount caps on the board for failure? Or even the solid metal can surface mount caps?
You need to look up the part numbers of the 2 large chips. Get their data sheets and look at the block diagrams for how they work internally.
Look for pin out information and trace where the PLL block comes out of the chip and look around the passive components of that circuit.

A oscilloscope would be really handy to debug this.

Quote
Since it takes a while for it to snap into place, perhaps thermal expansion creates a good contact?
Bear in mind that the fault might not be a component, it might be a dry joint.
 
The following users thanked this post: Obsidian_ShadowHawk

Offline Obsidian_ShadowHawk

  • Contributor
  • Posts: 10
  • Country: us
Re: Samsung LED Monitor (S27B350H) -- Bizzare Failure!
« Reply #18 on: September 07, 2016, 02:39:31 pm »
You need to look up the part numbers of the 2 large chips. Get their data sheets and look at the block diagrams for how they work internally.
Look for pin out information and trace where the PLL block comes out of the chip and look around the passive components of that circuit.
A oscilloscope would be really handy to debug this.
Bear in mind that the fault might not be a component, it might be a dry joint.

Will do. Next chance I will have to look at it will be Thursday evening. I'll get some good pictures of the caps and solder joints, too.  Sorry, those should have been in the original post.  I don't remember seeing any standard black electrolytics in the monitor, I suspect all those are in the external AC plugpack. Btw, I stopped at Staples yesterday and got a universal power adapter. Powered the monitor with that for a few hours (monitor takes 14v dc in, center positive) and got the same issues.

The main board only has some of the small surface mount caps, like these and these.  They look pretty well sealed, do they dry up in the same way? Can they be tested in the same way as regular caps? (Unsolder one side, measure capacitance and ESR). I have a budget ESR meter on order from an eBay seller, it's not good, but all I can afford right now. It will give a rough estimate, at least.

Btw, can anyone suggest an easy image hosting service? I'd rather post pics in line for you rather than linking to dropbox. Thanks!
« Last Edit: September 07, 2016, 02:49:26 pm by Obsidian_ShadowHawk »
 

Offline CJay

  • Super Contributor
  • ***
  • Posts: 3547
  • Country: gb
Re: Samsung LED Monitor (S27B350H) -- Bizzare Failure!
« Reply #19 on: September 07, 2016, 02:52:25 pm »
Those silver can electrolytics are evil.

They dry out and leak just like any other electrolytic capacitor which has a liquid electrolyte

have a look around 20:15 onward to see what they're capable of.



 

Offline Rasz

  • Super Contributor
  • ***
  • Posts: 2396
  • Country: 00
    • My random blog.
Re: Samsung LED Monitor (S27B350H) -- Bizzare Failure!
« Reply #20 on: September 08, 2016, 06:11:09 am »
tty well sealed, do they dry up in the same way? Can they be tested in the same way as regular caps? (Unsolder one side, measure capacitance and ESR). I have a budget ESR meter on order from an eBay seller, it's not good, but all I can afford right now. It will give a rough estimate, at least.

 |O :palm:
so you spend $20 to measure $5 of capacitors, instead of replacing all of them :clap:

Btw, can anyone suggest an easy image hosting service? I'd rather post pics in line for you rather than linking to dropbox. Thanks!

you click 'Attachments and other options' under post text window
Who logs in to gdm? Not I, said the duck.
My fireplace is on fire, but in all the wrong places.
 

Offline CJay

  • Super Contributor
  • ***
  • Posts: 3547
  • Country: gb
Re: Samsung LED Monitor (S27B350H) -- Bizzare Failure!
« Reply #21 on: September 08, 2016, 06:47:03 am »
|O :palm:
so you spend $20 to measure $5 of capacitors, instead of replacing all of them :clap:

Yeah, but test gear is an investment that should pay for itself many times over.

I do agree that replacing the caps might have saved a lot of time though.
 

Offline Obsidian_ShadowHawk

  • Contributor
  • Posts: 10
  • Country: us
Re: Samsung LED Monitor (S27B350H) -- Bizzare Failure!
« Reply #22 on: September 08, 2016, 10:54:55 am »
|O :palm:
so you spend $20 to measure $5 of capacitors, instead of replacing all of them :clap:

Yeah, but test gear is an investment that should pay for itself many times over.

Exactly. Some people just don't get it. For beginners, the main benefit of repairing things is EDUCATION, not monetary gain.  This Rasz guy is just being an asshole for the sake of feeling superior. I've noticed that it's kind of his MO on other threads.

Also, holy cow, the board in that video is in serious trouble with all that corrosion! I'll get a good look at mine tonight.
« Last Edit: September 08, 2016, 11:09:59 am by Obsidian_ShadowHawk »
 
The following users thanked this post: Srbel

Online MosherIV

  • Super Contributor
  • ***
  • Posts: 1373
  • Country: gb
Re: Samsung LED Monitor (S27B350H) -- Bizzare Failure!
« Reply #23 on: September 09, 2016, 09:36:01 am »
Hi

Randomly changing components may, or may not, fix the problem.

Good engineering practice is to find the root cause and fix that.

As TerraHertz said, he suspects the issue is to do with the video processor not corrrectly synchronising to the input video waveform, specifically not extracting the timing pulses. They normally use something called a Phase Lock Loop to adjust a local oscillator to synchronise with the incoming signal.
I suspected the same thing but I have not worked on that sort of thing before.

This may be caused by a bad component or it might be a solder joint gone bad and causing intermittent connection.

I suspect randomly changing capacitors will not fix the issue, but I wait to hear your results.
 
The following users thanked this post: Obsidian_ShadowHawk

Offline tatus1969

  • Super Contributor
  • ***
  • Posts: 1251
  • Country: de
  • Resistance is futile - We Are The Watt.
    • keenlab
Re: Samsung LED Monitor (S27B350H) -- Bizzare Failure!
« Reply #24 on: September 09, 2016, 10:52:06 am »
if the problem disappears after a time, the first thing I would do is to grab a can of ice spray and use it to spot the problem source.
We Are The Watt - Resistance Is Futile!
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf