Samsung LE40A536 constantly restarts itself

[Bambooz]

Hi! 

Stumbled across your forum while searching for the PSU/Inverter board (IP-211135A / BN44-00199A) for this hunk of junk.

TV: Samsung LE40A536 40" Full HD LCD
Mainboard: BN41-01145A ("530_NORMAL_idTV")

All PSU caps have been replaced with low-esr Panasonic/Matsushita and a few Rubycon, and some horrible soldering from a TV tech (?) before me has been fixed up. I even replaced every single tiny 10µF 50V cap (i was in there anyways.. so why not.. knowing what crap caps samsung always uses..)

TV turns on fine, picture is perfect, but it automatically turns off / goes into standby after 15 or so secs and restarts immediately.
Selected source doesn't seem to matter.
You can turn the set off (standby) within the ~15sec timeframe when it works normally.

In short: everything seems to work fine (functions wise), but it constantly reboots itself.

Read about the eeprom trick, and went hunting..
little 8pin sop chip on the solder side of the mainboard where the CI card slot is.

Chip markings Read:

ATMLH0834
2EB 1

google search lead me to this
http://akizukidenshi.com/download/at24c256b.pdf

Part marking scheme fits perfectly, so it's a AT24C256B Two-wire Serial EEPROM. bingo.
Soldered 2 wires to pin 5 and 6, reinstalled the whole mess, twisted wires together, plugged the set in and as expected, it didn't turn on. Removed the short on the two wired, got the TV out of standby aaaand.. reboots after 15secs 

Bad eeprom? Completely different problem?!
Or am i just doing the reset/delete procedure wrong?

*???*

I'm lost..

Any ideas?


Patrik

[fix2003]

i find it best to use new blank eeprom

[Bambooz]

That would probably be the best solution (if it works that is), but i'm having quite a few difficulties finding one locally.

Getting one from china via ebay (actually a 5 pack or more) is no problem. Getting one locally means paying 12eur a piece (wtf are they smoking?!) or having to try to kinda shove a dip package where a sop is supposed to go.. no fun.

I've put the set back together for now (need room) and left the 2 wires soldered to the eeprom hanging out of the side of the case. Playing with them while in standby (randomly shorting them) somehow got it to turn on and actually stay on in DTV/DVB-T/Freeview mode. It's been running for several hours now without a hiccup (apart from the usual crappy DTV picture and blocky-ness), but i fear if i turn it off and back on, the same stuff is going to start all over again :/

I'm gonna hunt around for an eeprom. I just hope that's all there is to it..
If not, I'm running out of options.. replacement mainboards are listed in several shops, but none of them has them in stock.

Patrik

edit: while playing with the 2 wires sticking out, i tried getting into the service menu once, and I noticed that I've somehow managed to reset the ~5900 power-on/lamp hours I saw previously to zero

[sparky]

hi
have you checked the tact switch on front of tv
standby button had a lot of problems you can disconect plug and try
dave

[Bambooz]

Tried that today (why didn't i think of this?! ). Disconnected the whole front panel button board cable. Same thing :(

As a little "update".. Youtube video (mainly to see if there's anyone else with that problem)
http://www.youtube.com/watch?v=xlfDREBcevs

[sparky]

have you checked the switch at the side
dave

[Bambooz]

All the buttons check out fine. Both on the little PCB on the side (menu, channel, volume, etc. buttons) as well as the power button in the center. No glitchy-ness whatsoever.
Kinda interesting that this one doesn't have a hardware power switch.

All those button boards are connected via 1 plug to the mainboard.
Can be seen in the video at around 0:25 at the bottom right.
Rightmost connector = speakers, the one to the left of that = all the buttons + IR sensor.

Disconnecting the whole plug has no effect on the problem. Disconnecting the side panel buttons from the power button and IR sensor board (in the center) doesn't help either.

New EEPROMs are on the way. If that doesn't fix it I'm pretty much lost.

I've also tried measuring the output of all the LDOs on the mainboard both while on and in standby, to see if any of them drop down too much. It uses a bunch of 1117's and a few 78Dxx's. All voltages are dead on, except for the two 3.3V ones which sit at 3.28V. Dunno if that's enough for the thing to go berserk, but it shouldn't IMHO as it's still within 1% of the nominal value.

What really confuses me is that the longer i poke aroung on the solder side of the mainboard, the more 2wire serial EEPROMs I find  

There's one very close to the CI Slot (the one mentioned in previous posts ; shorting SCL and SDA reset the power on counter to zero), another one close to the side panel HDMI port which i didn't notice before because it was covered in excess flux from the factory (was unreadable), and yet another one close to the LVDS connector to the panel.

Great. Now which one am i supposed to reset/erase or replace..!?

edit: forgot to mention that i know NOT to play with the main firmware EEPROM unless i want to brick it completely ;)

[sinclairtv]

This seems amazingly long winded. Change the eeprom first then worry about anything else. 5 gets you 10 that its the eeprom thats the problem.

[Bambooz]

This seems amazingly long winded. Change the eeprom first then worry about anything else. 5 gets you 10 that its the eeprom thats the problem.

Apparently not.
The EEPROMs arrived after more than a week (from a local seller on ebay). They're new cut-off-tape ST24C256B (marked ST 4256BWP).

Replaced the original with one of the new ones. After plugging it in, only the standby LED is on (doesn't turn on by itself). Trying to power it up with the power button in the front does nothing at all (LED stays solid, no reaction).
When you try it with the remote, the LED turns off (as if it's about to turn on), nothing happens (no backlight, no sound), and it goes back into standby (solid red LED) after ~10-20 seconds.

I've tried it multiple times, and let it sit there for a while just in case it was still doing something (re-writing the EEPROM perhaps).. but no. Same thing.

:(

[sparky]

are they soldered the correct way

[Bambooz]

Yep. Checked the orientation multiple times. With the original EEPROM back in there, it turns on and reboots as before.

[flyflytn]

So did you make any progress with this ? I have an LA40N81BDX doing the exact same thing.
Replaced all the caps (4 were definitely fried in the usual section), which got the TV to fire up for.....15 seconds. I can time it to be 10 seconds after the HV to the screen has fired up....so I can hear the HV inverter fire up in 3 or 4 stages in steps, gets to the final stage, wait 10 seconds and then it restarts. The restart will cause the screen to flicker off/on - sometimes once, sometimes 4 or 5 times very quickly.
I've fixed a few plasmas, but this is my first lcd so finding my way around.....
ciao

[flyflytn]

I managed to clear my EEPROM (or so I think I have) but still this restarting issue persists.
The service menu says zero hours use now, so I assume all the EEPROM was erased. Anyone managed to fix this at all ?

[watteau]

Hello,
Just a thought, are you certain it's the main board causing this problem?
I've had problems with the power-supply giving this effect.
You could try lifting the power-supply 'on' lead going into the power supply output socket
(normally marked PWR ON/OFF, or PS), coming from the main PCB, and supply a switch on voltage from the 5v standby supply via a 4K7, to that pin on the power-supply, and see if the set stays on.
Regards,
John.

[flyflytn]

Yes, thats what I was thinking too. I had already tried last weekend but with no load on the board - its only when the logic board is connected that the PSU goes down. So my new test is going to be holding that line high while the whole TV is going.....I'll stick a diode in there to stop any flow to the logic board and have a look with my CRO. But yes I still think its the logic board since the timing is always the same, nothing heat related.

Next stop after that I'm not sure....possibly tracing that on/off line all the way and check components along the way. Thanks for the suggestion, great minds think alike and so do you and I