Stumped on Samsung PN60E530A3F

[drussell]

Re: Stumped on Samsung PN60E530A3F

Have you swapped main boards to see if that's the problem?

You mean the "Logic Board" in Samsung parlance, I take it, not what they call the "Main Board" with the HDMI ports on it, but the "Logic / Buffer Board" center section of the three, the one behind the plastic stand holder thingy?

No, I haven't done anything with that one yet because it's such a PITA to get to.
I took one out previously to look for signs of what might be causing the sluggish turn-on and want to mess with everything else before I go in for that one again.

The "Logic Board" has nothing to do with the sluggish start-up because that problem followed the Y-Main board between sets but it is possible that the logic board is involved with the lack of Ve. I'm still trying to wrap my head around exactly what is going on here.

[drussell]

Re: Stumped on Samsung PN60E530A3F

LOL... Well, it's definitely Y-Main related...

I had the reference TV on for about an hour and it turned itself off while I wasn't even near it but I saw it go off. This time, however, I couldn't power it back up no matter what I tried. I was dismayed, thinking now I have two totally dead sets on my hands and was back to having no working reference. I even tried stiffening the Vs with extra capactitors, using an external supply for the Ve, and a bunch of other things, all to no avail.

While trying to figure out if it was even somewhat temperature dependent since I'd had the darn thing running a couple hours ago, I eventually decided to put one of the Y-Mains in the freezer for a few moments then put it back in the set and low and behold, the set powered right up on the first try!

I also realized while using the external power supply to force the Ve that I'm also losing most of my -Vscan (only gets to about -64 volts) which I hadn't noticed before but I can only probe so many places at the same time, or so it would seem.

Point being, next I'm going to concentrate my sleuting efforts in the -Vscan supply area and potentially the ramp generators, etc. to see if something is obviously flakey there since I believe that more stuff like ramp shapes on the Y-Main is fed back to the logic board rather than, say, the Ve directly. I need to probe some things while this one is actually up and running but I'm also puting the other Y-Main in the freezer right now to see if it still powers up also just as a test to be sure everything CAN still work.

The lack of Ve seems to be a symptom more than a cause.

At least I'm finally getting somewhere

Additional note: The reference set just shut down again while I was finishing writing this and, interestingly, the Yfr (forward side of the ramp, I assume?) heatsink is toasty while everything else is still cold after just this few moments. Perhaps I should stick a scope on there and see what things look like... Hmmmm...

[drussell]

Re: Stumped on Samsung PN60E530A3F

Yep, the second set powered up first try with a chilly Y-Main also...

[lookimback]

Re: Stumped on Samsung PN60E530A3F

Could the logic board be causing this? I had a very similar issue with an old Viore a couple months ago and I thought it was the main, so I parted it out and tossed the main in my scrap pile. Well, the logic board sold and my buyer informed me that he was having trouble getting the TV to turn on. And what was described here, having to cycle power several times before getting it to turn on, and that it would only do it after it had been off for a while, describes the issue exactly. I know it's a different TV, but thought this may help.

[drussell]

Re: Stumped on Samsung PN60E530A3F

I'm sure there are many faults in many TVs that will show a similar symptom set which is what makes it difficult and frustrating to track these kinds of problems down... That's also part of the fun of it, of course, making it somewhat of a personal challenge to try to figure out just what is going on.

I think I have this particular problem narrowed down to the -Vscan power supply, either the Infineon 3BR0665JZ PWM controller IC, one of the capacitors right by there or one of the surface mount components on the back of the board. I think I can rule out the through-hole photodiode optocoupler on the top, I thought it might be that but seems to be to the right of that. I'm still sleuthing, though...

I keep trying to just hit each area carefully with the freeze spray to see what it is but it doesn't seem to actually work until I blast the area too much to tell what all I'm really hitting for sure so it may well be something on the back of the board but there aren't many components right there so I'm still not sure. It's definitely something in that area, I can hear it start to buzz and fire up when I hit it with the spray but haven't been able to 100% narrow down which part it is yet and it's going to be tricky if it's something on the back of the board. Grrrr... I keep having to wait for it to warm up enough to fail when turned off again, of course.

I hope this is the same problem that the other people here are having since it at least certainly looks like both of the ones I'm working on have the same issue on the Y-Main board in the -Vscan supply and it would be nice if everyone's issue can all be sorted out at the same time.

Interestingly, Infineon discontinued a revision of this chip and introduced a new revision about a year ago. Not stocked at Digikey (of course) but seems fairly available elsewhere if it's the controller IC.

Shades of the TCZAA6755 in the Panasonic TC-P42X5 et al with no standby power., perhaps? It could be essentially the same type of stupid issue with a defective batch of controller ICs that nobody wants to "officially" admit to.

If I let it run until it shuts itself off the Yfr heatsink is quite warm and while it does heat up faster in operation, it seems like it really runs away at the end. It ends up much warmer than the other heatsinks and it is one of the smaller ones so that still doesn't seem right to me either but that could be due to the -Vscan starting to drop out or something. Of course it could be completely unrelated. I guess I should at least put the DSO in Transient Recorder mode again on the -Vscan and let it run until it goes wonky or shuts off. I'm just trying to not be too hard on it until I find out what's actually going on. I don't want to damage anything unnecessarily.

Getting deadly close, though, I think...
At least I have it localized to a small physical area in one circuit with only a few components to figure out what they're doing...

[drussell]

Re: Stumped on Samsung PN60E530A3F

Ok, so the problem with both of these sets here IS the Infineon 3BR0665J at U5014 right below the transformer for the -Vscan on the Y-Main board.

Using the datasheet and tracing the circuit, I verified that it is getting power by looking for +15v on pin 7 (VCC, also available on the right leg of C5162 and on the J5101 jumper above it), that the feedback voltage on pin 2 (FB) is up at close to 5v (saying, "Hello, Mr. Chip, we need more voltage", which is the inverse of how I've designed things in the past with a TL494 or SG3526, etc.), that the current sense pin 3 (CS) is not indicating an overload since it's showing no current flowing because it is at 0v drop across the current sense resistor, etc., so all look great, BUT...

Then we get to the BA, pin 1, which has two purposes. One purpose is to adjust the blanking time for the overload holdback but is not required. The other purpose is to manually put the chip into "Auto-Restart" mode essentially cutting it off and forcing a restart of the cycle from an external source.

This pin is supposed to be pulled up internally to 0.9v by a current source but shows 0v. The only circuitry connected to this pin in this implementation is a 0.1 uF capacitor to ground to add a little extra blanking time, there is no external circuitry that can be holding it down and, indeed, removing the capacitor made no difference. According to the block diagram, I'm guessing the current source must be defective, the zener diode is shorting or some other internal problem.

When I spray freeze spray on it, the voltage jumps up to 0.9v and it starts to oscillate, the output of the DRAIN pin looks like normal PWM on the scope and -Vscan comes up properly. Prodding it externally with 0.9v to try to make it come up does not work.

So, everyone who may have this problem, check the voltage on pin 1 (the lower right pin, indicated by the arrow on the PCB) with your meter's negative lead to ground and if you do NOT see the 0.9v when you power up the set, suspect this part. If you have freeze spray, give the chip a little squirt and you should hear it start to oscillate if the power is on (you'll have to then turn the set off and back on to get the high voltages to come up for a picture) or turn it on if it was off while you sprayed. Both of these ones here come right up and work fine until the next power cycle.

As a sidenote, the heatsink temperatures do vary with what is being displayed on the screen, some get hotter than others depending on the picture and while they're warm, they do stabilize so I guess that's just how they're sized. The sets here were eventually shutting down due to the inactivity timer even on an analog snow screen because they knew it was a weak/no signal condition. With a proper video signal coming in I ran them for several hours just fine. (This makes sense because when they were in operation they would both work fine if you could get them to actually power up.)

I ordered some of these ICs, I'll report back with final results when they eventually arrive but it certainly looks like this is the whole problem!

[drussell]

Re: Stumped on Samsung PN60E530A3F

Hmmm, I just read some of the other threads about this chip like https://www.badcaps.net/forum/showthread.php?t=38651 and had seen the "operating range of 10.5-25v" but hadn't noticed the VCC turn-on threshold of 18v typical at the bottom of the specification tables.

This seems strange to me as even with the zener diodes across the VCC capacitor and the optocoupler supply removed I still only get 15.something volts. I will have to look into this some more but I would appreciate any thoughts from anyone who has any experience with this ICE3BR0665J chip...

[MagicSmoke]

Re: Stumped on Samsung PN60E530A3F

I am so glad you found that. I put my Y main in freezer.....and lo and behold it came on. It stayed on long enough to do a channel search then remote became unresponsive but still had picture.

Off now to buy a Infineon 3BR0665J and change it out.

Many Thanks for taking the time to figure it out.

[drussell]

Re: Stumped on Samsung PN60E530A3F

It stayed on long enough to do a channel search then remote became unresponsive but still had picture.
...
Off now to buy a Infineon 3BR0665J and change it out.

The remote becoming unresponsive could be something simple like dirty power to or on the "Main Board" (the board with the inputs) since that's where it seems like the remote processing is done but it's SO much easier to find things like that when you can actually get the set powered up! Also, make sure all your screws are installed and nice and tight.

Let me know if the chip works for you if you find one before I get one to try. I'm not 100% convinced it is the chip itself anymore now that I see the 18v threshold thing, though it may well still be the chip as that is what is responsible for charging the VCC capacitor at 0.9ma so unless something else is holding it down to 15v but yet still works once it's running and it really won't start until 18v, changing the chip may well fix it.

I just find it strange that it's going right up to 15v and holding there if it's anything else except the chip itself but I'm going to pull one of these Y-Mains out again and take another look at the circuit in case I can spot something else that could be causing it since the chip does run fine once it starts!

At least we now know exactly what circuit is causing the problem!

Good luck!

[MagicSmoke]

Re: Stumped on Samsung PN60E530A3F

Found some on ebay in Florida http://www.ebay.com/itm/ICE3BR0665JZ...item5d45854d8a

Sent an offer for 2 as I have 2 Y mains with same problem. Both boards I have will come on after freezing. One of them works longer than the other. One only works about 15 min the other almost 30 min. Then no picture only sound.

Will let you know.

[drussell]

Re: Stumped on Samsung PN60E530A3F

Since they wanted all their TVs up and working for football, I did a temporary hack job yesteday to make these sets able turn on until the ICE3BR0665J chips arrive.

I made a voltage divider using about 120K ohm (needs to be at least a 0.5 W) from the Vs to a piece of wire and also to about 13K to ground (0.25 W is fine), heat shrinked it up and connected the wire to the VCC of the PWM controller. Check to be sure you don't get more than the 25V maximum rating of the controller chip before you connect it to the chip's VCC at the capacitor, the jumper or the VCC pin itself. It measured just over 20V here but will depend on your Vs voltage and the exact resistors used.

This charges up the VCC capacitor to above the 18v turn-on threshold regardless of the flakey precharge supply in the chip and makes the TVs come on first try. Not the best permanent solution but really shouldn't matter either.

At least it works.

[AFajitas]

Re: Stumped on Samsung PN60E530A3F

Since they wanted all their TVs up and working for football, I did a temporary hack job yesteday to make these sets able turn on until the ICE3BR0665J chips arrive.

I made a voltage divider using about 120K ohm (needs to be at least a 0.5 W) from the Vs to a piece of wire and also to about 13K to ground (0.25 W is fine), heat shrinked it up and connected the wire to the VCC of the PWM controller. Check to be sure you don't get more than the 25V maximum rating of the controller chip before you connect it to the chip's VCC at the capacitor, the jumper or the VCC pin itself. It measured just over 20V here but will depend on your Vs voltage and the exact resistors used.

This charges up the VCC capacitor to above the 18v turn-on threshold regardless of the flakey precharge supply in the chip and makes the TVs come on first try. Not the best permanent solution but really shouldn't matter either.

At least it works.

[drussell]

Re: Stumped on Samsung PN60E530A3F

I take it that means that this helped you out?

Just to be clear, in case someone new to electronics reads this, I mean this string:

[Vs] ---- [120K] ---- [wire to VCC comes off here] ---- [13K] ---- [GND]

[MagicSmoke]

Re: Stumped on Samsung PN60E530A3F

It seems that the chip is not the problem. Received 2 new chips today.

Still had to shoot it with freeze spray to get it to come on. Only showed picture for about 10 min then quit.

Is it possible that a resistor or cap has changed value enough to prevent it from working? I have changed all but the large green electrolytic caps.

[drussell]

Re: Stumped on Samsung PN60E530A3F

It seems that the chip is not the problem. Received 2 new chips today.

Is it possible that a resistor or cap has changed value enough to prevent it from working? I have changed all but the large green electrolytic caps.

Very strange... I can't see how it could be anything else since there is very little connected to that VCC pin. The photodiode optocoupler is connected to that supply also, I suppose it could be bringing it down somehow.

I tried removing the zeners thinking that they might be the problem but that doesn't change anything and interestingly the zener voltage must be above the 30v max of the power supply I was using to test them because they didn't pull it down at 30v input. You'd think the zeners were supposed to be 25v or less to protect the chip, otherwise what on earth are they for if not to prevent an overvoltage to the chip but they still test .7v in the other direction so they're not blown open either.

There's really only that and a few capacitors across that VCC and the chip is the one doing the precharge, so I was back to really thinking it would be the chip itself after checking basically everything else connected there but I don't have another of that kind of optocoupler here to try.

You could try changing the cap at C5162, it's a 10uf multilayer ceramic but an electrolytic should work, though I tried swapping one with one of the snubbers and it had no effect. SOMETHING is limiting the voltage there and now I really don't know what or why!

Still had to shoot it with freeze spray to get it to come on. Only showed picture for about 10 min then quit.

I'm guessing the fact that it shuts down means you still have another problem with that set. The ones here work completely fine and stay on indefinitely once they are actually started.

Very, very strange...

[MagicSmoke]

Re: Stumped on Samsung PN60E530A3F

Just tried hitting only the new chip w/ freeze spray, set did not come on.

Sprayed optocoupler with a quick burst and set came on. I was careful not to overspray so just the opto got cold.

I am going to check a few psu's I have to see if I can find a suitable replacement opto to try.

[MagicSmoke]

Re: Stumped on Samsung PN60E530A3F

Just figured out it is the 106 cap @C5162.

Froze it to turn set on set will run about 6 min then no pic.

Then while set had pic, put flame of lighter to cap and instantly no picture. Refreeze and it comes right back on.

[drussell]

Re: Stumped on Samsung PN60E530A3F

Just figured out it is the 106 cap @C5162.

Froze it to turn set on set will run about 6 min then no pic.

Then while set had pic, put flame of lighter to cap and instantly no picture. Refreeze and it comes right back on.

Intesesting. Here it seems to definitely be the chip freezing that works. I tried several times to spray things like the cap and it seemed to make no difference. It turns right on when hitting the chip, though and I did try the capacitor swap on one of them but maybe it is leaking above a certain voltage too.

I am just about to hook it back up to the DSO and watch the voltage closely in various configurations just to see if I can verify things. More sleuthing!

What voltage were / are you seeing on the VCC for that chip?

Does it come back on without having to power cycle or do you have to power cycle after the refreeze?

[MagicSmoke]

Re: Stumped on Samsung PN60E530A3F

Volts at VCC while running is 15.3v. Checked at leg of cap. When set looses picture volts go down to 9.6v. Hit cap w/ freeze and volts go back up to 16.1 before stabilizing at 15.3v, have to recycle power on to get picture. It does not come back on itself.


I changed the cap to an 10uf 50v electrolytic but have not installed the board yet to try it out.

[drussell]

Re: Stumped on Samsung PN60E530A3F

Volts at VCC while running is 15.3v.

That's about where they run here too but I've seen it vary between 13-15 volts. It will vary because once it is running it is being supplied by output from the transformer so is going to depend on the pulse width due to any varying load on the -Vscan which is being sensed for feedback.

I should have been more specific, though, I'm also particularly interested in how high it goes briefly at startup. Is your meter fast enough to monitor it at startup and see a reading or have a min/max mode you can use to watch for the highest voltage it gets to on startup?

Thanks for the info! Appreciated!