samsung un26 / un26d4003bd power supply failure

[cashkennedy]

I picked up a 26" samsung thin led tv only a year and 6 months old on craigslist. Owner had already taken it to the local factory service place, and they wanted 160 or something to replace the power supply board.

I took it apart and probed around the power supply. Getting 170 volts across the caps (there are 3 but i think they arent in series, just in parallel because this is 1 inch thick tv, so they had to use a lot of tiny caps). Im also getting 170 across the primarys of the transformers. I dont get any voltage on the secondary side, and when i test the secondaries , every pin is continuous to ground, and they are all very low resistance to ground and between each other. Specifically though the tranformer 2 for the LED backlight the output has 1.4 ohms vs the ground pins on the transfomer. On transfomer 1 (12v and SB), the all are essentially 0 ohms between every pin.

What parts should i check, that are common causes of a completely grounded secondary?

[selldoor]

Re: samsung un26 / un26d4003bd power supply failure

Sorry cashkennedy, I cant believe how poor the pics are from such an experienced member. Do you really think we can spot anything on them?

[tom66]

Re: samsung un26 / un26d4003bd power supply failure

There's no point in testing the resistance of SMPS transformers; that only works for inverters. I would test output diodes and replace the startup cap.

[ben7]

Re: samsung un26 / un26d4003bd power supply failure

Looks like there might be two bad caps on the PSU board, although the [last] picture is very fuzzy so they might look bad even if they are not.

[cashkennedy]

Re: samsung un26 / un26d4003bd power supply failure

Yes selldoor, my pictures arent very good. I know where to take readings / what parts are called / how to spot bad caps / and how to spot bad solder joints. Because of that i dont really need yall to examine the board yet in clear detail. (If it comes to that ill post better pictures)

I was just looking for someone like tom66, or budm to give me 2-3 parts that commonly fail on the secondary (other then filter caps obviously) that i should test first. Posted pictures cause not all power supply boards use the same methods to acomplish the same goal such as which ground they use.

The startup caps on the primary side are a little buldged (but a tiny buldge is kinda common for tiny little caps like that, which dont even have a vent), but that blurry picture makes them look much more buldged then they are. I've never really understood what the startup caps goal is in a circuit, so im not sure if it could be the culprit in this case. But im getting 170volts across the primarys of the transfomers / which i think is correct for an off state where only standby voltage is supposed to be on. And i believe the startup caps only come into play right when the set is plugged in / or turned on?

[tom66]

Re: samsung un26 / un26d4003bd power supply failure

From a previous post:

The controller IC (8 pins usually, sometimes more or less) needs about 10-12V DC of power, from 115/230V input.

In order to provide this a high value resistor charges up the cap and a zener regulates it. Once the voltage reaches a set point determined by the controller, the controller starts using the energy in this cap to run and turn the output on.

Once running an auxillary winding takes over and provides power to the controller IC, like an alternator in a car supplies power to recharge the battery.

The cap is most important in the brief time at which energy is being drained from it by the controller IC, but no energy is being provided to it. If its capacitance is too low (or in some cases if ESR is way too high) it will be like a car with a flat battery, unable to start. And you will hear a characteristic ticking from the PSU sometimes.

I say just replace 'em. They're dirt cheap. No need for a low ESR cap here, although there could be a reliability advantage achieved by using long-life general purpose caps.

[cashkennedy]

Re: samsung un26 / un26d4003bd power supply failure

Unfortunately this is a 1" thick tv so as you might have noticed you have to be quite carefull with getting the right size caps so ill probably have to order some and wait for them to come in, cause most of my panny fm's are larger then general purpose caps to begin with.

Ill eventually get around to replacing the caps on the primary if i dont figure anything else out, I just strongly doubt it could be causing my problem since i get voltage to the primary, but no voltage fromt he secondary side of the transformers.

[cashkennedy]

Re: samsung un26 / un26d4003bd power supply failure

How exact do you need to match the capacitance on startup caps.

The 3 small caps on the primary side are 2 47uf 50v, and 1 22uf 50v. I have some rubycon 47uf 50v's i can pop in, but the closest i have for the other is a 33uf 50v KMG series nippon chemicon.

I still dont think this will solve the problem, so reluctant to actually pay for any caps for this board. It appears the majority of those 3 small caps are just involved in a feedback circuit from standby and PSON. Perhaps thats what the 2 47uf ones are for and the 22uf one is the only startup cap. And from reading a description of the function of the startup cap , would going to a higher uF cause it to overvolt the ICC, or just give it more reserve to power the IC from.

[tom66]

Re: samsung un26 / un26d4003bd power supply failure

How exact do you need to match the capacitance on startup caps.

The 3 small caps on the primary side are 2 47uf 50v, and 1 22uf 50v. I have some rubycon 47uf 50v's i can pop in, but the closest i have for the other is a 33uf 50v KMG series nippon chemicon.

I still dont think this will solve the problem, so reluctant to actually pay for any caps for this board. It appears the majority of those 3 small caps are just involved in a feedback circuit from standby and PSON that uses a bridge (or whatever those 4 wire parts are that span the primary to secondary to report back to the primary. And since im not even getting any voltage on the standby i dont see how it could be using parts of the pson detection yet.

Those parts are optocouplers.

You must keep the capacitance the same or make it higher. I'm surprised you can't fit a 6.3mm(?) KMG.

It is likely the capacitor closest to the SMPS IC or the transformer. If you trace the circuit you will find a diode and sometimes a resistor leading to a tap off the transformer.

[cashkennedy]

Re: samsung un26 / un26d4003bd power supply failure

This board uses an odd so-224? (to-220 i think i mean) (i think thats the right name) package for the IC ... im used to seing 8 pin smd style ones. Ill have to look up the part number and see if i can find a spec sheet for it to figure out which is the vcc pin.

I didnt say i couldnt fit that 33uF KMG cap, that cap is freaking insanely small, smaller then what samsung used originally on this board, which kind of worries me. The original caps on the secondary are all samyoung (not a good sign), but the caps on the primary are some brand i dont recognize with a like triangle with a circle inside of it and the letters HXB.

BTW i edited my post while you were replying to sound more intelligent, lol but basically i figure the other 2 caps are involved more with the optocoupler.

[tom66]

Re: samsung un26 / un26d4003bd power supply failure

Think of the startup capacitor like a car battery - you could install a battery with higher cranking amps (higher capacitance for a capacitor) but not lower. The cap has to supply the energy during the brief ~50ms when the switcher is drawing power but not receiving anything yet from the transformer.

[cashkennedy]

Re: samsung un26 / un26d4003bd power supply failure

What is it(the IC) supposed to acctually do though? increase the voltage going across the transformers to higher then 170v? Or is there only supposed to be power across 1 winding on the transformer till it receives the on signal, then it powers the other winding?

[cashkennedy]

Re: samsung un26 / un26d4003bd power supply failure

http%3A%2F%2Fwww.allegromicro.com%2Fen%2FProducts%2FSanken-Products%2FSanken-ICs%2FSanken-Regulator-ICs%2FSanken-Off-Line-Flyback-Regulators%2F~%2Fmedia%2FFiles%2FSanken%2FDatasheets%2FSTR-W6051S-STR-W6052S-STR-W6053S-Datasheet.ashx&ei=QBPJT8eHF8nM2AWyrPHaCw&usg=AFQjCNGEeysCBm1MxHTj5nyVl8H-3pH79g

I doubt this link above will work, but its a str-w6052s IC so Vcc is on the 4th pin (pin 2 is missing).

Both caps appear to me to be possibly involved with the vcc pin ... so not sure if either is clearly the one for startup.

Picture is closeup on IC and surrounding traces / caps. Didnt realize they were hiding zener diodes in black silicon packages these days...

Damn is shrunk my picture since i uploaded it online first... might have to figure this one out myself, or post a higher res version later.

[tom66]

Re: samsung un26 / un26d4003bd power supply failure

What is it(the IC) supposed to acctually do though? increase the voltage going across the transformers to higher then 170v? Or is there only supposed to be power across 1 winding on the transformer till it receives the on signal, then it powers the other winding?

The switcher IC...

Any one of millions of possible topologies for driving the transformer to convert the 170V or 340-400V DC into low voltage DC for driving the rest of the TV. The LED backlight usually uses series strings and around 60 to 100V.

[cashkennedy]

Re: samsung un26 / un26d4003bd power supply failure

Anyways i just replaced all 3 caps, no change. As i said, I think if im getting 170v across the primary of the transformer and nothing on the secondary, then its most likely a problem on the secondary side.

Is there a way to test what the voltage of Vcc is while the power is connected. I tried vcc against ground and only get .14 volts or so but that quickly disipates to .02v, so im thinking the little bit of power in the startup cap gets drained by the multimeter testing it.

Whats suposed to power the Vcc after the startup cap? The diagram for common method to use the IC shows a winding on the transformer that would power Vcc? Are all the windings on the primary not suposed to be either - or + ?

[tom66]

Re: samsung un26 / un26d4003bd power supply failure

Anyways i just replaced all 3 caps, no change. As i said, I think if im getting 170v across the primary of the transformer and nothing on the secondary, then its most likely a problem on the secondary side.

Is there a way to test what the voltage of Vcc is while the power is connected. I tried vcc against ground and only get .14 volts or so but that quickly disipates to .02v, so im thinking the little bit of power in the startup cap gets drained by the multimeter testing it.

Whats suposed to power the Vcc after the startup cap? The diagram for common method to use the IC shows a winding on the transformer that would power Vcc? Are all the windings on the primary not suposed to be either - or + ?

You need to test Vcc across the cap - not relative to ground as the primary is isolated.

Nothing across secondary - your meter is probably not fast enough to pick up kHz level switching - but it could be primary OR secondary problems causing that.

The cap has quite a lot capacitance and the meter has a very high impedance so the cap won't discharge quickly. ~5RC = time to discharge. Typical R for meter = 10 megohms, typical C = 47uF, time = 2,350 seconds to discharge to ~3% of original voltage.

Once controller has detected cap has charged, it then starts switching. The auxillary winding then keeps the cap charged.

[cashkennedy]

Re: samsung un26 / un26d4003bd power supply failure

Tested the 3 small caps on primary for voltage (negative lead to negative side of caps ) and got 3v on 1 of the 47uf, 0v on the other, and i was getting nothing on the 33uF cap and it just popped and blew (i might have touched a lead to the wrong thing as there were some other parts quite close)

[tom66]

Re: samsung un26 / un26d4003bd power supply failure

Tested the 3 small caps on primary for voltage (negative lead to negative side of caps ) and got 3v on 1 of the 47uf, 0v on the other, and i was getting nothing on the 33uF cap and it just popped and blew (i might have touched a lead to the wrong thing as there were some other parts quite close)

I'm not sure how you popped the cap with the meter. Something else might be damaged.

[cashkennedy]

Re: samsung un26 / un26d4003bd power supply failure

There is a jumper / wire or whatever that connects to parts of the positive rail right next to the capacitors leeds. Im guessing i connected that to the positive on the cap (theyre within about 3 mm of each other) so 160v went across the cap...

The sheet on the IC says Vcc start is 30v, and Vcc on typical is around 15v, and says thats with terminals 4-5 (4 is vcc, and 5 is "ground"/negative). So is this basically saying that i could test from point 4 to 5 and should get 30v for a microsecond or something and then 15v afterwards?

[tom66]

Re: samsung un26 / un26d4003bd power supply failure

There is a jumper / wire or whatever that connects to parts of the positive rail right next to the capacitors leeds. Im guessing i connected that to the positive on the cap (theyre within about 3 mm of each other) so 160v went across the cap...

The sheet on the IC says Vcc start is 30v, and Vcc on typical is around 15v, and says thats with terminals 4-5 (4 is vcc, and 5 is "ground"/negative). So is this basically saying that i could test from point 4 to 5 and should get 30v for a microsecond or something and then 15v afterwards?

You should read an average of around 15V, there is a fairly wide tolerance on this. Since the cap has blown up, are you sure the SMPS IC hasn't been damaged by this 160V surge?