SMPS-TV Set

Discussion in 'The Projects Forum' started by pbones, Aug 18, 2014.

  1. pbones

    Thread Starter New Member

    Aug 18, 2014
    18
    3
    Hello I am new to the forum (please be kind):)
    Weeks ago my TV (Westinghouse) Internal PCB's stamped (Samsung). Except the power supply is made by Haider (stamped). The TV would turn on and stay on for 20-30 minutes then shut off (screen would go black) power light would stay the same color (blue) but the screen would not come back on. I checked the output of the power supply and when the screen would go out the output(s) would drop to zero volts. Everyone seems infatuated with changing the CAPS, not seeing any abnormal CAPS I changed them all anyway. Then the TV would come on for only 5 minutes shut off, after power cycling the tv it would work again for about 30-90 seconds then black out again. I probed around and found two shorted resistors (15.8Ω 1 watt 1%) I replaced them and now I am getting a weird sound from the large transformer, prior to the change I was getting a hum from only the smaller transformer, now it seems from both. Prior to the last change I had 12V, 5v and 24 Volts now the 12 volts is 8v, 24v is 17, 5v is 5v 5V sense is zero. This is very frustrating what should I be looking for? The problem appears to have changed after replacing the resistors, but for the worst. Although I know there must be more problems I do not know how to diagnosis this power supply. Please help. thanks:confused:
     
  2. ronv

    AAC Fanatic!

    Nov 12, 2008
    3,282
    1,236
    It is very unusual for resistors to short. Could they have been inductors? Sometimes they are marked in a similar manner.
    Without a schematic you are in for a rough time.
     
  3. takao21203

    Distinguished Member

    Apr 28, 2012
    3,577
    463
    Pictures? From the board and the resistors, the old ones you removed and the new ones.

    Caps are not always caps- the ESR could be different and the frequency response. Often they use a specific kind of capacitor and general purpose ones wont do properly or not at all.

    Have you probed the old ones?
     
  4. pbones

    Thread Starter New Member

    Aug 18, 2014
    18
    3
    I have include some pictures. One picture shows the six band resistors (x2) that are shorted. The DVM shows .2 ohms but my meter leads shorted together read s.1 ohms. The Color band is Brown,Green, Gray,Gold,Brown (15.8 ohms 1% 3/4 watt).
    I followed the 24V and 12 V circuits back to the Schottky rectifiers one for each voltage, and the low voltage readings are present there behind the low voltage output capacitors. The voltage at the Schottky is still low though. One of The Schottky Barrier Rectifiers where the low voltage is measured is also pictured. I assume the larger transformer is for 12 volts and 24 and the smaller transformer is for the 5 volt. 5 Volts seems OK.

    I removed the two Mosfets (on the high voltage side of the transformer) and attempted to test they seem OK with the diode test. Although, i did not heat sink them when removing or re-installing they might be bad now???

    If i understand the SMPS does not realy on transformer ratio to produce the low voltage output, there is a flyback ( or buckboost (like) circtuit) that turns the power on/off creating no wasted (theory) energy but increasing the frequency creates the proper output. Correct?:confused:
     
  5. to3metalcan

    Member

    Jul 20, 2014
    228
    23
    Nothing wrong with 'em...they're 0.15 ohm resistors. That's a silver band, not gray.
     
  6. pbones

    Thread Starter New Member

    Aug 18, 2014
    18
    3
    Are you confident, or does the color scheme seem to make more sense with silver band instead of gray? ( Im not being rude just asking).

    If those really are good resistors then placing them back into the circuit leaves me with the same question. What is causing the low voltage output to shut down after 5-10 minutes and not come back until it "cools" (sets in the off position) for 20-30 minutes? this happens with or without the TV display connected, and the only item connected at all times would be the TV output board. Any thoughts?
    I do not have an ESR meter which seems to be a problem, and not being able to rely on new capaitors to resolve the problem ( I am stuck reluctantly) putting all the old caps back in the circuit. each one tests good using the Cap selector on my DVM (within 10%) for all the stamped values. I knew this would be difficult as there is no schematic...any help or thoughts would be appreciated.
     
  7. to3metalcan

    Member

    Jul 20, 2014
    228
    23
    I am confident for a number of reasons. One is that immediately upon looking at them, I identified them as .15's given the color bands present, and atually had to look back at your post to see what you thought the were. Grey stripes are glossy, not metallic. And small-value, high-wattage resistors like this are extremely common at the output of solid-state amplifiers, which are my business.

    The second reason is that resistors don't fail short. They burn up and fail OPEN. Yes, there is a one in a million chance that a large resistor like that could melt down to the point that it would short across the leads, but A) I've never seen it happen, B) the resistor would be visibly physically damaged (read: charcoal) and C) the odds of it happening twice are about as good as your odds of flipping a quarter and having it land standing on the edge.

    I've fixed a few switching power supplies...a glitchy fault (as opposed to a massive failure) like what you're describing might be something in your power supply, but it's at least as likely that the supply is doing its job and shutting off because something on the output board is drawing too much current. To rule that out, disconnect EVERYTHING from the power supply board and see if it maintains voltage. Your meter is fine...I have one of those, too, and it might not be a Fluke, but it'll get the job done!
     
  8. to3metalcan

    Member

    Jul 20, 2014
    228
    23
    (Also, make sure all your electrolytic or other polar caps you installed are facing the right way!)
     
  9. pbones

    Thread Starter New Member

    Aug 18, 2014
    18
    3
    Thank you for taking the time to answer. I was/am not questioning your ability. You seemed certain about the size and needed to know how.
    As for the supply I have started measuring the current draw in series from the different boards. Although nothing jumps out. I assume that removing all the current draw will require a dumby current to fool the power into turning on like shorting 5v sense to ground using a resistor. Would you happen to know the resistor size to use?
     
  10. to3metalcan

    Member

    Jul 20, 2014
    228
    23
    Based on the specs, 5V / minimum current (.1A) = 50Ω. I'd give it a little more than the absolute minimum, though...a 33 ohm 1W resistor should work.
     
  11. pbones

    Thread Starter New Member

    Aug 18, 2014
    18
    3
    OK the test is complete. Everything disconnected from the power supply. Shorted the Standby output (5v) to the 5VSB pin through a resistor. I checked all the low voltage outputs at the beginning of the test, then 15 minutes in and then after 30 minutes. This is what I found.

    At the beginning of the test and 15 minutes in...the outputs measures 24=24.6, 12=12.7, 5v=5.01. The final test of the outputs after 30 minutes revealed less favorable results. The 24v=0, 12V=0, 5V=5, 5VSB=5V, STBY=5V. Basically the only voltage left operating was 5V. At this point the power supply is without a doubt, defective.
    I summarize that the 5 volt circuitry seems to be unaffected by the fault. The focus is the 24 V/12V components.
    I checked the polarity of the caps and they are correct. Where do I go from here. I assume I need to find components that are common to both 24/12 volts...I have been up and down this board. I would be grateful for any recommendations that you might have.
     
  12. to3metalcan

    Member

    Jul 20, 2014
    228
    23
    I'm not really sure about the test you've run, here...I think maybe we need to start a step back from here.

    According to the document you posted, each of the supplies has a minimum current-draw necessary to keep it from going out of regulation and turning off. For each of them, the resistor you would use for a test-load would be the voltage of the tap (ie, 12V) divided by the minimum current (ie, .3A) = resistance in ohms (in this case, 40 ohms.) You should then go one size smaller a resistor than that (ie, 27 or 33 ohms) so you aren't riding the edge of the minimum for regulation. These should be shorted to cold-side common (ground pin at the output jumper.) The 5V standby is a separate voltage rail that never turns off (it keeps the TV live so it can respond to signals from the remote, etc.)

    TL;DR: Supplies of this type are not out of spec if they turn off when there's no current draw on them. Your test has proven that the 5V tap is okay, but it's not necessarily weird that the others turned off.

    Ideally, you'd test them all at the same time to get a realistic idea of the thermal forces at work, but you don't necessarily have to...you can test one tap at a time and the unloaded rails will just turn themselves off.
     
    Last edited: Aug 23, 2014
  13. pbones

    Thread Starter New Member

    Aug 18, 2014
    18
    3
    I attempted to short 5VSB to ground using the resistor as the load. But the power supply did not turn on. The only way it would turn on is when Stby was shorted to 5VSB through the resistor. Then the outputs came on. Smashing it comes to mind...then I come around and realize I'm interested in finishing the repair. I used 32 ohm resistor to ground in both tests. If the power supply should come on with the resistor to ground it did not.
     
  14. pbones

    Thread Starter New Member

    Aug 18, 2014
    18
    3
    I get what's happening. I will test each tap and let the others shutdown. Probably finish this in the am. Thanks again.
     
  15. bwilliams60

    Active Member

    Nov 18, 2012
    722
    88
    pbones, what Westinghouse model are you working on? The tag will be on the back of the set on a tag. I may have a schematic for you depending which one it is.
     
  16. to3metalcan

    Member

    Jul 20, 2014
    228
    23
    Pbones, just to be clear, what you did as far as shorting the 5V to standby was exactly right! The unit would send the 5V standby to the TV, and then if it got it back at the STANDBY input it would turn on. There's just an extra step to keep the outputs running.
     
    pbones likes this.
  17. pbones

    Thread Starter New Member

    Aug 18, 2014
    18
    3
    The model # stamped on the back of the set is VR-4625
     
  18. pbones

    Thread Starter New Member

    Aug 18, 2014
    18
    3
    well, I do not have the correct (loads) /resistors..etc to draw the correct current off the 24V/12V outputs.

    I have mostly 1/4watt resistors, small case fan (only draws .12A) and some random 1/2watt, 3/4 watt, and some 1 watt but nothing that will work.

    I have some items on my work bench (obsolete parts) from older machines that i might be able to use for the test. I'll have to check.
    For the loads I plan to use 24 Ω resistor, 6 watt, in the 12V output, and 60Ω resistor, 9.6watt, on the 24V output. This should keep the current 100-200mA above the lower limits. ---More to come...:)
     
  19. pbones

    Thread Starter New Member

    Aug 18, 2014
    18
    3
    I finished the test. I loaded down all the outputs above there min. Requirements and after 4 min 56 seconds the outputs shut down. No other board was connected to the supply during the test, only the dummy loads. As for the loads, I used some old solenoid valves in parallel for the loads. Two for the 12 volt for a draw of .690 amps, two for the 24v output that draws .860 amps. I loaded down the 24volt video output along with the seperate 24 v output connector. After the power to the outputs goes out and on the 12 v and 24 v the 5 v stays live all of them 5v, STBY,and 5 VSB.
    I have 120 vac across the bridge after the outputs die off and 166 CDC on the output of the bridge and 166 vdc on the large filter capacitor. At this point I can safely assume the problem is on the power supply?! I then would ask for some guidance once the outputs die off can I or should I measure for voltages at other points, like the transistor or ????
     
  20. to3metalcan

    Member

    Jul 20, 2014
    228
    23
    You've already replaced the caps (which would usually be the next step), right? Do you have access to an oscillosocope?
     
Loading...