Samsung LN32B530P7F PSU issue

[b4ndit]

A coworker gave me a Samsung LN32B530P7F television for free because it would not power on. I suspected bad caps, but they all appear fine and none of them were shorted when I checked their continuity. I looked over the entire PSU and could not find any obvious problems.

Symptoms: No power on, no lights, no noise, nothing.

I got out my Fluke 114 multimeter and tried to help trace down the problem. I grounded the meter to one of the screws on the border of the PSU board (LUG4). I then used the other lead to check the AC side of the board. I'm showing ~67.5 volts AC when I touch it to the aluminum heatsinks on the AC side. Both fuses seem fine, they both have continuity and there is no voltage drop from one side to the other (of the fuse) when the board is given power.

When I go over to the DC side of the board, there is no voltage at all (same grounding point as before). The weird thing I noticed besides not having voltage is that it seems that most of the components are grounded. When I check continuity on a lot of the DC components (such as the striped side of CM871), I get 0 ohms to ground. This could be completely normal, but it just seemed odd to me.

I searched the forum and apparently another individual had the same problem with the same TV, but he never provided details so his request went unanswered. So, I'm going to provide pictures of what I'm looking at along with the board schematic I found online.

The board is a BN44-00261A.
Pictures of top: http://imgur.com/a/twA6U
Pictures of bottom: http://imgur.com/a/jzeKF
Schematic:

I am hoping you guys have a better idea of what the problem could be than I do. I'm at the limits of my knowledge on electronics as I've run out of things to check without help.

[budm]

Re: Samsung LN32B530P7F PSU issue

"I grounded the meter to one of the screws on the border of the PSU board (LUG4). I then used the other lead to check the AC side of the board. I'm showing ~67.5 volts AC " OK, when you make measurement in the primary side, you do not use the safety ground (chassis ground as the ground ref for the meter). For Primary AC feed reading you will measure between Hot and Neutral. The primary side DC circuit ground ref point in the primary is the negative leg of the main filter cap (that big yellow cap laying on its side). So check the DC voltage RIGHT AT THE TWO LEGS of that yellow cap, it should show around 165VDC, and it should go up to around 360~400VDC when PS_ON is present.

[b4ndit]

Re: Samsung LN32B530P7F PSU issue

I'm showing 168VDC on the yellow filter cap. Hitting the power button on the side of the TV does not seem to do anything and I still measure 168V on the filter cap.

When I've searched for similar problems, it seems that most of the time they indicate that their TV's red power LED is lit in some fashion. That is not the case here as the red LED is completely off.

EDIT: I checked the ribbon cable connecting from the PSU to the main TV circuit board. I used the GND at the far end of the ribbon cable (opposite of the power on wire) and I checked each of the pins on the cable. They all read 0VDC. I also checked with the ribbon cable unplugged (touching the red multimeter lead to each pin carefully as I hold the black lead to the labeled GND in the pin header) in order to rule out the ribbon cable. The pin I used as a GND has a 0.2 ohm resistance to LUG4.

I'm also going through your LCD TV troubleshooting guide in order to see if I can learn a little more about what is going on here.

From your guide, it seems that the red LED is not on because the standby 5VDC is at 0VDC. This seems to confirm the problem is located within the PSU itself. Your guide's common failures section indicates blown AC fuses (checked fine), bad capacitors (look fine, not shorted), and shorted diodes or transistors (have no yet checked). I'll see if I can check a few of these for a short today.

[tom66]

Re: Samsung LN32B530P7F PSU issue

What voltage do you measure across capacitors CM804 and CM808 on the primary side of the power supply?
Is fuse FM802S good?

[b4ndit]

Re: Samsung LN32B530P7F PSU issue

The fuse FM802S measures 0 ohms when I take it out and check continuity, so it seems fine to me. I also checked voltage from both sides of the fuse to the negative lead on the filter cap and it read 167VDC both times. If you know of a better way to test it, let me know.

CM808 measures between a range of 9 and 15 VDC. It seems to fluctuate pretty rapidly.

CM804 measures about 0.6 VDC.

[tom66]

Re: Samsung LN32B530P7F PSU issue

Ah, we had a power board like this on badcaps before, giving similar problems.

Try checking ZD803 and ZD804 for shorts.

If they are OK, replace UM801S (STR-W6252) and CM808.

[b4ndit]

Re: Samsung LN32B530P7F PSU issue

ZD803 and ZD804 both register 2.5 M ohms in one direction and OL (none) in the other. I would assume this would be fine (not a short).

I assumes this means I should look to replace UM801S and CM808, correct?

[tom66]

Re: Samsung LN32B530P7F PSU issue

One last check, forgot to mention. Check the following diodes for shorts:
DM857, DM852, DM851, DM801 (primary), DM802 (primary), DM803 (primary)

If no shorts replace that IC and capacitor.

[b4ndit]

Re: Samsung LN32B530P7F PSU issue

I had to Google how to measure the three prong diodes as I have never done it before and didn't know how to check it. I learned that the important bit is to measure from the center pin to the outside pins to ensure they don't have a short. I also measured the outside pins just to be thorough.

DM857 - Measured 18 M ohms with positive center and outside ground. -0.7 M ohms with ground center, positive outside. Outside to outside measured 0 ohms, but this should be OK (if I understand everything correctly).
DM852 - Meter registered OL and -OL
DM851 - Fluctuated between 2 K ohms to 18 M ohms with center positive outside ground. -1 M ohms measured middle ground, outside positive. Outside to outside measured 0.1 ohms.
DM801 - Measured 39 K ohms one direction and 28.6 K ohms the other. It also did not fluctuate at all, unlike some of the others. is this within spec?
DM802 - 18 M ohms and -6.3 M ohms
DM803 - OL and -OL

I haven't really checked diodes before, but it seems that leaving them hooked to the meter will increase their resistance. If I swap the leads, it reads a negative. If I leave it hooked up for long enough, then it will eventually read positive again and then switching the leads back then results in a negative. This is interesting to me because I thought a diode would have a low resitance in one direction and high in the other, but it doesn't seem to be the case.

Please let me know if I measured anything incorrectly.

EDIT: For the three prong diodes, I put one lead in the center and used the other on one outside pin and then moved it to the other outside pin. I seemed to get the same or a similar reading between the two tests. I forgot to mention this.

[tom66]

Re: Samsung LN32B530P7F PSU issue

Your meter will test diodes by applying a small current through the diode (the current is fixed and known.) It will then measure the voltage and display it as ohms (V=IR) or voltage drop. This current charges up the capacitors on the power board, so when you reverse the probes you may briefly see a negative reading, because you are picking up that old reading that was made. Your meter is almost always only measuring a voltage, just in different ways. Even in the current ranges, it's measuring voltage dropped across a shunt resistor.

None of the diodes show obvious shorts, so I think it is OK. Ideally for diodes you would use diode mode, however, all that matters in this instance is that the diodes aren't shorted.

I think you should be good on replacing that IC. Given the high failure rate of the zeners on this board and the low cost of them consider adding ZD803 and ZD804 to your order.

[tom66]

Re: Samsung LN32B530P7F PSU issue

Actually, scratch that. Check DM802,DM803 on the diode test mode. Completely open circuit is a little odd.

None of the output diodes are shorted, but it's theoretically possible a diode has failed open (that would be a first for me, I have only ever encountered shorted diodes.)

[b4ndit]

Re: Samsung LN32B530P7F PSU issue

Unfortunately, the Fluke 114 does not have a diode test mode. It's not a multimeter made for electronics, but it has worked well for what I've used it for in the past. However, I may be in the market for one that's made specific for electronics. Would you have any multimeter recommendations?

I retested DM802 and DM803. They both show OL and -OL. However, when I put the meter into audible continuity mode, it briefly beeps when I switch the leads on the meter back and forth on the leads of the diode. Because of this and the fact that it shows -OL (doesn't do this on an open circuit), I believe the diodes are not open, but they have a resistance higher than my meter can read. I could be wrong on this, however.

EDIT: I found a super cheap meter in the toolbox that seems to have a diode test mode. I'm going to play with it to see if I can get anything useful out of it.

Ok, I used the cheap meter to test the diodes. I watched a youtube video on how to utilize the diode function and he mentioned that 500-900 was an acceptable range in one direction and you want OL (just a 1 on this cheap meter) in the other direction.

All of the two lead diodes tested fine. I got a number between 500 (one came in at around 498, but I assume this is OK) and 900 on each of them in one direction and I got a 1 (OL) in the other direction.

The three prong diodes is what has me a little confused. I used the red lead in the center and put the black lead (label COM on the meter) on one of the side prongs. This tested as 1 (OL). I then put the black lead in the center and tested with the red lead on the outside. I got a reading of 241. Both of the three prong diodes tested this way. I know my knowledge is very limited on diodes (I'm going to Google around about these diodes after I post this), but I thought that they were supposed to work the other way. I thought that I'd get an OL reading with black on the center and a number reading with red on the center, but apparently I thought wrong. Anyway, I'm not sure if a reading of 241 (actually a range between 240 and 245) is OK or not. I would assume that it is since the diode isn't shorted or open, but I would like some verification that they're within spec.

I quickly looked around for a new meter and saw a Fluke 117. Seems to have all the features of my Fluke 114, but also includes a diode tester, a capacitor tester and a test for amperes. Seems reasonable at $146 on amazon.

[b4ndit]

Re: Samsung LN32B530P7F PSU issue

In my previous post, I posed the question of whether a reading of 241 is OK. I watched a few youtube videos on how to test different components and it seems that the reading I got is perfectly fine.

I've been testing different components on the board and I haven't found anything wrong yet. I retested each of the diodes (mostly for practice and to get used to this cheap meter) and they all seem fine. I checked continuity on the TM801S transformer and everything checked out fine as per the schematic. I'll keep testing things as I think of them and/or learn how to test them.

Is there anyway to test the IC to see if it's truly at fault? I'll also be sure to order replacements for the two zener diodes you mentioned, even though they're testing OK right now as I don't want further problems down the road.

[tom66]

Re: Samsung LN32B530P7F PSU issue

The low forward voltage diodes are OK, they are Schottky type so a forward voltage of around 150~450 is expected for them.

The only way to test the IC, at least easily, is to eliminate every other possible fault from the circuit, which has now been done.

[tom66]

Re: Samsung LN32B530P7F PSU issue

As for a recommended multimeter (I just saw your Q) I've been using a B+K 2709B for a few years, I recommend it quite well. Also I believe Fluke have a ~$100 multimeter out for the Chinese market that you can buy from DealExtreme. Fluke 17B. It's got basically the same features as the 2709B, I think it even uses the same multimeter chipset (a Cryustek one) but it will have better input protection and safety with full HRC fuses etc. Although that's only really important if you use it for proper electrical work, it's nice to know it's there.

[b4ndit]

Re: Samsung LN32B530P7F PSU issue

Thanks for all your help. I'll look at the different offerings from B+K and Fluke to see if I find a good meter I want to buy.

My plan is to order that IC (I'll probably get two, just in case) and those two zener diodes (also will probably buy a couple extra) sometime today. I started looking around the web last night and it seems that digikey only sells the IC in bulk. So it looks like my best bet is ebay.

[tom66]

Re: Samsung LN32B530P7F PSU issue

Don't forget the small capacitor as well.

[b4ndit]

Re: Samsung LN32B530P7F PSU issue

Good catch, I almost forgot about CM808.

[b4ndit]

Re: Samsung LN32B530P7F PSU issue

Ok, I have another question. The capacitor CM808 on the PSU is a SamYoung 50V 22uF. However, in the schematic it says it's a 50V 47uF. I'm leaning toward trusting the rating on the cap in the PSU, but I thought I'd ask to be sure.

[tom66]

Re: Samsung LN32B530P7F PSU issue

Probably a design change. In that position the value is not particularly critical, it just has to be larger than a set amount (probably 22uF in this case.) Could be cost saving, supply problems, etc.