NS-LCD32 - Power Cycling

[arunningpir8]

Ok guys, I have read nonstop for the past week about this TV i acquired secondhand from a guy I work with. Now first things first, the problem.

1. The problem with the TV is that as soon as you plug it into the wall, the TV will start to "Power Cycle" itself. What I mean by this is the power LED will turn on for 5 seconds, then off for about 2-3 seconds, and this cycle repeats nonstop until it is unplugged. Nothing else is responsive.

Now, my troubleshooting is as follows:

1. My coworker ordered a new power supply, thinking that would be it. The exact same symptoms surfaced after pluggin in the new power supply. I think it is somewhat safe to rule that out now (although there is a 1 in a million possibility he received another one with the exact same problem).

2. I disconnected EVERYTHING from the mainboard aside from the power supply, the problem still exists.

3. I measured all of the rails on the harness from the power supply to the mainboard. The order is as follows: 9V, GND, 5V, GND, 12V, 12V, GND, GND, 3.3V, GND, STY (Power Switch?). The ONLY rail that always has power is the 3.3v rail. when the power LED is on for the 5 seconds, I get 3.288V, when the power LED turns off for its 2-3 seconds I get a stable 3.304V on this same rail. All the other rails are only on when the power LED is on for its 5 seconds, and are lower than they should be (i.e. the 12V rail is only ~11.3-11.6v) This has me thinking it is something in the standby circuit causing the problem.

4. I tested all the voltage regulators. They all checked out as they should, but as expected only *work* when the power LED is on for its 5 seconds at a time.

5. I took out all of the caps and they all measured fine. I then replaced all of the caps.

So, I have done everything I could think of at this point (including the weeks worth of reading) and still feel like I have gotten nowhere. One other mentionable is that when the power LED is on for its 5 seconds, measuring between chassis GND and mainboard GND gets me -.02v, whereas when the power LED cycles to off for its 2-3 seconds, there is an almost perfect 0.00v between chassis GND and mainboard GND.

I am at the point now where I need to know what is wrong simply because of how I am. Being a recent EE grad from Purdue and not being able to figure stuff like this out really gets my spirits down. So if anybody out there could be of help please I am open to any suggestions!

Also, ignore the diode standing in the air by the standby circuit, I took the surface mount one off as it was a teensy bit out of spec but it did nothing to change the problem.

[arunningpir8]

Re: NS-LCD32 - Power Cycling

Also some more info, the 24V rails do turn on while the power LED is on. So it seems as if all the power supply turn on is going well, its just that for some reason after a couple seconds it turns itself off, my guess is a short somewhere but the question remains as to where. The bigger question is why does it try to turn itself on immediately after plugging it in?

[budm]

Re: NS-LCD32 - Power Cycling

Some TV turn on when the AC is applied to the set, not all model does that.
The STBY pin is the PS_ON (Power supply on control signal) signal which is generated by the main board. You should monitor the DC voltage on that pin to see if it changes state when you hit the power switch or when the TV power up by itself.
This TV uses 3.3V for the standby power supply (always ON power supply).
Do you have the pictures of all the boards? at this point it seems that you may have the main board problem. You should also try disconnecting the backlights inverter board, but I cannot tell right now since I do not see the pictures of the power supply board to know if it uses power suppply/inverter combo board or not.

[retiredcaps]

Re: NS-LCD32 - Power Cycling

Ok guys, I have read nonstop for the past week about this TV i acquired secondhand from a guy I work with.

Thank you for doing some homework first.

1. I think it is somewhat safe to rule that out now (although there is a 1 in a million possibility he received another one with the exact same problem).

Tom66, a regular here, thinks the number of replacement or "new" boards that are defective is almost 25% (purchased off ebay). I'm not saying this is the case, but not as high as you think.

[arunningpir8]

Re: NS-LCD32 - Power Cycling

Ok, I just measured the PS_ON pin and when the power LED is ON for its 5 seconds I get 3.267V stable, and when the power LED turns off for its 2-3 seconds it drops off quickly to 0V. Also, as you can see from the first picture of the mainboard I posted, it is connected ONLY to the power supply, and the power supply was also disconnected from the inverter board as well, so I'm 90% certain this is a mainboard issue.

A second note, when the power LED is ON for its 5 seconds, you get a "buzz" from the power supply, whereas when the power LED turns off for its 2-3 seconds you get a squeal from the power supply (kind of a "charging the lasers" type sound). If a sound clip is needed I could try and capture one to upload.

Here is a pic of the power supply, I'm glad you confirmed it has a 3.3v standby as every single service manual I have looked at for an NS-LCD32 have all been 5v standby. I guess I got the rare model of NS-LCD32.

Thanks for the initial help guys as like I've said I've been tearing my hair out over this one. I don't know why I can never get the easy fixes like a lot of other people get :P

[budm]

Re: NS-LCD32 - Power Cycling

I fixed two of this model with same bad main filter caps, but you should also replace the rest of the cap as a set. It was making buzzing sound due to bad main cap.
https://www.badcaps.net/forum/showth...+NS-LCD32_7D21

http://s807.photobucket.com/user/bud...?sort=3&page=1

[ben7]

Re: NS-LCD32 - Power Cycling

Ugh, disgusting crappy "Xindecon" capacitors are all over inside there. I see one is bulging, on the logic board!

[arunningpir8]

Re: NS-LCD32 - Power Cycling

Well I took a look at the original power supply and did find one of the output caps (47uf) to be at 16! Also, the main cap (150uf) was at 135 so I replaced both of those. Unfortunately I still can't get it to turn on as it still cycles when I plug it in. One thing that did change however is now the 3.3v standby is somewhat solid. When the power LED is on I get 3.304v and when it turns off I get 3.303v, which is a major improvement over what it was! The power supply also isn't making much of a buzzing noise anymore either.

I'm starting to get more discouraged now as it seems as if the power supply is working correctly now, however it still isn't allowing to the TV to enter a normal standby mode. As for the one bulging cap on the logic board, that was my idiocy in putting the caps back on the board as I put that one on backwards in the picture (perfect example of why you should take breaks on working on stuff like this after so many hours). I have since replaced it, but it is only for the 24v supply which I have left disconnected this entire time.

My next question is (budm maybe you can help me on this) will having ONLY the main power harness hooked up to the board allow it to enter standby? Or for it to enter standby do I need everything hooked up? I have been doing all my testing with only the one harness plugged in to isolate the problem, but didn't know if leaving everything else disconnected could be hurting me more than it is helping.

Thanks again for the help, at least I got closer!

[arunningpir8]

Re: NS-LCD32 - Power Cycling

Ok, so I am starting to feel more like this might be a power supply issue. As it is I already found those 2 caps that were bad and replaced them and now it sounds cleaner when its on.

I tried jumpering the PS_ON pin to the 3.3v standby and the power supply starts ticking and the voltage for the 3.3v standby goes up and down from 0-2 volts roughly. With absolutely no load shouldn't the power supply start outputting the other 12, 9 and 5v rails when i send it the PS_ON pin? If so then this is definitely looking like a power supply issue.

[budm]

Re: NS-LCD32 - Power Cycling

You have to cut or remove the PS_ON wire that goes back to the main board since the Main board is forcing the PS_ON LINE low through the transistor (transistor is used as a switch to ground the PS_ON line to LOW on the main board. I hope the transistor is not damaged.
Same for the BL_ON (INV_ON or ON/OFF) for the inverter board control signal.

[arunningpir8]

Re: NS-LCD32 - Power Cycling

I jumpered the PS_ON pin with nothing else connected to the power supply at all. Are you telling me that the PS_ON is active low? I thought the main board would output a 3.3v (high) signal via PS_ON back to the power supply to turn the whole supply on. Am I not thinking here correctly?

[budm]

Re: NS-LCD32 - Power Cycling

If you have nothing else connected to the power supply, then that 3.3V standby should not drop down to <1V. Try using 1K resistor between PS_ON and the 3.3V STBY.

[arunningpir8]

Re: NS-LCD32 - Power Cycling

Yep, I tried with a 1K resistor and nothing else connected and you can hear the power supply tick at about a 1 second interval with the voltage on both the 3.3v standby and PS_ON going between about 0-3v.

Also I measured earlier the voltage across the new main cap and now get about 350vdc with it dropping about 10vdc or so between cycles. Is 350 high enough? For some reason it doesn't sound like it.

[retiredcaps]

Re: NS-LCD32 - Power Cycling

Yep, I tried with a 1K resistor and nothing else connected and you can hear the power supply tick at about a 1 second interval with the voltage on both the 3.3v standby and PS_ON going between about 0-3v.

A ticking suggests one of the following:

1) bad startup capacitor
2) bad SMPS IC
3) shorted component like a diode on the output side

[retiredcaps]

Re: NS-LCD32 - Power Cycling

It looks like A6159 is your SMPS IC

https://cdn.badcaps-static.com/pdfs/...af0bce6e6c.pdf

1) Measure the VCC (pin 2) and see if it is steady and stable (pin 3 = GND)? Report measurement.

VCC should be around 17.5V for "typical" operation.

2) Measure the DC voltage on the startup cap (that is tied to pin 2)?

3) See if VCC (pin 2) is shorted to pin 3 (GND)?

[budm]

Re: NS-LCD32 - Power Cycling

The 3.3V standby was steady before when you have the main board which is the load attached to the power supply as per post#1. What happen if you hook everything back to the way it was before? It sounds like the 3.3V power supply has failed now.

[arunningpir8]

Re: NS-LCD32 - Power Cycling

So retiredcaps...I think you're on to something here. Pins 1-5 are all around 2.8VAC and pins 7 & 8 are ~8VAC. According to that datasheet that cannot be correct.

Bud, nothing has changed from what I've said. You now see my confusion. When hooked up to the mainboard I get 3.304V with power LED off, and 3.302 with power LED on. Yet, for some reason with no load when I jumper the PS_ON pin to 3.3V the power supply cycles on and off every second.

I feel like we are getting closer, something is obviously up with the SMPS. Note, all these readings were taken with the mainboard disconnected and the 3.3 standby stable and no power supply ticking.

Thanks so much for the help guys, you have no idea how much knowledge I'm gaining from this, which is always my favorite part in fixing things!

[retiredcaps]

Re: NS-LCD32 - Power Cycling

So retiredcaps...I think you're on to something here. Pins 1-5 are all around 2.8VAC and pins 7 & 8 are ~8VAC. According to that datasheet that cannot be correct.

You want to measure VDC on the SMPS IC and startup cap, not VAC.

[retiredcaps]

Re: NS-LCD32 - Power Cycling

And your GND reference is pin 3 (GND) or the negative leg of the large filter capacitor when measuring the SMPS IC (VDC). Remember you are on the hot side when doing these measurements.

[arunningpir8]

Re: NS-LCD32 - Power Cycling

Wow I feel like an idiot now I just had it in my mind DC on one side of the dotted line AC on the other and didn't even take a second to completely comprehend the datasheet schematic.

Pin
1........... 0.00
2.......... 17.78
4.......... 0.482
5.......... 177.5
7.......... 171.0
8.......... 171.0

So now seeing these numbers it looks like the SMPS is in fact working. So does that mean we can somewhat eliminate the hot side for problems? And does anybody understand why the supply switches on and off every second when I pull PS_ON up?