Sceptre X460BV-F120 2 seconds to power off.

[johnboy1313]

Just picked up this 46" X460BV-F120. When plugged in, there's an odd rumble from the speakers then the Sceptre logo shows and the picture turns off with no sound. With the back off, I immediately noticed a very loud whine coming from the large transformer in the center of the power supply. The lowest transformer between the heat sinks. The whine started with a low pitch that got higher and higher after a few seconds. With the unit unplugged from the wall, the whine continues for sometime as the caps discharge. I forced the PSU on and the backlights came on but strobed rapidly. At that time the whine from the transformer kinda went crazy.

At standby, the voltages are
1=.72V
2=0V
3=0V
4=0V
5=5.11V
6=5.11V
7=0V
8=0V
9=0V
10=0V
11=0V
12=0V
13=0V
14=0V
15=0V
16=0V

At power on the whining stops immediately and the voltages are
1=4.8V
2=0V
3=0V until relay clicks then jumps to 4.87V
4=4.27V
5=5.1V
6=5.1V
7=0V
8=0V
9=12.35V
10=12.35V
11=0V
12=0V
13=16.25V
14=16.25V
15=0V
16=0V

The backlights come on and then turn off and the voltages are
1=4.8V
2=0V
3=4.87V
4=4.27V
5=5.1V
6=5.1V
7=0V
8=0V
9=0V
10=0V
11=0V
12=0V
13=0V
14=0V
15=0V
16=0V

The set powers off with the power button and the whining immediately starts again until the power cord is unplugged.

At this point, I can't tell if the picture is on with no backlight or not. I'm not sure if I'm looking at a PSU issue or a main board issue.

[johnboy1313]

Re: Sceptre X460BV-F120 2 seconds to power off.

Any ideas?

[tom66]

Re: Sceptre X460BV-F120 2 seconds to power off.

Inspect the primary capacitor, and consider replacing it. I have heard of a PSU making such a noise when that cap fails, along with strobing backlights. You can try measuring the voltage across it. Other possibility is failed PFC circuit. If you don't have ~400V across it, but instead about 160V (USA), then the PFC is not working.

[johnboy1313]

Re: Sceptre X460BV-F120 2 seconds to power off.

I have the set plugged into a surge protector. With the probes of my DMM on the leads of the primary cap, when I turn the surge protector on, the voltage jumps to 394V immediately and slowly drops to 169V. The whine from the transformer gets higher at the same rate as the voltage drops on the primary cap. When the set is powered on, the voltage on the cap jumps to 394V and stays there until the set is powered off.

[tom66]

Re: Sceptre X460BV-F120 2 seconds to power off.

So when it is working, do you see any drop as the PSU whines? Try to locate the PFC section. It'll have an inductor (marked Lxxx) and a MOSFET (marked Qxxx) and a diode (marked Dxxx) in the same area. The capacitor near it is probably faulty, try replacing that.

[johnboy1313]

Re: Sceptre X460BV-F120 2 seconds to power off.

So when it is working, do you see any drop as the PSU whines?

No, when the set is on, that cap holds steady at 394V and the whining stops. The transformer only whines while the set is in standby.

Try to locate the PFC section. It'll have an inductor (marked Lxxx) and a MOSFET (marked Qxxx) and a diode (marked Dxxx) in the same area. The capacitor near it is probably faulty, try replacing that.

On that note, can you recommend a decent but inexpensive ESR meter? I will look into that cap here in a minute.

I'm forcing the PSU on with two 220 ohm resistors from each 5V to LCD-ON and PWR-ON. When I unplug everything from the PSU and force it on, the transformer still whines, but I get steady voltages at all rails. The LCD-ON and PWR-ON rails each measure 5.11V. Is that too high? Should I try two AAA batteries?

With the inverters connected to the PSU and the PSU forced on using the method described above, the 12V and 16V rails measure (11.8-11.9) and (15.5-15.6) respectively.

Is it possible I'm looking at a bad inverter board?

[tom66]

Re: Sceptre X460BV-F120 2 seconds to power off.

My ESR meter I use is a Peak Electronic Design Atlas ESR60, but it's around $110.

Those voltages look good.

I think the problem is with the PSU. The buzzing/whining is not a good sign.

I would suggest you scope the outputs of the PSU. Do not try to measure the PFC cap with the scope, doing so will result in blowing up the scope. Just see if the outputs have ripple on them. Set scope to line trigger to see if there's any 120Hz component to the noise, if there is, consider the capacitor.

[johnboy1313]

Re: Sceptre X460BV-F120 2 seconds to power off.

I'll have to borrow the scope again tomorrow. I'll keep you posted.

What do you think of this? I know it's coming directly from China.

http://www.ebay.com/itm/Transistor-T...item27ceb788e3

[tom66]

Re: Sceptre X460BV-F120 2 seconds to power off.

Some people use that one and find it works fine. I have no experience with it though. Hard to go wrong with $20 if it does what it says it does.

[johnboy1313]

Re: Sceptre X460BV-F120 2 seconds to power off.

Out of curiosity, do you suspect the transformer is whining because it's bad or because another component is bad?

[tom66]

Re: Sceptre X460BV-F120 2 seconds to power off.

SMPS transformers "don't" go bad. They can fail open, or shorted, but those are extremely rare failure scenarios, and then the PSU wouldn't work at all.

A transformer is a coil of wire. When it has a current passed through it, it expands. When the current stops, it contracts. This makes them audible. Most power supplies operate way outside audible frequency range -- more than 20kHz -- so you don't heard this coil whine. (I don't know if dogs hear it though!) When a PSU can't properly regulate the outputs it often drops in frequency and becomes audible.

Several causes of regulation failure include bad output capacitors, bad input capacitor(s), feedback failure and control IC problems.

[johnboy1313]

Re: Sceptre X460BV-F120 2 seconds to power off.

Any idea where I might get a hold of a service manual for this set?

When I get the scope, what point am I taking the waveform from?

http://www.shopjimmy.com/sceptre-3bs...upply-unit.htm

[tom66]

Re: Sceptre X460BV-F120 2 seconds to power off.

I'll keep an eye out for you, but I doubt it would be of much use.

I'd suggest you set the scope to AC coupling, line trigger, 5ms/div and 200mV/div, and measure each of the 24V, 12V and 5V outputs.

[johnboy1313]

Re: Sceptre X460BV-F120 2 seconds to power off.

I'll keep an eye out for you, but I doubt it would be of much use.

I'd suggest you set the scope to AC coupling, line trigger, 5ms/div and 200mV/div, and measure each of the 24V, 12V and 5V outputs.

I set the scope per your instructions, but I couldn't figure out how to set to line trigger. With everything disconnected from the PSU and the PSU forced on, the PSU whined like before but every ~3sec there was a brief click and then the whine. This cycle was continuous.

The pictures below, in order.

1) 16V rail while whining
2) 16V rail at click
3) 12V rail while whining
4) 12V rail at click
5) 5V rail while whining (used autoset here, the wave just looked like a thick black wavy line with the settings above)
Not Pictured, 5V rail at click looked nearly identical to 5V while whining but higher voltage.

[tom66]

Re: Sceptre X460BV-F120 2 seconds to power off.

5V is the standby line, and it's likely going to run off a separate supply.

We need to figure out why the rail starts becoming unstable but that's definitely your problem right there.

To set up line trigger on the TDS series of scopes, you need to select "Line" as the source, I think -- however, we can clearly see this isn't related to the line frequency, as it's decreasing in frequency as the problem develops. That start frequency though is way too small. I don't know why yet...

I'd suggest you find the cathode of the diode for one of the outputs. We're looking for the negative pulse on this. It'll probably be around twice the normal output, but it should vary with the line voltage. Try to get a capture on 1sec/div; use of the RUN/STOP button will help here. (I can't remember if this scope has a SINGLE button, but if it does, use that instead.)

[johnboy1313]

Re: Sceptre X460BV-F120 2 seconds to power off.

5V is the standby line, and it's likely going to run off a separate supply.

We need to figure out why the rail starts becoming unstable but that's definitely your problem right there.

The 5V rail is the problem?

To set up line trigger on the TDS series of scopes, you need to select "Line" as the source, I think -- however, we can clearly see this isn't related to the line frequency, as it's decreasing in frequency as the problem develops. That start frequency though is way too small. I don't know why yet...

I'd suggest you find the cathode of the diode for one of the outputs. We're looking for the negative pulse on this. It'll probably be around twice the normal output, but it should vary with the line voltage. Try to get a capture on 1sec/div; use of the RUN/STOP button will help here. (I can't remember if this scope has a SINGLE button, but if it does, use that instead.)

The diode for either 12V or 16V? Do you want to see a capture from both or just one?

[tom66]

Re: Sceptre X460BV-F120 2 seconds to power off.

The 5V rail is the problem?

Hmm. The 12V and 16V are both completely unstable, but the 5V isn't.
What do you think? The 12V & 16V are the fault. 5V is fine. Just noisy, but that's normal, for any SMPS.

The diode for either 12V or 16V? Do you want to see a capture from both or just one?

Only one is necessary. One end with have a "nearly" DC voltage, the other end will be the wave we are looking for.

[johnboy1313]

Re: Sceptre X460BV-F120 2 seconds to power off.

Hmm. The 12V and 16V are both completely unstable, but the 5V isn't.
What do you think? The 12V & 16V are the fault. 5V is fine. Just noisy, but that's normal, for any SMPS.

I figured that was what you meant, I just wanted to be certain.

Here's a pic of the buttons on the scope.

[tom66]

Re: Sceptre X460BV-F120 2 seconds to power off.

OK, no SINGLE button, but you can recreate the effect with RUN/STOP. Just press RUN/STOP, then again RUN/STOP on the frame you want.

[johnboy1313]

Re: Sceptre X460BV-F120 2 seconds to power off.

Couldn't get a decent pic. Not sure if the settings I used were correct. Here's a video I made. I hope it shows you what you want to see.

http://i177.photobucket.com/albums/w...CCEB5501CE.mp4