Just broke a monitor PSU and have some questions

[wkx]

I have a Samsung P2770H monitor, the power supply is an external one, 14v 3.5amps.

I was having this issue where the monitor would turn off and back on after 1-2 seconds, some days it would work fine, some days it would do it a lot of times. Mostly happened when watching videos full screen, but also a few times when just browsing a website or coding in a virtual machine.

Anyway, since the psu is external I said what the heck lets open it, and found these:



Two were a bit swollen and had the rubber seals pushed out, the other looked fine but I replaced it anyway. I replaced a 820uf 25v with a 1000uf 36v because I couldn't find the 820uf anywhere, and since these caps have a tolerance of like +/- 20% I thougth it would be ok, and two 470uf 25v with the same, all with same temp rating.

I thought this would fix the issue but the monitor would still occasionally turn off and back on by itself. So I opened it again and removed the two heatsinks, one with a rectifier diode (Y2010DN) and a mosfet (FQP 8N6OC), to remove some of the white adhesive and have a better look.



I didn't find anything wrong visually so I put it back together, but before doing that, I removed the diode and mosfet from the heatsinks (for some reason I still don't know). In that process, I broke (what I now know are) the insulation pads. I thought it was just for thermal conductivity so I replaced it with some cpu grease. And when I plugged it in, a spark noise and some smelly smell came out of it. Now it's dead.

Lesson learned. I'm going to buy a new one but may want to try to repair it some day to use for leds, an arduino project or something.

I have two questions:

1) Could I be so lucky that only the rectifier diode broke and replacing that would make it "work" again? (with the same issue I described at the beggining of the monitor turning off/on itself)? Or it is likely other components went down? Nothing looks burnt visually, the smell went away in seconds and the board has no visible marks. The spark was in the 220v plug.

2) Is there a component in these switching power supplies that normally causes the issue I described at the beggining?

[stj]

Re: Just broke a monitor PSU and have some questions

if you want to learn something, learn to upload images with the [Go Advanced] button

[CapLeaker]

Re: Just broke a monitor PSU and have some questions

Yeah... these insulator pads are there for a reason. same as the plastic screw insulator that's with it. Other than that, things go poof!
Well, I guess the damage is done and no it is not going to be easy as just replacing a diode. The Fet is gone and the gate drive circuit and the gate drive IC. Maybe some more smaller stuff like some small diodes and resistors on the back side of the board as well. It's not looking too good.
This happening occasional again thing... would have probably something to do with the backlights.

[Topcat]

Re: Just broke a monitor PSU and have some questions

if you want to learn something, learn to upload images with the [Go Advanced] button

I took care of that....I attached the images.

[stj]

Re: Just broke a monitor PSU and have some questions

thanks.

[momaka]

Re: Just broke a monitor PSU and have some questions

Two were a bit swollen and had the rubber seals pushed out, the other looked fine but I replaced it anyway. I replaced a 820uf 25v with a 1000uf 36v because I couldn't find the 820uf anywhere, and since these caps have a tolerance of like +/- 20% I thougth it would be ok, and two 470uf 25v with the same, all with same temp rating.

I thought this would fix the issue but the monitor would still occasionally turn off and back on by itself.

It's also possible that the small electrolytic cap between the transformer and primary bulk cap is bad as well, given that the others are. This small e-cap is the "start-up" cap and also what maintains power to the IC/controller going for the PSU to work. So when this cap goes bad, it's possible for the monitor to not start reliably every time. Thus, you should replace that one as well.

As far as replacements... I can't see what brand and series you used for your new caps above. Generally, we only recommend quality Japanese caps here (like Panasonic, Rubycon, Nichicon, and United Chemicon), because other brands (especially cheap caps from China/Taiwan) have proven to be very unreliable, even when new sometimes.

That said, it is unlikely that your new caps are already bad... but it is worth noting what they are anyways - should give us give you an idea of how long your repair will last (that is, if we fix the rest of the PSU, of course).

1) Could I be so lucky that only the rectifier diode broke and replacing that would make it "work" again? (with the same issue I described at the beggining of the monitor turning off/on itself)? Or it is likely other components went down? Nothing looks burnt visually, the smell went away in seconds and the board has no visible marks. The spark was in the 220v plug.

If you removed the insulating pads both from the MOSFET and the diode, chances are the MOSFET is likely blown now. Along with it, the fuse and possibly the MOSFET's Source resistor and controller too.

Your best bet here is to start removing parts again and check what has gone bad. Simply putting new parts in without checking the old ones will very likely leave you disappointed with more sparks and "kabooms".

On the positive side, this is now a good time to further develop your component testing/troubleshooting skills, if they aren't yet. On that note, you can measure some parts in circuit. But if the reading seems suspicious, it's best to take them out and measure them that way.

Anyways, to get this PS fix, report back what parts are blown. There should be plenty of info online showing you how to check various parts, but here's a quick "guide" for how various components should read *out* of circuit.
--- with resistance test ---
- Resistor: resistance value printed on resistor within 1/5/10% tolerance (whichever is indicated on resistor). E.g.: a 100 kOhm 5% resistor might read 103.5 kOhms... which means it's good, because 103.5 kOhms is still within the 5% tolerance spec.
- Capacitor: open-circuit (but large electrolytics over 2000 uF might show a value for a few seconds before the meter resistance climbs to show open-circuit)
- MOSFET: Drain-Gate and Source-Gate always reading open-circuit regardless of multimeter's probes orientation. Drain-Source should *not* read short-circuit. However, it may or may not show a high-resistance reading, depending on your multimeter. (typically if you have the red probe on Source and black probe on Drain and multimeter provides sufficient voltage to activate the Source-Drain diode junction).
- Inductor: always a short-circuit
- SMPS transformer (between various pins/windings): short-circuit or very low resistance, typically, but not necessarily between all combinations.
- Diode: does *not* read short-circuit or low resistance regardless of probes orientation. May show a fairly high resistance when diode is forward-biased (just like MOSFET Source-Drain resistance).

2) Is there a component in these switching power supplies that normally causes the issue I described at the beggining?

Sometimes that small start-up capacitor can cause a similar issue. However, if that's the case, typically the whole monitor's power should be intermittent.

Other than that, it could also be a problem with the inverter, as CapLeaker mentioned. But this we will find out when you get a new SMPS (as you plan to) and report back if it works.