Using Polymers to replace 2nd stage lytics (Samsung Syncmaster 223BW)

[Uranium-235]

This board has bloated Samyoung NXE's (16v 1000uf). They're 8mm and 19mOhm. Mouser has the equivalent Panasonic out of stock. The bitch is these are 8mm. Only ones in stock that come close to the ripple (but not beat it) are 30mOhm.

Polymers I find that work are Nichicon 16v 820uf and really low ESR (0.008), 7A. I've heard even on SMPS boards this can cause issues. Now I'm replacing the NXB's with Kyocera RPF polys that are just one mohm under and 3200mArms (vs 2000 mArms which is just perfect). I'll also be replacing the inverter caps with these too, and replacing the other NXB (yellow writing) with a more than sufficient lytic (matches ohms and ripple almost exactly with a slight bump up on the ripple)

I've read that using poymers with such a decrease in impedance can cause high frequency voltage spikes in these types of circuits. Mind you I'm not sure if they're in the VRM in or out.

edit: oh yeah the symptom is hitting the power button turns the power led on for two seconds, then it turns off, and starts cycling (no backlight fires during this)

[stj]

so,
polymers have 3 potential problems.
1: they have high leakage current.
so fine for smoothing power but very bad for circuits that put a cap in series with audio/video signals to block DC.

2: the impedance is so low that some psu circuits think the output is shorted and go into a shutdown-restart loop.
this can probably be eliminated by changing feedback components if you have a schematic - but we dont need to get that deep into it.

3: *if* you have a group of several on a power rail the initial inrush with such low impedance could blow a fuse.
but thats only on like 1970's or 80's industrial psu's that would use a bank of 4 or more regular electrolytics on a single rail.
inrush limiting could fix that anyway.

[Uranium-235]

I haven't put them in yet. The restart loop is from the bad NXE's. But I'm honestly out of options on those. Lytics are too high ESR, polymers are too low.

[stj]

when i recap monitors i just stuff them with FR's
not had a problem yet.

[Agent24]

One option is to go a little ghetto with it... single sided board, likely large traces in that area. You can probably drill some extra holes, scrape away some solder-mask, and position the capacitors further away from each other so you can squeeze in some 10mm diameter parts.

Or just go with the Panasonic FR as suggested by stj, the ratings are close enough they should work for ages. Remember there will be some (or should be) derating of the parts in the design spec, so they won't be running at max ratings generally. So a slightly lower spec'd part will often work just fine.

Probably the backlights or something else will fail before the new capacitors wear out.
(Although backlights can be replaced, too)

[stj]

actually, after the caps the thing that fails the most is the mosfets in the ccfl invertor circuit

[Uranium-235]

motherfucker. I couldn't find my ruler so I eyeballed the sizes. These are 10mm. A FR would have worked and mouse had them in stock. In fact I found lytics for everything in 10 minutes. Well I already put the nichicon polymers in so I can't take them back.

I have backlight turns on and off with the button, but no logo. Video board is probably damaged from continual use with the original bad caps. I'll have to take a closer look at that (I never detached the LVDS cable, so I know it's not that)

[lotas]

I have backlight turns on and off with the button, but no logo. Video board is probably damaged from continual use with the original bad caps. I'll have to take a closer look at that (I never detached the LVDS cable, so I know it's not that)

The firmware on the SPI Flash 25l1005mc has crashed. A common problem is swollen capacitors (unstable, pulsating power supply).
Also check the electrolytics on the motherboard and whether 5V, 3.3V, 1.8V are present (without pulsations)

[Uranium-235]

The firmware on the SPI Flash 25l1005mc has crashed. A common problem is swollen capacitors (unstable, pulsating power supply).
Also check the electrolytics on the motherboard and whether 5V, 3.3V, 1.8V are present (without pulsations)

I think you're right. I think the eeprom is corrupted. Anyone know where to find the .bin? Board is BN41-01075A/2008.5.16/Rev MP1.0

[Agent24]

I'd like to know too, I have a monitor the same model or very similar that also has similar problems, that I never fixed.

The power supply had all dead capacitors, but upon replacing them, the mainboard still wouldn't start.

Voltages were all OK on PSU and mainboard, but the microcontroller won't respond to the power button input signal and the monitor never turns on.

[lotas]

Try

[Uranium-235]

shit I have to be a premium subscriber to download? I've been here longer since anyone in this thread (though one not by much), and am a mod.

[Tesla]

sh*t I have to be a premium subscriber to download? I've been here longer since anyone in this thread (though one not by much), and am a mod.

Yeah, it just means you pay a little for a month of downloads. I think I read it's so he can keep the website online (without all that ad nonsense).

[Agent24]

shit I have to be a premium subscriber to download? I've been here longer since anyone in this thread (though one not by much), and am a mod.

Yes it's a bit ridiculous when you've been here almost 20 years but the site changed ownership and apparently needs to make money for itself now...

[Uranium-235]

Well I got the flasher. Flashed it with the bios. Wouldn't even turn on, no backlight at all. Tried to flash it back to original bin and the clip came off and I tried to put it back on while still connected to the programmer. Pretty sure the chip is dead. I tried.

[lotas]

Board is BN41-01075A/2008.5.16/Rev MP1.0

Did you indicate the motherboard number correctly? This number points to the monitor keyboard.

[Agent24]

Well I got the flasher. Flashed it with the bios. Wouldn't even turn on, no backlight at all. Tried to flash it back to original bin and the clip came off and I tried to put it back on while still connected to the programmer. Pretty sure the chip is dead. I tried.

Get another EEPROM? It's only a small capacity one, they're cheap.