So a friend gave me an ASUS monitor, it's a 27in job, MA279 something I believe. The backlight circuit has failed -- he claims that there is a particular MOSFET that needs replacing. It's an unobtanium part, # BA5N10. There's two eBay listings and both are super sketchy Chinese sellers from overseas. Normally I don't mind that but I've gotten enough defective silicon from eBay in the past to arguably make a really bad solar roof of it all, and I'm not interested in adding to that here.
However, in the course of trying to determine if any of the stuff in my eWaste bin (aka parts scavenge bin) has anything useful for me, I've discovered that (a) if the circuit works the way I think it does, the way this MOSFET is being used is REALLY janky at best, (b) it's really not a suitable use of a MOSFET at all as I understand it, and (c) I'm starting to think it's not the actual failed part after all.
So here's the thing. The BA5N10 is driven off an MP3389EF 28pin PWM controller for LED buck-boost supplies via a chip resistor marked R100. The Gate pin of the MOSFET is tied high; it's got continuity with the center pin of the 19v barrel jack power input! (Oy.) They're actually toggling it with the *Source* pin... the Drain pin is the tab which is being used as an output. This drives a large surface-mount inductor marked "330" on top and a bank of three 33uF 100v electrolytic can caps in parallel, with the output going through a component marked "BD1" on the silkscreen. More on this in a moment. The circuitry layout appears to suggest a second-stage converter, but this is actually most likely for the sound driver daughterboard, as the setup *there* is driving an AX10008 regulator component.
Panelook says nominal backlight voltage is 43.8vDC (+/- 3.0vDC) for this panel at roughly 65-70mA current draw; I am habitually wary of sites that require one to have an account for documentation access alone, and IIRC Panelook requires a paid subscription for datasheet access (although this recollection is years old, literally, and may now be incorrect!) so I don't have a proper datasheet at the moment.
OK, so remember I mentioned a "BD1" part? Going by the way the silkscreen is labeled, backlight parts are labeled with a B in front of them. So capacitors are BCx, inductors BLx, etc. The PWM chip is BU2; oddly, BU1 is the BA5N10 MOSFET! (Other transistors are BQx in certain spots so I'm not sure why they made that distinction.) I *suspect* BD1 is a diode; however, it conducts in both directions at present. It also looks like it's in pretty awful shape. I suspect that it's this part and NOT the power transistor that's at fault, not least because, in addition to being connected to the PWM drive pin on the MP3389EF, it's shorted straight to power as well. There's a dead short tying those pins high. There's also an adjacent pin that's tied high -- the COMP pin, as marked on the chip's datasheet, which, from the description, is used as a sense pin to control regulation.
I can sort of make out markings from a high-magnification photo on my phone taken with flash at an angle, but quite honestly this part is pretty baked. It *looks* like it reads
B8CD
540.
but I can't be for sure. I've attached the photo I took to help. Whatever it is, I don't think it works any more.
It seems likely to me that this is the point of failure; the regulator circuitry sounds to me a bit like what I've heard described as a "charge pump" configuration, I think, as a subtype of buck-boost regulator -- it works a bit like BigClive's Joule Thief circuit, using inductive kickback and capacitors to boost voltage from a lower source voltage to a higher output, and if so, the part in question is a diode that is supposed to help control that, preventing a sort of 'backflow' condition from the capacitors after the coil has discharged its kickback into them. I'm... not very good at power supply circuitry, though, so I may be misunderstanding something.
Ultimately, I need to know what that baked part in the photo actually *is*, and I need some idea from someone smarter and more experienced than I am (that's a low bar lol) as to genuinely what it does, because I'm not quite sure... then I need advice on whether replacing it will likely solve the whole problem or whether I need to look at substitutes for that unobtanium BA5N10 MOSFET.
Whatever help or suggestions you can give me in regards to that, I appreciate.
Home
Help
Search
Login
Register
