macdan replace a toasted chip on a TiBook 667 DVI I've got a 667 DVI that outputs nothing on its LCD, just a blank whiteness. I examined the MLB for damage and found this mess at Q23:

Yikes! Q23 is a SOT-23 6 pin package and it's fried. What exactly is the chip? I can swipe one from a dead MLB but I'd rather not if I can just get a new part. I don't even know exactly what sort of component Q23 is, except that it's some sort of transistor. I looked at a couple of other TiDVI MLBs I have and found these numbers on the chips: 43FDG or 43FD6 - can't tell for sure- with the G (or 6) underlined? and 431DA - with the A underlined? What caused the chip to burn up like this? Dunno. I acquired this TiBook with broken hinges which naturally points to cable damage causing a short, but I can see no problem with the cable. AFAIK the LCD itself still works fine. The solution! Well, I ended up swiping a similar Q23 from a dead MLB. Removing the dead chip isn't difficult, just snip leads and unsolder one at a time. Pulling the replacement from the other board is more difficult. Since it's got only 3 legs per side I was able to move a pair of irons quickly enough to get each side free. I also grabbed the cap at C224, dunno if it was needed or not, but since I already had it . . . Result was what I expected, LCD display now works fine. The broken hinge was the primary cause, the LVDS cable got cut or scraped at some point afterwards. I more carefully reexamined the cable and found a tiny nick in the insulation on one of the lines fed by Q23. I presume Q23 supplies power to the LCD's onboard controller PCB, so it makes sense a shorted power feed would kill any coherent output. Part ID and source? This fault is easily fixable, it's not even necessary to pull the MLB from the case. I imagine it's not uncommon amongst the many broken-hinge TiBooks, and for those similarly afflicted I'd still like to be able to point to a correct replacement part. If you have any idea of the part ID and a source please contact me and I'll add the info to this page. TIA :-) [macdan at comcast dot net]

For reference here's some values from a working 'Book with a known good Q23, which looked like this to my VOM, measured to ground (use pic below as reference:) (this end closest to J17) 3 6 2 5 1 4 #s 1,2,4,5 = +3.29V via a black something (L43? inductor??) which jumps to J17 LVDS connector pins 1 + 2 (red stripes on cable) # 3 appears to be the supply(?), +3.33 v # 6 +1.7 mV (to ground? with circuit off it measures ~200K ohms to a "real" ground.) #s 3 and 6 are connected by C225 - what's this do? On examining the remains of the old chip I noted that 1, 2, 4 and 5 are all connected internally to the same 'slab' of metal, explaining why they all read the same voltage.

back to macdan revhist 2004.12.20 - added solution 2004.09.16 - new page