| Electronics > Repair |
| LG 3D 4K TV repair - UB980 range 65UB980 79UB980 84UB980 |
| << < (4/7) > >> |
| 3DMovieList:
So I was able to desolder a good version of the "J516T" chip and do some tests. Buzzing out the chip with a PEAK diode/transistor/semiconductor tester indicated that the chip contains 4 diodes, as indicated in the diagram below. The configuration is (to me) unusual and the forward voltage of the diodes falls into two groups (two are 0.65V and two are 2.8V) as indicated in the diagram. For the bad "J516T" chip, the diode on pin 1 was open circuit, but otherwise the same as the good chip. I also tried testing various combinations of the SMD legs to see if there was a transistor or FET present, but nothing came up - just independent diodes. So... the plot thickens. I didn't spot any SOT-23-5 devices with two sets of different diodes in the package when I was looking around. Any ideas? |
| 3DMovieList:
I just had a look through a range of 4 diode SOT-23-5 packages available on Mouser and couldn't find anything remotely similar to the diode configuration measured on the good J516T. The two different forward voltages, and the alternating directions of the diodes I found seems very unusual. I did test for transistor and FET modes too using the semiconductor tester, but the device always identified just four separately diodes. I did some quick tracing of connections to the chip today and found just bypass capacitors so nothing special to suggest this chip is anything other than ESD protection. |
| projectgus:
Is it possible you're measuring pin protection circuitry here? Bit of a wild guess, but could it make sense for a differential line driver or receiver? Something like DS90LV011A (but not that exact one). Many SOT-23-5 all with the same basic pinout for driver or receiver. Pins 3 & 4 are the differential pins so you may be measuring overvoltage for an LVDS pin relative to GND (pin 2)? (The LVDS maximum voltage is not referenced to VCC in these chips, although I don't know how they are actually implemented.) However if this was the case then you'd also expect to be able to measure some protection circuitry directly between pin 1 (VCC) and pin 5 (the LVTTL single-ended pin), and probably also undervoltage protection between 3 & 4 and GND. Also, I have no idea why these boards might need this. The only similar (equally wild guess) option there would be an op amp, which are available in similar SOT-23-5 standard pinouts and pins 3 & 4 are typically the op amp inputs. |
| 3DMovieList:
Thanks for the suggestion projectgus. At this point anything is possible. ;-) I did some further testing on the board yesterday to check the chips connections to other points on the board. Pin 1 and pin 4 of the J516T chip are connected together, and together connect to a zero ohm resistor (R394) (I think effectively a fuse), but it doesn't seem to connect to anything else on the board, which is really weird. The only thing they might connect to is a connection on the thin ribbon cables that run to the LCD panel, but I'm unable to probe those. Pin 3 goes to one pin on the input connector CN2 (pin 3) (which I think is a power supply pin, but I'm yet to confirm) on the other end of the board and there is one bypass capacitor (C305) on the line near the input connector. Pin 5 goes to three sequential pins on the input connector CN2 (pins 68,69,70) on the other end of the board and it also connected to 8 bypass capacitors (C304,C316,C323,C326,C333,C336,C343,C346) going to ground along the way. Pin 2 is ground. I've captured a high-resolution image of the board, which helps understand the configuration. The J516T chip is removed in this image. Unfortunately, it seems the board is multi-layer, so this top view doesn't tell the full story. But... given it seems there are only bypass capacitors connected to the J516T chip, to me it seems likely to me that this is a Transient Voltage Suppression (TVS) chip. Given it's only one of the diodes (pin 1) that is malfunctioning on the bad J516T chip (which should have a forward voltage of 0.7V) I might just solder the chip back with pin 1 disconnected, and see if this allows the LCD panel to behave correctly. I could solder a small external signal diode between pins 1 and 2 to replace the missing diode in J516T, but I wonder if that is absolutely necessary. I was thinking of drawing up a simple schematic to illustrate what the circuit looks like. But as you can see from the circuit board image, the J516T chip is the only chip on the board. Everything else is just capacitors and resistors. All comments, ideas and theories very welcome! Thanks for reading this far! |
| 3DMovieList:
--- Quote from: sinewave on October 25, 2022, 02:06:56 am ---3DMovieList, can you measure the height/width/length of the package? --- End quote --- I've measured the J516T chip with a vernier and the measurements are: 3.0mm (L) x 1.6mm (W) x 0.75mm (H) |
| Navigation |
| Message Index |
| Next page |
| Previous page |