SAMSUNG SPINPOINT HD103UJ - Testing TVS etc

[kalasa]

Hi,

I recently had this drive die on my (not spinning up at all). It has been acting kinda wonky for a while, so it wasn't that unexpected. I have changed PSU though (a new, high quality one) so I would like to know if it might be related to that. There are two other drives on the same lane and they all work fine. I found this image in another post:



I'm kinda noob at electronics, but I measured the 0Ω resistor which showed 0, so that's good. The 12V TVS shows 0.64 one way and .0L the other which also seems fine. The 5V shows 0.4 the "correct way" but 0.8 the other. So it doesn't seem busted since that would say .0L both ways, right? But why is it open in both directions? I read somewhere that measuring diodes that are connected to other stuff might make it seem open both ways. Is this correct? Do my measurements indicate death by overcurrent or something else?

Thanks!

[kalasa]

Ok, looking closer at the 5V diode it looks like it's a zener diode (it says 6V8A on it, some googling leads me to this datasheet) which I guess means that it's working as intended since it should allow current in both directions?

[colanco]

TVS 5V appears damaged.

Remove and measure on the pads.
If still short , the pcb has another problem, if the measure is ok , the disc should work.

if you have another pcb, change the TVS , no good work without TVS

[kalasa]

Before pulling out my soldering iron I would like to understand what's happening. Sorry for making this into an electronics course.

Based on what I've read my understanding of the zener-resistor pair used for "fuses" is that when the voltage goes above the rating of the diode it will open when reverse biased to "let out" some of the current into ground. This protects against small, short spikes. If the voltage is really high then the "fuse" will blow. This can lead to one of three things:
1) If the diode dies and closes none of the current will leak out to ground which will melt something else in the circuit.
2) The diode dies in an open/semi-open state. This should make the PSU shut down since it's basically shorted.
3) The resistor dies because of the high current. This should be the preferred scenario since it cuts off the connection to +5V.

In my case (2) can be ruled out since the computer starts and I have 2 other disks on the same lane that work great. (3) isn't the case either since the resistor looks fine. (1) doesn't seem very likely since the diode is indeed open if my measurements are correct.

Have I understood this correctly? What am I missing?

[fzabkar]

I recommend testing the diodes on the 200 ohms resistance range rather than the diode test. That's because all we want to know is whether the diodes are shorted. By quoting numbers in the absence of units, you have confused us. Also it is not ".0L", it is "OL". Your numbers would suggest that you performed a diode test. Testing a diode in-circuit is not always conclusive since you would also be testing everything connected to the diode.

Did the zero-ohm resistor measure 0.0 or OL?

[colanco]

Remove a TVS and measure, are 3mtos and it is not difficult.

TVS should not change much (or anything) the resistance between + 5/GND. Fzaback can correct me on this if I'm wrong (he knows much more than I of these issues)

So, if you remove the TVS and the resistance between +5/GND not change, then you probably have another problem on the PCB.

If removing TVS and the resistance values ​​change substantially, I think the TVS may be the problem. Change it. PCB test isolated, if not damaged TVS again, should be sure to test it on the disk.

If damaged again by placing it on the disk, it is likely that the damage is in the preamp (or -5V circuit)