We've been using TLV62130 for many years now in multiple devices, not giving problems.
The protoboards of our latest devices all worked fine.
Validation of the final PCB was done, all OK.
Now after putting the first batch of functionally tested devices into the field, multiple devices are being returned, all with the same root cause: no output on the TLV62130.
Replacing the TLV makes the devices operational again.
We have been able to reproducing the defect, by disconnecting and reconnecting the device. After between 50x - 100x replugging cycles the devices die.
Schematic and powering is identical to the 1000s of other devices we have designed without issues: they get 15V max at the input.
PCB layout is significanlty different --> PCB could be a root cause.
However, when putting a TLV from an older PCB on the new PCB, we cannot kill the TLV anymore --> indicates PCB layout is OK, but we are facing a bad TLV part.
We've sent 'good' and 'bad' TLVs for XRay, with no differences to be seen --> genuine parts
PVIN and SW have been monitored using a 1GHz tek scope, using <1cm leads from probe to PCB (both signal and GND): all voltages, under- and overshoots within spec of the datasheet.
It is as if this batch of PCBs is using TLVs from a 'bad batch', is this a possibility?
