IC Chip Pad without buffer

[goatmxj666]

Hello,

I'm testing my chip right now.

But it doesn't work ideally, so for the operation, I connect the external source to sepecific pad.

I connected function generator and applied pulse (D=50%, 10kHz, 0V-5V, supply=5V circuit) signal.

But at some point, the chip stops working.

Inside the chip, there is no buffer and ESD between the pad and the output node. It's just a direct metal-to-metal connection.

And it works well at first, and then it doesn't work anymore.

After a few iterations of swapping out the probe that connects to the pad, I think there's something physically wrong with the pad.

But is this really possible?

The pad is connected to the drain of the front-end MOSFET and the source of the back-end MOSFET.

 

[dick_freebird]

Pulling an I-V curve at the pad can show you some things.
Especially if you have some virgin units to test. Loo at
repeated pulls to see if there is "creep" indicating latent
damage or charging. Run some past the rails to a destruct
event (open or short) to see how close to the edge you may
be.

If you rescrub the probe does it ever get better? Is your
probe station on a good isolation table so that floor noise
can't kick probes off?