it turns out the process of debugging a semiconductor die is *fascinating* and also what we've been doing at work the past couple weeks
-F
The hands-on bit of debugging an IC
So we had a problem with an IC our company is in the process of designing, where you could only apply 18 volts across its power rails before it started to break down and conduct huge amounts of current (it's supposed to be powered by 25 volts)
this is a relatively complex device, so it was important to find out *where* on the die the anomalous current was flowing, so we could match that up with the layout and identify what part needs work
conveniently, however, all that current generates heat.
-F
re: The hands-on bit of debugging an IC
So, based on the principle that the heat is generated where the current is flowing, right at the point of the breakdown, we can find where the problem is if only we have a method of identifying where the heat is located.
One's first thought may be a thermal camera, and that would indeed work, but three problems make it relatively impractical:
- Thermal cameras are expensive
- You need special microscope lenses designed to be apochromatic down to near infrared, which are also expensive
- The materials that make up an IC are relatively thermally conductive, and so the temperature differences we need to look at are quite small
So a thermal camera isn't the best option. But then, what else could we use?
-F
re: The hands-on bit of debugging an IC
As it turns out, liquid crystals have an interesting property that can be exploited here.
We placed, using an ordinary paintbrush, a very thin film of specially-formulated liquid crystal over the surface of the die. Then, by illuminating the die with polarized light, and putting a cross-polarized filter in the optical path of the microscope, the film of liquid crystal turns rainbow-colored as it rotates the polarization of light passing through it.
Then we apply power to the die, carefully adjust the voltage until the breakdown just starts.... and as if by magic, black spots appear in the rainbow pattern right on top of the hot spots.
-F
re: The hands-on bit of debugging an IC
@Felthry that's extremely fuckin' cool
re: The hands-on bit of debugging an IC
@starkatt it really is! and it's remarkably precise; depending on how close to the surface of the die the fault is located, and how well the liquid crystal is applied (a thicker film means worse resolution), you can pinpoint the location of the fault to within a micron or so
-F
re: The hands-on bit of debugging an IC
@Felthry frick!!
re: The hands-on bit of debugging an IC
@starkatt oh yes something i forgot to mention is that you can actually make this even more precise--if, instead of having the wafer under test at room temperature, you use a heated chuck to control the entire wafer's temperature to *just* below the crystal's clearing point (here that would be like, 27 or 28 degrees), you can roughly double the sensitivity
again, though, that's still subject to whatever's in between the actual hot spot and the surface of the wafer where the liquid crystal is; metal layers especially can act like heatsinks and pull heat away, leading to the hot spot appearing to be wherever there happens to be less metal near the actual hot spot
-F
