Vibration & Electrical Probe Conductivity

[shadowg]

Hello,

Has there been any study on the vibration of probe pin (used in semiconductor wafer testing) and its effect to electrical conductivity (maybe the contact area change)?

If there is significant impact of vibration on electrical conductivity, how should I model the system to find the optimum damping?

Any reply/replies are very much appreciated.

 

[IRstuff]

Not sure what they do now. But 15 yrs ago, most probing was done either with relatively high scrub force or on vibration isolated tables. The convention probe cards at that time used a cantilever shank with a bend near the tip. The probe station was adjusted to provide a visible scrub mark on the order of about 50 microns as well as uniform scrub across the die.

TTFN

 

[shadowg]

You mean the vibration isolation of the table can help in minimizing the vibration of the probe?

What about electrical conductivity of the probe? How to model the vibration of the pin as a function of time? Is is similar to sine wave? Any study on this? Ideally no test signal should be transmitted through these pins before the amplitude of the vibration becomes zero.

Also does the contact surface of the bond pads have any damping property? They consider this in silicon wafer fabrication?

TIA

 

[IRstuff]

I'm not sure where you're going. Truth be told, in over 16 yrs, the only vibration I had to contend with was my hand shaking while I adjusted manual probe needles.

The probes are usually something like 10 mil tungsten or tungsten carbide, although the low capacitance probes tend to be significantly slimmer. As these are usually used for monitoring, there is no issue with conductivity. Higher currents are run through the larger probes with several in parallel, so again there is no problem.

TTFN

 

[shadowg]

I'm involved with Wafer Testing. We have many probers that are used for this purpose in large production scale. One of the effort for engineer like me is to increase the yield. So, I'm looking for opportunity in this area.

We have many products that are being tested with more than 20,000 touch downs on each wafer in roughly 2 hours of test time.

So, at 10,000 touchdowns an hour, I'm beginning to think of the vibration's impact (as little as can be) on contact of the probe and bond pads. The vibration of the chuck that holds the wafer, and the stiffness of the probe pins/probe cards, etc., all these I'm considering. As the bond pad size keep decreasing (about 3 mils), whereas the probe tip size is at min 1.5 mil (the one that we have), vibration's is becoming critical IMHO.

From your reply it sounds like your usage of probes is in lab work, am I right? Is so, then you manually start or stop the signals on the probes right? I apologize if this is not the right forum to ask these questions.

Again TIA.

 

[IRstuff]

I've been involved in both production and lab probing, albeit a while ago. I would guess that you're barking up the wrong tree on this one, since the amount of touchdown force is sufficient to scuff the bond pads and that force should be more than enough to get rid of most common vibration levels seen on the production test floor.

Not sure what your product is, but production wafer probe generally does not get into measurements that could be affected by your concerns. At wafer test, you should be primarily interested in gross functionality and gross performance grading.

TTFN

 

[shadowg]

Thanks IRstuff,

Well, I guess it's better be well-informed than making speculations. Thanks for your input on vibration (negligible) of probe pins. I'll do a few more check ups on this topic before calling it goodbye to vibration.

Thanks for the replies, this is a good site. I've book marked it in my favourites.