aflannery at invensense.com
Thu Sep 23 12:30:36 PDT 2004
I did some work on the newer bonding tool. There were a couple of things
wrong with it.
1. The center chuck was loose inside the frame. This lead to gross
misalignment as the wafer position would shift during clamping on the
2. The flags had become loose over time. They would deflect enough that they
came into contact with the stainless steal of the frame and stick. I
adjusted these so that they do not deflect with moderate pressure and move
3. It looks like the pivot pins on the flags may have become worn over time.
4. The flag arms may have developed a slight bow.
I replaced a couple of the pins and arms.
To all the bonding tool users.
1. If you are doing a manual alignment with no flags, please use the older
tool (the darker one).
2. Aligned bonds should use the newer (brighter) bonding tool. Verify that
the center piece of the tool is still tight before bonding with it.
3. The flags are not supposed to be loose. They are supposed to glide with a
small amount of resistance. If you loosen them because you think that makes
them pull easier, you are doing a bad thing. Please don't do that. If you
decide that you need to adjust them, for whatever reason, please let me
know. I would prefer that you didn't.
4. I had some feedback from Chad Brubaker about two things which should
improve both slipping and flag pull.
A. Make sure the wafer bow is set correctly. If you are concerned, you
can subtract 100 - 200 um from the stack height. This increases the force of
the wafer bow slightly.
B. Pull the flags one at a time and insert a wait state of 10 seconds
between each pull. This works better than pulling the flags simultaneously.
I am hoping this solves the problem with flag pulling getting stuck. If you
do an aligned bond, in the next couple of weeks, please drop me a line and
let me know what your experience is (i.e. did the flags pull).
Anthony F. Flannery Jr.
Director of Development
2900 Gordon Ave., Suite 203
Tel: 408-720-8482 ext. 203
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the evbond