Smoke from chimney when catalyst engaged - Update

Our Buck Model 80 stove seems to be bypassing quite a bit of smoke around/through the catalyst when it is engaged. There are four possible ways this can be happening (that I've thought of at least):

1) I've had the cat housing down several times and had to fabricate a replacement gasket from fiberglass insulation. Am ordering a factory gasket to eliminate this as a bypass route. This seems the most likely.

2) The gasket between the cat and housing could be bad, though when I had it down in the spring it appeared to be fine.

3) The sliding bypass damper could be warped, allowing smoke to get by it. Not sure how to test this except to check around the perimeter with a feeler gauge.

4) The catalyst itself might need replacing. This is the start of its 4th season, so I wouldn't expect it yet. It does climb to a little over 1000° F with a small fire, though it seems to take longer to get there than when new. Can a cat fail in sections, where part is working and other parts not? Not sure how to test this other than replace it.

Any suggestions?

I received the new cat housing to stove body gasket yesterday. The gasket I had made appeared to be all there, though some was stuck to the housing and some to the stove. The new gasket is quite thin and appears to also be made of a white fiberglass.

While cat was out of stove, I checked the gap around the closed damper. A .011" feeler gauge would slide in easily all the way around, but a .025" would not. Did notice that the damper would sometimes get cocked a little bit upon closing and one corner would be open slightly. The push-pull damper rod is attached in only the center of the ~ 14"x4" sliding damper plate with a pivoting connection. Unfortunately, there is no way to see the damper when the cat housing and flame shield are in place.

Also while cat was out of the stove I wet cleaned it with the hot vinegar and distilled water method described here: http://www.condar.com/combustorcleaningmanual.pdf.

After baking the cat for at least an hour in the oven to dry, I reassembled everything and fired it back up tonight. I think the cleaning was the answer. It went active much quicker and was glowing by the time the probe reached 800° F. Couldn't see the exhaust well due to darkness, but it appeared to be pretty much smoke free again and the typical "burnt smoke" smell was back.