Nate,
After taking a look at your code I notice a few things. These are in no order.
1) you have no wait in your acquisition loop. It might be nice to put a wait in there.
2) try graphing in your producer loop as well as your consumer loop. See if there is a difference between the two.
3) is this actualy a response rate time or a difference in amplitude? It doesnt look like the second screenshot ever gets above 30.
4) how are you heating the thermocouples? is it consistent?
5) if you still think its a response rate problem it would be nice to see an X axis that has more than 2 points.
Let us know if you discover anything else.