Weird SENER output in visco-hyper-elastic model

The inp file is at:

https://docs.google.com/open?id=0B-ZhJmggjeGSNjJFYWxPZHo2TFk

 

Hi all,

 

I ran my model and sampled the SENER at element 2601 in SKINSUBSTRATE-1 (integration point), and see this weired behavior: the SENER goes to absolute zero and then comes back up. 

 

This looks weird to me, but the energy balance of the whole model seems reasonable. Could someone help me with this issue?

 

Thank you very much!

 

-Shawn