I think this has been resolved.  Let's close the issue for now.
If something comes up, we'll reopen.

I believe this is fixed as discussed in email (but put here to make it public).
 There was a call to substitute() in the R code to get the name of the eSet, but
I had put an as() call before that in case an exprSet was being passed in.  In
that case the substitute would return a blank name, so you likely were running
into having multiple datasets named "".  I submitted a fix, see if that helps
things.

I am seeing this:
> eset2pgsql(teuffel2004, db)
[1] "eSet  is already in the database"
NULL
>                                       

But it isn't there and this happens even when it is the first time
I have tried for a given eSet.?