I had a painful experience with this a while ago.
I thought it would be in my best interest to “tack” the two pieces
of fine silver together with hard solder, then count on the enamel
to act as glue for most of the plaque. Two things went wrong: 1)
Because I was afraid of damaging the enamel with solder
contamination, I didn’t use enough, and what little I used, most was
lost in clean-up. 2) The COE (expansion coefficient) of the silver
was significantly different of that of the glass.
As I was doing final assembly, the plaque couldn’t take the strain
and “popped”. The enamel sheared off and I was quite distraught.
My new way of doing this (in silver) is to use eutectic solder and
depletion gild until the copper is removed. That way I can do a
decent job soldering, but still not damage the enamel with zinc,
copper, or any other nasties. Besides counterenameling, I might
anneal the enamel in my kiln, to reduce strain (champleve-so much
silver, not much glass, uneven strain).
Hope this is helpful to you,
Margaret