I made a ring with a setting that essentially consisted of one 6mm round bezel cup set within a larger outer bezel cup. The inner bezel cup was prefabricated, the outer was made from a strip and base plate.

The inner bezel was to have a cabochon, the resulting outer channel was to have a "ring" of twisted wire soldered into it, serving as a decorative border around the inner stone, to be set after soldering was completed.

All the soldering went ok until it came to the last soldering task: soldering the ring of twisted wire into place. I placed the ring of twisted wire into position within the outer hoop, liberally applied flux around it and on the floor of the outer bezel, and wedged four small pallions of easy solder at regular intervals around the ring of twisted wire, wedging them between either the inner or outer bezel walls. I heated the piece from above, the solder flowed, and, although it secured the twist of wire ok, you've probably guessed that some of it flowed upwards on to the wire, ruining its clean lines. I tried filing away the excess solder with bits of emery paper held edge on, but this only ended up making things look more unsightly. I soldered the whole setting off the ring shank and binned it.

Can anyone suggest how I should have gone about things to avoid solder getting on to the upper, visible side of the twisted wire? Was getting flux on to the upper surface of the wire my downfall (on the basis that what the flux reaches the solder will probably reach too)? Should I have sweat soldered the outer hoop (i.e. the floor of the outer bezel) then put the twisted wire into position and then simply re-heated the sweated-on solder? Would this have reduced the tendency for the solder to flow upwards?

I did think about sweat soldering (and have done it successfully already on two other different projects), but my main concern (apart from the risk of re-opening one or more of the 5 previous solder joints) was that if I used the sweat soldering technique I wouldn't be able to see when the solder flowed as the floor of the outer bezel was entirely hidden from view by the twisted wire placed over it. I figured that wedging pallions at intervals around the twisted wire would at least allow me to see the solder flow. I did see it flow; the only problem was that some of it flowed where I didn't want it to.

Before I re-attempt the same project any advice would be appreciated.

Mark.