Flexible PCBs are neat. It seems like it shouldn't be that hard to make them with the usual hobbyist techniques. Has anyone tried this?
The main problem I foresee is that the flexible material is not as heat-resistant as the normal PCB materials (fr4, phenolic, or whatever). This shouldn't be a problem for toner-transfer or photoresist etching, but it would be a problem for attaching components and for rework. Maybe a carefully-controlled hotplate reflow setup would work.
The data sheets for Dupont Pyralux (polyimide) have a specification that mentions 10 sec. at 288 C for soldering, and a few hours at ~190 C for lamination, but no mention of what the maximum temperature for the material is. Rogers rates its R/flex 1000 to an operating temperature of 150 C.
The other problem is finding the flexible copper-clad material in hobbyist quantities ... and at hobbyist prices :-) Farnell sells it, as does