Yet another one, an unsupported 3W resistor , with a fractured wire.
Doesn't matter what maker, none of them seem to take this failure mode into
consideration. The common factors bass combo (so far only bass ones ), large
Rs suspended off the board, can be mounted vertically or horizontally, the
fracture always at the pcb and looking at the cross-section of the wire and
a copper colour then presumably work-hardening via vibration of wires with a
high copper content.
I'd never seen Rs supported on ceramic bead insulators with this failure so
used them on replacements. Then i thought a rigid pillar could exaserbate
the problem by a fulcrum effect plus chance of them vibrating / rattling in
use. So later I used beads or PTFE vias for tin-plate (discarding the pins )
for most of the stand-off and finishing with a blob of RTV silicone.
This one I'll try blocks of that orange silicone rubber cut from the
pressure roller at the output of a junked photocopier. Small hole "drilled"
through each so making fuzzy contact with the R wire and theoretically damp
any tendency to vibrate. Incidently anyone know what the incredibly firm
adhesion system they use for bonding that rubber to the steel shaft running
through the rubber.
Has anyone seen the pro solution to this problem by any maker or any other
ideas?

With a number of such large Rs in a line then I tie each to adjascent with
stout silicone sleeving with slight tension in the sleeving and anchor to
the pcb at the ends.

--
Diverse Devices, Southampton, England
electronic hints and repair briefs , schematics/manuals list on
http://home.graffiti.net/diverse:graffiti.net