View Single Post
  #12   Report Post  
Posted to sci.electronics.design,sci.electronics.repair
Tim Williams Tim Williams is offline
external usenet poster
 
Posts: 257
Default CFL ballast design, and using dead lamps for repair

wrote in message
...
On Jan 21, 8:43 am, "Tim Williams" wrote:
Besides the electrolytic and bulb, the starter cap is very
failure-prone.


FWIW, of the modest pile of CFLs I autopsied from my own house, 100%

of the failures were from failed heaters.

This may be atypical. I get good service from CFLs running them

without enclosures; the enclosed, base-up fail modes may be different.

I don't know if any do fail specifically due to the start cap -- as you
note, the filament tends to fail first. I haven't measured any of the
caps to see the capacitance loss, but the color change is reason enough.

EPCOS makes capacitors specifically for this service; their graphs of peak
voltage vs. capacitance are reminiscient of the Write-Only Memory's
"insertions vs. pins remaining" graph. ;-o

Base-up tends to cook the capacitor; better ones don't mind so much. My
parents have a GE "long life" model on their front porch, an enclosed
(vented enough that bugs find their way in, but not so much that they find
their way out before dying and making a pile...), base-up fixture.

I installed them something like four years ago; the other outdoor lights
(upright, even) have since failed, but the front porch, surprisingly,
still works.

I think. Come to think of it, I didn't look closely last time I was by;
the front porch may've been replaced, most likely under pressure, the
other "low priority targets" simply being "left in the dark"... :-)

Tim

--
Deep Friar: a very philosophical monk.
Website: http://seventransistorlabs.com