BulbCollector Forums
BULB DISCUSSION BOARDS => Christmas Lighting => Topic started by: glimmerglow on January 21, 2004, 11:34:00 am
-
I recently acquired about 12 miniature based carbon,u-shaped filament colored bulbs. Used a variac and slowly raised the voltage until some bulbs lit and others just barely glowed. Problem is, several lamps flashed out in a very short period of time, these being the ones that lit very dimly. What is the safest way to light these bulbs? I read that carbon filaments have a negative temperature coefficient, i.e. they will lower their resistance and draw more current as they heat. One of the lamps that still works shows a fairly steady current draw within a range of several volts. Should I use a constant current supply??
Any help would be greatly appreciated. Thanks
-
I wonder if the ones that failed had air in them. If they were only glowing dimly, and brighter ones survived, this could be the case. I can not think of any reason for using other powering methods; a variac should be fine.
-
quote:
Originally posted by Ross:
I wonder if the ones that failed had air in them. If they were only glowing dimly, and brighter ones survived, this could be the case. I can not think of any reason for using other powering methods; a variac should be fine.
Ross,
Thanks for the reply. I believe that at least one (of the three!!) that failed had air in it, as the filament sputtered like a firecracker fuse for a brief second. Glad to know it wasn't the lack of a current controlled supply that caused the problem.
Ed
-
quote:
Originally posted by glimmerglow:
I read that carbon filaments have a negative temperature coefficient, i.e. they will lower their resistance and draw more current as they heat.
If this were true, all carbon filament bulbs would require ballast resistors. It's possible for a small part of the temperature curve to have a negative coefficient, but it must be positive in the region of normal use or the filaments would always self-destruct when first energized.
------------------
Alan "A.J." Franzman
Email: a.j.franzman at verizon dot net
------------------