Fritz Schlunder wrote:
wrote in message
ups.com...

I am using the 120V LED bulb to light my computer keyboard. Since the
bulb is fed by the UPS, I prefer the LED bulb in order to reduce the
load on the UPS during power failure. Since I did notice a voltage peak
in the UPS when it turns on, I wait for it to stabilize before turning
anything else on, including the light. Still, two bulbs did not survive
a month each. I believe this is due to poor workmanship. Trial and
error at the prices of these bulbs is not good enough. I am hoping for
someone to report good experience with some brand and/or vendor.


Ah hah! There is a fairly high probability the cause of your LED bulb
failure is not due to any inherent manufacturing flaws or low quality
workmanship or anything of the like really. The problem is most likely an
incompatibility between your specific LED bulbs and UPS power source.

Typical LED bulbs designed for operation from 120V AC often use a capacitive
ballast (with a small inrush limiting resistor) to limit the current through
the LEDs. This is a very cheap, efficient, and easy to implement solution,
so it is very commonly done.

The problem is, a capacitive ballast must be operated with a truly
sinusoidal power source (like the mains are). The current through a
capacitor is I=C*dV/dt where I is current in amps, C is capacitance in
farads, and dV/dt is the time rate of change of voltage per unit time in
volts per second. A clean sinusoid has a very slow maximum rate of change
of voltage, so the peak current that the LEDs normally are subjected to is
fairly low. In other words, the current crest factor of the capacitive
ballast when powered from a sinusoidal source is fairly modest.

Your UPS most likely does not produce anything at all like a clean sinusoid
(especially when powered by battery, although depending upon it's design it
may never produce clean sinusoids even when the AC mains are fully
energized). If you look at the output on an oscilloscope you may find that
it is rather a square wave output that goes between 0V (with significant
dwell time), 150V, 0V, and to -150V or something like that. Or it could be
some kind of stepped square wave that more closely approximates a sine wave,
but is still very much not a sine wave since it has high dV/dt edges, or at
least steps. These high dV/dt edges will cause very high peak currents
through the capacitive LED ballast, and the current crest factor will be
outrageously high, likely much higher than the LEDs are specified to be able
to handle reliably. Hence your LED lamp failed in very short order.

Had the LED lamp been operated from a truly sinusoidal power supply, they
would likely have been much more reliable, quite possibly meeting the
claimed life expectancy ratings.


I greatly appreciate your detailed explanation, as well as Noozer's,
for the reason of my problem. However, I need a solution. I do not need
LED light bulbs for general lighting, since I prefer the compact
fluorescent bulbs for this purpose. I need LED light to illuminate my
keyboard because of its very low power consumption. Its being
directional is also a boon.

Noozer points out that the electronic components within the LED bulb
are cheap, implying that with better quality electronic components the
bulb might last longer. Would it be possible to place between the UPS
and the bulb a filter, made of quality components, that would smooth
down those sharp edges? If so, where can I find a good design for such
a filter and specifications for its components? I am willing to
consider building one myself.