Re: Biax lamp query

In article <43CFCFFD.4C894D0F@xxxxxxxxxxx>, Jack wrote:
>Gentle people ?
>
>I have a number of GE 2 pin double biax compact fluorescent bulbs that I
>can?t get to work. P/N F13DBX/840 F13DBXT4-SPX41.
>
>The GE spec *** lists them with a G24d 2 pin base, internal starter,
>91 v. 0.17 a., but I?ve been unable to find further information on
>circuitry.
>
>I?ve tried them direct on 120 v. with a dropping resistor, with no
>result. No current draw ? the voltage at the pins remained at 120v.
>
>I made the assumption that since they contained an internal starter that
>they may have been self-contained, and this may be incorrect. Most
>fluorescent bulbs require a ballast, but there is no information
>supplied on this item.
>
>Can anyone shed some light on the subject (!) for me?

I have tried these with homebrew electronic ballasts.

In my experience, these need:

1. AC, or id DC is used, it may start only if one particular polarity is
used since the starters are at least sometimes not symmetric.
Effects here may vary from one manufacturer to another and from one
production run to another.

DC may also cause the mercury to drift towards the negative end, making
the positive 1/3-1/2 or so dimmer than normal.

2. In my experience, starting works better if there is inductance in
series with the lamp. Sufficient inductance may limit the frequency that
can be used.
Alternatively, especially with AC and especially with higher
frequencies, starting has some tendency to be easier if a metal object is
close to both electrodes. This affects electric field distribution within
unionized gas nad can help it ionize.

3. Sufficient current must flow while the starter is closed to get the
electrodes hot enough for starting. It helps if sufficient current flows
through the electrodes to make arcs form across them - this takes about 11
volts across each filament (electrode), maybe peak as high as 15 volts (in
my limited experience when I had a voltmeter across a filament). In my
limited experience in this area, this tends to require peak current
through the filaments at least around .25-.3 amp (in the regular 9 watt
twintube, which has similar current requirement). I have seen starting
occur without signs of an arc first forming across one of the filaments,
but starting occurs more easily when one does form.
However, current must not be so high as to burn out the filaments for
for starting to be ecessively hard on them. With DC, an arc that forms
across a filament during starting might be harder on the filament since
more average current must flow to get one started than with sinewave AC.

4. In my experience, twintubes and doubletwintubes intended for use with
120 volts have operating voltages around 60 volts or less. In my
experience, hot cathode discharge lamps require line voltage at least
about 30% above the arc voltage to keep an arc going with not really good
stability, and I doubt you will start with 120V and a resistor or reactor
ballast a fluorescent lamp with a nominal arc voltage around 90 volts. I
suspect this lamp is designed for use with 220-240 volts or with ballasts
that boost voltage if 120V is used.

- Don Klipstein (don@xxxxxxxxx)
.

Loading