Maker Pro
Maker Pro

AF fluorescent overloading

C

C. Jensen

Jan 1, 1970
0
I am trying to determine the frequency characteristics of an 8W
fluorescent tube. For this purpose, I am driving it with a commercial
signal generator via a 12W audio amp and step up transformer.

Before the tube ignites, I get get a nice clean sinewave. After it
ignites, the waveform diminishes in amplitude by half and exhibits
distortion proportionate with the brightness of the tube (as I
increase the signal voltage).

I assume this is due to loading, but I am unsure of what approach to
take, since the amp seems suitably rated. Series resistance does
nothing.

Can anyone offer any suggestions?

Claus Jensen
 
P

Phil Allison

Jan 1, 1970
0
"C. Jensen"
I am trying to determine the frequency characteristics of an 8W
fluorescent tube. For this purpose, I am driving it with a commercial
signal generator via a 12W audio amp and step up transformer.

Before the tube ignites, I get get a nice clean sinewave. After it
ignites, the waveform diminishes in amplitude by half and exhibits
distortion proportionate with the brightness of the tube (as I
increase the signal voltage).

I assume this is due to loading, but I am unsure of what approach to
take, since the amp seems suitably rated. Series resistance does
nothing.

Can anyone offer any suggestions?


** I suggest you are just about to destroy that fluoro tube or your amp or
both.

A fluoro lamp must not be operated without some form of current limiting or
it will soon fail - seems that the limited power output ability if your
12W amp and the winding resistance and inductance of the "step up "
transformer are providing this now - by accident.

If you care to reveal details of what you are actually doing - ie what
amp, what "step up" transformer & what frequencies you are trying - then
your Q will make more sense.




......... Phil
 
T

Tim Williams

Jan 1, 1970
0
First of all, you cannot run a fluorescent tube from a constant voltage
source! You must have some resistance present, or else the tube will
explode or the amplifier will be shorted.

A gas discharge tube is best measured in the time domain. The two
characteristics are ionization time and deionization time. This is
comparable to a bipolar transistor without any negative base current turning
it off. It turns off eventually as charges recombine, but it doesn't go
very fast. Most tubes ionize (turn on) pretty fast compared to turn-off
time, with typical time for a xenon or mercury tube around 10us ionization
and 1000us deionization. The fastest controlled gas tubes are hydrogen
thyratrons, used for radar pulse generators (switching near megawatts in
several nanoseconds).

As long as the tube has ions in it, it'll conduct, given voltage. Over a
short period (a few microseconds perhaps), a slow tube might behave
resistively. For longer times, the plasma will respond by producing more
charge carriers (ionization), and longer still, deionization will take place
inbetween.

A glow discharge's V-I characteristic isn't very nice. Terminal voltage can
be quite high before breakdown (i.e., near zero current), while it maintains
itself pretty well at some voltage when conducting. That alone is negative
resistance, but the discharge itself also has a negative resistance
characteristic (voltage falls off by a few volts as current is increased).
This is why you cannot run a discharge tube from a constant voltage source.

Tim
 
Linear Technology LTC publish some excellent application notes on
flourescent drivers using ICs and transformers.
As sais time domain is the better way.
 
Top