Flourescent bulbs

This is the place for any magazine-related discussions that don't fit in any of the column discussion boards below.
Robert Reed
Posts: 2277
Joined: Wed Nov 24, 2004 1:01 am
Location: ASHTABULA,OHIO
Contact:

Flourescent bulbs

Post by Robert Reed »

I have a problem with the flourescent fixtures in my home in the heat of summer. House is not air conditioned but generally stays cool as its in the shade. However during long periods of stuffy weather, the humidity eventually filters in . This is when these bulbs give me problems. They either come on half lit or won't start at all. Sometimes repeated switching cures the problem, sometimes not.
My limited knowledge of these devices are that upon power up, a high voltage spike is fed to them to "ignite" it. This is fed from a starter circuit or some sort of "ballast" transformer. Once ionized ,normal line voltage is enough to sustain illumination. I'm not even sure if this is correct, but anyhow, could humidity be loading hi impedance and high voltage start circuitry preventing turn on. Have any of you experienced this problem or know of a fix?
We dont see tropical weather here in northeast ohio, but dew points can reach 70 plus for a period of time
User avatar
Chris Smith
Posts: 4325
Joined: Tue Dec 04, 2001 1:01 am
Location: Bieber Ca.

Re: Flourescent bulbs

Post by Chris Smith »

On some bulb assemblies, the “Ground” or reference plane is the light shield.<p> When humidity or other conditions interfere with this “reference” point or the solder Joints, or if the ground is not present, the bulb will fail to ignite, stay lit, etc. <p>Also some cheaper models have paper wound Caps to drop the ballast voltage, and may absorb water into the paper. <p>NOT good for a cap? And there are other causes. <p>As to the high voltage spike, this is true only on some models. <p>Most like the twin 40s have heaters in the ends, and the ballast pre-heats the gas, with a ballast dropping the 120V to 5.6 volts or so, at .43 amps to these heaters. <p>This is like a tube radio, it pre heats the gas, and its also a “emitter” of electrons to ignite the bulb, requiring no “spike” to ignite the gas. <p>There are many ways to “Run” a fluro tube.
Enzo
Posts: 276
Joined: Tue Sep 14, 2004 1:01 am
Location: Lansing, Michigan, USA
Contact:

Re: Flourescent bulbs

Post by Enzo »

Also, measure the line voltage when this happens. Your hot humid weather is also when draw from the power grid is highest. This will frag the AC line voltage down. You might be suffering from lower line voltage making it harder for the tubes to strike an arc.
User avatar
jwax
Posts: 2234
Joined: Mon Feb 09, 2004 1:01 am
Location: NY
Contact:

Re: Flourescent bulbs

Post by jwax »

Do ALL the fluro fixtures have the same problem in the house, some of them, or just one or two?
After verifying OK line voltage, you need to replace fixtures.
WA2RBA
User avatar
sofaspud
Posts: 531
Joined: Wed Feb 13, 2002 1:01 am
Location: San Antonio, TX
Contact:

Re: Flourescent bulbs

Post by sofaspud »

Could it be similar to what you might encounter with a car battery (where climate creates a high impedance at the connection), of tube and socket in this case? Maybe a twist of the tube next time your fluoros act up will allow you to experiment for this cause.
Ed446
Posts: 98
Joined: Sun Jul 21, 2002 1:01 am
Location: Burns, TN
Contact:

Re: Flourescent bulbs

Post by Ed446 »

You need to also clean the pins on the lamps that make contact in the fixture.
Will
Posts: 310
Joined: Tue Feb 11, 2003 1:01 am
Location: Katy Texas
Contact:

Re: Flourescent bulbs

Post by Will »

I have been trying for some time to determine (By reading these threads) how modern fluorescent lamps work - So far unsuccessfully because, among other things, the contributors who seem to know are too brief in their descriptions.
Let me tell you precisely how the old ones work then I may precipitate some enlightenment.
Rough descriptions of a 'ballast' don't really help - the ballast consisted of a choke, a capacitor and, sometimes a starter switch. If the starter switch was not in the so-called 'ballast' then it was furnished separately - usually plugging into tht e side of the fitting so that ir was easily replaced. The normal 120 vac mains voltage was not enough to ionize the (Mercury/Mercury Vapor ?) at room temperature so the electrodes at each end of the tube were furnished with heaters. The 120 vac ground was connected to one heater/electrode connection at one end of the tube and the 120 vac HOT to one connection at the other end of the tube. The two remaining connection were connected to the 'starter switch' so that, when 120 vac was applied current would flow in series through the two heaters and the starter switch (SS). The SS was simply a timing device (Ususally a heater heating a bi-metallic switch) which, after a short period of heating, switch OFF. At this point, with the tube contents heated, the tube would fire from the 120 vac (No high strike voltage needed) and be ON. Unfortunately, once the tube contents were ionized and current flowing they represented a near short circuit so that, all other things being equal, an abnormally high current would have persisted and may have damaged the tube. This is where the choke/inductor came - it would reactively limit the tube current without using energy as would have a series resistor (Which would cause real energy/heat loss which would have to be paid for) . This meant that the tube current would be, near enough, 90 degrees out of phase with the applied 120 vac. Since utility watt-hour meters work near properly and register only true kW this did not cost the user too much but, the utilities did not like to be stuck with the voltage drop which this, 'wattless' current effected in their distribution lines so, at their insistence, capacitors have to be furnished in such fittings. The purpose of the capacitor is only to correct the power factor of the fitting - The combination of capacitor and choke became popularly known as the 'ballast' _ I still don't know how modern ones work ?
BB
User avatar
Chris Smith
Posts: 4325
Joined: Tue Dec 04, 2001 1:01 am
Location: Bieber Ca.

Re: Flourescent bulbs

Post by Chris Smith »

The modern fluro is different slightly from the old “Sparky” type assembly. The modern ballast is a “Dropping Resistor” usually formed from a Current transformer which lowers the 120 volts down to 5.6 volts, at ½ an amp to prewarm the gas through the heaters. Any higher voltage to the heaters burn them out. Each heater is like a 2.5 watt light bulb element burning or glowing at each end. <p>The heaters are also “emitters” in that they pass all the voltage/current [120v] to the tube gas through one pin each. The other pin forms a loop and has the dropping elements in it to drive the heaters at less than 6 volts.<p> Generally speaking, no spark or boost is needed for a tube but that is in a perfect world. <p>Some ballasts are a “power convertor”, much like caps are for a Cap Type motor. <p>The current and voltage are brought into phase, thus the power is in sync with the voltage at the same time. This is more efficient than ones that are out of phase, like current high, voltage low, the way AC is normally. Reactance timing does this.<p>Still other ballasts can give a buck or a boost signal at the right time to jump start things several ways. <p>The start button, the auto start ballast, or just a electronic circuit that charges up and then “breaks down” at the right moment. <p>The “ Neon” starter for the Fluro tube charges up a voltage until the tiny neon bulb conducts and lights up, and this causes the circuit to the fluro to have a small charge in power through the tiny neon, but only after it conducts and creates a path This build up and then passage through the bulb is a surge. Even as tiny as it is, it fires off the fist buck or boost to the main bulb. <p> Dragging the 120 volts leads across the terminal in most cases for a slight arc, will also light up most fluros. <p>Still other ballasts use a CAP of the right Reactance value to drop the heater voltage to the right value, and just the delay surge through the cap and the cap it self, cause a buck to start the tubes.<p> The use of paper caps over expensive caps and the transient loss of the cap charging add in a great amount of boost to the circuit. I had one that made a “snapping noise” each time the power was turned on, and I assume it was the magnetic properties that cause the cap to expand or contract over a few milli seconds, and this acted like a whip. As noisy as it was, that tube always lit on time. When I replaced it with a more expensive cap, it was quiet but hesitant to start at times, except in warm weather. <p>Common Fluro tube don’t like cold weather, and give all sorts of start and run problems in the 40 degree temperatures or lower. Modern ballast and other methods such as high speed switching take care of this. Twist Fluros work just fine around here in winter, and have been tested to minus 20, with out a flicker. <p>If you crank out a high power, high voltage short pulse to the tube, you don’t need a ballast, pre-heaters, or anything except one pin at each end of the bulb. With out the wasteful heater current, which is the majority of the total current used, the same or less current over all is needed with high frequency, and the tubes last longer, shine brighter, and take all weather conditions in their stride.
Robert Reed
Posts: 2277
Joined: Wed Nov 24, 2004 1:01 am
Location: ASHTABULA,OHIO
Contact:

Re: Flourescent bulbs

Post by Robert Reed »

Jwax, Enzo
All flouros in home suffer, but to different degrees. I am fortunate to be blessed with line voltage that doesn't vary more than 2 volts all year long.
Sofaspud, Edd
I will check contamination on contacts.Although they work fine again when humidity clears out.
Chris
Not sure about your radio tube analogy. Radio tubes are gasless, actually in a vaccuum-hence vaccuum tube. Their heater is used to super heat a kathode by boiling Thorium to its surface, thereby allowing a free flow of electrons to the plate element.However I think the old Ignitron tubes were gas filled. As to hi frquency operation and efficiency--My first job in the electronics feild was in mobile radiotelephone. As proof to the occasional cranky operator, I would hold a 40 watt flouro near the antenna and key up the transmitter. The bulb would glow at near full brilliance. In a semi darrk garage, this was an amazing site and the operators jaw would drop in awe. I too was in awe (but never showed it) as I never quite understood it. Just that RF enegy was being absorbed by the bulb and transferred to light energy. My point being that your input of efficient high frequency operation must be right on the money.
To all
Thanks, I now have a much better understanding of how these bulbs work.
peter-f
Posts: 114
Joined: Fri Nov 26, 2004 1:01 am
Contact:

Re: Flourescent bulbs

Post by peter-f »

(hey- a really useful thread- unlike that 11 pager!)<p>I find the cheaper bulbs (Ikea imports) need to be in DRY areas to work right.<p>My garage is usually dry, but up to 70%RH will drive the 1 bulb there nuts... it crackles like a high-voltage arc... and that remark about cheap paper-wound caps seems to supplt the answer.<p>(Too bad opening it up will likely destroy it)
User avatar
Chris Smith
Posts: 4325
Joined: Tue Dec 04, 2001 1:01 am
Location: Bieber Ca.

Re: Flourescent bulbs

Post by Chris Smith »

Robert,.....<p>Think of your hot “Boiling Thorium ” and Replace it with the "Mercury gas” for a “analogy” conversion. <p>Both need the preheat to work. <p>With out the pre-heater / heater, neither tube conducts or operates well? <p>And Tesla had it all down pat with his “wireless” transmission of power, sort of? <p>My Tesla coil would light a Twin Forty at over eight feet away. <p>About 200,000 plus volts, and when It landed on me one day after falling over, I thought fleas had invaded me. <p>No shock or burn, just fleas crawling all over my skin. Ill bet my eyes glowed?
User avatar
dacflyer
Posts: 4748
Joined: Fri Feb 08, 2002 1:01 am
Location: USA / North Carolina / Fayetteville
Contact:

Re: Flourescent bulbs

Post by dacflyer »

i used to have the same problem with lights not lighting right..especially in cold damp or very humid weather.. and the fixtures i have are H.O.
type. 800Ma.
i thought i had a good ground out to the shop.. but i didn't.. so easiest fix i had was to tie the neutral to the ground...since they was connected this way in the breaker box anyway.
and this fixed the problem..
so you might be able to tie the ground of your fixture to the neutral and see if this fixes your problem, good luck..
yesiamwds
Posts: 5
Joined: Sat Jul 23, 2005 1:01 am
Contact:

Re: Flourescent bulbs

Post by yesiamwds »

<blockquote><font size="1" face="Verdana, Helvetica, sans-serif">quote:</font><hr>Originally posted by Chris Smith:<p>My Tesla coil would light a Twin Forty at over eight feet away. <p>About 200,000 plus volts, and when It landed on me one day after falling over, I thought fleas had invaded me. <p>No shock or burn, just fleas crawling all over my skin. Ill bet my eyes glowed?<hr></blockquote><p>You didn't say that you used 6' of your body to light those lamps! :D
User avatar
Chris Smith
Posts: 4325
Joined: Tue Dec 04, 2001 1:01 am
Location: Bieber Ca.

Re: Flourescent bulbs

Post by Chris Smith »

Only once! :eek:
User avatar
haklesup
Posts: 3136
Joined: Thu Aug 01, 2002 1:01 am
Location: San Jose CA
Contact:

Re: Flourescent bulbs

Post by haklesup »

I've learned a few things in this thread but I would like to point out another possibility.<p>The symptoms you describe are also indicitive of end of life of fluorescent bulbs (especially the 99 cent variety). In locations where many of these lights are used, failures often occur in waves because bulbs are replaced in groups from the same box (lot) and experiance similar usage so tend to fail at the same time.<p>Changes in temperature (often associated with changes in weather and humidity and usage of A/C etc) can push a bunch of marginal bulbs over the edge since the ignition characteristics of the gas are at least in part influenced by temperature. (case in point, my garage lights barely start below freezing)<p>In any case, if the bulbs are of the sort with heaters at the end, check for darkening of the glass at the ends, this usually indicates it is close to dead.<p>You say the phenominon is reversible with varying humidity but have you done a control experiment on a dry day by cranking up the AC to see if it might be temperature instead?
Post Reply

Who is online

Users browsing this forum: No registered users and 31 guests