Deleted

i do care for my stuff, for example my westinghouse econ-nova CFLs (one of the very first CFLs sold in the usa). i lit them once and properly then i put them away, still as good as never used. i do use a lot of lamps in the house but i always operate them properly, and the rare, special stuff i do not use. just for my collection and to show to other collectors.

i do admit that i have done experiments by overdriving lamps etc but i only do it on spent lamps i got from work, especially if they are modern and cheap stuff.

the glow starter by the way is hard on lamps. i heard the arlen pulsestarter is a much better choice but i have not tried one yet.

@imj:
Well, "profit first" is, what you are following as well, you only may not name it that way.

If such contraption have it's benefit, there would be plenty of users, who would pay for that.
In fact there were even way simpler designs allowing the same degree of electrode treatment for starts: Manual push-button fixtures. They are even less prone to the "operator errors" (permanent preheat mean no light...), frequently made with push button control (one button to switch them ON, second to OFF). And they were cheaper than the automatic starters.
But they disappeared - people simply don't want to fiddle with the fixture to start it, they want only to flip a switch to get the light, nothing more.
Therefore the way more complex and expensive systems, sacrificing the lamp life and fixture reliability (more components) just to need no more than simple power ON to start the lamps. And they won on the market.

Even the instant start, harsher on the electrodes, are valued for their ability with instant light to compete with the incandescents, even when that mean shorter lamp life...


I would like to see, how would you explain to your grandparents, how to use such fixture and why to prefer that over the automatic start commercially available systems.



And lamp abuse: It is something like fireworks: You expect it would do something spectacular. But what it does, it teaches, how laps may behave, when a failure or error (leading to similar conditions) happen in a real installation.

I am with Imj on this

I put the common stuff in the shops in normal use on normal Switch Start, or for "high wear" effects (strobe, lightnings simulator etc). Stuff beyond it (even fairly modern which is out of production anymore) use sparingly, and stuff which is rare (T12s etc esp older ones) very sparingly when needed for deco for special events, "haunts" and so on, with no much use hours and on good gear (Perfect Start ballasts)

With EOL lamps it splits :

Not "cheap new" ones i keep well and the most i ask from them is a couple starting attempts and thats it, or to glow dimly on low power (Switch Start with starter removed which makes the lamp glow dimly at low current, or very low power HF which is to weak to do damage) but still makes great "haunted" effect

"Cheap new" ones do get to flash on startres or strobe on Perfect Start B when i need a haunted show but thats about it - i dont do "cooking" or other stuff like that which i consider sick as well

Looks like some members here have fairly easy access to older tubes (Lifelines, Westys etc) and they go and "cook" EOL ones and tip them in the recycle centers, while i am in search for getting one in any condition just for the sake of keeping it

The starting voltage either does not ignite the lamp, or it is clamped by the lamp. There is nothing else.

The open circuit pulse voltage could be whatever high, it get clamped by the lamp to whatever low voltage the lamp suffice with for the ignition.

What causes the wear is the high voltage drop between the cathode and the discharge plasma.
This voltage drop is high, when the electron emission does not suffice to handle the current coming from the ballast. And as the emission is strong function of the electrode temperature, it require the temperature to be sufficient whenever there is current flowing through the plasma.

During an inductive kick ignition the current equals the ballast current, so it is the same as when the lamp normally burn. So the cathode wear rate is the same as well, the ignition only set the moment, since the current flow. So igniting the lamp by lower current does not make sense for lamp life at all. The only purpose for a "soft start" would be to ease the light shock the people (and other animals) have to handle when there suddenly appear bright light from complete darkness. But for that the startup would have to be way slower than the limited current starter would allow.
By the way it is easier to modify the electronic starter to behave in such way: Instead of instantly giving way when detecting the lamp ignition, it would gradually lower the phase angle of its conductivity, so leaving larger portion for the discharge, so light. For your 58W tubes it only mean using the "4..18W series" S2e (according to Philips nomenclature) type - the transition from "preheat" via "ignition" to "Fault" mode is such gradual there (it is given by the circuit simplicity), where the starter's "fault" mode would be the mode, where the lamp would light normally. The only drawback would be the need of given OFF time (~10..20seconds) after switch OFF allowing to reset the starter from the "fault" mode, so it would be able to restart the lamp.

The electronic timer you described is the same complexity as the complete programmed start electronic ballast, so I do not see any advantage in the prospect of reliability, it only remain bulkier than the magnetic gear...

Yes, but only to the extent it make sense.
So when the extra lamp treatment would mean too much complexity, while because on how the lamp is operated it mean only marginal improvement, I better stay with as simple gear as possible.

That include the simple selfoscillating electronic ballasts even without the PTC (when not cooled down properly before start, it clamp the voltage, so slow down the electrode heating in the glow mode, so make the starting even more damaging).
After some experiments I gave up the "Rapid start" modifications, as the lamps tend to start prematurely anyway and the required extra ballast protection clamping slowed the heating up as well, so bottom line it was rather worse than better compare to the "instant" start original (well, it is not the true instant start, as the electrodes are mainly heated up by the external circuit and not so much by the glow discharge).
So the only true programmed start is the dimmable ballast based on this, where it came for free as part of the controller.

But when some feature is wanted for the sake of comfort, I accept it's cost on the lamp life, as the top priority is how well the fixture fulfill it's job, what include the operation economy. The lamp life is after that in it's priority.

I run life tests. I am running a pair of F30T8 GE lamps in a dual lamp RS high power factor fixture. Yes GE F30T8s start fine in this service. Sylvania will not, nor will the Duro Test F30T8 Optima 50s I got a case of.  One of the GE lamps is Hungarian made the other is Indonesian made. The Hungarian lamp has a few thousand more hours on it than the GE. The Hungarian made GE has brown aging on it, that is tan and equally brown, sorta smoothly smeared. The Indonesian lamp has end banding and a pronounced but not dark brown spot that eventually will be a black tear draped spot.

I have a Sylvania F14T12 lamp in my bedroom ( master bedroom) that stays on 24/7 when I use a diffferent bedroom, which I normally do.  This lamp likely has over 20,000 hours on it.  There is a picture of it in my gallery, and it has not changed any.  The "night" light is an old desk lamp that was converted to a starter years ago. I got it at a thrift store free because it wouldn't light. Well, the wires were wrapped around the starter pins and taped and the starter was slow. I put in a starter socket, and later had to replace the switch. I have a Dura FS-2NA (automatic reset) and it glows the ends usually once and well. Right now I have a cheap Sunlite F15T8DL ( daylight ) in it, and I thought at first it would not last very long at all as within a week there was substantial "end banding", but that after a few more weeks went away, and now many many months later there is light tan on both ends.  In the kitchen I have one fixture ( both dual 40s high power factor rapid start) 2 Sylvania D865 lamps.  The other one has a Feit DX lamp and a GE new 6500K lamp.  I was going to use both GEs but I dropped one.  There were originally CXL Topaz lamps in there, but one got black spots and wouldn't start unless I regularly gave it a few starts in a preheater. One of the Feit bulbs did the same thing. I currently am running 2 of those CXL lamps in the 1946 Half Piper at the radio station.  I am running a found in the trash close to new cool White Mainliner next to an ACE CWX. Right now the ACE is a bit brighter, but that won't last. Also I am running a GE that had been in a fixture for a short time that abused it, and it had some dark almost black spots, but they have almost gone away, and I am getting some nice ( short ) end banding started. I put in a new ACE bulb to go with it. The GE is the old but outside etch. It now is brighter than the ACE bulb. When bulbs reach EOL I bring them home and finish them off on the 40 watt solid state test unit. Often, these lamps will run for hours in preheat service before they start blinking. Once that happens it goes on the solid state ballast. 

i do preserve very old,rare,historical items.
i tend to use older well made items if safe/practical to do so.if not i upgrade them.if i modify i try to make the mod 100% reversable.
and tend to only mod stuff that is trashed anyway.like a 1939 fluorescent desklamp that is now 12v with led nightlights.it was scrap metal when i got it.
sandblasted,painted,and modded.i would not do that to a similar unit in presentable original condition.everything else is use up and pitch.vintage lamps are set aside for collectors and those that want period correct lamps for their fixture.in my shop areas i have overdriven(2x) t8 950's in use.
i have plenty of spares and dont care about the shorter life.i need lots of light!
the beforementioned rare items are safely packed and stored in a controlled environment.these items range from original audions to more modern ham gear.
i tend to hang onto very rare items and stuff i will restore/use.

My latest life test:  2 lamp fixture intended for F17T8 with electronic ballast. One lamp GE F17T8 SPX41, other F18T8 CW Indonesain made GE. Brightness is close, the color on the SPX41 is of course better.  NOW, which bulb will last longer? We should all know that answer!

@Medved - About the clamping voltage during starting I don't fully agree. During one of the experiments with the 2x58w lamps I modified it to be normal starter connection and it was a Osram starter but China made so it was very fast to close. The first start where the starter closed only once and the ends glowed and lighted the tube in one start but the ends got sputtering marks and stayed. This did not happen when used on the preheat circuit with the incandescent bulb and starter even when the filaments stopped heating the tube was still 'new' when running. It happend to both tubes how would it be explained?

When the cathodes are only partially heated, their emission is limited in current. Now as the inductive kick have the current strength exactly as the coil current before the kick, limitting the curent by the incandescent may lower the current below the limit of the partially heated filaments, so they could handle that. The full current then exceeded the limit of the partially heated filaments, so the cathode fall become way higher, so damaging.
Now the correct fix of this problem is not reducing the ignition current, but maing sure the filaments are FULLY heated before theignition attempt. The reduced ignition current may help on some tube types, but would be of no effect on other design (different maker,...), while the way of fully heating the cathodes is effective to protect cathodes of all lamp designs (meeting the standard). And when the cathodes are fully heated, the ignition current reduction have no effect anymore.

As I already asked somewhere else, wasn't the banding only a mercury deposit? In my experience this frequently form, when you heat up electrodes on lamp, which was longer time sitting somewhere (or in a colder environment), so the mercury got attached to the cathode assembly metal work. After you heat that up, the mercury evaporate and sits onto the glass wall. And it form there a gray coat, but after the glass warm up, this coat evaporate (but it could take sometimes a few hours)...

I try to look after my lamps as much as possible. The lamps in my galleries, particularly the vintage incandescent, are close to 100 years old. All are in cardboard boxes and wrapped in newspaper, stowed out of harm's way. My fluorescents are up on shelves in the shed, away from harm. Most in sleeves or at least lying on sleeved lamps. Haven't had a breakage yet, but it will happen I suppose.