Greetings to you all.

As some of you may know, I wish to acquire 6 watt preheat fluorescent light fixture. I've been planning, unless I run across a whole fixture locally, to acquire the required individual parts to build such a fixture myself.

For the sake of creativity (not to mention saving a little bit of cash), I've been thinking recently of making said fixture incandescent-ballasted.

I know such a system is definitely possible and works well in most situations, but I was just wondering if there are any other downsides, particularly regarding lamp life, compared to a standard ballast? I once read that an incandescent ballast accelerates degradation of the cathodes compared to a magnetic ballast due to a high current crest factor, which I have no knowledge about. Can anyone confirm this for me? I plan to use whatever I acquire/build to run my vintage GE F6T5/CW, so I won't do it if it will shorten lamp life.

Also, while I'm here I figure I'll ask another question I've been wondering about - has anyone ever heard of the existence of an electronic preheat ballast? Basically, a high-frequency ballast that uses a standard glow starter? I'm as big of a fan of preheat as the next guy, but I'm slightly sensitive to the mains flicker of a magnetic ballast, so if such a thing were to exist, that would be awesome.

Thanks.

I have a Incandescent-fluorescent mini troffer. That had a bad fluorescent choke
 so I just used A15 lamps to ballast my lamp. one thing it hell on the starters
as it waiting for the tube to warm up to temprature, as it warming up it can sputter the electrode
 but runing cheap lamps that dont mean anything. it can bre fun to watch them start up.

 As for Electronic Preheat, Sound like a Program start ballast. It heat the electrode at predetimin time
before the lamp stikes the arcs.

funkybulb: Preheat is different from programmed start in two thing:
1. Preheat: After the ignition, the lamp operates like instant start (Slimline)(No current flow throguh the electrodes to heat them and they are heated by the discharge itself).
2. Programmed start: After the ignition, the lamp opeartes like rapidstart (Electrodes continues to flow current through them, to heat them, rather then heating by the discharge itself).
So even a magnetic ballast + an electronic starter would described as preheat, even if it will have the same long life as program start.

@dor123:
The heating power of electrodes after ignition on all HF electronic ballasts is negligible, so it could be well treated as no power at all.
The difference between the electronic vs preheat is rather in the fact, than the programmed start continue to apply heating power during an ignition attempt (after successful ignition and arc buildup the heating power drop to the low level), while the "preheat" apply the heating power alternatively with the high voltage, so only one of them at a time.

Indeed, program start is an option, but I was thinking of electronic ballast that makes use of a standard glow starter. I think I've heard of an electronic adapter for 2-pin PL lamps before, so it sounds like the concept is possible.

What about the incandescent ballast concept? Any problems with that? I'm wondering if anyone can confirm or debunk the current crest thing I once read about.

Indeed, program start is an option, but I was thinking of electronic ballast that makes use of a standard glow starter. I think I've heard of an electronic adapter for 2-pin PL lamps before, so it sounds like the concept is possible.

This ballast does not use the starter, but rather the ballast only tolerate the starter presence. During normal operation the starter doesn't do anything useful (only somehow obstruct the ballast operation), only as a side effect it help the ballast to survive the lamp EOL by shorting it out.
In this configuration the ballast make use of the capacitor connected parallel to the starter (it act as the resonant capacitor), so it does not work with lamps without this cap.

What about the incandescent ballast concept? Any problems with that? I'm wondering if anyone can confirm or debunk the current crest thing I once read about.

Manual (push-button) start: On 230V OK with total lamp voltage up to 160V, so I would expect on 120V it would be OK with two
F8T5 in series, when the ballast efficiency would be a bit above 50% (something comparable with really low wattage ballasts, like F4T5 or so).

But glowbottle starters tend to be unreliable, so need the lamp voltage to be way below the mains voltage, so the ballast efficiency would be very low (single F8T5 on 120V would mean only 25% ballast efficiency).

Electronic starters do not work (they need the inductive component) and very frequently they get killed by the incandescent inrush current (they are very sensitive on that, they really count on the series inductance in the circuit)

Assuming starting wouldn't be a problem, would the lamp last as long on an incandescent ballast as it would on a regular ballast as long as it got the correct current?

Join the question, also what about life of 2 in series vs 1 alone, on incandescent ballast at right current ?

There is a bit higher crest factor, so a bit higher peak current load of the cathodes. On the other hand the filament is resistive for the DC as well, so in case of asymmetry causes any rectification, the consequent DC current component would be still low, compare to the magnetic ballast.
So the life may be a bit shorter, but I don't think it would be too significant. Capacitor based ballasts are way worse...
Two lamps would become a bit worse on the crest factor, but become twice as efficient system...