@funkybulb
That is a very good idea, common 18-20W chokes provide 240 ohms at 118V, while (according to my calculations) the tube needs ~125 ohms. Two 18-20W chokes in series gives 120 ohms which is close enough. Plus preheat is a lot cooler than RS in my opinion.

a 20W *Reference ballast* is 240 ohms, but many production ballasts are not, especially not your common 14W-20W choke some of which dont even drive a F15T8 at full power, just keep that in mind when trying to work things out like this

@funkybulb
That is a very good idea, common 18-20W chokes provide 240 ohms at 118V, while (according to my calculations) the tube needs ~125 ohms. Two 18-20W chokes in series gives 120 ohms which is close enough. Plus preheat is a lot cooler than RS in my opinion.

 If u  have two  15  to  20 watt  preheat  chokes  and  combine the  two  in  perallel.  It will run a  F30T12
High output quite easly just unsing  two  preheat chokes
 And FS 2 starter.   

I like how he blows on the electrical code! 🤣, bet that infuriates the time served old boy electricians 😫

I was looking at Lamptech and some other resources at gas fill types for fluorescent tubes. All I can consistently find is that normal-output T12 tubes have approximately 3 torr of argon inside. Well that is all fine and dandy, but not all tubes are T12 NO. Are the gas fills for HO and VHO the same as NO? How about krypton tubes, are they pure krypton or a mix (and at what pressures)? I hear there is xenon and neon in some other tubes, what pressures are those at?

Does anybody have a chart or resource (or just know off the top of their head) what fills are in different diameter and different output level fluorescent tubes? I would like to add this to my public lamp spec sheet.

@Ash
Thanks so much! I am certainly not in a rush.

I have a few Eltam SR40 on hand and could make some measurements if you would like, though you will have to wait until i have a window of free time

@Medved

So you have a transformer with two identical windings (X-Y and Y-Z) connected in series so that their total measured inductance is effectively zero across X-Z. When you apply current across X-Z, the center node Y will gain voltage relative to the points X and Z (assuming X and Z are at equal voltage)? That sort of makes sense, but anything after that I am lost.

Thanks for sharing that @LightBulbFun!

So a series resonant circuit can have higher voltages across the inductor and the capacitor, but when because they are phased differently they add up to the voltage put across them. That makes plenty of sense. But there isn't just a capacitor across the tube or just an inductor, there are both in series, so shouldn't it add up to the supply voltage and no higher? How does it make higher voltage across the tube?

I know it works, and I know higher voltages are produced, I just don't get how. No resource that I have read has explained that.

if I am reading the book right it mentions that effectively the voltage of the capacitor gets added onto the voltage of the secondary winding

When the lamp has no discharge, the capacitor current flows via one winding one direction and via the other the opposite direction. So their magnetic fields cancel out, except a small magnetic leakage between the windings.
So at the end it is only the small inductance, related to that leakage, which can boost the voltage at least a bit. The thing operates way below the resonance of the capacitor and the leakage inductance (which is the only inductance effectively in series with the cap), so the voltage boost barely makes for the resistive drop on the winding wire resistance. So it is not wrong to say the capacitor has just the mains voltage on it. And that includes also the state when the lamp has ignited, the coupled windings are enforcing that.

But for T12 tubes no voltage boost is actually needed, these ignite on the bare 220V (and above) mains, so the only thing needed is to heat up the cathodes. And the capacitor current does just that, until the lamp ignites.
Interesting thing happens after the lamp ignites: The capacitor has still the mains voltage on it (plus minus some small vlotage boost on leakage inductance and resistive drop), similar voltage being on the mains side. That means to quite a big extend the total lamp current splits to these two components, so big part being pushed from both ends of the filament to the arc itself, so quite reducing the resistive heating the lamp filaments experience. It does not cancel out completely, the 90deg phase shift and the different current in each branch won't allow that, but still the current on each of the filament end is less than the total arc current.
Compare to the preheat or even the very common current mode preheat HF output (mean electronic ballasts) circuit, where all the current flows from just one filament side.

there is certainly some above-mains-voltage produced by an SRS ballast on a couple 4ft 40W examples I have tested their OCV is about 280V on a Parmar 5ft-8ft 65W-85W one I tested it was about 315V

When the lamp has no discharge, the capacitor current flows via one winding one direction and via the other the opposite direction. So their magnetic fields cancel out, except a small magnetic leakage between the windings.
So at the end it is only the small inductance, related to that leakage, which can boost the voltage at least a bit. The thing operates way below the resonance of the capacitor and the leakage inductance (which is the only inductance effectively in series with the cap), so the voltage boost barely makes for the resistive drop on the winding wire resistance. So it is not wrong to say the capacitor has just the mains voltage on it. And that includes also the state when the lamp has ignited, the coupled windings are enforcing that.

But for T12 tubes no voltage boost is actually needed, these ignite on the bare 220V (and above) mains, so the only thing needed is to heat up the cathodes. And the capacitor current does just that, until the lamp ignites.
Interesting thing happens after the lamp ignites: The capacitor has still the mains voltage on it (plus minus some small vlotage boost on leakage inductance and resistive drop), similar voltage being on the mains side. That means to quite a big extend the total lamp current splits to these two components, so big part being pushed from both ends of the filament to the arc itself, so quite reducing the resistive heating the lamp filaments experience. It does not cancel out completely, the 90deg phase shift and the different current in each branch won't allow that, but still the current on each of the filament end is less than the total arc current.
Compare to the preheat or even the very common current mode preheat HF output (mean electronic ballasts) circuit, where all the current flows from just one filament side.