If you remember that correctly, then your starter that you observed was not a usual glow switch starter.
There is the possibility in a few cases that a glow switch starter may glow while in parallel with the lamp, since the glow has positive resistance between a few milliamps and an amp or so. But I have observed a lot of glow switch starters and never seen that. That would be waste of some of the power that would otherwise go into the lamp.
I have even connected a few glow switch starters to ~150-300V limited to about a milliamp - they glow right off the bat. FS-4 usually needed more than 150V, but FS-2 and most others did not. On an ohmmeter, every one I tried came up open.
No. The starter is normally open. The starter is connected between the open ends of the lamp filament (the other ends go to the ballast an neutral.) The sequence at start up is:
1) Power applied.
2) Since the discharge in the lamp has not started the current through it is low and there is no voltage drop across the ballast.
3) The high voltage across the starter causes it to break down and have a glow discharge.
4) The heat from the glow discharge heats up the starter electrodes until the bimetallic strips cause the starter to short itout.
5) Current now flows through the ballast through the live end heater to the starter, through the closed contacts to the other filament and then to neutral - this causes the lamp filaments to heat up and since the discharge in the starter is stopped it cools down.
6) After a second or so the starter bimetallic strips cool down and the contact is broken.
7) The breaking connection causes a high voltage back-emf from the ballast which ideally will start the discharge in the lamp now the filaments are hot.
8) Since the lamp is now glowing it passes enough current through the ballast so that voltage across the lamp is only 80v or so (depending on lamp). This is not enough to break down the gas in the starter so there is no glow discharge in the starter and it cools down and remains open circuit.
9) The start sequence is now finished and will remain in this state while power is applied.
10) If the lamp did not start the voltage across it will be high enough to restart the glow discharge in the starter and the sequence repeats. This can also happen when the lamp is nearing the end of its life and the voltage across it when operating is enough to trigger the glow in the starter (the voltage across the lamp tends to increase as the lamp ages).
Take the cover off one, and watch it from the side when starting. You can actually see the contacts bend and move to each other, in that purple glow. you might have to extend it out of the fitting to get a good look in it.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.