Hard circular fluorescent lamp revusured

I replaced the lamp with a new one.So far everything seems o. I did take some measurements.
The circular lamps are mad with two filaments at end. That enables them to start with old fashioned glow starters. In my case, the quick start method is used. The filaments act like electrodes. If you can start them by hook or crook you do not need intact filaments.
What appeared to have happened in my lamp was that one of the filaments opened up. Apparently the filament was still connected at one end and that was good enough. I just checked the continuity of the filaments. One was ok and the other not.
I hung the new lamp by its connector. It did not start spontaneously.I think that having the lamp in its fixture helps starting. It is held in place by three spring clamps. That extra capacitance to ground is probably important. For the few dollars a new lamp costs it is not worth tracking down the failure any further. I continue to have starting problems, I will post again.

  Quoted Text Here  
Fluorescent lamps are intended to be mounted near a grounded surface for starting. A fixture that is not "grounded" may have starting problems. Not surprising the loose lamp did not start. I believe you can also have problems if H and N are reversed.
"Quick start" lamps, which replaced "preheat" (with a starter), have the filament at each end heated by windings in the ballast. Open filament means no electron emission at that end for starting. When operating I believe the filaments can be heated by electrons bombardment. If I remember right, with an open filament the electrons have a longer path before hitting the filament, so they hit with higher energy, and sputter off material causing a black ring at that end of the tube.
Don't know what fixture you have, but the old Circleline (sp?) fixtures had the balast right under the electrical box and are notorious for overheating the wires in the box. You move the wires and the insulation falls off.