Is the value of the capacitor correct for the size/type of tube?

It's the capacitor installed at the factory - decades ago. Always worked before. It's 8 microF.

But as I said, removing it has no effect.  I cut the lead for this test, so it wasn't there.

After a decent interval, this thread, if not entirely appropriate for this  forum, could be  deleted.

Thing is, I've never come across a fault like this..
I'm sure it's the choke as disconnecting the capacitor has no effect - I think it's only there to minimise power factor, and I suppose, limit interference.
The starter is also not responsible - it does its job of getting the tube to strike, and can be removed while running with no effect - it shouldn't be drawing any current while tube is conducting - surely?

The tube comes on, works fine for a short time - say, a few minutes.  Then ripples start gently and then violently, and have an effect something like waves of light down the tube. They are not like the normal tube failing  with momentary intermittent light & dark bands.

I wonder if any of you engineering types have come across the sort of fault before. What should I look for if I decide to test the choke?

I am annoyed because the existing tube was acting up, and knowing how hard it is to find obsolescent T12 8' tubes, I actually found a local supplier and bought two. This set me back £22 each plus vat, and then to fit one and find it was the fitting - well- you can imagine.

In the end,  as I did have another fitting in another room, I mounted that in place of the faulty one, so there is no urgent need to do anymore. But my curiosity remains..