Why isn't my lighting working?

[iann]

I have lots of different ones, Tridonic, 3AAA, ELT. I've never had one that didn't work.

You shouldn't really need any metal next to the tube with an electronic ballast, so I can't say how much. Just a suggestion of something new to try. There's all sorts of fun tricks people used to try, rubbing tubes, or heating them, but it just shouldn't be necessary to do these things any more. You've probably tried with all three tubes wired in? Again, shouldn't matter with most ballasts.

Here's the one in the first picture (four tubes, connectors to both pins at both ends):

lights3.jpg (62.09 KiB) Viewed 1249 times

And here's a different one (talk about keeping the leads short!):

wiring-281010.jpg (36.38 KiB) Viewed 1249 times

Neither of these ballasts are exactly the same type as yours.

[Sharpy]

Just a little interjection here. I have a long background in electronics, and the length of the wire will not make a difference. The measurable resistance between a piece of wire, of that small gauge, 2cm long or 2 meters long is almost unmeasurable.

If it had anything to do with frequencies then the manufacturer would need to state that it has to be exactly 7.345cm long, or whatever.

[iann]

Just a little interjection here. I have a long background in electronics, and the length of the wire will not make a difference. The measurable resistance between a piece of wire, of that small gauge, 2cm long or 2 meters long is almost unmeasurable.

If it had anything to do with frequencies then the manufacturer would need to state that it has to be exactly 7.345cm long, or whatever.

Nothing to do with resistance. It is the capacitance of the wire that is important, due to the high frequencies used in modern ballasts. All electronic ballasts will state a maximum cable run, but in practice, the cables should be kept as short as possible without doing anything stupid. A typical spec for a 4 foot tube will state a maximum wiring run of around six feet, but it varies.

I've never seen a tube not light because of this, although I've never tried to push the limits. Tube lifetime or performance could still be compromised by long wiring even if it lights. In this case there is a problem and this is one possible, if unlikely, reason. Specifications for instant-start ballasts also sometimes state a maximum distance between the wire split (because of the single wire running to multiple pins) and the tube. In one spec I have to hand, the distance should be no more than four inches. Again, might not be relevant, but when all else fails it is time to eliminate every possible source of error.

[Sutremaine]

I tried running some steel wire into the ground connector and then wrapping it around the tube.

Thanks for the suggestions, but I think I'll head on over to the BCSS forums and see if anyone wants to pick up the ballast and everything at the next meeting and try and get it working.

[iann]

Probably not relevant but worth mentioning: steel is an awful conductor of electricity, about a tenth as good as copper. Stainless steel is about 50 times worse than copper. I bet Sharpy could measure the resistance of your length of steel wire

[Sutremaine]

Probably not relevant, but since nothing at all is working it's a little hard to tell. If I don't get any breakthrough on this, I'll just take everything to the next meeting and see if anyone has any ideas.

[Sharpy]

Probably not relevant but worth mentioning: steel is an awful conductor of electricity, about a tenth as good as copper. Stainless steel is about 50 times worse than copper. I bet Sharpy could measure the resistance of your length of steel wire

Dang right I could

Thanks for the info on the new florescent lights. Did not know that. Learn something every day

[iann]

Dang right I could

Thanks for the info on the new florescent lights. Did not know that. Learn something every day

Those touchy new-fangled electronic gadgets dagnarn-it! A couple of pounds of copper wrapped round an anvil wouldn't have any trouble, if you don't mind the humming and flickering