Fluorescent Tubes

[Chris Hogg]

We have three very old fluorescent strip-light fittings that don't use 'starters', probably early 1960's era. I like them because when switched on, they light instantly just like a filament light with no delay and none of the flashing typical of modern fittings on start-up. But I seem to be having difficulty getting tubes that will work in them. The existing tubes, 4 ft, are approx 1.5 in diameter. A new tube bought recently but only about 1 in diameter, just glows pink at each end but won't strike, although it works OK in a more modern fitting. So what's so special about the fatter tubes, and can they still be obtained, or am I going to have to sacrifice my old fittings and buy ones that modern tubes work with?

[Andrew Gabriel]

Are they really instant full-on? They would be instant start ballasts, but they are virtually unheard of in the UK. If they are, it sounds like they don't have a high enough open-circuit voltage to strike T8 tubes. It's worth making sure the fitting is correctly earthed which will help, but otherwise I suspect they aren't going to work.

You can still see T12 tubes occasionally in shops. They will be banned in the EU sometime soon though. London Underground seems to still use loads of them, even in some fittings which don't look particularly old. Interestingly, London Underground also had the first fluorescent tube installation in the UK in the late 1930's.

[John Williamson]

Have you tried running your finger along the outside of the tube? If that gets it to start, then the foil dodge mentioned elsewhere may work. Otherwise, you'll need a high frequency ballast, which will let the tubes work in cold cathode mode. They are a straight drop in replacement for standard ballasts, though you often have to rewire the tube connections.

[Harry Bloomfield]

It worked in the days when the end cap holder would have an earth terminal, the modern ones don't have one.

[John Williamson]

London Underground use enough of them that they probably get a batch made every so often. Some of the Undergound light fittings still in use probably date back to the 30s, too.

They get the economy of scale from only stocking one type of tube and ballast for all uses, too.

[Mike Tomlinson]

En el artículo , Andrew Gabriel escribió:

Why are they being banned?

On a slightly tangential topic: I was on a train recently lit by tubes. One caught my attention: the light only went halfway along the tube, the other half was in darkness. Since I understand the arc has to strike from one end of the tube to the other, how is this possible?

[NT]

T8s are more energy efficient, T12s are dropping off the radar now.

Mercury migration due to dc current in the tube

To get T12s, if you cant find them new, look for batch replaced tubes, you'll get a fair %age work fine. Life expectancy will be less and disposal costs, pay nothing.

NT

[The Natural Philosopher]

Its not an arc.

[Andrew Gabriel]

Capacitive coupling through the tube wall to the metalwork behind. It may be that the connection at the other end of the tube is broken.

The other variant of this which I used to see quite often on underground trains is where one end is bright as normal, and the other end is a dim pink. This is what happens if the tube is run from DC for a long time, as it causes the mercury ions to gradually migrate to one end of the tube. Where the mercury has gone, the arc current forms in the argon base gas, which generates only a dim pink discharge, and there's no UV to excite the phosphor. I presume the control gear they use had a common failure mode resulting in DC or heavily unbalanced AC current in the tube.

[NT]

Failure of emission at one end of the tube results in partial rectification.

NT

[Andrew Gabriel]

That's true, and that's also how some modern electronic ballasts detect lamp end-of-life. However, the tubes on the underground trains did not show any signs of sputtered electodes (dark stains at the tube ends), so I was more inclined to blame a ballast fault.

[Andrew Gabriel]

I took the opportunity to read some of the T12 tube ratings in the underground stations today.

The colours were a mixture of 740 and 840. I would guess the 740's are hallophosphate tubes and 840 are triphosphor. T12's are not normally triphosphor, so those may be specially made for them, and might even meet ongoing EU requirements for tube efficiency, but a T12 triphosphor tube is going to be a serious price premium over T8 due to needing twice as much phosphor as a T8 where triphosphor is normally used.

(The '7' means a relatively low colour rendering index between 70 and

79, which is at the higher end of common halophosphate tubes, and the '8' means a colour rendering index between 80 and 89, which is normal for modern triphosphor tubes. The '40' means a colour temperature of 4000 kelvin.)

The 5' tubes were all rated 65W. If they are correct for the fittings, that makes them no older than late 1950's, as 5' tubes were 80W before then (originally designed to run from an 80W mercury vapour lamp ballast). These were very old fittings though, and it might be that they really had 80W ballasts and wrong tubes, or that the ballasts may have been changed during their lifetime. I couldn't see if the tubes were bi-pin or bayonet cap (5' 80W tubes were originally BC, but later ones were bi-pin, and through to about 1980, 5' bi-pin tubes were often dual rated 65W/80W).

There were a mixture of different makes of tube.

[NT]

Its relatively easy to get 80+ CRI with triphosphors. And of course it makes mroe sense to use tripes in T8 size. FWLIW halophosphate exist in upto the low 90s.

65w tubes on 80w ballasts would run at nearly 80w, which is near to 23% over rated, which doesnt have a lot of effect on T12 life expectancy.

I gather fl lighting wasnt rolled out across the network all at the same time, but hopefully someone else knows the history more precisely.

Its curious theyre still using T12s, perhaps a very long term contract, or maybe SRS ballasts? Last time I was on the underground they were still running quite a lot of filament lighting, so I guess financial efficiency isnt the priority.

NT

[TKM]

Andrew Gabriel

Interesting. Transit systems seem to be a repository of old fluorescent technology in many places.

Are the T12 lamps being operated on dc power? The T12 lamp was extensively tested on dc and, with certain restrictions, performed well. It was found that switching the polarity of the applied power at switch-on minimized single-end blackening, for example. The use of smaller diameter and longer tubes is still common; but I don't think the modern T8 lamp would perform well on dc since it's been optimized for high-frequency electronic ballast operation.

Terry McGowan

[Chris Hogg]

Many thanks for all your replies. I'll try and find a source of T12 tubes first, and failing that I'll think about an HF ballast.

[Mike Tomlinson]

In article , The Natural Philosopher writes

Oh really? What is it then?

[F Murtz]

If you can get a tube with a metalic stripe the full length use that or put a metal earthed reflector within an inch of the tube and it will work. As an experiment hold one hand on the case and run the other hand wrapped round the tube along the length of the tube when switched on and it will probably light.

[The Natural Philosopher]

Plasma discharge

You wouldn't call an old style TV tube an arc either.

Or a neon tube (which is essentially what a fluorescent is..

Or a radio valve. All have currents passing through gases Or a partial vacuum..

Nor will you strike an arc more than a few mm with 240VAC.

It takes about 25KV to strike an arc across a spark plug....

[Steve Firth]

Arc discharge and plasma discharge are synonyms.

[Mike Tomlinson]

In article , The Natural Philosopher writes

I know what you're saying, that the current passing through the gas in a tube isn't the same thing as drawing an electrical arc (I'd be inclined to call that a spark), but it's still an arc.

The Wikipedia entry on fluoro tubes refers to 'striking the arc' several times in the context of starting the lamp.