I fitted a five foot fluorescent unit from Wickes around a year ago. It's had very little use - I'd guess less than 24 hrs on in total. It comes on OK but after a couple of minutes, it goes off and then enters a cycle of humming whilst struggling to relight/few seconds on/off again.
I'm assuming because it comes on OK from cold, the starter's OK. Can you test them? Is it just a duff tube? Or is it likely the electronics in the unit have gone phutt?
I haven't another unit, tube or starter so I'd like to know what's most likely before ending up with a load of spares to store.
Electronics ? If it has a starter, there are no electronics!
Most likely it it the starter which has failed, but starters cost 50p and tubes around £1.50 - Hardly worth not getting one of each to be certain. If the tube has suffered lots of flashing due to the starter being faulty, it would be worth replacing it anyway.
You could try an electronic starter (which in aus is 5 times more costly but seem to work better)although it could be the tube or ballast. There are other things to consider as well, the longer tubes sometimes like earthed metal in close proximity along their length,
After it has initially come on OK, remove the starter (without switching off). Does the light now stay on? If so I'd suspect the starter first. The suggestion to fit an electronic starter, like
Thank you all for the replies. Took the starter out and it did indeed stay lit. So now I have another question. How does the connected starter cause the problems described? I thought once it had done its job in starting the tube, it just sat there doing nowt.
Also, yes, it is in a cold place. What influence does that have?
About a year ago, I bought two presumably very similar (could even have been identical) lights from B&Q, for the garage. After a couple of days, first one wouldn't start, followed soon by the other. Replacing the starters (a pack of 2 which I happened to have - from a Pound Shop, I recall) fixed them.
It should, but if the tube voltage gets too high the glow starter can re-strike, effectively shorting the tube and re-starting the switch-start cycle. Check that the starter fitted is compatible with the tube. You could also try re-fitting the starter after the tube has thoroughly warmed-up - after half an hour, say. What happens then?
How cold? Starting in cold conditions is always more difficult, especially with the longer tubes (8 ft. in particular). The voltage drop across the tube is greater, with the consequences you have discovered. The light output is less too, as the phosphors have an optimum working temperature which is optimised for normal room temp conditions.
I suspect that fitting an electronic starter or converting to an electronic ballast will overcome your problems.
I did that in the utility room. The trouble is that it took so long to strike you were out of the room again before the light came on. I replaced the fitting with a daylight CFL from CPC and have had no more complaints from SWMBO.
The short answer is that the starter will strike at 240V (but the tube won't). After the tube has struck (due to the back-emf of the choke), it has a running voltage of about 100V. So, once the starter has struck and caused the tube to strike, it (the starter) goes out of play. As the starter deteriorates, it gets to the point where it will strike at 100V, so it keeps re-striking ad infinitum.
The long answer is that the standard switch-start fluorescent fitting is a marvel of cleverness.
The tube has four pins. Between two pins at each end there is a filament.
Mains live goes to an inductor (called the choke) then to one end of the tube. The other end of the tube goes to mains neutral. The starter connects the other two pins at the end of the tube.
formatting link
you apply power to the fitting, the contacts in the starter are open. There is 240V across the tube, which is not enough voltage to strike the tube.
The starter contains a fixed contact, a bimetallic strip contact and neon gas.
There is 240V across the contacts in the starter (less the volt drop in the filaments, natch), so the neon strikes and the starter glows. The hot neon gas warms up the bimetallic strip and the contacts in the starter close. This has the effect of connecting the choke between live and neutral via the filaments in the tube. Meanwhile, the starter is shorted, so the gas stops conducting and cools down. Consequently, the bimetallic strip cools down and the contacts in the starter open. At this point, the choke, being an inductor, tries to maintain the current flowing in it by raising its potential difference. Consequently, the voltage across the tube rises sharply until there is enough voltage to strike the tube. The tube strikes and the resistance of the running tube drops about 100V. Now 100V is not enough to re-strike the neon in the starter, so the starter is out of play and the circuit comprises the tube in series with the choke.
If the fitting is power-factor corrected, there will be a capacitor between the live an neutral.
The fitting may also have a fuse fitted (usually a 1" plugtop fuse).
There are (therefore) five things that can go wrong with a fitting.
The fuse pops.
The capacitor fails (usually causing the fuse to pop).
The inductor burns out (may or may not cause the fuse to pop).
The tube dies (usually black at the ends, very dim light and won't stay on).
The starter dies (usually the starter will begin to strike at 100V, so it keeps cycling when it shouldn't). Usually, if you let a dodgy starter continuously re-strike the tube, the tube will quickly die as well.
No, the bimetallic contacts are initially closed. It's heating from the glow discharge that opens them. The starter is essentially a heater in series with an NC thermal switch.
The starter discharge strikes immediately, being part of the series circuit supply L -> ballast inductor -> tube preheat filament 1 -> starter -> preheat filament 2 -> supply N.
No, they open, giving the necessary 'inductive kick' to strike the tube.
Indeed, although the running voltage of the tube depends on its length and type.
The starter is out of play at this time because its contacts are open. As it cools the contacts will re-close. If the tube has successfully struck the tube running voltage will be less than the voltage required to re-strike the starter and the latter remains out of play. If the tube has not struck the starter will see full mains voltage and it will re-strike, repeating the whole cycle.
So initially the contacts are open and a low glow discharge current flows in the series circuit. Although this passes through the tube heaters it is too small to heat them significantly. The contacts are heated though and close allowing the main pre-heat current to flow - this will be the mains voltage less a few volts for the tube heaters divided by the impedance of the choke. The starter contacts cool and open, giving the tube a kick, etc., etc. etc.
I've understood this wrongly for donkeys' years. You live and learn. 'umble pie for dinner tonight...
They will get some heat from the glow current. The bimetallic contact warms up pretty quick, so the glow current will not be insignificant.
The contacts are
Yes, but the point here is not so much that the "main pre-heat current" flows, but that a 'significant' current flows in the choke, so that when the contacts open, the p.d. across the choke rises to a voltage high enough to strike the tube.
for the tube heaters
Yes.
The starter contacts cool and
Yes, but it's the back e.m.f. from the choke which generates the "kick".
etc., etc. etc.
I have to say, I never really understood the point of the preheat in the tube nor how the tube actually makes light (which is actually quite subtle in itself), which is why I completely glossed over that part of the explanation...
It's both, surely. You need enough current (and time) to heat the tube filaments sufficiently to get plenty of thermionic emission to lower the striking voltage of the lamp, AND enough for a good inductive kick.
HomeOwnersHub 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.