Fluorescent Behavior

A suggestion: could the connections between the vent fan and the light be crossed over? Many fans have two live terminal connections: one runs the fan on the switch and the other sustains the fan after the run connection is de-energised, using a timer inside the fan. If the light feed and the fan feeds had become crossed over at some point then that would explain it, particularly if the timer had a changeover contact and the light feed was derived from it rather than having its own separate circuit (unlikely though perhaps possible in non-UK wiring). Did someone twiddle with it about a year ago and leave it like this?

"A suggestion: could the connections between the vent fan and the light be crossed over?"

No wiring changes since installing vent fan. Both fan and light are controlled via low voltage relays. A big thing when the house was build in '53. The fluorescent strip line was replaced completely due to a bad ballast. Easier & no more cost than ballast alone. Everything works as it should at all other times than shortly after a shower. tube is warm so no condensation there. ???

Very perplexing!

Does renewing the fluorescent starter unit make any difference? And is it correctly sized?

What happens when the fluorescent is changed for a filament lamp temporarily?

Stan

Make sure your low voltage relay is rated for an inductive load like this fluorescent light; many are rated for resistive loads like incandescent bulbs only

A fluorescent fitting would not be allowed in a bathroom in the UK, unless classified for that location, but I doubt if condensation is the cause of problem.

In the absence of a physical wiring mix up, the most likely cause is a loose connection in the light fitting - either a wire, an end cap, starter, or overheating component.

Vibration from the fan might be enough to cause a faulty connection to make or break

Condensation is the correct answer. The moisture on the outside of the lamp makes a current path across the outside of the lamp and current does not flow inside the lamp so it does not light. When the condensation evaporates the lamp is able to light again. Fluorescent lamps are coated with silicon on the out side so that any condensation on the lamp beads up and helps to prevent this problem, but as you have found out it does not always work.

Mike Love

Thanks for suggestions to check using a incandescent lamp and condensation.

I had thought of condensation but then the lamp was warm/hot from being on for some time. Only a short time after being turned off it would not come on.

Will try to remember to report findings.

Mr Roboto has probably nailed it. Also, a dusty tube will present the same problem. Most of the time simply wiping the tube off will cause it to lite.

Fluorescent lights require the presence of a grounded surface parallel to the tube, along its length. This gives a slight capacitive effect which helps the arc to initially strike. If the change of the fixture was fairly recent, it is possible that you also changed from a core & coil type of ballast to an electronic one. With this change, the heat loss within the ballast compartment and at the ends of the tube would be a lot less than it was previously.

I would not look for problems related to condensation along the tube itself, because it is sufficiently warmer than the surrounding surfaces. This leaves two possibilities -- condensation at the sockets on either end (cooler fixture than before) or elsewhere near the fixture, or a lack of a ground. The house was built in 1953, so I believe that the problem is a lack of a ground to the fixture. Often, this can be quickly checked by touching the metal fixture housing when it is refusing to light, or temporarily testing with a wire from the housing to a nearby ground (a water pipe in a house of that age would probably be sufficient unless plastic lines were installed in the last 15-20 years). The low-voltage relay systems used then, and still available today, (usually Touchplate--spst, or GE--spdt) are very unlikely to be at fault with this type of load.

Fixes--1) ground the fixture. 2) Use an enclosed fixture listed for damp or wet locations.

Another possibility -- double check the ballast and the tube because many new electronic ballasts are supposed to be used with F32T5 bulbs instead of the older F40T8 bulbs. These two tubes are the same length and have the same pin spacings but one is 5/8" diameter and the other 1" diameter. With the dry surroundings, a wrong tube could light, but fail to do so when the condensation is nearby.

A modest correction- T8 and T12 lamps have identical pin spacing, the T12 being the more mature.

Thanks, Guest. I should have said: F32T8 and F40T12. Feel welcome to become a member. --jmm