Electronic Domestic Instant Start Ballast Keeps Blowing

[RS2OOO]

Hello.

I'm no electrician (well, previously an auto-electrician but no expert on domestic electrics) and do have an electrician lined up to look into this problem next Week but would like to present the problem here for the sake of interest and for gaining some knowledge as to possible causes.

My Kitchen has 2 light fittings operating from the same supply through 2 switches.

Up until 1 year ago I had 3x 50w GU10's in each fitting.

I replaced them with a pair of (high end) fluorescent lamp fittings, 2 x PL-L 4 pin 36w tubes in each fitting.

Each fitting has an instant start electronic ballast powering the 2 tubes.

After 1 year (to the exact day!) one of the fittings flickered a few times and the 2 lamps went out and would not switch back on. Took the cover off, felt the ballast which was hot but not beyond the 70c operating temp.

The unit was exchanged under warranty.

Fitted the replacement unit and after around 30 mins of use the lamps went out (new bulbs), and would not switch back on. Replaced the bulbs (correct spec), but still would'nt work.

Now Realising the ballasts were failing I replaced the ballast and re-fitted. But again, 30 mins after use the lamps went out, this time both brand new Philips tubes were completely blackened at the ends.

So, ahead of calling in an electrician and to give him/her something to work from, I fitted new tubes and another new ballast (all correct spec etc) and put the fitting back up. Unfortunately the missus forgot and switched the light on, but this time the ballast had blown within 5 minutes and again both tubes were blackened at the ends.

The unit in question is the 2nd and final fitting in the loop, the first fitting from the same supply and using the same lamps/ballast etc works fine and continues to do so.

Polarity is correct, the ballast is earthed correctly, I have no other issues with power surges etc that I know of and all visible and accessible wiring connections are good. I've not checked actual voltage at the supply as this was something I'll let the electrician do. Haven't had any issues with blown fuses etc.

If I replace the suspect fitting with a normal 100w tungsten bulb it works and operates fine.

So my question is, what could cause an electronic ballast to blow so often and so quickly, particularly when an identical fitting just 6' away works and operates fine?

Thank you in advance for your time reading and answering this. If the problem is finally sorted I will repost to confirm what the eventual cause was.

 

[Moses]

Could be a wiring fault in the fitting

 

[RS2OOO]

Hmmm.... I never thought of that! Seems an obvious thing to check now you mention it, thanks.

When there's daylight tomorrow I'll compare the wiring to the other fitting and report back.

 

[Top Cat]

Could be a wiring fault in the fitting

But why did it last for a year?

 

[RS2OOO]

The fitting that lasted for a year was replaced under warranty.

The replacement fitting failed within 30 mins. New ballast and bulbs, then failed again within 30 mins. New ballast and bulbs again, and failed in around 5 mins. No big event or funny noises, lights just went out and wouldn't come back on.

Whilst the first fitting failed after 1 year's use, the Wiring fault could be with this replacement fitting.

In fact I may have enough light here to check the wiring now....

 

[Farmelectrics]

Think one needs an electrician

 

[RS2OOO]

I've just compared the 2 fittings, and they are definitely internally wired up slightly differently but at this stage I don't think it is the cause of the problem.

Outlets 1 & 2 of the ballast both go to the same "side" of the 4 pin tombstone but on one of the fittings they are swapped around, i.e, on 1 fitting wire 1 goes to pin 1 and wire 2 to pin 2, but on the other fitting wire 1 to pin 2 and wire 2 to pin 1.

Wiring diagram is on the image, from what I can make out, it doesn't matter which way around wires 1 and 2 go as long as they are both connected to the same "side" of the 4 pin tombstone, but please correct me if I'm wrong.

 

[Moses]

I've just compared the 2 fittings, and they are definitely internally wired up slightly differently but at this stage I don't think it is the cause of the problem.
View attachment 12593
Outlets 1 & 2 of the ballast both go to the same "side" of the 4 pin tombstone but on one of the fittings they are swapped around, i.e, on 1 fitting wire 1 goes to pin 1 and wire 2 to pin 2, but on the other fitting wire 1 to pin 2 and wire 2 to pin 1.

Wiring diagram is on the image, from what I can make out, it doesn't matter which way around wires 1 and 2 go as long as they are both connected to the same "side" of the 4 pin tombstone, but please correct me if I'm wrong.

You to ensure that the wires are connected exactly as on the unit that is working.

 

[RS2OOO]

Got an electrician coming round later today (if he can make it), so I'll report back on his findings.

At the end of the day, this isn't my trade and if people who've been doing it for years can get it wrong (having read the cock-up's sticky thread), then it'll put my mind at rest to have a pro in. Furthermore, he's got a few spare ballasts to play around with so no waiting for parts either.

House is 52 years old, on the original wiring and has never seen a sparky (other than a new meter when the original one stopped)... be interesting to see what he's got to say.

I'll update as and when it's sorted.

 

[RS2OOO]

The electrician came round tonight...not long left.

First thing he did was install new ballast in the fitting. It's a higher quality Philips warm start ballast as opposed to the Chinese instant start ballast previously fitted so he had to rewire a couple of the terminals.

Plugged it into the wall and left it for an hour or so whilst fiddling with other stuff, and it worked fine.

So, I've asked what's causing these ballasts to keep blowing after 30 mins usage?

He checked the connections at the open rose, L+N = 250v and L+E = 250v.

He said that could be the problem, electronic ballasts can have fairly low voltage tolerance, and the excess voltage could be the problem, or, his meter might be over reading.

He checked a wall socket to compare, also 250v.

I asked him to run his meter over a car battery I had sitting around, it read 13v. (Should be 12.6v).

So, I pulled out my own multimeter (it's donkeys old from when I was an auto-electrician). Ran that across the battery and got 12.59v, probably accurate.

So he ran my multimeter at the rose and got 246v and again 246v at the wall socket.

Now he's thinking 6v excess is probably not enough to cause these ballasts to keep blowing, especially as I've got an identical light on the same kitchen circuit that's working fine.

So he suggested I pull the other light down (the 1st one in the ring of the 2 lights) and check all connections within the loop are tight and do the same at the wall switch. He advised that a loose connection somewhere could potentially be causing an extra current draw causing these ballasts to fail.

If all connections are sound, then he has no idea why the ballasts keep blowing, only to conclude they are of poor quality. The light itself is a fairly high end unit from Endon lighting, retailing well over £200 in shops. Endon lighting themselves advised me they'd never previously had a warranty claim on this light model relating to failed ballasts.

So, still no definitive answer.

I'll check all the connections and re-fit the light tomorrow and report back on any findings.

 

[Marvo]

Hello.

.....felt the ballast which was hot but not beyond the 70c operating temp.......


.....Fitted the replacement unit and after around 30 mins of use the lamps went out (new bulbs)......


......I replaced the ballast and re-fitted. But again, 30 mins after use the lamps went out, this time both brand new Philips tubes were completely blackened at the ends......

........Unfortunately the missus forgot and switched the light on, but this time the ballast had blown within 5 minutes and again both tubes were blackened at the ends.......

You're ploughing your way through ballasts and lamps like they're going out of fashion. Sounds like a ballast/lamp mismatch to me. To blacken tubes or blow ballasts in 15 minutes of use it doesn't sound like a marginal overvoltage supply problem. Could also be the way the system is wired, check the wiring again to the diagram on the ballast and check the wattage ratings of everything matches.

 

[RS2OOO]

Definitely not ballast/lamp mismatch - 100% certain.

The wiring within this fitting differed very slightly to the wiring on the working fitting that's still in use. Personally, I don't think this was the cause - check out the photo in previous post and imagine swapping round wires entering terminal 1 and terminal 2. That was the only wiring difference between the 2 fittings.

However, the new ballast supplied by the electrician had a different wiring diagram, so he completely ripped everything out of the fitting and re-wired from scratch (although effectively only a couple of wires ended up differently).

In the morning I'll check all other connections within the supply circuit...half hoping to find a loose one somewhere just so I have an answer.

Failing that all I can do is stick it back on the ceiling and see what happens. Maybe I'll check the current with it switched on to see what it's drawing, but I don't have the knowledge to really know what it should be drawing.... If it's over 5 amps I guess my main fuse should have already blown, but it never has. I'm guessing "In (A) = 0.319" printed on the ballast refers to the current draw.

Fact is, plugged into the wall the fitting was fine for an hour. If when back on the ceiling it blows again in under that time, then the lamp itself can be completely excluded from diagnostics.

There are odd stories around the web relating to ballasts blowing through high frequency interference. One story I read was about half the street losing their Cable TV and the Cable guy knocked on this blokes door and said the cause was coming from his house....that turned out to be an electronic ballast.

 

[RS2OOO]

May have finally found the cause of all these blown ballasts...not 100% sure, but appears likely.

As advised by electrician I removed all switch plates and other downstairs ceiling roses in search of a loose connection.

No loose connections found, but in the first ballasted light of the circuit I noticed 2 blacks looped together with brown sleeve going to the live terminal of that fitting. Only 1 black neutral going into the neutral terminal.

Got me thinking, where the hell is the neutral coming from for the final light in the circuit (the one that keeps blowing ballast).

Continuity test showed polarity in the first fitting was wrong and the 2 "lives" were in fact neutrals, despite the brown sleeving.

The "neutral" was the switched live.

I corrected the polarity and have had both lights on for an hour or so and all seems fine thus far (although early days).

Therefore my amateur conclusion is that for years using GU10's and Bayonet 100w bulbs the polarity wasn't a major issue, But with incorrect polarity, the first ballasted light in the circuit worked fine, but was sending an excess voltage to the 2nd and final light causing it's ballasts to blow. (No idea why the orignal ballast lasted a year though).

Does that sound feasible?

Previous owner of the house was something of a cowboy and I had to re-do a lot of his DIY, including replacing the leaking brand new garage roof that he was in the process of fitting when I bought the property! My guess is he'd sleeved the neutrals as live in that rose, and of course, when changing fittings I'd always trusted them as being correct.

 

[Marvo]

Not sure I fully understand the wiring issue you found but I don't think reversed L/N polarity would cause premature ballast failure. That said, with electronics anything is possible.

 

[RS2OOO]

Yeah... I do tend to turn a sentence into an essay which doesn't help.

Basically, polarity at the prior fitting in the circuit was wrong.

5 hours passed now, and still working fine, temperatures OK.

With cars, get the polarity wrong on something that has windings, i.e Coil, Alternator, and things go bang.

Ballasts I assume have windings, therefore for now, I'm happy to conclude the problem was solved by correcting polarity at that other fitting.