(Not Strictly DIY) Lighting

[Iggifer]

So at my day job (until I am competent/qualified enough to work as a spark full time) we have 6 x 70 CDMT light fittings fed from a 5 way IEC block connected to a 16a SP+N busbar.

The lights are connected to the block in pairs so occupy 3/5 ways, way 4 is taken up by a double MR16 LED 12v 2x11W downlighter.

The problem is that all 6 of these lights will go off at the same time simultaneously at random, I've witnessed it happen at least 4 or 5 times over the past couple of days.

If it was just these 6 lights on the block I would assume it was a L-N short or a loose connection in the cable run but the MR16's do not seem to go off when the CDMT's do.

I think I can rule out any sort of short as it's protected by a 6.3a FF glass fuse (or it should be 6.3aFF) in the block and then a 10 or 16a in the busbar tap off.

Our spark (brilliant guy, very knowledgeable) has been out already and meggered (IR, continuity), he replaced all the ignitors and lamps and tightened the connections etc but this hasn't solved it. He is stumped.

We're thinking the best plan is to replace the block, feed and busbar tap off but a) this is going to be very expensive (we have a crap maintenance contract) and b) if it doesn't solve it it will be a very expensive waste of time as the cable run is pretty long and very tangled.

Anybody got any suggestions?

The best I can up with is a loose connection and I'm just not noticing the MR16s go off as they fire up so fast compared to the CDMTs which take about 10-15 minutes to fire.

It seems to happen when the current draw of the building is at it's highest but it's a pretty well designed install, and the total calculated draw of these lights is only 3.58a including the MR16s and I would expect to see other peculiarities if we were exceeding our supply capabilities rather than just 6 lights failing

 

[Spark123]

What sort of control gear do they have? If it is electronic it could be as simple as a power dip and the electronic ballasts locking out.
If you power down and then power back up again do they re-start?

 

[Iggifer]

not sure about the control gear, I will check later, my spark has left me a spare ignitor somewhere so it's easy enough

Yes, I I throw the main LTG breaker on / off they will come back on on their own with no help

 

[viewer]

When you say "random" is it really random? Is there a voltage drop when something apparently unrelated to that circuit starts?

 

[ChrisRogers]

I'd stick a power logger on them for a week or two....

 

[AdrianUK]

I'd look harder at the connections if I were you. I think that you'll find a bad connection to be at the bottom of this.....

The CDMT's are discharge lamps - any break in the supply, however short, will cause them to go out & they cant re-strike until they have cooled down enough for the pressure to drop inside the tube, allowing the ignitor to restrike the lamp - this takes several mins.

The LED MR16s however, contain a small SMPS and have the ability to 'ride through' a small break so you may not see them go out. Even if the do go out, they'll come on again immediately

 

[333rocky333]

If they have ignitors then your using magnetic ballasts, check the tapping on the ballast, measure voltage at the fitting and match the tapping to the voltage, cdm lamps can sometimes be sensitive to incorrect voltages
Also check they are actually 70 watt cdm - m/h ballasts

 

[RF Lighting]

Definitely sounds like a VD issue to me. Possibly a big inductive load being dropped onto the supply if the demand is already quite high.

 

[Iggifer]

Power logger sounds like a good idea to me as does rechecking tightness of connections - however this is just going to be the block to the busbar - as the all go off immediately and simultaneously I highly doubt the fittings are at fault

As for it being VD related, I cannot think of anything that could cause it - suspect more likely it's a summation of various pieces of equipment / machinery.

 

[Iggifer]

Spoke to my spark about this at length today, and gave me a few suggestions to check out. The one I suspect that may turn out to have cured it is:

The busbar tap-off had one of the clips snapped off it (they're a good few years old so are starting to show signs of age) also the busbar is mounted upside down so the tap offs are on the underside rather than on the side.

On top of this there was a piece of 5 core 50mm YY over the lead from the busbar which was pulling the cable tight.

When I pushed the tapoff onto the busbar it made a lovely buzzing sound as it snapped home. So I put a couple of cable ties round it to hold it in nice and tight and untangled it from the YY.

Fingers crossed this will have solved it. Cost all of about 3p instead of £200+ to re-do the whole run.

It's always the simple things!