I am trying to find out the starting characteristics of Metal Halide Lamps. Does anyone know where I can find manufacturers technical information? I have found a manufacturer here in Australia, Sylvania lighting they have technical information but their links are messed up and the files cannot be opened. I am trying to find typical run, start, and inrush currents for lamps and the start up times.

Thanks
Gray

Hi there Gray,
Try these links here:
Venture Lighting Technical Center
GE Lighting
Philips Lighting
Thorn Lighting

Not sure that you'll find what you are after there, but it's a start.

Thanks for the information Mike,

The GE Link provided all the data I was looking for, in regard to lamp starting currents run-up time etc.
http://www.gelighting.com/eu/litlib/techbrochures.html

That is the link on their site if anyone is interested.

I am trying to find out why lighting contactors, a contractor has installed are failing.

The contractor is using contactors with a resistive rating of 40 Amp and and AC 3 rating of 25 Amps.

These are protected by a 20 Amp c/b

The contractor told me that the load on the different lighting circuits varies between about 13 to 17 Amps.

Probably six of these contactors have now had a failure, The Neutral link terminals have also failed due to heat.

The contractor also said the metal halide fittings are fitted with PF correction.

I guess it could be that the terminal strips with integral linking bar are under rated and since they are situated close to the contactor this has taken the coil out.

any ideas?

Is there any site that recommends the number of fittings on a contactor?

I have read that pf corrected fittings can have an inrush current of up to 16 times the full load current, which is similar to incandescent lamps.

Gray,
I'd like to help.
What bulbs/Control gear are you after the figures on?.
I use quite a bit of MH gear with the local Power Board here in Ashburton.
We are changing from the old Sodium lamps to MH, in a retro-fit situation.
There are bulbs that can do that.

OK Gray,
How many sets of control gear are on each contactor?.
I'm not that conversant with the AC ratings, but AC 3 sounds a little low for an Inductive load like that.
I'm told that the AC 22 or 33 is the one to use with Inductive loads.
And another thing with respect to the resistive rating of Contactors, at laest halve the current rating for Inductive loads, where a contactor could carry 10A resistive, it only as a rule will carry 4A Inductive.
Ask the guy to get a heavier contactor, say 50A or run 2 in parallel.

Dapo I think the AC3 rating is for motor loads. Contactor manufacturers usually publish their ratings in their catalogues with helpful tables on light fittings per contact and switching frequencies etc. You should look at the number of operations of the faulty contactors before buying replacements.

Yes, AC3 ratings are for squirrel cage induction motors, AC1 are resistive loads. Power factor correction doesn't necessarily reduce inrush current.

These lamps when struck from cold use mercury vapour to warm the halide, when warm the potential on the electrodes pushes away the halide plasma so they need a very high strike voltage to get them started over again. But even that would be only wery short current spikes.

Squirrel cage motors present a heavy inductive load only during startup for a few seconds at most, whereas these lamps could be highly inductive for literally minutes as they warm up. I'm surprised they don't have guidance on what contactors to use in the installation paperwork?

Do they have a simple ballast or more sophisticated control electronics?

Gideon.

It is most likely the power factor correction that is giving your contactors such a hard time.
Depending on whereabouts in the mains cycle the contactor closes, PFCs can and do generate savage inrush transients that chew contactor contacts.
Contactor manufacturers are aware of this, and do publish tables giving the derating factors for different combinations of contactor and lights.
I advise you to seek the wisdom of your contactor manufacturer.

Thanks for all your input,

I think they were 250W metal halides with standard ballasts (non electronic) and fitted with PF correction capacitors.

The AC3 rating is indeed a motor current rating. Someone said there is a rating for lighting contactors and it is designated as AC21 though I have not seen this rating personally.

I am not sure how many lamps are on each circuit only that the total load per circuit is between 13 and 17 Amps
You are right Mike if I were the contractor I would have looked at a larger size contactor before now.
I still find it disturbing that the neutrals on the terminal strips, are getting hot enough to melt the plastic of the strips. I was told that the rating of these strips was matched to the contactor ratings.

Kiwi if you have any links to manufacturers which have data on numbers of light fittings for contactor sizes it would be appreciated.

There shouldn't have been too many operations on the contactors as this all part of new work at an Abattoir boning room which was commissioned only three months ago December last year. I would think there would be only two operations per day, one at the start of the shift the other at the end of the day. This would be far less than the thousands of operations a correctly installed contactor should be capable of.

gideonr, looking at the manufacturers tables, most metal halides seem to draw about 120% of their full load currents between 1 and four minutes.

Paulusgnome, I have read in one cataloger that the contactor must be able to withstand the inrush current for power factor corrected fittings and they that the inrush for these fittings is almost the same as incandescent lighting.

Thanks your help guys.

Dapo,
Try Square D Webb site, their UK loadcentre KQ + Distribution boards catalogue contains all the info you may need.
Regards Alan

Yup, Dapo AC21 is a fluorescent load rating. G.E. has very comprehensive tables of loading and switching specs for their contactors in their contactor catalogue. In my opinion you should just get contactors that are well over-rated for the task ( provided they fit in the board). If your contactors are for a boning room then there could be a considerable "down-time" cost to your customer. An extra couple of bucks for over-size contactors may be economically viable.

Gray,
Just a small Health and Safety question if I may?.
Are the light fittings spread over 3 phases?.
Reason I ask that is because before I started my time as an Electrical Apprentice, I worked in a Meatworks as a Slaughterman and if all of the lights went out at once, the last thing you would want to be doing is using a razor-sharp knife.

Yes Mike, they are across three phases, and they are in mixed rows to eliminate the strobing effect I am told.

Did the contractor actually _measure_ the load on the various phases, or was this calculated?

What is the THD of the ballasts on these fixtures?

Given your description of the neutral bar heating, I'd urge checking for harmonic neutral currents. From your description, these ballasts are fitted with power factor correction capacitors, which won't do much for harmonic content. Any triplen harmonic content will show up on the neutral; and it is possible that the neutral current exceeds the phase current.

If this is the problem, then upsizing the contactors may simply mean that the next problem seen will be neutral conductor overheating.

A meter used to measure the neutral current should be known to read correctly at 180 and 360 Hz. (Or 150/300 Hz, as appropriate)

-Jon

Thanks for the reply Winnie.

You have raised some good questions. Which will lead me to ask more questions, since I know very little about harmonics on neutrals (other than I have read about it occurring).

1. I don't know whether the contractor measured the neutral load, can a normal clamp tester measure the harmonic current? For example a Fluke true RMS eg a model 33 or 35.

2. THD ? I assume is total harmonic distortion, do fittings have this marked on them?

3. I always thought that non linear loads such as switch mode power supplies on computers caused harmonics, can normal ballasts and capacitors cause harmonics as well?

Triplen Harmonics is definately not a term I had heard before, though a quick search on Google showed it is a common term. Are there sites which explain harmonics, triplen etc.




[This message has been edited by Dapo (edited 03-12-2005).]

If the meter is a Fluke 33 as described here: http://www.metrictest.com/product_info.jsp?mfgmdl=FLU%2033 then it should be _perfect_ for determining if harmonics are the problem. Apparently the device can measure true RMS current up to 10000 Hz, and can measure crest factor, which is a measure of harmonic distortion. I've never used one of these myself, but take a look at the datasheet on the page mentioned above.

Get measurements of the crest factor on each phase, and the neutral current, and that should tell you the story.

THD is total harmonic distortion. I doubt that fittings have this marked. Some devices might have power factor and crest factor marked.

_Any_ 'non-linear' load will have a certain amount of harmonic current flow. Computer switching power supplies are one example, because they have input rectifiers that only conduct at the peak of the AC waveform. But transformers are also non-linear, especially when close to saturation. Arc lamps are also non-linear, because the voltage has to get high enough each cycle before they start conducting.

Triplen harmonics are harmonics which are third multiples of the fundamental. In three phase systems, triplen harmonics flow through the neutral, since they are not balanced phase to phase.

I don't really have a list of web sites that would help you; I had to read up on this stuff about a year back, but all I could do now is more google searching. If you search on the term 'IEEE 519' and look for 'tutorial', you'll probably find some good stuff.

-Jon