We've got four 400W 220V metal halide pole mounted fixtures in our church parking lot that currently are being switched by a 15A 110V photocell (switching one side of the 220V). After 10 months of working, one of the ballasts has blown. I thought you should get several years out of these ballasts. My suspicion is that since the photocell is switching only 110, there is still 110V at the ballast all the time and thus leading to premature end of life. The original electrician who installed says it shouldn't matter and claims he checked with the ballast manufacturer on this. Would only switching one leg of the full voltage lead to a short ballast life?

I've got a few problems with your arrangement. The short version? Fire the handyman, and hire an electrician!

As a general principle, you should open both of the "hot" wires. While I don't think switching only one side caused the failure, switching both sides of the circuit does make for a much safer installation. It's hard to remember that everything in the fixture is "hot" even though the light is out.

Which leads to the next detail ... that photocell is switching 220, not 110. I suspect a 220v rated photocell was used; a 110v version wouldn't last very long.

800 watts of MH lighting is pushing the limits of the photocell; this kind of lighting is much more demanding of the switch than ordinary light bulbs; photocells are typically rated at much lower wattages for "tungsten" uses.

A preferred way to instal this system is to have the photocell operate a contactor, and let the contactor actually operate the lights. This arrangement, would also alllow you to add a timer, or a manual bypass switch (for testing). (Do you really want the lights on all night?)

While the ballast is being replaced .... there seems no pattern to their lifespan .... you should have all the components replaced ... that is, the ballast, the igniter, and the capacitor. They're sold as a set.

The electrician should have a one year guarantee on his work.

If the bulbs are more than three years old, you might as well change them all at once. While you're up there, check the ballast and fixture for marking indicating what sort of bulb should be used ... there ought to be a marking along the lines of "M400."

If the ballast is a quad-tap ballast, it will have a 120 volt lead, a 208 volt lead, a 240 volt lead, a 277 volt lead, and the common.
There is no such thing as 220 volt (or 110 volt, for that matter)

If you can determine the operating voltage, make sure that you don't have it inadvertantly hooked up to the wrong tap. It can cause premature failure, and I'd check all the other one's while you're at it.

For the proper photocontrol, I'd recommend that you use one of these:

http://www.intermatic.com/Default.asp?action=prod&pid=9154&sid=236&cid=119&did=45

It's rated 208-277 volts, at 8.3 amps with a ballast.

Thank you, Scott. I used "220" because the OP did.

That Intermatic unit has a white wire, does that mean it requires a neutral? (the installation manual on the page linked doesn't say).
That would be a show stopper in most 240v installations since they probably did not bring a neutral to the pole.
I am still not sure why the 2 wire PC in the top note is not acceptable (assuming it is listed to open a 240v circuit). A photocell is not a disconnect and should never be confused with one.

I suppose it depends on whether a photocell is a "switch".

Greg, you may technically be correct ... but a lot of less trained, less paranoid folks are not clear on just how 240 works, and are likely to think "off" means "dead."

I agree that, while a photocell is certainly a switch, it isn't a DISCONNECT switch.

Here, again, is where good design / trade practice differs from "code." You switch ALL the hots, fuse ALL the hots, etc ... even if 'code' doesn't strictly require it.

As for wiring the photocell ... the Intermatic wiring diagrams are extremely misleading as to wiring for 208, 240, and 277. There is at least one error in the instructions. It's best to leave 'murphy' off the job, and stick with a contactor.

Asking a photocell to switch that much inductive in-rush current is just false economy, sooner or later the contacts in the photocell will wear off their coating and will weld together, leaving you with a large power bill, if not noticed.

I would tend to agree with Reno, in saying that if you are going to switch this sort of load at the current it will require on start-up, get yourself a decent contactor and a licensed electrician to install it.
It is not a job for a handyman!

I've used that particular photocontrol in a variety of different applications, up to it's rated load (the 4- 400 Watt @ 208 Volts included), and never had a problem with its longevity.
It's MADE to take the load, otherwise Intermatic wouldn't make it, and rate it as such. Contacts weld on contactors, too.

It's a switch, nothing more, nothing less. In that mode of thinking we'd all be installing oversize switches, or contactors, on everything.

Yes it has a white wire, it's not a neutral (except for 277 volt applications) It's listed as a unit.



I agree with the the fact that the OP needs to hire a qualified electrician to service the installation, and the electrician needs to turn off the power at the circuit breaker. I don't think anybody qualified would consider the P/C to be a disconnecting means.

I too am surprised that the 15A photocell hasn't welded yet from the inrush current.

I'm in with the rest here - 120V photocell (or 240V if you prefer) operating a contactor. Switch BOTH legs of the power with the contactor - fixtures should not be hot (AT ALL) when deactivated, they should be dead.

There is no reason to have this installation any other way, but intermatic does make a 25A photocell that would give you more headroom on an installation like this.

Good Luck!

thanks for the info from everyone .. I'm sorry for my confusion in the original post. We are using quad tap ballasts at 208V. I can only hope that the electrical contractor is actually checking the correct tap is being used. At 30', we have no easy way to do this ourselves. The original installer was a large electrical contractor co. that the general contractor had used as a sub. They keep saying they've never had this type of problem on any other job. So much for that. We'll try someone else.

One last question ... just how much life should you expect from these types of ballasts? and bulbs?

Ballasts seem to have no set 'average.' Some seem to die after a year; another put in, right next to it, on the same circuit, will last over a decade.

If I were to try to guess at a pattern, I'd look back at one job I maintain .... 13 years, 18 ballasts like yours, and I've had to replace two.

Bulbs seem to last 2-3 years.

Thanks - again - electre! Not having a photocell box handy, I wasn't sure just how close this installation was to the capacity of the photocell.

I assumed these things were triac based, the ones I have used were. There are no "contacts" to burn. At that point there is really no advantage to a 2 pole device since they all leak some current. The voltage on the lamp (to ground) won't be full line voltage but it won't be zero either.
In the equipment I have worked on with solid state switching they don't even bother switching all legs on a l/l load. Single phase gets one device and 3p gets 2.
Triac switches don't really like 3 phase since there is no real "zero crossing" but they do work. Typically they use something several times the actual current.