Multi-tap ballast question

[PetrosA]

I had a service call yesterday where a wall mounted 250W HPS was causing a short. It turned out that the 40+ year old fixture, a nice Hubbell back in the day, had one of the two screws missing that holds the faceplate in place and enough water had gotten into the fixture during a recent storm to submerge the ballast, cap and ignitor. The receptacle for their telephone system shares the same breaker that feeds the light and whatever happened took out a few PCBs in the tel. system as well as the UPS it was on (I am recommending they install a dedicated circuit for the tel. system obviously...).

I'm not convinced that the building took a lightning strike and I'm wondering if the other taps on the ballast could have backfed some kind of "dirty" voltage through the neutral or ground to the telephone recept. when they were submerged or that the cap or ignitor could have been discharging into the water sending large and repeated spikes through the neutral or ground. Is there any validity to my theory? I've never actually tested the other leads on a multi tap ballast to see what their output is, but since they need to be capped off, there must be some. In this case it was a 120/208/240/277 MT being fed on the 120V lead.

 

[kwired]

If proper voltage is applied to proper tap then you should have marked voltage on the other taps to the common. A short to a low impedance ground should not be a problem with other equipment. Your ballast is just an autotransformer with multiple taps to the primary side winding. Shorted primary leads will result in higher current between the shorted leads, Connected voltage to common short should be nearly instant trip of overcurrent protection. Shorting of other leads will depend on impedance of the portion of the windings involved to determine how much current will flow. If submerged in water the water will likely have high enough impedance that you will not have a high current.

If you have a UPS on the phone equipment most of them should have surge protection and over/under voltage protection and should have protected the equipment plugged in. If you have a poor equipment ground this protection could be compromised.

 

[mxslick]

It is certainly possible that the submerged parts did indeed kick enough voltage to ground to spike the telco system. It would not take much current to cause damage, and remember to keep in mind one major thing: electricity WILL seek ANY and ALL paths back to its source, and can do multiple paths at the same time, with the current and voltage dividing according to Ohm's Law.

We had an incident in a cinema where a faulty triac for the dimmer shorted to ground, which also fried the power supply to the automation AND the LV control card to the dimmer system. (All of these components are grounded according to Code and share the same subpanel.)

Put out a lot of smoke in the process.