I have a ballast that we are replacing in a church. There are approximately 12 dual lamp T12 Fixtures in parallel drops. The fixtures are all in line -- but L1, Neutral and Ground are parallel drops.

After 40 years some ballasts have failed - they are 4-wire magnetic type units. We have purchased 5 wire electronic ballasts to replace some. Upon replacing the first ballast, a GE magnetic ballast, with a Philips QuickPro 60 Electronic Ballast I managed to get a shock installing the lamp. I measured 50 VAC between the FRAME of the Fixture (my ground wire is attached to the frame), and the LAMP GLASS ITSELF - note the GLASS which is what I was touching when I got shocked - I was touching the middle of the lamp 6 feet in so I know I was not mistaken in what I touched.

I know I wired the new Ballast Correctly.

I have attached a picture showing the new Ballast wiring at the TOP of the image and the original magnetic ballast at the bottom of the picture.

My Question is two fold:

Since I have installed an electronic ballast and the other units in parallel have magnetic ballasts could this be causing an issue ?

What could be the problem that I am measuring this 50VAC between the glass and the frame of the fixture? Example of one Fixture NEW Ballast To Old Ballast Bottom - there are 12 fixtures of the old ballast style in a row - parallel drop

My examples of wiring are correct and I did find the culprit - I will post my answer.

  • Ate you sure you got shocked? 50v is pretty harmless. The telephone system uses 90V to ring
    – Kris
    May 5, 2017 at 20:38
  • @Kris oh I' am pretty sure I got shocked - I would almost do it again just to be sure (I am not stupid - although it might be contested somewhere down the line) - but I will leave that to someone who is electric shock qualified :-) . For me I then used my meter - 50VAC between frame and Glass Tube, the other fixtures with the Old Magnetic Ballast 1.7VAC ~ 2.5VAC - no shock when I touch frame and bulb. POTS uses 48VAC - and yes you can get shocked form that too. I have done enough ballast changes and lamp work to know this is not correct - I just do not know why.
    – Ken
    May 5, 2017 at 20:45
  • Do the fixtures have shunting or non-shunting lampholders? May 5, 2017 at 21:52
  • @Harper the tombstones are just tombstones with wire connectors - 1 for black hot and one for neutral white. I am thinking maybe I need to pull that assembly off and see if someone has a wire attached that I can not see. As I said the fixtures are butted up against each other in a long line. It may be someone did something sneaky on me.
    – Ken
    May 5, 2017 at 23:12
  • 1
    @Mazura Yeah, Tesla did a lot of really fun stuff with wireless electric fields and making fluorescent tubes light up that he's holding in his hand with no wires connected at all. Well, not fluorescent tubes exactly, but things so similar that a fluorescent tube would work. Dec 28, 2018 at 3:55

1 Answer 1


After scratching my head - I decided to pull the tombstones out and inspect the wiring to make sure there was no sneaky connections and also nothing out of the ordinary. The older ballasts (original diagram are Magnetic Ballasts with PCB) the new ballast are electronic. I contacted the manufacturer and they suggested to verify my grounds for the ballast assembly. Since I had several that were fine and all inline - I verified ground from good unit to ground on unit that I was shocked with. Grounds were good.

So upon pulling the tombstones out and checking the wiring - I tugged on each of the wires, one of the wires just slid out of the connection. While it was inserted in the tombstone connection just fine; it just kind of slid out as if the spring clamp had not locked it in. I removed the wire and installed a new wire into the tombstone making sure it clamped down on the wire. After I had done this there was no longer an issue. I assume the wire was close enough to conduct but far enough away to cause the issue.

I do not understand that engineering wise - but maybe it has something to do with RF, High Voltage and an Air gap.

  • So, a break in a conductor on a properly grounded balast can present a shock hazard through the lamp's glass?
    – Mazura
    Jul 13, 2017 at 14:33
  • @Mazura -- it was probably a capacitive "zing" from wire to Ken with the lamp glass in the middle -- as thus, it was impedance limited to not pose a major shock hazard, but still could give you a good nip. Jul 13, 2017 at 22:09
  • @Mazura I physically measured the voltage between glass of the bulb (8 feet bulb) and the fixture I had 50VAC. After fixing the wire that reading went down to just about 1 VAC. No shock. I never experienced this before and I have changed plenty of ballasts and bulbs. I am not sure of the reason this happened but electronic ballasts operate in the Khz range and so I suspect the HV and the RF along with the 'loose connection -AirGap' of the wire to the lamp pin allowed this to occur. I searched briefly to try to understand it; all I had found so far was a spark gap transmitter.
    – Ken
    Jul 14, 2017 at 21:01
  • 1
    Tubes interact with the grounded metal fixture. Some types of ballasts (i.e. Instant) rely on this capacitive coupling to strike the lamp, so you can get a big surprise if you use that ballast type for back illumination in a lit sign where there's only plastic on either side of the tubes. Dec 28, 2018 at 3:39

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.