I have a double circline light fixture in the kitchen. I replaced the ballast resistor 3 years ago. The lights went out again a day ago, so I replaced the bulbs. Still no light.

I figured that left the ballast resistor, though it seems strange it went so fast when apparently the original from the seventies lasted until 3 years ago. Suspecting something was up, after I took out the ballast resister I measured the power line with a digital multimeter. It showed 15 volts with the switch on and off.

If one of the readings is stray voltage, how do I correct for that with a digital voltmeter. I think the light is on a 30 amp line, and my digital voltmeter only measures AC current up to 10 amps.

  • 2
    Put any load on the circuit and it will remove stray voltage. A night light, even a lamp based circuit tester. However if the reading is the same with the light switch in both positions, something is wrong, your testing method is faulty or the problem isn't the ballast! Commented Jan 26, 2017 at 2:09
  • Thanks for answering. I put an old night light on the circuit, flipped the switch, and nothing happened. Several other ceiling lights on the same line, (I checked by turning off the circuit breaker), are functioning. So I assume it's the wall switch. I'll take care of that tomorrow. Thanks for the advice. Commented Jan 26, 2017 at 4:26
  • You seem to be mixing up voltage and current measurements. 1) Don't do that 2) Choose your tools carefully Commented Jan 26, 2017 at 14:58
  • Thanks for answering. I used only the AC voltage measurement on the multimeter, since the AC current measurement had a limit of 10A and I thought the circuit was more. Turned out the line actually was 10A. However, I think I'm just going to install a new wall switch and test the leads to the now-removed ballast resistor for 115 or 120 volts. I will not do any wiring up without flipping the circuit breaker to OFF. I haven't taken the new ballast resistor out of the box yet, in case the old ballast resistor still works and I have to return the new one. Commented Jan 26, 2017 at 16:38

1 Answer 1

  1. Are you sure that it's a ballast resistor and not a transformer? A resistor would be extremely inefficient, obviating any benefit in efficiency of a fluorescent lamp.

  2. A DVM has a high input impedance and will pick up stray voltage coupled by wiring capacitance even when the switch is off. There is no need to measure the voltage when the light is switched off.

  3. N.B. Do not measure the current across the mains with the meter. If you put the meter in series with the lamp, it should draw no more than 1 ampere, but do not put the meter, in current mode, directly across the line!

If you're having frequent issues with the fixture, replace it with a surface-mount LED fixture. You can find theses in different color temperatures from 2,700 K (co0lor similar to incandescent) to 6,000 K (actinic blue-white). LED lamps should (at least theoretically), last many years more than fluorescents. (The link is just an example.)

  • Thanks for your advice. I think you're right, it's a transformer. It just says GE Magnetic Ballast on the unit. I removed the old ballast and measured the leads to it-the unit says it's supposed to work on 120V. The night light connected across the leads did not light. Since the circuit is 10A, I could use by AC current setting on my multi, but I decided to just turn off the circuit breaker, change the wall switch, turn the circuit breaker back on and measure the voltage. Then go from there. Commented Jan 26, 2017 at 16:47
  • 1
    What do you mean "the circuit is 10A"? You wrote it was protected by a 30 A circuit breaker!? If you put the VOM in the current mode in series with the light fixture you will not read 10 A but 1 A or less. On the other hand if you put the meter in current mode in parallel with the light fixture (i.e., across the neutral and hot) then the current will rapidly rise (milliseconds) to more than 30 A before the breaker in the panel would trip. Your VOM prolly has its own breaker and it might trip first and save the meter, but there would be no point in doing this. Commented Jan 27, 2017 at 0:56
  • Exactly... you would most likely see a bright blue flash of vaporised metal from the test probes before the fuse or breaker kicks in. Commented Jan 27, 2017 at 17:35

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.