Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Strange Voltages to ground on control circuit

  1. Apr 30, 2009 #1
    Hi all,

    One of my guys was working on a control ciruit for a fan bank. The control circuit in powered by a 480-120 V control xfmr. His voltages from x1-x2 were 123 Vac. Across all energized relays: 123 Vac. But when he took voltages to ground he got around 40 Vac from the X1 bus to ground and 80 Vac from the X2 bus to ground.

    I asked him if the transfromer's X2 terminal was properly bonded to ground. He said he could not definetly find where the control circuit was grounded.

    I told him there is a reference to ground somewhere in the circuit. My guess is that the XFMR is not grounded and he is reading a voltage divider from the high resistance (megaohms?) leakage to ground form each leg of the circuit.

    Unfortunately we cannot tag out this circuit to do furthur testing so this all the info I have right now.

    Can anyone give their thoughts and the theory behind these strange voltages? Remember I THINK the ground is not connected to X2, but I don't know if it is for sure, I haven't seen it myself.

    thanks in advance for your theories.

    Danny
     
  2. jcsd
  3. Apr 30, 2009 #2
    I agree, the circuit probably is not grounded.
     
  4. Apr 30, 2009 #3
    Make absolutely sure the case of the transformer is grounded. Also, with the power off, check resistances between all transformer taps and taps to ground.
     
  5. Apr 30, 2009 #4
    For industrial control panels, it's common practice to have the 120vac control wiring isolated from ground (floating). At least it was back in the 70's and 80's. But like Bob said, the transformer case and panel should be grounded.
     
  6. May 1, 2009 #5
    Like I said this circuit cannot be tagged out right now due to its importance. so I can't take resistance readings.

    I am looking for the ENGINEERING THEORY or formulas of why I am getting 80 v to grd from one leg and 40 V to grd from the other leg. We all kind of agree the reason is becuase X2 in probably not grounded.

    My question is WHY doe these voltages occur? And what does the term "floating ground" mean? I have heard it before but don't quite know the true definition.

    thanks Danny
     
  7. May 1, 2009 #6
    TurtleMeister, perhaps you can enlighten me as to why a control circuit would be ungrounded. As I understand it one of the purposes of grounding is to prevent accidental bypassing of control functions. For instance, imagine a boiler with a simple thermostat. When the water in the boiler gets up to the set point, the switch opens and the boiler shuts off. If after years of use the insulation frays and the conductor on one side of the switch touches to ground, what happens? Nothing, the circuit can continue to operate for years with no one noticing. So at some point should a conductor on the on the other side of the switch contact ground, the switch that controls the boiler is bypassed and the boiler can continue to heat up until it explodes.

    If on the other hand, the return conductor were grounded, as soon as the first instance of the control circuit touching ground, the fuse would have blown and the defect corrected.
     
  8. May 1, 2009 #7
    Because the circuit you're measuring is high resistance, your meter will be sensitive to anything in the panel. It's probably caused by whatever magnetic fields are present in the panel. Try this - take a 120 volt light bulb and connect it between x1 or x2 and ground. Nothing should happen. It will not light. Now measure the voltage across the bulb. It should be zero. I don't know where the term "floating ground" came from. We use to call it "floating neutral" because the neutral wire in the panel, usually labeled a small number such as 1 or 2 and usually having white insulation, would be isolated from ground.

    Correction: Rather than it being magnetic fields causing your strange voltage readings, it could also be leakage through the transformer due to capacitance between the windings. But this is still a high impedance path, so placing any kind of load on it should cause it to disappear. But it still may provide enough current to give you a shock.

    Sorry, I don't know. I've worked with lots of industrial control panels and that's just the way they were wired. However, these were all machines used for industrial manufacturing. I don't remember ever seeing one that had the 120v neutral grounded.
     
    Last edited: May 1, 2009
  9. May 1, 2009 #8
    From Rockwell Industrial Automation Glossary fifth edition page 49:

     
  10. May 1, 2009 #9
    I worked in a lab at MIT many years ago that had the standard 3-pin grounded 120V AC plug strips. I measured the voltage out one day, and to my amazement, between every combination of sockets the voltage was 120 volts. I asked the electrician why we had 3-phase 120 volt power, and he told me the step-down transformer had a 3-phase delta, and not a 3-phase wye, output. So they chose one of the wye legs as ground.
     
  11. May 5, 2009 #10
    I wanted to test the theory of capacitive coupling, so I had a control XFMR laying around and wired up on the bench. I plugged in 120 from the wall outlet and set it up to put out 60 on the secondary winding. I did NOT ground the secondary output.

    I took a fluke DMM and went from X1-X2 got 65 Vac.
    Went X1 to ground got 29 Vac
    Went X2 to ground got 30 Vac

    I then took a wiggins tester and got nothing to ground on either leg.

    Then I bonded X2 to ground and got:
    Went X1 to ground got 65 Vac
    Went X2 to ground got 0 Vac

    I am very happy with the results and this proves the theory to me

    (attached my xfmr diagram)
     

    Attached Files:

Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: Strange Voltages to ground on control circuit
  1. Circuit ground (Replies: 3)

Loading...