Odd Voltages on secondary of ungrounded control transformer

[Rem1061]

Got a question for all of you. I have a ungrounded (on the secondary) transformer, It's primary is 4160, secondary is 120. The "Neutral" is not grounded. The control system works fine, If you measure across the ends of the secondary, you will get 120V. If you should measure from one end of the secondary to ground (someone else had started troubleshooting this system and did this) you will get 460Vac. If you measure from the other end to ground, you get 401Vac. With it being ungrounded, I understand that the way the person was checking was wrong, however, for the life of me I cannot figure out why he/I would see that type of voltage. In the past when I measured an ungrounded transformer to ground, I saw a much lower voltage (if any) than expected. Thoughts?

 

[jim dungar]

Ungrounded systems can have anywhere from 0% to 100% of the line to line voltage appear as steady state Line to Ground voltage depending on the amount of coupling capacitance in the system. During certain fault conditions this voltage can reach as high as 200%.
Voltages above 200% would be rare, but because there is no true reference point I guess anything is possible, depending on the internal resistance of the measuring device.

 

[LarryFine]

You have discovered why we do ground such circuits, and a problem with using high-impedance meters.

 

[winnie]

In the situation described, a tiny amount of capacitive coupling or leakage from the 4160V primary to the 120V secondary could cause the high voltage to ground.

This might actually be a problem, because the insulation system on the 120V side is being subject to high voltage.

Can you ground the 120V system? If not, is there a 'shield' in the transformer that is supposed to be grounded?

Jonathan

 

[zooby]

You have discovered why we do ground such circuits, and a problem with using high-impedance meters.

I had that thought on the type of meter used and what the "low-impedance" reading may have been? Guess it would not have mattered here but just curious. After reading ALOT of posts on here, I have a ?Fluke solenoid type meter arriving today.

 

[ptonsparky]

I wasn't aware Fluke made solenoid type testers.
Maybe one of their affiliated brands.

 

[zooby]

I wasn't aware Fluke made solenoid type testers.
Maybe one of their affiliated brands.

Crap... I hope I did not misread this.... I did.

Fluke T+PRO Electrical Voltage Tester | Fluke

The Fluke T+PRO Electrical Tester complies with NFPA 70E regulations by providing rapid indication of voltage levels, faster and safer than old-fashioned solenoid testers even without batteries. Tester helps troubleshoot three-phase systems.

www.fluke.com

 

[ptonsparky]

Crap... I hope I did not misread this.... I did.

Fluke T+PRO Electrical Voltage Tester | Fluke

The Fluke T+PRO Electrical Tester complies with NFPA 70E regulations by providing rapid indication of voltage levels, faster and safer than old-fashioned solenoid testers even without batteries. Tester helps troubleshoot three-phase systems.

www.fluke.com

That is what I use and supplied to my help so we were all on the same page when testing. Not the solenoid tester others use but it is good. Worth the money. Learn it and you won't be sorry.

Solenoid testers can be found at Menards and other stores like it.

 

[winnie]

I had that thought on the type of meter used and what the "low-impedance" reading may have been? Guess it would not have mattered here but just curious. After reading ALOT of posts on here, I have a ?Fluke solenoid type meter arriving today.

The reading with a low impedance meter would almost certainly have been much lower.

The most likely scenario is that there is a very high impedance leakage path between the 4160V winding and the 120V winding (what I described in post #4). When you connect a meter you form a voltage divider with the impedance of the leakage path and the impedance of the meter. The lower the impedance of the meter the lower the voltage reading.

If you have 2 meters with 2 different but reliably known meter impedances, then by measuring both voltages you can figure out the impedance of the leakage path.

Thinking further on this, there is another possibility: there could be an actual insulation failure between the 4160V coil and the 120V coil. If this insulation failure is at a single point of the coil, you create an autotransformer, and it is possible that this fault just happened to be at the 460V 'tap' of the 4160V coil.

-Jonathan

 

[ptonsparky]

As a for instance between Hi impedance and Lo.

VFD while powered, was Off. Load side to EG was 150 volts DC with a Fluke 87.
Adding the T+ would drop it to 15. I didn’t add another T+ but suspect the voltage would have dropped again.

I am slow but, No, I did not do a finger test.