Stray voltage and bundling issues

[JonNye]

Hey guys, I have a stray voltage scenario that you might find interesting. I’m getting between 5V to 20V on a thermostat cable to an air handler. I removed the cable from system completely (AH, HP, and tstat) and still had that voltage.
I also have same voltage on the ground of the branch circuit coming in even with the power disconnected.
There is an excessive amount of bundling of romex cables on the ground level. This is a clubhouse with maybe 40-50 golf cart chargers running where these cables are bundled as well. Well over 12-15 12/2 and 12/3 cables bundled with the 6/2 romex I’m testing.
Only solution I can find is to pull a new circuit and new tstat cable. Any suggestions?

 

[tom baker]

Use a low impedance meter.
What issues is the induced voltage causing?

 

[JonNye]

The low voltage would get high enough to pull in the contactor for the heat strips when the unit was turned completely off. Voltage is always present even without wires connected to any equipment at all.

 

[JonNye]

Voltage also fluctuates greatly depending on the demand on the building. Roughly from 5V up to 20+V

 

[Elect117]

I also have same voltage on the ground of the branch circuit coming in even with the power disconnected.

What are you measuring between? Phase to equipment ground? Phase to neutral? Neutral to equipment ground?

If you amp-clamp the equipment ground what do you get?

What are the actual voltages phase to phase, phase to neutral and neutral to equipment ground?

Is the circuit 3ph? wye or split phase delta? Single phase?

Voltage induction from one circuit to another is exceptionally rare with lower current and voltage values like you have. Especially with insulated wiring.

 

[don_resqcapt19]

Hey guys, I have a stray voltage scenario that you might find interesting. I’m getting between 5V to 20V on a thermostat cable to an air handler. I removed the cable from system completely (AH, HP, and tstat) and still had that voltage.
I also have same voltage on the ground of the branch circuit coming in even with the power disconnected.
There is an excessive amount of bundling of romex cables on the ground level. This is a clubhouse with maybe 40-50 golf cart chargers running where these cables are bundled as well. Well over 12-15 12/2 and 12/3 cables bundled with the 6/2 romex I’m testing.
Only solution I can find is to pull a new circuit and new tstat cable. Any suggestions?

That would require an ampacity adjustment of 45% for the conductors, so the maximum breaker for the 12 AWG conductors would be 15 amps.

It would be very rare that the capacitively coupled voltage on the thermostat wire would have enough current to pull-in a controller, unless it is some type of electronic controller and not an actual electromagnetic device. Even with an electronic controller it would not be common.

The easiest solution is to reroute the thermostat wire. The voltage on the power conductors is not an issue.

A second solution would be to add load in parallel with the controller so that the capacitively coupled current cannot pull in the controller.

 

[winnie]

Before running a new circuit, I'd investigate the termination of the existing circuit.

My guess is that you have a low voltage cable connected to a high impedance input, perhaps some sort of isolator with a logic level input rather than driving a contactor coil directly. The cable is possibly driven by a relay contact in the thermostat, and high impedance when the relay is open.

The miniscule current induced in the low voltage cable is sufficient to create a high voltage because there is no significant load. A small 'pull down' resistor is probably all that is needed.

As a first step, do what @tom baker suggests, measure the voltage with a low impedance meter. If the voltage is present with a high impedance meter and goes away with a low impedance meter, then using a pull down resistor is likely the solution.

-Jonathan

 

[garbo]

Years ago heard that was called transformation. Bundles cables will induce a voltage to other wires. If you wired a 24 volt low wattage light bulb to it and see if it produces much light. Using a typical digital VOM to measure just one lead of a 480 volt circuit will often show close to 100 volts expect in damp locations while the other test lead is not connected. That throw a lot of guys a wild goose chase when they switched from low impedance analog meters to high impedance digital meters.like Winnie stated install a resistor .If you have a 5 k ohm pot handy temporarily install it and while voltage is high turn the pot until voltage drops to zero volts then remove pot and measure the resistance and use a similar ohmage resistor.

 

[Elect117]

Years ago heard that was called transformation.

Transposition? Used to reduce the impedance due to mutual inductance.

Before running a new circuit, I'd investigate the termination of the existing circuit.

My guess is that you have a low voltage cable connected to a high impedance input, perhaps some sort of isolator with a logic level input rather than driving a contactor coil directly. The cable is possibly driven by a relay contact in the thermostat, and high impedance when the relay is open.

The miniscule current induced in the low voltage cable is sufficient to create a high voltage because there is no significant load. A small 'pull down' resistor is probably all that is needed.

As a first step, do what @tom baker suggests, measure the voltage with a low impedance meter. If the voltage is present with a high impedance meter and goes away with a low impedance meter, then using a pull down resistor is likely the solution.

-Jonathan

I guess I don't spend enough time with controls / contactors. I wouldn't have thought the input resistance of the circuit would be that high. I always thought the circuit was built to have a higher output resistance and a gain of less than or equal to 1 and a feedback loop to reduce the effects of noise.

But then again, it has been awhile since I have looked at that stuff.

 

[don_resqcapt19]

Transposition? Used to reduce the impedance due to mutual inductance.



I guess I don't spend enough time with controls / contactors. I wouldn't have thought the input resistance of the circuit would be that high. I always thought the circuit was built to have a higher output resistance and a gain of less than or equal to 1 and a feedback loop to reduce the effects of noise.

But then again, it has been awhile since I have looked at that stuff.

When we first started switching over 3 phase 480 volt motors with across the line starting and the control conductors run in the same conduit with the motor conductors we has issues. The original motor controls were 120 volts from a CPT in the starter. We used a 120 volt control interface card on the VFD, but had to add relays because they would never respond to a stop command. The capacitively coupled voltage on the control conductors would keep the drive running. With larger motors, the coupled voltage would even hold in an ice cube relay. For those we went to a larger relay to solve the problem.

Where new motors were installed, we started installing two conduits...one for the motor and one for the controls to avoid this issue.

 

[jim dungar]

This is definitely a voltage coupling issue not a stray voltage one. Look into solutions to minimize "capacitive coupling" issues.

 

[garbo]

When we first started switching over 3 phase 480 volt motors with across the line starting and the control conductors run in the same conduit with the motor conductors we has issues. The original motor controls were 120 volts from a CPT in the starter. We used a 120 volt control interface card on the VFD, but had to add relays because they would never respond to a stop command. The capacitively coupled voltage on the control conductors would keep the drive running. With larger motors, the coupled voltage would even hold in an ice cube relay. For those we went to a larger relay to solve the problem.

Where new motors were installed, we started installing two conduits...one for the motor and one for the controls to avoid this issue.

At least for the last 35 years VFD manufacturers stated to use a conduit to feed drive, a ( preferably steel ) conduit with a ground wire to the motor and a conduit for controls. Had a few starter to drive change overs where conduit was buried in concrete and control wires ran in same conduit as motor so we used 24 volts DC for controls.

 

[don_resqcapt19]

At least for the last 35 years VFD manufacturers stated to use a conduit to feed drive, a ( preferably steel ) conduit with a ground wire to the motor and a conduit for controls. Had a few starter to drive change overs where conduit was buried in concrete and control wires ran in same conduit as motor so we used 24 volts DC for controls.

That was not going to happen when upgrading an existing motor from across the line starting to a VFD. The cost to add an additional conduit for the controls in a Class I, Division 2 location would far outweigh any benefit from installing the VFD. If we could not make a VFD work there are a number of other methods to accomplish the same thing...one example would be a flow control valve on a pump discharge in place of a VFD on the pump motor.
The relays provided an interface between the 10 volt DC controls, on the original Lovejoy VFDs and the existing 120 volt motor control circuits, so the actual VFD control circuit was not in the conduit with the motor leads.

Later when they switched to Yaskawa drives, with a 120 volt interface card, we still used the relays for isolation when an existing motor was upgraded to a VFD.

It seems that when the switching frequencies went up in the drives, we even started having issues with the coupled voltage holding in the run relay so the stop command from the PLC did not stop the motor. This was solved with a larger relay where the coil had a larger burden on the control circut, a burden that the coupled current could not over come.

As I said in the previous post, this was only done when an existing motor was switched to a VFD. Any time a new motor was installed and was to be supplied by a VFD, two conduits were run.