ECN Forum Forums Technical Discussion Electrical Theory and Applications Elevator feeder circuit

I have recently come across a 480 volt elevator being fed by an ungrounded system. This was the most competitive way to feed the elevator - delta to delta - but my concern is that without a grounded leg or midpoint tap, to create the return path for faults, is this a proper way to connect an elevator to a power source?

As long as the equiment is properly grounded I don't see any problem with a delta-delta set-up....in fact it's common for motor circuits.

If you have a ground fault no current should flow. The delta is simply referenced to ground at the point of the fault.

GJ

[This message has been edited by golf junkie (edited 04-02-2002).]

As GJ mentioned, an Ungrounded Delta 3 wire system will function safely as long as there is proper Grounding done.

This may sound incorrect because the system is Ungrounded, but the metallic enclosures, raceways and such are bonded to Ground [Earth, Structure, etc.] to drive the Voltage To Ground on them [in the case of a fault to the metallic items] to a very low value.

There will not be a high Ground Fault current flow in this event unless one "Phase" has previously faulted to the metallic items.

If the metallic items are not bonded to ground as mentioned above, if one "Phase" faults to the metal it will setup a shock hazzard to someone touching the metallic equipment - yet will not draw enough current to trip an OCPD.

Scott SET

I appreciate your input, but my concern is the equipment running and utilizing the ungrounded feeder. Without a reference to ground through the separately derived system you must rely on a ground established at the service equipment that may be a long distance away. It will take two faults to clear the breaker because there is no current flow back to the transformer. A phase to phase fault must occur to clear the original fault. If the distance between the phase to phase fault is lengthy the O/C device may consider this part of the load. I understand about the use of this in industrial applications where warning lights and horns go off when a fault occurs and can be cleared by maintenance personnel, but I am concerned about the elevator being fed by this ungrounded system and all electrical equipment within the elevator relying on protection from an overcurrent device that may not operate under fault conditions. Are these elevators listed (tested) using both grounded and ungrounded systems to check for safety conditions that may occur during operation? The grounded system in this application is by far a safer system and will clear the original fault when trouble occurs but I cannot be as confident utilizing the ungrounded system. The price of the transformer should not be the determining factor when life safety issues could be compromised.

Distribution at 240-600 volts in other than a wye, solidly-grounded configuration isn't necessarily undesirable, although sometimes it is chosen for apparently having the least initial cost. Phase-to-ground fault current will typically be low—well below the hundreds or thousands of amperes seen with arcing faults in solidly-grounded systems.

An ungrounded system can resonate across the natural capacitance formed between ground and the three energized phases. An intermittent failure of insulation can incite phase-to-ground overvoltage with magnitudes well above system ø-ø voltage. One apt recounting of a decades-ago occurrence is http://67.115.161.42/dat/beemaIPSH6.doc

An acceptable tradeoff is to convert an ungrounded system to a high-resistance grounded system. A slightly more complex circuit than that of a ground-detector can be effective in damping ground overvoltage. This will limit ø-g overvoltages, both transient and steady-state. Ground-fault currents in properly compensated systems are usually below 5 amperes. Where the serving transformer is 4-wire wye, a single resistor can be used to make a system having significantly limited transient and steady-state overvoltage incidents above the ø-ø level. For a 3-wire delta source, three small auxiliary transformers are also required.

The petrochem industry actively promotes low-voltage {typically 480V} systems that are high-resistance grounded. They advocate installation of ground detectors so that a single ø-g fault can be located prior to a second, more debilitating fault occurring on another phase.

It is a serious misapplication to apply 480Y/277V molded-case circuit breakers or 300V class-T (or G) fuses to other than a wye, solidly-grounded system.

Question to orignal poster johngeorge: What is the highside voltage of the subject Δ-Δ transformer?


[This message has been edited by Bjarney (edited 04-17-2002).]

This particular installation used the Delta to Delta transformer to transform 240 volt to 480 volt, feeding a hydralic motor and other components. I have recently talked with a U/L engineer and he related to NEC 250-30, 20 and 21 in describing his concern over the 120 volt circuits derived from an internal transformer 480V to 120V without proper grounding of the lower voltage side of the transformer. This second transformer, within the elevator, supplies lighting and possibly outlets with the cab as well as control circuitry, and if not grounded properly may create a hazard. The 240 to 480 Delta to Delta was within code requirements but possibly poor judgment on the part of the installer.
I appreciate your description of the proper uses of the ungrounded systems and the reasoning behind their existance. I do not believe this installation is a proper use of this type system.
I will have to check the type overcurrent device used on the secondary of the transformer, and that is a good point to keep in mind.

Minor note: Based on 99NEC240-3(f), the 240V overcurrent device may be permitted to serve as protection for the 480V side.

[This message has been edited by Bjarney (edited 04-18-2002).]

The idea behind 480V ungrounded systems is that the first phase-to-ground (ø-g) fault will not usually operate an overcurrent device, but if it’s ignored, the second ø-g fault will cause somewhat greater problems, and likely be harder to isolate with possibly two overcurrent devices operating at the same time. If the ungrounded system is not monitored by even simple means, then this advantage cannot be realized.

In jg’s case, the 240∆-480∆ transformer availability may have “won out” over a 240V∆-480Y/277V unit with a typically long non-stock lead time. [Interestingly, several sizes of 208V∆-480Y/277V models are shown available as catalog items.]

Electricians need to be aware that the usual ~280V ø-g reading may shift, with some measurements found significantly above or below this reading. {In an oversimplified sense—with nothing else served ø-g or ø-n, who cares what it reads?} Determining whether a real problem exists with goofy ø-g readings {where ø-ø readings are normal} is part of being able to troubleshoot something beyond plain-vanilla problems. A skilled troubleshooter can be a lot more valuable [and make more money] when he understands which acceptable variations are perfectly fine and others indicate a true problem.

Bjarney:
Thank your for your input into this problem. I now understand, more clearly than before, the uses and of course misuses of the ungrounded electrical system.