I'm working in a building with a 240V 3Ø service for 'power', and a 120/240V 1Ø service for 'lighting'.

There are welding plugs (240V 1Ø) run off the 'lighting' service. The strange part is I get ~130V from one leg to ground, and ~140V from the other leg to ground.

I speculating the POCO is using a delta 4-wire arrangement to supply the two services.

Is my voltage-to-ground reading indicative that the welding plug is not properly bonded? (i.e. measuring delta-system capacitive 'ground' instead?)

I ran out of time on this visit... I'm going to delve into it further next time.

Hello, u2slow;

I have dealt with many Dual Metered Services, such as the ones you are describing.
One Gear Section (w/ KWH Meter) for 120/240 VAC 1Ø 3W "LIGHTING",
and another Gear Section + KWH Meter for 240 VAC 3Ø 3W "POWER".

Most likely, the PoCo's Transformer bank is a 4 Wire Delta - could be an Open Delta, or a Closed Delta configuration used - I have seen them both used, depending on the location.

I have seen a few locations with 3Ø 3W "Power" as a 3 Wire Closed Delta arrangement, and the 1Ø 3W "Lighting" from a single phase Transformer, for Customers with dual Services.

These were in areas which were primarily Commercial / Light Industrial, whereas the 4 Wire Deltas were in Commercial / Residential type areas, but seen plenty of 4 Wire Deltas in Commercial / Light Industrial areas too.

The 4 Wire Deltas were Pole Mounted Configurations, consisting of:
<OL TYPE=A>

[*] On Open Deltas: very large 1Ø Transformer, with all secondary taps used, and a smaller 1Ø Transformer used to create the 3Ø "C" leg,


[*] On Closed Deltas: Center Xformer used for 1Ø 3W is maybe 200% + the size of one of the two Outer Xformers.
</OL>

The 130-140 VAC reading may be either from the type of Meter used, or simply what is being delivered from the PoCo at that time - like at a low usage time, or maybe the P.F. is Leading (Capacitive Reactance).

The 140 VAC is reaching the "Excessive" limits, but should not be too concerning unless there is Equipment which does not like that high of Voltage - and this is also measured at the Equipment during operation.

Let us know what you find after the next visit.

Scott35

I am not an electricity engineer or any expert.

I live in a place where the mains voltage is 220 ac. Sometimes I have a device requiring 110 ac. I learned this trick from the local electrician, namely, use one of the two wires from the mains and substitute a connection to a metal rod driven into the ground for the other wire from the mains outlet, thereby acquiring 110 ac for the 110 ac device.

How does this setup work?

Suppose you get one wire from the mains and touch it with the metal rod driven to ground, will you set off a short?

gerry

Bill, Sorry to sound like I am answering a DIY question, but this post has been up for a couple of days, and it is just too darn dangerous to let slide.

gerry,

The 'solution' that you just described will not work in many cases, will be unreliable in those cases where it does work, and is extremely dangerous. Depending upon the circumstances you will do nothing more than kill a few worms and damage your electrical equipment, to triggering ground fault protection equipment, to getting someone killed.

It might work in some circumstances, but far more likely it will damage the connected electrical equipment, because the '120V' is not well regulated. You are connecting electrical power to a non-bonded grounding electrode, which means that any current flow will have to somehow get through the earth back to the source transformer. This current through the Earth could do it's own damage, and also present a considerable shock hazard.

Just don't do this.

-Jon

To add to Jon's post, many (if not all) 220v power areas have one end of the secondary grounded, meaning the ground rod would be using the earth as a voltage-divider resistor.

If this is true, your resulting voltage could be anywhere between 220 and 0. If I am wrong, you'd be getting an unreliable voltage anyway, because earth resistance is not a constant.

In either case, the voltage applied to the load will vary with the current usage (or, more accurately, the load impedance as it attempts to vary the current). You can also create dangerous voltage gradients in the earth.

Bottom line: it's a bad, bad idea.

U2, are your voltage measurements to earth or the system neutral? The should be the same, of course, meaning zero difference between ground and neutral.

It sounds like you're getting a high voltage from the utility, which will accelerate the power meter fro most load types. ($)

I still haven't been back to investigate.

I checked from hot to ground (metallic oulet box) at the 6-50R welding receptacle and got ~140V. Same on each leg. I confirmed it was fed from a single-phase panel. I got ~120V to ground (panel chassis) on each leg at the panel itself.

The walls are finished in between the outlet and panel. Judging by the connector in the box, it is piped with EMT - and there's no bond wire in the pipe.

Supposing the EMT is not providing a bond path anymore - perhaps through mechanical damage or dissolved in the slab - isn't that the same as turning your 4-wire delta back into a 3-wire (ungrounded) delta? And isn't ~140V (240/(root 3) what you should read from phase-to-ground on an ungrounded delta system?

Is it a true RMS testor? My bro went head to head with PoCo at mt uncle's shop over a funky voltage read. When it turned out it was his cheap testor, he tucked his tail between his legs and scurried over to the supply house to drop $500+ on a nice new Fluke.

The presence or absence of a feeder/branch grounded conductor does not change the system. The creation of an ungrounded or grounded system can only take place at the source of the system which is either a transformer secondary or a generator output.

What are your measured line-line and line-ground voltages at the "power" service and the "lighting" service. Ideally you would take these mesurements as close to the same time as possible.

Can you put a recording volt meter at the load to see how the voltage changes over the course of the day?

Remember that 'ground' potential is _not_ the same everywhere. The two supply legs might be at 120V relative to the supply ground reference, but at a different voltage relative to a different ground reference.

This would require both 1) bad bonding between the supply and the point that you are measuring and 2) something introducing current into the ground so that there is different ground potential at your measurement point. Perhaps there is a high impedance ground fault on the third leg, dumping current into some grounded electrode, presenting a shock hazard and killing a few worms, but not anything that anyone has yet noticed

Perhaps the panel at the source is properly bonded to neutral, but the grounding electrodes have dissolved, and the bond to the receptacle box is also toast. In this case your source is only kinda-sorta grounded, through the impedance of whatever ersatz 'grounding electrodes' happen to be bonded to the panel. In this case, 'ground' at the panel will be right at the center tap potential, but elsewhere in the system, the potential to ground might not be as well defined.

-Jon


If there is bad ground bonding between the source and the point of measurement, then 'stray voltage' could

Ahhh, an old and forgotten thread!
Need to get a "Heads-Up" out to u2slow, and find out just a little bit more regarding this scenario; such as:

<OL TYPE=A>

[*] What type of Volt Meter was used - Low Z input or Hi Z Input,


[*] Any "Free-Wheeling" Motors running during the Voltage testing - i.e. Induction Motor(s) running with no loads connected to shaft,


[*] Are both Services fed from the same Transformer Bank - as in a 4 wire Delta, or is there a 3 wire Delta and a separate 1Ø 3 wire Transformer,


[*] Any Incandescent Lamps in the location "Burning Bright" or failing rapidly,


[*] Any Resistance Heating Equipment, which runs exceptionally hot,


[*] Any way to connect the 2nd tester lead to a "known to be solidly bonded" item, or maybe the Grounded Conductor side of a nearby 125 VAC Receptacle,


[*] Can you place a moderate load - say possibly 10-15 amp load, on the Circuit at the test point, then check Voltage,


[*] If using a Hi Input Impedance Meter (Hi Z Input) - such as a DVM with a very low AC Voltage range - have you shunted the meter with a lower Impedance item (like a "Wiggy" in Parallel).
</OL>

Per "A", Would like to know what type of Meter is being used, which may / may not be of any influence. Also, is there any "Highest / Peak" switch on the Meter?

Per "B", Unloaded Induction Motors on the same Circuit(s) or Line(s) may have caused a higher Voltage reading at one point, and maybe by the time the Voltage was tested again (at the Panel), the Motor(s) were loaded up or turned off.

Per "C", maybe if there are two separate Transformer banks feeding the dual services, somehow something got mixed up???

Per "D", looking for an overall Voltage increase... as would be explained by Incandescent Lamps which are "Brighter Than Really Should Be", or "Full-Time Job Replacing Lamps".
Similar in respect for item "E", except the Heating Elements really kick Butt!

As to "F", could "The Other" lead of the Tester be connected to something known to be Ground Bonded - or the "Larger" tab side of a nearby 125 VAC Receptacle?

"G" would see what effects placing a load on the circuit would have on the Voltage - at that point (the 6-50R Receptacle).

And for "H", try testing Voltage with both a "Wiggy" and your Meter at the same time, then see if any difference in Voltage readings. Alternative to the "Wiggy" might be an Incandescent Lamp (something safe please!!!).

Will check back later for replies.

Scott35