3 phase leg problem

What are you calling ground? The only ground in a system without multiple in plant distribution transformers should be at the service entrance panel. Are you running a wire back to that point to serve as your meter's ground reference?
There is a service called corner grounded delta which has one of the phase conductors bonded to ground at the transformer secondary. But that should show ground on that conductor all the way back to the distribution transformer secondary.
Note, of course, that readings to ground are irrelevant in so far as 3 ph is concerned. Only the phase to phase measurements matter to equipment using the 3 ph power.
Gary

No, everything is 3 wires run in conduit. I was referencing the conduit as "ground" I will have to check inside the main box after shut-down to see if there is a true ground connected to the box or is it so much conduit connected to stuff cemented into the floor that is creating the "ground". It worries me that it's not consistent in this part of the plant and I never would have noticed if I wasn't wiring in a new machine. I don't want anyone shocked or a fire so I thought I had better track it down.

Could be an isolation transformer someplace, and its outout has been corner grounded.

A splice or connection to that third dead leg has failed. Likely inside a plugin box if they daisy chained.
Only way I know of to find the break in the dead leg, is to work backwards from the first place the dead leg appears, follow the conduit backwards to the panel, and open each box, testing each leg.
The break or open will be between the last known dead one and the first good one.
Sigh..if its been wired by a half century of good and bad electricians and those that claim to be...it may or may not be easy to track down.
If the hot leg was shorted to the conduit..it would have tripped that particular breaker, so you have an open, rather than a short.
Its pretty common for a lot of guys to stick the incoming and outgoing wires (in a daisy chain) under the screws of female receptical instead of pigtailing and wirenutting . Shrug. If one of the outgoing wires broke or came loose..there is your open.
Gunner
That rifle hanging on the wall of the working-class flat or labourer's cottage is the symbol of democracy. It is our job to see that it stays there. - George Orwell

>I'm not sure what you call it but I have a problem with 1 leg of 220v 3 >phase in one area of the shop. Everything runs fine. With a meter, I get >@220v from 2 of the three legs to ground and nothing on 1 leg to ground. >Grounded leg? I do have @220v on each leg at the mains. This is going to >be a bitch as Ben Franklin wired this place. The only thing I can imagine >is all the conduit on this side is shorted to a leg and the conduit is >broken away from the rest of the building and thus "floating" Does this >make sence? I'm trying to think of a strategy for tracking it down that >doesn't involve licking my fingers and touching stuff while standing in a >puddle. We have replaced lots of stuff over the years with better than code >but it's a HUGE project, we just keep pecking away at it. > >-- >

You neglected to mention if there is 220 volts between the two hot wires in your circuit. If there is 220 vac between each of the three wires, then you are wired oddly, but may have legitimate three phase that will run things.
You must also be alert to the possiblility of having a "wild leg" three phase. This was used in residential areas. The standard 220 with center tap ground was wired into the house. A third wire (the wild leg) was connected to provide three phase using the +/- 110 legs (220 volts) as the other wires. On such a circuit a meter to ground will show 110V, 110v, and 190V. Between each of the wires (not to ground, but from hot wire to hot wire) you will measure 220 volts. Thus three wires provides standard 220 volt house wiring and three phase. These voltages do not match what you have quoted though...
If someone has installed an isolation transformer incorrectly in your circuitry, all bets are off. You could easily see the voltages you measure, and motors would work fine.

The "wild" leg should be 208 to ground, not 190.
There is another way of providing three-phase which uses only two transformers, and has few of the limitations of the standard "open Delta" installation.
This is the "Scott T".
A "Scott T" installation has much better regulation than an "open Delta" installation.
The disadvantage is a transformer with an unusual ratio must be used: one with a conventional 0.5-0-0.5 "per unit" ratio, and another with a 0-0.867 "per unit" ratio.
Another advantage of the "Scott T" is a Wye service can be provided. 277/480 would be one such example.

Finally a question within my area of expertise! You have described a three phase delta power connection , since you have 220 volts phase to phase on all combinations ( a-b, b-c,c-a) and you are showing a voltage from each phase to the equipment ground that is close to line voltage at the open mains your ground fault is not on the utility side and every thing on the incoming is fine .
As I understand by your note when you close the main switch you get good phase to phase voltage but your phase to equipment ground voltage is 220/220 and 0 , in actuality the phase to ground voltage is not likely a full 220 ,but a voltage that we in the industry call " indeterminate" which means it exists but can fluctuate due to a variable phase to ground reference point. The important reading is not the ~220 volts, it is the fact that one phase is reading 0. The 0 reading phase is your grounded phase if a second ground is introduced on another phase in your shop or on a neighboring shop that is connected to the same transformer bank your first grounded phase and the second grounded phase will experience severe fault currents that will likely trip your fusing and in some cases take out the transformer fuses on the utility pole.
To locate this I recommend that you drop all of your sub panels and any other connected loads , if your leg that was reading 0 to ground now reads close to line voltage you have isolated the problem to a sub panel or a piece of connected equipment, if the phase is still at 0 then the problem will be between the mains and the subs. If your voltage was high (~ 220) start adding the subs back in if the voltage drops to 0 you have your ground fault. Even when you find the first one carry on testing the rest. I have found as many as three faults on one leg in the past ... two of which were lighting ballasts in a machine shop and one in a neighboring building that cycled on and off. All of these faults co-existed for years until a second leg was grounded by a motor failure...... that was easy to find since it let out lots of smoke!
Good luck , the satisfaction of finding an evasive grond fault that everyone else missed is is akin to catching and landing the "big one" . ( come to think about it maybe I've chased too many of these!)
> I'm not sure what you call it but I have a problem with 1 leg of 220v 3 > phase in one area of the shop. Everything runs fine. With a meter, I get > @220v from 2 of the three legs to ground and nothing on 1 leg to ground. > Grounded leg? I do have @220v on each leg at the mains. This is going to > be a bitch as Ben Franklin wired this place. The only thing I can imagine > is all the conduit on this side is shorted to a leg and the conduit is > broken away from the rest of the building and thus "floating" Does this > make sence? I'm trying to think of a strategy for tracking it down that > doesn't involve licking my fingers and touching stuff while standing in a > puddle. We have replaced lots of stuff over the years with better than code > but it's a HUGE project, we just keep pecking away at it. > > -- >

The most obvious sign of an intentional corner-grounded Delta service is the panels have only 2-pole breakers. The grounded leg is essentially the same as any neutral conductor in single-phase service, and needs no breaker. So, you have 3-phase machinery with only a 2-pole breaker in the panel.
If the breakers are 3-pole breakers, then you have an UNINTENTIONAL ground on one of the 3-phase legs, and that should be investigated. A 3-phase Wye service will generally not permit a grounded leg, and will blow fuses somewhere right away. 3-phase Delta-only service that is intended to be balanced (ie. not corner-grounded) may have a ground-fault protector that should shut down the feed if a ground develops. But, many smaller delta-only services are allowed to float, for cost reasons. That's a bit scary, as you can have excessive voltages develop on the other legs when one leg grounds out. Not a problem for equipment designed as 220/440 and running on the 220 setting, but it could be a problem for equipment already running on the high voltage setting.
Jon

Not necessarily so, Jon. The particular case I worked on had a standard 3-phase panel (I know, one breaker was essentially at ground) and everything was run like it was a wye 3-phase (maybe because someday the corner grounded Delta would be changed over and the electrical work would already be done... dunno... shrug... Ken.

So the implication is that the original service (here I am figuring that it is not corner grounded delta) has the low voltage side of the transformers *floating*? That sounds like a very bad practice to not have a dedicated reference to ground, someplace.
Of course this would allow a single fault to occur such that one phase gets pulled all the way down, but I would think that it would be better to have the overcurrent protection trip out under those circumstances.
Jim
================================================== please reply to: JRR(zero) at yktvmv (dot) vnet (dot) ibm (dot) com ==================================================

Sorry for being vague but I think you have the answer. I tried to clarify things in a new post but I will start troubleshooting your way today...THANKS!!!

Ungrounded delta motor power systems were once common in industrial plants that required maximum reliability such as chemical plants. The idea is to prevent an outage on a single line to ground fault. However, there was always a ground fault alarm to alert the operating staff that a fault had occurred and SOP was to find the fault ASAP and repair it when operating procedures allowed the faulted circuit to be shutdown. Over the years, gov. regulation and better understanding of transient voltages sources made such systems undesirable because they can cause high magnitude transient voltages when the fault is intermittent or arcing.
Nowadays, the systems are usually wye and are grounded thru a resistor that allows a maximum of 3-5 A to flow on a line to ground fault. A monitor watches the voltage drop across the resistor and alarms when the voltage goes out of spec. Then the faulted circuit is located and repaired as above with no unscheduled outage.
Randy

That all makes eminent sense. Thanks for the explaination.
Jim
================================================== please reply to: JRR(zero) at yktvmv (dot) vnet (dot) ibm (dot) com ==================================================

Tom if you have any further questions email me directly. Pete

Ungrounded delta motor power systems were once common in industrial plants that required maximum reliability such as chemical plants. The idea is to prevent an outage on a single line to ground fault ... Sure, and at my former employer (this nation's largest municipal utility) we operated all of our lines ... 4,800 volts to 500,000 volts ... ungrounded, for the very same reason: you can *continue* to deliver three-phase power to the customer while the protective relays continue to do their job: identifying and clearing the root-cause of the fault.
It is not an accident that our reliability figures were an order of magnitude better than those of our "investor owned" counterparts.