Feeders in conduit

[mull982]

Why is it not good practice to run the same phases for 480V or 4160V in a single conduit. For a parallel circuit for example, each of the three phases is run in on conduit and each of the three phases is run in another conduit and so on. Why do you not want to have say three "A" phases in one conduit, three "B" phases in another and yet three "C" phases in another. What is the point of seperating out the phases in different couduits? I have a feeling it has something to do with induction, but I was hoping that someone else could give a more scientific explanation.

Thanks

Mull982

 

[haskindm]

You've got it. If only the conductors on a single phase are installed in a single metal conduit a currrent will be induced on the conduit. You have essentially created a transformer. It may be desireable to separate the phases for ease of connecting the conductor at the terminations. If so then you must use non-metallic conduit. See 300.20.

 

[charlie b]

If you put a clamp-on ammeter around all three phases (and the neutral, if there is one), you will read a total of zero amps. The three (or four) currents are going different directions and at different magnitudes, at each and every point in time, with the result that they balance each other out. Therefore, the magnetic field that surrounds each conductor will be balanced out by the magnetic fields from the other conductors. The total magnetic field, as seen by the conduit or by the outside world, will be essentially zero.

But if you put only phase A into a single conduit, there will not be a cancellation effect from the other two phases. The result is that the conduit, and the outside world, will see a significant magnetic field. That will cause current to flow in the conduit, and perhaps in other nearby metal objects. Not good. That is the reason that putting only conductors of a single phase into a conduit is allowed only in limited circumstances, and only if the conduit is non-metallic.

 

[eprice]

mull982,

The others have pretty well covered it, but here is a code reference: 300.3(B)(1) (2005 NEC)

 

[mull982]

Is this also the reason why you want to have your current balanced in all three conductors. If I was running conductors of each phase throug a conduit, I would want to have the current draw on each phase pretty close in magnitude correct? If they were different in magnitude say A=45, B=70,
C=100, then I'm assuming that those magnetic fields wouldn't totally cancel out and could induce some small magnetic field in the couduit?

Also if the current draw was different on each phase the b/c of the impedance in the cables would't that create voltage variations at the bus on which the cables were connecting to? Could this lead to problems with motors?

I have a feeder breaker that is feeding an MCC with several three phase motors and several single phase loads. I notice from a metering device that is hooked up, that there are current variatinons in the three phases as noted above. Is this a problem?

 

[GH_Vegas]

Un-balance Currents

Un-balance Currents

To mull982:
The unbalance is typically measured with just a current meter (clamp-on or owtherwise). This only indicates the magnitude of the current. Not the actual phase angle relative to the other currents. In a 3-wire 3ph system, the currents you provided as an example could indeed 'cancel' by including the actual phase angle of the current for each phase(A,B,C) when summing.


In a 4-wire Wye system, the neutral will carry the true unbalance as computed using the magnitudes and phase angles. This is why besides the 3 phase conductors, you must run the neutral also in the same conduit.

Regarding the unbalanced currents...Quoting: "there are current variations in the three phases as noted above. Is this a problem?"
This is not a problem in itself. The problem is in the voltage inbalance created at the source. This leads to the 3ph motor heating when the phase voltages are not balanced and should be avoided.

To haskmdm:
Plastic conduit will work, but there also must be NO metal bushings, locknuts, etc anywhere in the conduit run. Otherwise there will still be a current induced in the metal bushing(it will get REAL hot) as Charlie_b indicated.

 

[mull982]

To GH Vegas:

When you mentioned above that the clamp meter was only reading the magnitue and not the phase angles, dont you assume that the current phases are all 120 degress out of phase with each other? Are you saying that if you know the phase angles, then then those unbalanced current readings may indeed cancel? Are the current phases not all the same number of degrees out of phase with each other?

When you mentioned the current unbalance being a problem at the "source" what are you referring to as the source. Are you referring to the source where the motor is being fed from, or are you referring to the motor leads as the source?

For example: I have one MCC with a 300A feeder breaker feeding another MCC, and I measured the current loads a mentioned above. It appears that the phases are unbalanced to to various single phase loads on the second MCC. Because of this unbalance could three phase motors on the second MCC which require all three phases off the second MCC bus be affected by voltage variations?

Thanks for the help

 

[GH_Vegas]

This is kind of long winded but here goes....

To respond to your example. You are correct in that the unbalanced currents are due to the single phase loads. Where as 3 phase loads will have currents 120 degrees apart, the single phase loads, when added to the 3 phase loads, will change the phase angle according to the magnitude and phase angle of the individual single phase loads.
Assuming a 3 phase load with 20 degree phase shift(pf = 93%) in current the representative angles would be 20(A), 140(B), 260(C) degrees. However, single phase(line-line) loads are themselves both +30 degrees and -30 degrees 'out-of-phase' with the 3 phase currents. The (+) would be added to C phase and (-) subtracted from B phase. So depending on the relative magnitude of the single phase load vs the 3 phase load, the angles might measure at 20(A), 135(B), 265(C) As you can see, these are no longer the nice 120 degrees apart from each other and you also realize that the currents of B and C are larger than A (due to the 1 phase load). When you think about this, you realize that this unbalanced line currents results from a balanced 3 phase load and a balanced 1 phase load (line to line). Thus no current will flow in the neutral even though the currents are not balanced. Similar arguments can be made for the phase angle effect from line-neutral 1 phase loads, however, there will be neutral current in this case.

It is this phase angle AND magnitude that I was referring to in my first response where I said you should use both to accurately determine if there is neutral current. Practically, you do assume that the currents are 120 degrees apart. It gives a good starting point, even ending point, for balancing the loads, but you may never be able to get it perfect due to the 'assuming' 120 degrees.

Now to the meat...the unbalanced currents will:
1) Produce unbalanced voltage drops in the ALL the cables/bus feeding the 'effective' unbalanced loads.
2) Present different loads to the individual phase windings of the utility transformer serving the site, or customer transformer, as the case may be, producing unbalanced voltages due to individual phase winding losses.

It is this voltage unbalance that has a detrimental effect on motors. Each MFG should be able to tell you how much unbalance can be tolerated by their motors, but know that it is best to minimize the imbalance.

This leads to the definition of "source". Certainly the transformer(customer or utility) is a source. Another would be the MCC 1 and MCC 2 in your example. Wherever the motor is connected to and getting it's power. Both your MCC's will have unbalanced phase voltages due to the unbalanced currents as you described your example. MCC 2 will probably be worse than MCC 1 for voltages but this may NOT necessarily be true. I would not consider the motor terminals a 'source'(although it is) here because there is nothing you can do about the voltage from that point. I consider the MCC's to be sources because that is where you have some control over balancing loads to 'fix' the voltages.

"As clear as mud now?"