Calculation of neutral current

Status
Not open for further replies.

Smart $

Esteemed Member
Location
Ohio
... where did it come from an what was its application. ...
The basic formula, as I call it, is derived using the Law of Cosines. Since there are four vectors involved forming a quadralateral, of which one side (the neutral) is unknown, initially apply the Law to two sides to determine an intermediary side to use with the third known side for determining the length of the fourth (the neutral).

Note the basic formula assumes a 120? angle between known sides (when placed head to tail they actually form an inside 60? angle). As such the 2 cancels out the cosine function in the latter part of the equation. However, one could use the 'full version' to arrive at the same result as using vector math.
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
100524-1244 EST

Smart $:

Upon redoing the problem I get an answer that is 1.89761 different than your value at an angle of 126.63757 from a typical math or physics 0 angle.


him76er:

Is this a class problem? It looks like one.

.
 

Smart $

Esteemed Member
Location
Ohio
100524-1244 EST

Smart $:

Upon redoing the problem I get an answer that is 1.89761 different than your value at an angle of 126.63757 from a typical math or physics 0 angle.

.
Sent you my vector math solution via PM.
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
100524-1522 EST

Smart $:

I used the opposite phase rotation. Flipping two wires I approximate your value.

I have slightly different numerical values than yours. My result is 58.62990 and 308.77567 or 51.22433 . I used 0.70710678, 0.50000, 0.8660254, 0.250819, and 0.9659258.


him76er:

Tell us what you have learned. We are assuming this is a class problem. Do you know how to do the calculation from this discussion?

.
 

Smart $

Esteemed Member
Location
Ohio
100524-1522 EST

Smart $:

I used the opposite phase rotation. Flipping two wires I approximate your value.

I have slightly different numerical values than yours. My result is 58.62990 and 308.77567 or 51.22433 . I used 0.70710678, 0.50000, 0.8660254, 0.250819, and 0.9659258.
Well of course you're going to get a different result using different numbers :roll:

As I mentioned in my PM, I made assumptions as to the voltage vectors being 0?, -120?, and -240?, respectively as listed. They could just as easily been reversed in rotation... but I still get 56.97A.
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
100524-2140 EST

him76er:

Do you know why we think this is a test or homework question?

Suppose the question is changed slightly to:

The loads of a 4-wire, 3-phase system are,
red to neutral current - 50A pf .707 lagging
yellow to neutral current - 40A pf .866 lagging
blue to neutral current - 40A pf .866 leading
determine the value of the current in the neutral wire

In one case you should be able to do this by inspection without any substantial calculation.

If you do not understand the ramblings of this thread, then ask questions. It should be a good learning experience.

.
 

rattus

Senior Member
No, it is based on purely resistive loading.

Equal power-factor loading also works, but is provisional on the power factors all be leading or lagging, and not a combination of leading and lagging.

Does not have to be resistive, just 120 degrees separation between phase currents.
 

him76er

Member
It is an exam question and all through my notes, i have not come across any formulas as to how to solve it. And ofcourse the details of this thread is a little too advanced for my current competence.

A simpler approach will be much appreciated.
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
100525-0704 EST

Ham:

For a non-linear load and assuming you have a stationary process, an invariant periodic signal, then you need to sum the instantaneous values at each point in time for one full cycle. Then process each of these point values to obtain the desired averaged measurement. It might mean you want to calculate the RMS value, or the full-wave rectified average value, or the probability distribution curve, or some other measure.

.
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
100524-0711 EST

Smart $'s calculated value is probably more correct than mine because he used a canned program that took the individual power factors, applied the arc cosine operation to get the angles, used the angles to get the individual components, and then summed the component vectors, and last calculated the resultant vector magnitude and angle.

I looked at the problem and it was obvious that the writer of the problem had assumed 30 and 45 degree phase shifts and then I used accurate values of the sin and cos of these angles to calculate the vectors.

In a real world problem it is unlikely you would have some nice round phase shift angles and thus Smart $'s approach would be used with or without a canned program.

.
 

him76er

Member
Yes it is an exam problem and the level of the discussion here is a little above my competence. It would be great if the solution could be simplified for me.
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
100525-0833 EST

him76er:

You need to provide us with information on your background. What math courses have you had? Do you know how to work with vectors? What kind of a class is this problem from? What school? Have you worked with vectors in mechanical problems? Do you know what an arc cosine is? What is your trig background?

What is the course that this question is from?

.
 

rattus

Senior Member
100524-0711 EST

Smart $'s calculated value is probably more correct than mine because he used a canned program that took the individual power factors, applied the arc cosine operation to get the angles, used the angles to get the individual components, and then summed the component vectors, and last calculated the resultant vector magnitude and angle.

I looked at the problem and it was obvious that the writer of the problem had assumed 30 and 45 degree phase shifts and then I used accurate values of the sin and cos of these angles to calculate the vectors.

In a real world problem it is unlikely you would have some nice round phase shift angles and thus Smart $'s approach would be used with or without a canned program.

.
gar,

As you already know, this problem is easy enough to do with a calculator or even a slide rule--if I still had one and if I remembered how to use it!. I got 58.86A using Excel.
 

winnie

Senior Member
Location
Springfield, MA, USA
Occupation
Electric motor research
him76er,

We won't be able to give you a simplified answer, but I think we could break it down into simpler steps. You will need to ask questions/research the steps that you don't understand.

1) You start with 'Kirchhoff's current law' which basically says that any current flowing into a point needs to be balanced by current flowing out of that point. Think of a set of loads connected 'wye' to a neutral...all of the current flowing into the center point of the 'wye' needs to flow out somewhere. We know the currents on the legs of the wye, so we sum those three up to figure out how much current must flow through the neutral to balance this. http://en.wikipedia.org/wiki/Kirchhoff's_circuit_laws

2) Now you have the problem that you are adding up AC currents which are constantly changing. The trick here is that sinusoidal AC current may be represented by a 'vector'. A vector is simply a way of describing a line segment that has both magnitude (length) and direction (angle). So for each of your three phases, you represent the current flowing by a vector, where the length of that vector is the RMS current, and the angle of that vector is the phase angle of the current flow. There are a bunch of different ways to add vectors; the easiest to understand is graphically; you just stick the vectors head to tail and draw a new vector from the tail of the first to the head of the second.
http://hyperphysics.phy-astr.gsu.edu/hbase/vect.html#vec7

3) The next step is to figure out the phase angles of the currents. For a 'unity power factor' load, the currents are in phase with the applied voltages. The three phase voltages are 120 degrees out of phase, so if you have three unity power factor loads you get three currents that are 120 degrees out of phase. In your example your loads have different power factor, so you need to figure out the 'current angle' from the power factor. http://en.wikipedia.org/wiki/Power_factor#Definition_and_calculation Power factor is the cosine of the angle between the voltage and current,so if you know the power factor you can use the inverse cosine to get the phase angle difference. You have to add this phase angle difference to the phase of the applied voltage (since each phase has a different voltage phase angle) to get the current phase angle.

So once you know the current in each phase, represented in both magnitude _and_ phase angle, you simply add up all three currents, and this will give you the net current flowing on the neutral.

Consider balanced loading, where the current on each phase is exactly the same, and the power factor is the same. You represent this with three vectors, all of the same length, but with three different angles, all 120 degrees apart. If you draw this out, and add all three vectors 'head to tail', you will get an equilateral triangle; the head of the last vector touches the tail of the first, so the sum of these three vectors has length zero. Thus with balanced loading you have zero neutral current.

As I said, there are a number of different ways to add up vectors, including methods that are purely calculation based (no graphics at all). The formula that we've been discussing is one such method, though it assumes specific phase angles. Once you understand the basics of what is going on (the use of vector representation and vector addition) you can figure out the formula, but if you don't have these 'background' concepts, then the formula will be meaningless.

Heh, a GE book from 1911 has this spelled out quite nicely: http://books.google.com/books?id=QB...mR9Ll43yV&dq=AC+current+vector+representation

-Jon
 

him76er

Member
Thank you Jon,

That was quite nicely explained. I am a marine engineer and I am updating my license, so I have been out of touch with formulas and method to solve typical questions.

Thank you everybody for your time and patience
 

mull982

Senior Member
I'm using phasors based off of the following:

Va @0deg
Vb @ -120deg
Vc @ -240deg

From this and the given information I come up with current phasors of:

Ia = 50A @ -45deg
Ib = 40A @ -150deg
Ic = 40A @ -210deg

Adding these phasors I come up with about 48A neutral current. Am I doing something wrong?
 
Last edited:
Status
Not open for further replies.
Top