parallel feeders

Status
Not open for further replies.
Re: parallel feeders

I can not answer the reactance question but what you describe is a violation.

Part of 310.4
Where run in separate raceways or cables, the raceways or cables shall have the same physical characteristics.
Click to expand...
 
Re: parallel feeders

Hi Bob. That is the answer that I gave. I consider the impedance of the raceways to be one of the "physical characteristics" addrssed in 310.4.

I wonder if someone out there could enlighten me on the effects of the reactance though.....
 
Re: parallel feeders

It would be good to learn that for myself also.

Until you brought it up I never thought of reactance as one of the characteristics that article is referencing, although now that you bring it up it sure makes sense. :)

When I read "the raceways shall have the same physical characteristics" I just think of size and material.

A 3" and a 4" or RMC and RNC have different physical characteristics even if they somehow did provide the same reactance .
 
Re: parallel feeders

The only thing that the laws of Physics would care about is the reactance. If I could build two raceways, one with 3? conduit and one with 4?conduit, and if I chose the materials and the thickness of each in such a way that I could create the same reactance the conductors installed in each, then the circuit would operate with perfect equality. But I could not ask an Inspector to buy my proof that the two were the same. The easy way to assure the same reactance is to built both conduits out of the same materials, and with the same internal diameter, and with the same thickness. In other words, the code requirement is worded in such a way as to make it easy to verity compliance. I think that is a good thing.

Now a short statement on the nature of ?reactance.? It is a combination of ?inductive reactance? and ?capacitive reactance.?

?Inductive reactance? is a measure of a component?s ability to store energy in a magnetic field. If you have two conduits, and one is made of a magnetic material and the other is not, then the reactance of the conductors within the two conduits will be different. If one has a diameter of 3? and the other has a diameter of 4?, then the reactance of the conductors within the two conduits will be different. Note that most fields get smaller as you get farther from the source. In this case, the source of the magnetic field is the current flowing through the conductors within the conduit.

?Capacitive reactance? is a measure of a component?s ability to store energy in an electric field. The value of a parallel capacitor gets larger as the cross sectional area of the two parallel plates gets larger. A 4? conduit has a larger area than does a 3? conduit. But the value of a parallel capacitor gets smaller as the distance between the two parallel plates gets larger. A 4? conduit is farther from the conductors within it than is a 3? conduit. So which has the higher capacitance? I?m not sure, but it is likely that they will be different.

That?s about as far as I can delve into reactance without (1) Getting into math, and (2) Digging out my books, so that I can remember the math. So I hope this is of some help, and that you will not ask for more math.
 
Re: parallel feeders

Charlie: Would you say the only way to determine the conductors are the same impedance is to install, load them, and measure the current?
 
Re: parallel feeders

Charlies answers with out the math books amaze me!

[ April 11, 2004, 05:36 PM: Message edited by: tom baker ]
 
Re: parallel feeders

if i had three sets of parrelle conductors set 1 10' set 2 9' set 3 8' what difference would it make,would it be safe,would it over heat,this is a 90.4 call i would like to know if you were 90.4 would pass this, i would.
 
Re: parallel feeders

A 2' difference in a 10' run would be a 20% difference in length, I would not ask for it to pass inspection, and depending on the load could overload the short conductors.

Make it a 2' difference on a 100' run and you would have a 2% difference, that would be reasonable.

JMO Bob
 
Re: parallel feeders

Originally posted by domnic:
if i had three sets of parrelle conductors set 1 10' set 2 9' set 3 8' what difference would it make,would it be safe,would it over heat,this is a 90.4 call i would like to know if you were 90.4 would pass this, i would.
Click to expand...
Not a chance would I pass it. Read my reply above, where I did fail one like this not more than a week ago.
 
Re: parallel feeders

Originally posted by bennie: Would you say the only way to determine the conductors are the same impedance is to install, load them, and measure the current?
Click to expand...
You would have to load them with a load that has a known impedance, and you would have to measure the current and voltage angles, as well as magnitudes. No simple task, and not a convenient in-the-field test. But I submit that no test is needed. The code does not say that the cables have to have the same impedance. It says that they have to have the same length. It is my belief that the only reason the code requires the same length is to get the same impedance. But the rule only says to have the same length.

The problem is that the code gives no acceptance criteria for ?close enough to the same length.? Absent such a standard, all Inspectors are essentially free to set their own standards. Here is my view of what constitutes ?close enough?:
</font>
  • <font size="2" face="Verdana, Helvetica, sans-serif">Less that 5% difference should be acceptable.</font>
<font size="2" face="Verdana, Helvetica, sans-serif"></font>
  • <font size="2" face="Verdana, Helvetica, sans-serif">More than 10% difference should not be acceptable</font>
<font size="2" face="Verdana, Helvetica, sans-serif"> </font>
  • <font size="2" face="Verdana, Helvetica, sans-serif">If the cables all enter the box and go directly to their termination points (i.e., without obvious extra coiling or bending), then it should not be necessary to even think about measuring length.</font>
<font size="2" face="Verdana, Helvetica, sans-serif">
 
Re: parallel feeders

A few years ago, I changed out a 100 foot long, over the roof feeder for a service at a large cold storage plant.

This service feeder was 480/277 4 sets of 500 Kcmil Al. per phase. I had to change these aluminum conductors to copper while system was energized and under load.

I had to connect the copper conductor at both ends before I could remove one of the aluminum cables, due the the load on each cable was already peaking at near maximum on some cables.

I read each cable, in a phase, with a clamp on ampmeter.

When the copper cable was connected, it immediately picked up the largest amount of current for that phase. Some times the copper cable went over the rated ampacity.

There was a lot of difference in current on the aluminum feeders. Current was a bit better balanced on the copper.

The aluminum cable was actually failing due to overcurrent on some cables in the phases.

A clamp on ampmeter is all that is needed to determine the impedance imbalance of individual cables in a phase.
 
Re: parallel feeders

Originally posted by bennie: A clamp on ampmeter is all that is needed to determine the impedance imbalance of individual cables in a phase.
Click to expand...
I certainly agree that nothing more is needed. But I think something less might be enough.

If a quick look shows that the cables don?t have an obvious major difference in length, then I say don?t even bother to get out the ammeter. I am not an Inspector, and I have not seen thousands of field installations, so I may be missing something here. But I am inclined to believe that this is a non-problem, and that it therefore is not in need of an ?easier solution.?
 
Re: parallel feeders

Charlie: I understand your stance, but as an inspector it is sometimes very difficult to determine visually if ythe conductors are the same length. One of the things that I do is try to remember what the underground conduit runs look like. If I inspect an underground with a lot of 90 degree bends and things in it, I try to make a note of it to remember at the power to panel inspection. If the pipe run has a lot of bends in it and I come back for power inspection, I expect to see some of the excess length consumed in hte panel(s). If there are many bends in the pipe and the wiring looks very clean, Istart to wonder about the length of the wires.

For example, on my city hall. They had a few bends in the pipe, and it was obvious that the pipe lengths were different. When I inspected it I started in the MDP. All of the conductors just went straight into the lugs because the panel was very tight. I started to get concerned until until I went inside to the panel it was feeding. In that panel the wiring was a mess! They had about 8' feet to make up on that side, so they had the 600's wrapped around the inside of the panel to make up the difference. When I saw that, may worries were gone.

I understand your stance Charlie, but understand mine. From an inspection point of view, it is very difficult to determine the length of conductors.
 
Re: parallel feeders

This only becomes a problem when the load is near the maximum ampacity, and other contributing factors are present.

The fire at the First Interstate Bank building in Los Angeles, was determined to be caused by loosing one cable of paralleled conductors from corrosion on the lugs. The load shifted to the remaining cables, causing overheating and igniting the building.

With aluminum wire, the cable with the most current will fail first.

Length is not the only factor to consider. Testing under load, should be in the job specs.
 
Re: parallel feeders

The reason for 1/0 minimum for paralleling is because resistance change in smaller sizes will have greater effect when there is a difference in length.
 
Re: parallel feeders

the reason for paralleling wire 1/0 and bigger is the skin effect,wire 1/0 and bigger has more impedance per CM. in a ac. circuit and will have less ampacitie per CM.
 
Re: parallel feeders

Domnic: Please explain this "skin effect" phenomenon for 60 Hertz power.

[ April 12, 2004, 08:56 PM: Message edited by: bennie ]
 
Re: parallel feeders

Domnic: I was hoping you would clarify the skin effect statement.

Skin effect is predominant after 1 MHz. It is insignificant at low frequencies, including the audio band.

At 60 Hertz, it can't be measured.

May I ask where you heard about the "skin effect", and it's impact on parallel conductors.
 
Status
Not open for further replies.
Top