Let's discuss the subject of nonlinear loads defined in Article 100 as:


Discussion on "the major portion of the load" as it relates to this subject when dealing with wye systems.

I run lots of this cable and double netual panel feeds. I really never knew why, other then someone saying harmonics.

Can someone explain it to me?
Bob

In my opinion a "major portion of the load" is when the nonlinear load exceeds 50% of the branch circuit or feeder load. This would be based on the load on the phase conductors.
I have to ask the question that I always ask when this subject comes up. Why do we continue to install 3 phase wye systems for these types of loads? There is a problem with harmonics and excessive neutral currents when these types of loads are installed in wye systems. This problem is addressed by the use of oversized neutrals, K rated transformers, harmonic filters, panels with double neutral bus and other expensive "fixes". Why don't we just use single phase transformers with a 120/240 volt secondary to feed these types of loads. The neutral problem does not exist on 120/240 volt single phase systems.
Don

I'm sure others can provide better insight, but heres my take:
In a balanced 3-phase, 4-wire, wye system, the neutral currents cancel each other out and the result is zero current flow on the neutral conductor. Increased neutral currents can result from equipment that draws an asymmetrical current wave, such as computer equipment or SCR driven equipment. These currents are additive as opposed to canceling out and can exceed the ampacity of standard sized neutrals. The same problem may arise with certain ballast loads due to 3rd-order harmonics, which also are additive.
The solutions can include individual neutrals for each circuit, or a single oversized neutral with adequate ampacity.

Go to http://www.ecmweb.com/ and type three words in the search box
harmonic double neutral
and stand back for the 17 articles to read.

http://www.fluke.com has an application note (PDF format) entitled In Tune with Power Harmonics found via this link .

You'll need to register to download the PDF.

Gentlemen:

I'm not up on this subject at all, but I see that Redsy mentioned separate neutrals for each circuit.

In the photo, the cable on the right appears to have a striped wire (red and white), while the center cable has a larger neutral. Does derating become an issue when more conductors are added, as opposed to one larger conductor?

Good thinking, Thinkgood.

Where a common neutral would only carry the unbalanced load current, and therefore not be considered a current carrying conductor(per 310.15 (B)(4)), individual neutrals would each carry their respective load current. The total number of current carrying conductors will increase and derating factors will apply.


[This message has been edited by Redsy (edited 01-17-2003).]

I can forgive and overlook the thin pinstripe of red that the double neutral NM Cable has, but the MC cable's red striped neutral (as far as I can see) has a much larger stripe and would cross the line for me on being compliant with 200.6(A) unless there are three red stripes and subsequently three white stripes present.

Overkill?

Folks,

This is an interesting one. I have found the key to understanding nonlinear loads centers on how one describes the nonsine wave current. It helps to start by looking at each half-cycle of voltage that the power company provides, that is put across the load that is connected to one hot conductor of a branch circuit. The voltage can be thought of, essentially, as a clean sine wave with a frequency of 60 Hz. 60 Hz is the base frequency. . .the fundamental, or the "first harmonic".

Now, if the load on my branch circuit conductor is an incandescent light bulb, or a resistance heat element, as the voltage rises and falls through the sine wave, the current in the bulb or the element will rise and fall exactly the same way, in the shape of a clean sine wave. The current, at any instant in time, will be exactly that calculated by Ohms Law. The resistance of the load stays constant, and, as the voltage rises, the current rises in direct proportion. I=V÷R.

Set inductive and capacitive load to the side for the moment.

Let's jump straight to loads that receive their voltage, not directly from the branch circuit voltage, but rather from the output side of a silicon controlled power supply (SCPS). The SCPS, during any one half cycle, turns on and off (in its simplest form), and generally turns on and off a bunch of times in a half cycle. When the SCPS is off, no current flows, when it is on, the current that flows is a function of the logic chip(s) of the SCPS. The half cycle wave form of the current going through the branch circuit to the SCPS is a bunch of pulses more like square waves than any thing else.

This half cycle of branch circuit load current is now very complex, compared to the simple sine wave of the resistive load current. With the help of good math, electrical engineers learned to think of this half cycle of pulses of current as the sum of a lot of currents in the shape of pure sine waves that are multiples (harmonics) of the fundamental frequency (the Fourier Series, so named for Joseph Fourier near the turn of the 18th century, a French mathematician and friend of Napoleon).

In fact, the pulses of current that a silicon controlled power supply draws during a half cycle of voltage are made up of an infinite number of current sine waves of every multiple of the fundamental frequencies of the voltage.

When the neutral current of the branch circuit is then combined with the current of other branch circuits in a multiwire branch circuit neutral, some of the currents add to each other.

OK. Enough. I've probably really confused it.

The math really helps, the words are hard.

Al

These sketches might help illustrate what Al said.




Ed

[This message has been edited by electric-ed (edited 01-17-2003).]

I am learning, thanks ElectricAL and electric-ed.

How do I recognize the loads that cause this?

For a while I used to hook up concert power feeds, one night during a show I was checking the 200 Amp 208/120volt 3p 5w disconnect for the sound system feed.

I felt the cables near the terminations (before I new about PPE) the 3 hots and ground where fine but the neutral was buzzing/vibrating.

Amprobe showed aprox. (This was 8 years ago I might be mistaken) 70 amps on the hots and over 210 amps on the neutral I remember thinking at the time it seemed the current had added up, I did not put much thought into it at the time, other then hoping the neutral feeding the disconnect would not fail.

Was this caused by a Nonlinear Load?

Another question, a lot of the data centers I do work in, run the servers at 208 instead of 120. Does this take care of nonlinear load problem or just move it from the neutral to a Hot?

As resqcapt19 said why not avoid the problem and go 120/240 single phase?

I will read the articles Ron linked to.

I hope these questions are not too basic, I feel like I should know this as I install this stuff (super neutral) all the time.

Thanks Bob

Equipment that uses switch-mode power supplies produce harmonic currents. Most electronic equipment, such as computers, printers, photo-copiers, some electronic ballasts, etc.

Ed

Don,

When using the SDS 1Ø 3W system(s) from a 3Ø 4W source, would the 3Ø transformer(s) need any special consideration?
i.e. If a premise system is 3Ø 4160V then at a point a 3Ø 480Y is used for that building and then 1Ø 3W 120/240V system is used for these non-linear loads.
Is there any non-linear load reflected 'up-stream' though the transformers?

Gwz,
My understanding is that there are very few harmonics on the primary side of the single phase transformer used to supply nonlinear loads.
Don

brave soul Don.

considering all the markets opened up in recent years due to harmonics ,and the simplicity of a single ph fix putting the kebosh to them.....

rescapt — The description of using 3-wire 1ø-secondary transformers is not unlike a description of {something like} a media-production studio {previously described at this board?} that contained some fairly exotic grounding and a specialized delta/dual-wye transformer; possibly a 480/dual-208Y drytype intended to limit high-side line harmonic currents.

If one were to employ three 1ø 2-wire-primary to 3-wire-secondary transformers with primaries served in delta, with the center taps of secondaries bonded and grounded in a conventional manner, this would effectively mimic the earlier-described “6ø”-secondary application furnished for the studio electrical system.

A description of the transformer connections could be, using three similar 1ø transformers—rated 240x480 - 120/240 Volts, with each 1ø transformer fed phase-to-phase from a 480V 3ø 3-wire source—1H1 and 1H4 connected to 480V Aø and Bø, 2H1 and 2H4 connected to 480V Bø and Cø and 3H1 and 3H4 connected to 480V Cø and Aø, comprising a delta arrangement. For the secondaries—1X2/1X3, 2X2/2X3 and 3X2/3X3 are tied together and grounded; treated as a separately-derived system. This comprises a staggered or “asterisk” configuration of ungrounded terminals sequenced 1X1 - 3X4 - 2X1 - 1X4 - 3X1 - 2X4, for serving various ø-n 120V loads.

This is somewhat similar to moderator Scott35’s drawings at https://www.electrical-contractor.net/ubb/Forum15/HTML/000052.html

sparky,
I think that is exactly why we don't hear much about the use of single phase 120/240 volt systems to supply these loads. There is more profit to be made by all involved using K rated transformers, harmonic filters, oversized neutrals and other treatments for the symptoms of the problem.
Don

I understand that if there are too many conductors in a box, this can become a problem because of fill harmonics.

Sorry, I just couldn't resist

<comedy club drummer>
ba-dump bump *crash*
</comedy club drummer>

Very "punny" TG!

Harmonic filters and the like seem to be the cure all. Went on a svc. call to a resturant today, it's served by 2 200a single phase services.Seems they had problems with light flicker and dimming/brightening. owner said one of the larger ec's in the area told him he needed a harmonic filter among other things. A few thousand dollars worth of "improvements". Did a little snooping and discovered a loose neutral connection at line side of service disco. Problem solved. Instant Hero! (and a free meal to boot).